Acta Periodica Technologica 2009-40

FACULTY OF TECHNOLOGY – NOVI SAD
ТЕХНОЛОШКИ ФАКУЛТЕТ – НОВИ САД
ACTA PERIODICA
TECHNOLOGICA
APTEFF, 40, 1-220 (2009)

ACTA PERIODICA TECHNOLOGICA - Novi Sad (formerly Zbornik rado-
va Tehnološkog fakulteta and Proceedings of Faculty of Technology) publishes
articles from all branches of technology (food, chemical, biochemical, pharmace-
utical), process engineering and related scientific fields.
Articles in Acta Periodica Technologica are abstracted by: Chemical Abs-
tracts, Columbus, Ohio, Referativnii zhurnal – Khimija, VINITI, Moscow, listed
in Ulrich’s International Periodical Directory, and indexed in the Elsevier Biblio-
graphic data bases – SCOPUS.
YU ISSN 1450 – 7188 CODEN: APTEFF

UDC 54:66:664:615
Publisher
University of Novi Sad, Faculty of Technology
21000 Novi Sad, Bulevar Cara Lazara 1, Serbia
For Publisher
Prof. Dr. Zoltan Zavargo, Dean
Editor-in-Chief
Prof. Dr. Sonja Đilas
Editorial Board
From Abroad
Prof. Dr. Živko Nikolov
Texas A and M University, Biological and Agricultural Engineering
Department, College Station, TX, USA
Prof. Dr. Erika Békássy-Molnár
University of Horticulture and Food Industry, Budapest, Hungary
Prof. Dr. Željko Knez
University of Maribor,
Faculty of Chemistry and Chemical Technology, Maribor, Slovenia
Dr. T.S.R. Prasada Rao
Indian Institute of Petroleum, Dehra Dun, India
Prof. Dr. Đerđ Karlović
Margarine Center of Expertise, Kruszwica, Poland
Dr. Szigmond András
Research Institute of Hungarian Sugar Industry, Budapest, Hungary
Dr. Andreas Reitzmann
Institute of Chemical Process Engineering, University Karlshruhe
From Serbia
Dr. Ratko Lazarević, Academician Prof. Dr. Petar Dokić
Prof. Dr. Slobodan D. Petrović Prof. Dr. Spasenija Milanović
Prof. Dr. Erne Kiš Prof. Dr. Vladimir Srdić
ACTA PERIODICA TECHNOLOGICA
APTEFF, 40, 1-220 (2009)
CONTENT
FOOD TECHNOLOGY
Biljana R. Cvetković and Marija R. Jokanović

EFFECTS OF PRESERVATION METHOD AND STORAGE
CONDITION ON ASCORBIC ACID LOSS IN BEVERAGES 1
Gordana R. Dimić, Sunčica D. Kocić-Tanackov, Dušanka J. Pejin,

Jelena D. Pejin, Ilija J. Tanackov and Danijela Tuco
ANTIMICROBIAL ACTIVITY OF CARAWAY, GARLIC AND
OREGANO EXTRACTS AGAINST FILAMENTOUS MOULDS 9
Ljubica P. Dokić, Marija I. Bodroža-Solarov, Miroslav S. Hadnađev

and Ivana R. Nikolić
PROPERTIES OF EXTRUDED SNACKS SUPPLEMENTED
WITH AMARANTH GRAIN GRITS 17
Aleksandar Z. Fišteš and Đuro M. Vukmirović

REDUCTION OF WHEAT MIDDLINGS USING A CONVENTIONAL
AND EIGHT-ROLLER MILLING SYSTEMS 25
Gordana B. Koprivica, Nevena M. Mišljenović, Ljubinko B. Lević

and Vjera S. Pribiš
CHANGES IN NUTRITIVE AND TEXTURAL QUALITY OF
APPLE OSMODEHYDRATED IN SUGAR BEET MOLASSES
AND SACCHAROSE SOLUTIONS 35
Radomir V. Malbaša, Eva S. Lončar, Spasenija D. Milanović

and Ljiljana A. Kolarov
USE OF MILK-BASED KOMBUCHA INOCULUM FOR MILK
FERMENTATION 47
Anamarija I. Mandić, Sonja M. Djilas, Jasna M. Čanadanović-Brunet,

Gordana S. Ćetković and Jelena J. Vulić
ANTIOXIDANT ACTIVITY OF WHITE GRAPE
SEED EXTRACTS ON DPPH RADICALS 53
Spasenija D. Milanović, Mirela D. Iličić, Katarina G. Duraković
and Vladimir R. Vukić
TEXTURAL CHARACTERISTICS OF FERMENTED
MILK BEVERAGES PRODUCED BY KOMBUCHA 63
Dragan V. Palić and Sophia E. Coetzee

PROTOCOL FOR USING PROTEIN SOLUBILITY AS AN INDICATOR
OF FULL-FAT SOYBEAN HEAT TREATMENT 71
Dragan V. Palić and Klaas-Jan Leeuw

COMPARISON OF THREE IN VITRO METHODS FOR DETERMINING
AND PREDICTING THE ORGANIC MATTER DIGESTIBILITY OF
COMLETE DIETS FOR RUMINANTS 79
Dragana Pešić-Mikulec and Gordana B. Niketić

COMPOSITIONAL CHARACTERISTICS OF COMMERCIAL YOGHURT
BASED ON QUANTITATIVE DETERMINATION OF VIABLE LACTIC
ACID BACTERIA 87
Slađana M. Savatović, Aleksandra N. Tepić, Zdravko M. Šumić

and Milan S. Nikolić
ANTIOXIDANT ACTIVITY OF POLYPHENOL-ENRICHED APPLE JUICE 95
Mirjana A. Vasić, Biserka L. Vujičić, Aleksandra N. Tepić,

Jelica M. Gvozdenović-Varga and Zdravko M. Šumić
DIETARY FIBER CONTENT IN SOME DRY BEANS 103
Tanja D. Žugić-Petrović, Nataša M. Joković and Dragiša S. Savić

THE EVOLUTION OF LACTIC ACID BACTERIA COMMUNITY
DURING THE DEVELOPMENT OF MATURE SORDOUGH 111
CHEMICAL TECHNOLOGY AND PROCESS ENGINEERING
Eva S. Lončar, Miroslava M. Radeka, Snežana B. Petrović,

Andrea S. Skapin, Ognjen Lj. Rudić and Jonjaua G. Ranogajec
DETERMINATION OF THE PHOTOCATALYTIC ACTIVITY OF TiO2
COATINGS ON CLAY ROOFING TILE SUBSTRATES – METHYLENE
BLUE AS MODEL POLLUTANT 125
Nataša Lj. Lukić, Svetlana S. Popović and Jelena Dj. Marković

MATHEMATICAL MODELLING OF FLUX RECOVERY
DURING CHEMICAL CLEANING OF TUBULAR MEMBRANE
FOULED WITH WHEY PROTEINS 135
Nevena M. Mišljenović, Gordana B. Koprivica, Ljubinko B. Lević,
Bojana V. Filipčev and Tatjana A. Kuljanin
OSMOTIC DEHYDRATION OF RED CABBAGE IN
SUGAR BEET MOLASSES - MASS TRANSFER KINETICS 145
Slaviša S. Putić, Marina R. Stamenović, Branislav B. Bajčeta
and Dragana D. Vitković
DETERMINATION OF TENSION STRENGHT IN THE
LONGITUDINAL AND CIRCUMFERENTIONAL DIRECTION
IN GLASS-POLYESTER COMPOSITE PIPES 155
Aleksandar R. Stanić, Saša I. Jovanić, Nikola J. Marjanović

and Zvonimir J. Suturović
THE USE OF L-ASCORBIC ACID IN SPECIATION
OF ARSENIC COMPOUNDS IN DRINKIG WATER 165
Marina B. Šćiban, Mile T. Klašnja and Mirjana G. Antov

TREATMENT OF SUGAR BEET THICK JUICE SPENT
WASH BY CHEMICAL AND NATURAL COAGULANTS 177
Ivana M. Šijački, Radmilo R. Čolović, Milenko S. Tokić and Predrag S. Kojić

SIMPLE CORRELATIONS FOR BUBBLE COLUMNS
AND DRAFT TUBE AIRLIFT REACTORS WITH
DILUTE ALCOHOL SOLUTIONS 183
BIOCHEMICAL AND PHARMACEUTICAL ENGINEERING
Marija M. Škrinjar and Nevena T. Nemet

ANTIMICROBIAL EFFECTS OF SPICES AND HERBS
ESSENTIAL OILS 195
Dušanka J. Pejin, Olgica S. Grujić, Jelena D. Pejin, Irena S. Došenović

and Sunčica D. Kocić-Tanackov
THE INFLUENCE OF CARBOXYMETHYLCELLULOSE, XANTHAN
AND GUAR-GUM ADDITION IN BREAD DOUGH BEFORE FREEZING
ON METABOLISM AND VIABILITY OF Saccharomyces cerevisiae 211
IN MEMORIAM
Prof. dr Nikola J. Marjanović 223
INSTRUCTION FOR MANUSCRIPT PREPARATION

ACTA PERIODICA TECHNOLOGICA
APTEFF, 40, 1-220 (2009)
САДРЖАЈ
ПРЕХРАМБЕНА ТЕХНОЛОГИЈА
Биљана Р. Цветковић и Марија Р. Јокановић

УТИЦАЈ МЕТОДЕ КОНЗЕРВИСАЊА И УСЛОВА СКЛАДИШТЕЊА
НА ГУБИТАК АСКОРБИНСКЕ КИСЕЛИНЕ У ОСВЕЖАВАЈУЋИМ
БЕЗАЛКОХОЛНИМ ПИЋИМА 1
Гордана Р. Димић, Сунчица Д. Коцић-Танацков, Душанка Ј. Пејин,

Јелена Д. Пејин, Илија Ј. Танацков и Данијела Туцо
АНТИМИКРОБНА АКТИВНОСТ ЕКСТРАКАТА КИМА,
БЕЛОГ ЛУКА И ОРИГАНА НА ФИЛАМЕНТОЗНЕ ПЛЕСНИ 9
Љубица П. Докић, Марија И. Бодрожа-Соларов, Мирослав С. Хаднађев

и Ивана Р. Николић
СВОЈСТВА ЕКСТРУДАТА СА ДОДАТКОМ КРУПИЦЕ
ОД СЕМЕНА АМАРАНТУСА 17
Александар З. Фиштеш и Ђуро М. Вукмировић

ЕФЕКТИ МЛЕВЕЊА ПШЕНИЧНОГ ГРИЗА У КЛАСИЧНОМ
И ПОСТУПКУ СА ОСМОВАЉНОМ СТОЛИЦОМ 25
Гордана Б. Копривица, Невена М. Мишљеновић, Љубинко Б. Левић

и Вјера С. Прибиш
ПРОМЕНА НУТРИТИВНОГ КВАЛИТЕТА ЈАБУКЕ
ПРИ ОСМОТСКОЈ ДЕХИДРАТАЦИЈИ У РАСТВОРИМА
САХАРОЗЕ И МЕЛАСИ ШЕЋЕРНЕ РЕПЕ 35
Радомир В. Малбаша, Ева С. Лончар, Спасенија Д. Милановић

и Љиљана А. Коларов
ПРИМЕНА МЛЕЧНО-ФЕРМЕНТИСАНОГ ИНОКУЛУМА
КОМБУХЕ ЗА ФЕРМЕНТАЦИЈУ МЛЕКА 47
Анамарија И. Мандић, Соња М. Ђилас, Јасна М. Чанадановић-Брунет,

Гордана С. Ћетковић и Јелена Ј. Вулић
АНТИOКСИДАТИВНА АКТИВНОСТ ЕКСТРАКАТА
СЕМЕНА БЕЛОГ ГРОЖЂА НА DPPH РАДИКАЛЕ 53
Спасенија Д. Милановић, Мирела Д. Иличић, Катарина Г. Дураковић
и Владимир Р. Вукић
ТЕКСТУРАЛНЕ ОСОБИНЕ ФЕРMЕНТИСАНИХ МЛЕЧНИХ
НАПИТАКА ДОБИЈЕНИХ ПРИМЕНОМ КОМБУХЕ 63
Драган В. Палић и Sоphia E. Coetzee

ПОСТУПАК ЗА КОРИШЋЕЊЕ РАСТВОРЉИВОСТИ
ПРОТЕИНА КАО ИНДИКАТОРА ТЕРМИЧКОГ ТРЕТМАНА
ПУНОМАСНЕ СОЈЕ 71
Драган В. Палић и Klaas-Jan Leeuw

ПОРЕЂЕЊЕ ТРИ IN VITRO МЕТОДЕ ЗА ОДРЕЂИВАЊЕ И
ПРОЦЕНУ СВАРЉИВОСТИ ОРГАНСКЕ МАТЕРИЈЕ У
ПОТПУНИМ СМЕШАМА ЗА ПРЕЖИВАРЕ 79
Драгана Пешић-Микулец и Гордана Б. Никетић

КВАНТИТАТИВНО ОДРЕЂИВАЊЕ БАКТЕРИЈА МЛЕЧНЕ
КИСЕЛИНЕ КОМЕРЦИЈАЛНИХ УЗОРАКА ЈОГУРТА 87
Слађана М. Саватовић, Александра Н. Тепић, Здравко М. Шумић

и Милан С. Николић
АНТИОКСИДАТИВНА АКТИВНОСТ СОКА ОД ЈАБУКА
ОБОГАЋЕНОГ ПОЛИФЕНОЛНИМ ЈЕДИЊЕЊИМА 95
Мирјана А. Васић, Бисерка Л. Вујичић, Александра Н. Тепић,

Јелица М. Гвозденовић-Варга и Здравко М. Шумић
САДРЖАЈ ДИЈЕТЕТСКИХ ВЛАКАНА У НЕКИМ СОРТАМА ПАСУЉА 103
Тања Д. Жугић-Петровић, Наташа М. Јоковић и Драгиша С. Савић

РАЗВОЈ ПОПУЛАЦИЈЕ БАКТЕРИЈА МЛЕЧНЕ КИСЕЛИНЕ
У ТОКУ ФОРМИРАЊА ЗРЕЛОГ КИСЕЛОГ ТЕСТА 111
ХЕМИЈСКА ТЕХНОЛОГИЈА И ПРОЦЕСНО ИНЖЕЊЕРСТВО
Ева С. Лончар, Мирослава М. Радека, Снежана Б. Петровић, Андреа С. Скапин,

Огњен Љ. Рудић и Јоњауа Г. Раногајец
ОДРЕЂИВАЊЕ ФОТОКАТАЛИТИЧКЕ АКТИВНОСТИ
TiO2 ПРЕВЛАКА НА ЦРЕПУ КОРИШЋЕЊЕМ
МЕТИЛЕН ПЛАВОГ КАО МОДЕЛ ПОЛУТАНТА 125
Наташа Љ. Лукић, Светлана С. Поповић и Јелена Ђ. Марковић

МАТЕМАТИЧКО МОДЕЛОВАЊЕ РЕГЕНЕРАЦИЈЕ ФЛУКСА
ТОКОМ ХЕМИЈСКОГ ЧИШЋЕЊА ТУБУЛАРНЕ МЕМБРАНЕ
ЗАПРЉАНЕ ПРОТЕИНИМА СУРУТКЕ 135
Невена М. Мишљеновић, Гордана Б. Копривица, Љубинко Б. Левић,
Бојана В. Филипчев и Татјана А. Куљанин
ОСМОТСКА ДЕХИДРАТАЦИЈА ЦРВЕНОГ КУПУСА
У МЕЛАСИ ШЕЋЕРНЕ РЕПЕ – КИНЕТИКА ПРЕНОСА МАСЕ 145
Славиша С. Путић, Марина Р. Стаменовић, Бранислав Б. Бајчета

и Драгана Д. Витковић
ОДРЕЂИВАЊЕ ЗАТЕЗНЕ ЧВРСТОЋЕ У УЗДУЖНОМ И ОБИМНОМ
ПРАВЦУ У СТАКЛО-ПОЛИЕСТЕР КОМПОЗИТНИМ ЦЕВИМА 155
Александар Р. Станић, Саша И. Јованић, Никола Ј. Марјановић

и Звонимир Ј. Сутуровић
ПРИМЕНА Л-АСКОРБИНСКЕ КИСЕЛИНЕ ПРИ ОДРЕЂИВАЊУ
РАЗЛИЧИТИХ ОБЛИКА АРСЕНА У ВОДИ ЗА ПИЋЕ 165
Марина Б. Шћибан, Миле Т. Клашња и Мирјана Г. Антов

ТРЕТМАН ЏИБРЕ ОД ГУСТОГ СОКА ХЕМИЈСКИМ
И ПРИРОДНИМ КОАГУЛАНТИМА 177
Ивана М. Шијачки, Радмило Р. Чоловић, Миленко С. Токић и Предраг С. Којић

ПРЕДВИЂАЊЕ ОСНОВНИХ ХИДРОДИНАМИЧКИХ И
МАСЕНОПРОЦЕСНИХ КАРАКТЕРИСТИКА У БАРБОТАЖНИМ
КОЛОНАМА СА И БЕЗ УНУТРАШЊЕ ЦЕВИ СА РАЗБЛАЖЕНИМ
РАСТВОРИМА АЛКОХОЛА 183
БИОХЕМИЈСКО И ФАРМАЦЕУТСКО ИНЖЕЊЕРСТВО
Марија М. Шкрињар и Невена Т. Немет

АНТИМИКРОБНО ДЕЛОВАЊЕ ЕСЕНЦИЈАЛНИХ УЉА ЗАЧИНА
И ЛЕКОВИТОГ БИЉА 195
Душанка Ј. Пејин, Олгица С. Грујић, Јелена Д. Пејин, Ирена С. Дошеновић

и Сунчица Д. Коцић-Танацков
УТИЦАЈ ДОДАТКА КАРБОКСИМЕТИЛЦЕЛУЛОЗЕ, КСАНТАНА
И ГУАР-ГУМЕ У ХЛЕБНО ТЕСТО ПРЕ ЗАМРЗАВАЊА НА
МЕТАБОЛИЗАМ И ВИЈАБИЛНОСТ Saccharomyces cerevisiae 211
IN MEMORIAM
Проф. др Никола Ј. Марјановић 223
УПУТСТВО ЗА АУТОРЕ
FOOD TECHNOLOGY
APTEFF, 40, 1-220 (2009) UDC: 663.86:577.164.2:663.053
DOI: 10.2298/APT0940001C BIBLID: 1450-7188 (2009) 40, 1-7
Original scientific paper
EFFECT OF PRESERVATION METHOD AND STORAGE CONDITION ON

ASCORBIC ACID LOSS IN BEVERAGES
Biljana R. Cvetković and Marija R. Jokanović
Global market is flooded with vitamin-enriched foods, mainly beverages. Major vita-
mins for enriching beverages are the antioxidant vitamins A, C and E. Ascorbic acid is
readily oxidized and lost during storage of the beverages, at rates depending on the con-
ditions of storage. This fact is of great importance for the consumer who must know how
to store beverages and when to consume them in order to get the maximum benefit of ad-
ded vitamin C. The objective of this paper was to determine the amount of ascorbic acid
lost in beverages applying different preservation methods and storage condition. Beve-
rage was made in laboratory conditions with synthetic L-ascorbic acid added according
to the national legislations. After 30 days of storage at 4-8oC ascorbic acid overall loss
was from 81.01% to 90.27% in thermally pasteurized samples and from 97.83 % to al-
most complete loss in samples preserved with sodium benzoate.
KEY WORDS: L-ascorbic acid, colorimetric method, beverage, storage
INTRODUCTION
L-ascorbic acid is largely accepted as additive in human diets because of its anti-
oxidative potential. The richest natural vitamin C sources are fruits and vegetables like
pepper, rose hip, citrus fruit, and green vegetables. Fruits and vegetables supply more
than 90% of vitamin C in human diets (1).
A high recommendation of daily intake for humans has been suggested, since stress in
modern life is known to increase the requirement for vitamin C (2). L-ascorbic acid is
nutrient that besides its vitamin action is valuable for its antioxidant effect, stimulation of
the immune system and other health benefits, such as prevention of scurvy and mainte-
nance of healthy skin, gums and blood vessels. Vitamin C also reportedly reduces the risk
of arteriosclerosis, cardiovascular diseases and some forms of cancer.
Vitamin C is a generic name for all compounds exhibiting the biological activity of L-
ascorbic acid (AA). AA is the principal biologically active form but L-dehydroascorbic
Biljana R. Cvetković, B.Sc., biljana.cvetkovic@fins.uns.ac.rs, Institute for Food Technology, Bulevar Cara La-
zara 1, 21000 Novi Sad; Marija R. Jokanović, M.Sc., Assist., marijaj@tf.uns.ac.rs, Faculty of Technology,
Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia
1
APTEFF, 40, 1-220 (2009) UDC: 663.86:577.164.2:663.053
DOI: 10.2298/APT0940001C BIBLID: 1450-7188 (2009) 40, 1-7
acid (DHA), an oxidation product, also exhibits biological activity (3). Addition of syn-
thetic ascorbic acid increases content of vitamin C, influences maintenance of colour, fla-
vour and universal stability of the food products (fruit juices, beverages, baby food, etc.)
(4). Natural and synthetic L-ascorbic acids are chemically identical and there are no
known differences in their biological activities or bioavailability (5).
Based on available biochemical, clinical and epidemiological studies, the current
recommended daily acceptance for ascorbic acid is suggested to be 100-120 mg/day to
achieve cellular saturation and optimum risk reduction of heart diseases, stroke and can-
cer in healthy individuals (6).
As a consequence of the common man’s increasing awareness regarding the im-
portance of vitamin C, the global market is flooded with vitamin-enriched foods, mainly
beverages. Major vitamins for enriching beverages are the antioxidant vitamins A, C and
E. Vitamin C is usually added as ascorbic acid (7, 8). Fortification is a growing trend in
soft drinks and in the dairy sector, and played a role in 8% of all new food and drink
products introduced in 2003 (9).
L-ascorbic acid application in the food industry increases quality and technological
properties of food, as well as nutritional value (10).
Ascorbic acid is highly sensitive to various modes of deterioration. The main factors
that can affect ascorbic acid loss include temperature, salt and sugar concentration, pH,
oxygen, light, metal catalysts, initial concentration of ascorbic acid, the ratio of ascorbic
acid to dehydroascorbic acid, microbial load and protection by the container (11).
The loss of nutritional quality during processing and storage of food has become an
important problem. Since the discovery of the basic vitamins and their many forms, ef-
forts have been made to retain them in foods during post-harvest, commercial processing,
distribution, storage and preparation. Vitamin C is usually selected as an index of the
nutrient quality because of its labile nature as compared to the other nutrients in food (1).
The term beverages-soft drink originally applied to carbonated and non-carbonated
drinks made from concentrates, although it now commonly refers to almost any cold
drink that does not contain alcohol (12). Ascorbic acid added to beverages is readily
oxidized and lost during staying, at a rate depending on the conditions of storage. This
fact is of great importance to the consumer who must know how to store the beverages
and when to consume them in order to get the maximum benefit of vitamin C content.
(13). Determination of the nutrient content of foods is becoming extremely important as
researchers learn more about the relationship between dietary intake and human health
(14).
Various methods have been reported in the literature for the quantitative deter-
mination of vitamin C in foods or biological fluids. The usual methods include titration
(AOAC), colourimetric, spectrophotometry, fluorometry, electrophoresis, and high per-
formance liquid chromatography (HPLC).
Objectives of this paper were:
a) preparation of non-alcoholic beverages in laboratory conditions with addition of
synthetic ascorbic acid and two methods of preservation;
b) analysis of the amount of ascorbic acid loss in samples during 30 days under
different storage conditions, in closed glass bottles, storage in the refrigerator,
and in the dark at room temperature and in a thermostat at 37oC.
2
APTEFF, 40, 1-220 (2009) UDC: 663.86:577.164.2:663.053
DOI: 10.2298/APT0940001C BIBLID: 1450-7188 (2009) 40, 1-7
EXPERIMENTAL
Materials
Beverage preparation. Beverage was prepared by diluting commercial beverage
concentrate (Aretol no. 72 408, Celje, Slovenia) in water in a ratio 3:97. Sugar was added
like inverted syrup (dry matter content in the final product about 9.5 %), and then citric
acid was added to get appopriate sensory characteristics. There were alltogether prepared
12 samples. Ascorbic acid was added in two concentrations: 100 mg/l and 150 mg/l. For
thermal pasteurisation were transferred into 200 ml glass bottles, second part of samples
were first preserved with sodium benzoate and than transferred into 200 ml glass bottles.
Preservation. Twelve samples (with both AA added concentrations) were divided in
two parts:
a) One part was thermally treated (pasteurisation). Thermal pasteurisation condi-
tions (85oC, 15 minutes) were selected to be the same as in a conventional
pasteurisation of industrially produced beverages.
b) The other part was preserved with sodium benzoate in concentration of 130 mg
/l, in the final product, which is in accordance with national legislations (15).
Shelf-life study. Samples of beverages thermally pasteurised or preserved with sodi-

um benzoate were stored in three different temperatures conditions: refrigerator (4-8oC),
room temperature (20-22oC), and in a thermostat at 37oC. Samples were evaluated after
30 days of storage, by measuring ascorbic acid content.
Method
Determination of L-ascorbic acid. Ascorbic acid content was determined using co-
lourimetric method. L-ascorbic acid reduces the tetrazolium salt MTT (3-(4,5-dimethyl-
thiazolyl-2)-2,-diphenyltetrazolium bromide) in the presence of the electron carrier PMS
(5-methylphenazinium methosulphate) at the pH 3.5 to a formazan (MTT-formazan),
which is determined by measuring absorbance at 578 nm. Under the conditions stated in
this procedure, assay is specific for L-ascorbic acid. The L-ascorbic acid content of these
clear solutions was determined without any sample treatment. The detection limit of the
method was 0.175 mg/l.
Each value was measured in triplicate and averaged with standard deviation.
RESULTS AND DISCUSSION
According to our results pasteurisation method had high influence on vitamin C

content. Thermal pasteurisation gradually decreased L-ascorbic acid content. Ascorbic
acid contents immediately after thermal pasteurisation in samples with 150 mg/l and 100
mg/l of added vitamin C, were 37.66 mg/l and 25.84 mg/ respectively (Table 1).
3
APTEFF, 40, 1-220 (2009) UDC: 663.86:577.164.2:663.053
DOI: 10.2298/APT0940001C BIBLID: 1450-7188 (2009) 40, 1-7
Table 1. Average ascorbic acid content with standard deviation and it loss in samples
immediately after pasteurisation and preservation
Ascorbic acid content (mg/l)

Samples
Pasteurisation Sodium benzoate preservation
100 mg/l added ascorbic acid 25.84 ± 3.58 71.19 ± 8.02
150 mg/l added ascorbic acid 37.66 ± 2.98 113.40 ± 8.35
According to Blasco et al. (2004), there are two different rates of ascorbic acid degra-
dation observed during the heating process: an aerobic degradation followed by an anae-
robic degradation. In the beginning of the heating process oxygen remains in the bottle
and therefore aerobic degradation of the ascorbic acid with oxygen in abundance takes
place. With prolonged time of heating the atmosphere in the bottle becomes saturated
with vapour, so that the oxygen concentration is minimal and the ascorbic acid is degra-
ded anaerobically (16).
During the preservation of beverages with sodium benzoate the loss of added ascorbic
acid was much lower than in thermally treated samples. Immediately after preservation,
the loss of ascorbic acid in samples with sodium benzoate was 24.40 % and 28.81 % for
samples with 150 mg/l and 100mg/l added ascorbic acid, respectively.
80 74.16 74.89 100 mg/l added vitamin C

70 150 mg/l added vitamin C
Ascorbic acid loss (%)
60
50
40
28.81
30 24.4
20
10
0
Thermal pasteurisation Sodium benzoate preservation
Fig. 1. Ascorbic acid loss immediately after the two methods of preservation
Table 2. Average ascorbic acid content (mg/l) with standard deviations in the beverages
after 30 days of storage at 4-8, 20-22 and 37oC
Pasteurisation Preserved with sodium-benzoate

Storage
temperature AA AA AA AA
150 mg/l 100 mg/l 150 mg/l 100 mg/l
4-8oC 28.37 ± 1.26 2.17 ± 0.30 14.6 ± 1.99 0.43 ± 0.08
20-22oC nd* nd nd nd
37oC nd nd nd nd
*nd - not detected; AA- ascorbic acid
4
APTEFF, 40, 1-220 (2009) UDC: 663.86:577.164.2:663.053
DOI: 10.2298/APT0940001C BIBLID: 1450-7188 (2009) 40, 1-7
100 mg/l added vitamin C 150 mg/l added vitamin C
120
Ascorbic acid loss (%)
97.83 99.57
100 90.26
81.01
80
60
40
20
0
Thermal pasteurisation Sodium benzoate preservation
Fig. 2. Ascorbic acid loss after 30 days of storage in a refrigerator (4-8 oC)
Storage temperature had a great influence on ascorbic acid loss. After 30 days of sto-
rage at room temperature (20-22oC) and in thermostat at 37oC ascorbic acid was not de-
tected in any sample. After 30 days of storage at 4-8oC and thermal pasteurisation the
overall loss of ascorbic acid was 81.01 % in samples with 150 mg/l added ascorbic acid,
and 97.83 % in samples with 100 mg/l added ascorbic acid. In the beverages preserved
with sodium benzoate after one month of storage at 4-8oC ascorbic acid overall loss was
from 90.27 % in samples with added concentration of 150 mg/l to almost complete loss
for samples with 100 mg/l of added ascorbic acid (Fig. 2). Heating method had a definite
influence on the retention of ascorbic acid. According to Vikram et al. (2005), by each
heating method of orange juice, temperature had a greater influence and the degradation
was rapid at higher temperatures (17).
The decrease of ascorbic acid concentration to levels unacceptable by declaration or
industrial practise often defines the product shelf life. During storage, the vitamin C con-
tent gradually decreases at a rate depending on the processing and storage temperature.
The more rapid decrease of ascorbic acid concentration at the beginning of the storage
can be attributed to the immediate reaction of an amount of ascorbic acid with the dissol-
ved oxygen (18).
Degradation of ascorbic acid both by aerobic and anaerobic pathways depends upon
many factors such as oxygen, heat, light, storage temperature and storage time. Oxidation
of ascorbic acid occurs mainly during the processing, whereas, anaerobic degradation of
vitamin C mainly appears during storage, which is especially observed in thermally pre-
served juices (10).
CONCLUSION
The decrease of vitamin C content to levels unacceptable by declaration or industrial

practise often defines product shelf-life. During storage, the vitamin C content gradually
decreases at a rate depending on processing and storage temperature. The more rapid de-
crease of ascorbic acid concentration at the beginning of the storage can be attributed to
the immediate reaction of an amount of ascorbic acid with the dissolved oxygen (18).
5
APTEFF, 40, 1-220 (2009) UDC: 663.86:577.164.2:663.053
DOI: 10.2298/APT0940001C BIBLID: 1450-7188 (2009) 40, 1-7
According to Blasco et al. (2004) there are two different rates of ascorbic acid de-
gradation observed during the heating process: an aerobic degradation, followed by an
anaerobic degradation. In the beginning of the heating process oxygen remains in the
bottle and therefore aerobic degradation of the ascorbic acid with oxygen in abundance
takes place. With prolonged time of heating, the atmosphere in the bottle becomes satu-
rated with vapour, so that the oxygen concentration is minimal and the ascorbic acid is
degraded anaerobically (16).
The experiments have shown that L- ascorbic acid added like additive in non-alco-
holic beverage is extremely unstable in water solution. The samples with a lower initial
content of ascorbic acid lose it faster than those with a greater content. Ascorbic acid loss
was greater in preserved than in pasteurised beverages. It should be recommended that
beverage with ascorbic acid added should be consumed after preparation with no long
time of storage.
ACKNOWLEDGEMENT
The authors gratefully acknowledge the financial support from the Ministry of Scien-
ce and Technological Development of the Republic of Serbia (Project TP-20068).
REFERENCES
1. Erentuk. S, Gualaboglu M.S. and S. Gultekin: The effects of cutting and drying
medium on the vitamin C content of rosehip during drying. Journal of Food Engine-
ering 68 (2005) 513-518.
2. Olivier Fain: Musculoskeletal manifestations of scurvy, Review, Joint Bone Spine
(2004).
3. Lee S. K. and Adel A. Kader: Preharvest and post harvest factors influencing vitamin
C content of horticultural crops. Post harvest Biology and Technology 20 (2000) 207-
220.
4. Del Caro A.: Changes of flavonoids, vitamin C and antioxidant capacity in minimally
processed citrus segments and juices during storage. Food Chemistry 84 (2004) 99-
105.
5. Lee H.S. and G. A. Coates: Vitamin C in frozen, fresh squeezed, unpasturized, poly-
ethylene-bottled orange juice: storage study. Food Chemistry 65 (1999) 165-168.
6. Klimezak I., Malecka M., Szlahta M. and A. Gliszezynska-Swiglo: Effect of storage
on the content of polyphenols, vitamin C and the antioxidant activity of orange juices.
Journal of Food Composition and Analysis 20 (2007) 313-322.
7. Arya S. P., Mahajan M. and P.Jain: Non-spectrophotometric methods for the deter-
mination of Vitamin C. Analytica Chimica Acta 417 (2000) 1-14.
8. Rodriguez-Comesana M., Garcia-Falcon M. S. and J. Simal-Gandara: Control of nu-
tritional labels in beverages with added vitamins: screening of β-carotene and ascor-
bic acid contents. Food Chemistry 79 (2002) 141-144.
9. Prieto P. S., Grande C. B., Falkon G. S. and S. J. Gandara: Screening for folic acid
content in vitamin-fortified beverages. Food Control 17 (2006) 900-904.
10. Burdurly H. S., Koca N. and F .Karadeniz: Degradation of vitamin C in citrus juice
concentrates during storage. Journal of Food Engineering 74 (2006) 211-216.
6
APTEFF, 40, 1-220 (2009) UDC: 663.86:577.164.2:663.053
DOI: 10.2298/APT0940001C BIBLID: 1450-7188 (2009) 40, 1-7
11. Zerdin K., Michael L. R. and J.Vermue: The vitamin C content of orange juice
packed in an oxigen scavenger material. Food Chemistry 82 (2003) 387-395.
12. www.wikkipedia.org
13. Kabasakalis V, Siopidou D. and E. Moshatou: Ascorbic acid content of commercial
fruit juices and its rate of loss upon storage. Food Chemistry 70 (2000) 325-328.
14. Gokmen V., Kahraman N., Demir N. and J. Acar: Enzymatically validated liquid
chromatographic method for the determination of ascorbic and dehydroascorbic acids
in fruit and vegetables. Journal of Chromatography A, 881 (2000) 309-316.
15. Bylaw of quality and other conditions for aditives and their mixtures for food pro-
ducts (Službeni list SRJ articles no. 56/2003, 4/2002 and 16/2005).
16. Blasco R., Esteve M. J., Frigola A. and M. Rodrigo: Ascorbic acid degradation kine-
tics in mushrooms in a high-temperature short time process controlled by a thermo-
resistometer. Lebensm.-Wiss. u.-Technol, 37 (2004) 171-175.
17. Vikram V. B., Ramesh M.N. and S.G. Prapulla: Thermal degradation kinetics of nu-
trients in orange juice by electromagnetic and conventional methods. Journal of Food
Engineering 69 (2005) 31-40.
18. Polydera A.C, Stoforos N.G. and P.S. Taoukis: Comparative shelf life study and
vitamin C loss kinetics in pasteurized and high pressure processed reconstituted oran-
ge juice. Journal of Food Engineering 60 (2003) 21-23.
УТИЦАЈ МЕТОДЕ КОНЗЕРВИСАЊА И УСЛОВА СКЛАДИШТЕЊА НА

ГУБИТАК АСКОРБИНСКЕ КИСЕЛИНЕ У ОСВЕЖАВАЈУЋИМ
БЕЗАЛКОХОЛНИМ ПИЋИМА
Биљана Р. Цветковић и Марија Р. Јокановић
Светско тржиште је преплављено витамински обогаћеним производима, углав-

ном освежавајућим безалкохолним пићима. Најчешће се производи обогаћују
антиоксидантима, односно витаминима А, Ц и Е. Л-асакорбинска киселина у осве-
жавајућим безалкохолним пићима врло брзо оксидише и разлаже се током скла-
диштења, у зависности од услова чувања. Ова чињеница је од велике важности за
потрошаче који треба да знају како да чувају и када да конзумирају освежавајућа
безалкохолна пића (која су обогаћена аскорбинском киселином) у циљу максимал-
не користи од додатог витамина Ц. Задатак овог рада је био мерење опадања кон-
центрације витамина Ц у узорцима током различитих услова скледиштења. Осве-
жавајуће безалкохолно пиће је припремљено у лабораторијским условима у складу
са важећим Правилником, уз додатак синтетске Л-аскорбинске киселине. Током
складиштења од 30 дана на температури од 4-8оС губитак аскорбинске киселине је
81.01- 97.83% у пастеризованом узорку пића, а у конзервисаном 90- 99%.
Received 17 November 2008

Accepted 26 February 2009
7
APTEFF, 40, 1-220 (2009) UDC: 664.5:66.061.3:582.28
DOI: 10.2298/APT0940009D BIBLID: 1450-7188 (2009) 40, 9-16
ANTIMICROBIAL ACTIVITY OF CARAWAY, GARLIC AND OREGANO

EXTRACTS AGAINST FILAMENTOUS MOULDS
Gordana R. Dimić, Sunčica D. Kocić-Tanackov, Dušanka J. Pejin, Jelena D. Pejin,

Ilija J. Tanackov and Danijela Tuco
Inhibitory effect of caraway, garlic and oregano extracts (0.07, 0.1, 0.5, 1 and 2%),
against four moulds species was investigated. The caraway extract had the strongest in-
hibitory effect by inhibiting the germination of Emericella nidulans, Penicillium com-
mune and P. implicatum at the concentration of 0.1% and Aspergillus tamarii at the con-
centration of 0.5% during 7 days of incubation at 25oC. The extract of garlic only parti-
ally inhibited the growth of A. tamarii and P. commune. However, it inhibited completely
the growth of P. implicatum and E. nidulans at the doses of 0.5 and 1%. Oregano parti-
ally inhibited all mould species, significantly reducing the growth of colonies, especially
of E. nidulans (93.3%).
KEY WORDS: Spice extracts, antifungal activity
INTRODUCTION
Moulds highly prevail in nature and frequently contaminate human food. Some of
them produce secondary metabolites such as aflatoxins, ochratoxin A, stergmatocystine,
which are cytotoxic and carcinogenic, and as such present a potential health hazard for
humans (1). Medium moisture food (0.75-0.90 aw), low moisture food (< 0.75 aw) and
sour food are especially susceptible to the presence of moulds. The development of mo-
ulds on food can be expected in cases when inappropriate sanitary practice is applied in
production plants. Moulds occur more frequently than other microorganisms on food pro-
ducts during storage and distribution as a consequence of inadequate conditions.
Essential oils extracted from spices and other herbs, as well as their biologically
active components, have been intensively investigated for their potential role in the pro-
tection of food from microorganisms, especially the foodstuffs with short shelf-life, such
as bread, bakery products, cakes, salads, fresh fruits and vegetable, fish, etc. which are
the most susceptible to microbial spoilage. Being natural antimicrobial agents, their usage
Dr. Gordana R. Dimić, Assoc. Prof., gordanad@tf.uns.ac.rs, Sunčica D. Kocić-Tanackov, M.Sc., Assist., Dr.
Dušanka J. Pejin, Prof., Dr. Jelena D. Pejin, Assist., Faculty of Technology, University of Novi Sad, Bulevar
Cara Lazara 1, 21000 Novi Sad, Serbia; Dr. Ilija J. Tanackov, Assoc. Prof., Faculty of Technical Science,
University of Novi Sad, Trg Dositeja Obradovića 6, 21000 Novi Sad, Serbia; Danijela Tuco, B.Sc. Etol JVE
d.o.o., Bulevar Vojvode Stepe 40, 21000 Novi Sad, Serbia
9
APTEFF, 40, 1-220 (2009) UDC: 664.5:66.061.3:582.28
DOI: 10.2298/APT0940009D BIBLID: 1450-7188 (2009) 40, 9-16
can minimize the application of synthetic preservatives and additives, preserving simulta-
neously food freshness and sensory quality. Preserving properties of spices and their
extracts have been recognized long ago; their residues have been found on old Egyptian
mummies (2) and there are evidence of their usage as antiseptic agents (3).
Studies have shown that some spices like vanilla do not possess antifungal activity (4)
whereas some of them have a stimulating effect (5, 6, 7, 8).
This study was aimed at investigating the antifungal potential of caraway, garlic, and
oregano extracts against some food-borne fungi.
EXPERIMENTAL
Materials
Commercially available food grade ethanol extracts of caraway, garlic, and oregano
were provided from Etol, Celje, Slovenia. Test cultures for antifungal investigations, As-
pergillus tamarii, Emericella nidulans, Penicillium commune and P. implicatum were ta-
ken from the culture collection of the Laboratory for Food Microbiology, Faculty of
Technology in Novi Sad, isolated from food. The cultures were maintained on potato-
dextrose agar (PDA) slants at 4º C.
Preparation of inoculum
Prior to the experiment, moulds were cultured on PDA slants for 10 days until fully
sporulated. Spores were taken by adding 10 ml of medium which contained 0.5% Tween
80 and 0.5% agar in sterile distilled water (4), scraped with sterile loop and aseptically
transferred into sterile test tubes. Spore suspension obtained in this way was adjusted to
final concentration of 2×106 spores/ml using the hemocytometer, and used for further
work.
Antifungal test
The inhibition of mould growth was determined by performing daily measurements

of the radial growth of colonies cultured on PDA medium which contained spice extracts
(each plate separately) in the following concentrations 0.07, 0.1, 0.5, 1 i 2% (v/v). For
test moulds, PDA plates without any added material were made and used as control pla-
tes. The solid plates were inoculated with spore suspension containing 1µl (103 spo-
res/ml) in the centre of the medium and were incubated for 7 days at 25oC. Diameter of
the growth was determined by averaging the radial growrth of the colony in two ortho-
gonal directions. Each test was run in triplicate.
Inhibitory concentrations for caraway, garlic and oregano extracts against A. tamarii,
E. nidulans, P. commune and P. implicatum are presented in Table 1.
10
APTEFF, 40, 1-220 (2009) UDC: 664.5:66.061.3:582.28
DOI: 10.2298/APT0940009D BIBLID: 1450-7188 (2009) 40, 9-16
Table 1. The inhibitory activities of spice extracts against moulds
Inhibition colony growth (%)

Extract Conc. (%)
A. tamarii E. nidulans P. commune P. implicatum
Caraway 0.07 12.7 20.0 14.8 11.1
0.1 22.2 33.3 37.0 77.8
0.5 47.6 100 100 100
1 100 100 100 100
2 100 100 100 100
Garlic 0.07 1.6 22.2 11.1 11.1
0.1 14.3 31.1 18.5 33.3
0.5 25.4 73.3 26.0 100
1 30.1 100 33.3 100
2 33.3 100 59.2 100
Oregano 0.07 12.7 11.1 14.8 5.5
0.1 17.5 15.5 18.5 11.1
0.5 30.1 31.1 22.2 16.7
1 47.6 48.9 29.6 55.5
2 79.4 93.3 74.1 66.7
As can be seen from data presented, the caraway extract exhibited the strongest inhi-
bitory activity that was particularly expressed against E. nidulans and both Penicillium
species (P. commune and P. implicatum) which did not grow at extract doses over 0.1%.
At this level, the growth of P. implicatum was markedly reduced (77.8%). The level over
0.5% was needed to completely inhibit A. tamarii.
The garlic extract at 0.5 and 1% concentrations inhibited only partially the growth of
A. tamarii and P. commune and completely the growth of P. implicatum and E. nidulans.
If compared to A. tamarii, stronger antifungal activity against P. commune was observed
in all applied concentrations. The level of growth reduction in the presence of garlic ex-
tract for A. tamarii ranged from 1.6 to 33.3% and for P. commune between 11.1 to
59.2%.
Although none of the tested species was completely inhibited by the oregano extract,
high concentrations were found to significantly inhibit the growth of colonies. The 2%
extract inhibited almost completely (93.3%) the growth of E. nidulans. Against A. tamarii
and P. commune, the same extract concentration exhibited approximately 70% inhibition
rate (79.4 and 74.1%, respectively), whereas P. implicatum was found to be the least sen-
sitive species.
The effect of caraway, oregano and garlic extracts on the germination and growth rate
of moulds is presented in Figures 1-3.
At the concentrations 0.1 and 0.5%, the caraway extract delayed the beginning of ger-
mination of A. tamarii by two and three days, respectively, as compared to the control.
The appearance of the growth of E. nidulans and P. commune was not under the influ-
ence by the increased extract concentration; however, the differences in their growth rate
were noticed during the next days. At 0.1% concentration, colonies of P. implicatum be-
came visible only on the sixth day after the inoculation of agar plates.
11
APTEFF, 40, 1-220 (2009) UDC: 664.5:66.061.3:582.28
DOI: 10.2298/APT0940009D BIBLID: 1450-7188 (2009) 40, 9-16
Fig. 1. Effect of caraway extract on the growth of moulds
Fig. 2. Effect of garlic extract on the growth of moulds
12
APTEFF, 40, 1-220 (2009) UDC: 664.5:66.061.3:582.28
DOI: 10.2298/APT0940009D BIBLID: 1450-7188 (2009) 40, 9-16
The growth rate decline with increase of garlic extract contents in the agar medium
was especially pronounced in the case of P. implicatum and E. nidulans, pointing to the
greater sensitivity of these spesies (Fig. 2). Higher concentrations did not significantly
influence the appearance of A. tamarii, whereas P. commune did not grow in the presence
of 2% garlic extract until the fourth day.
At lower levels of oregano extract (Fig. 3), the begining of germination was delayed
by two days only in the case of P. implicatum, at the concentration of 0.5%. The 1 and
2% concentrations delayed the growth of P. implicatum by two and four days and E. ni-
dulans by two and five days. The growth of A. tamarii and P. commune was delayed
only at the concentracion of 2%, for three days. Stronger inhibitory effect on the growth
rate of E. nidulans was noticed at the concentrations over 0.1% for A. tamarii and P.
implicatum over 0.5% and at P. commune over 1%.
Fig. 3. Effect of oregano extract on the growth of moulds
The increasing concentrations of caraway, garlic and oregano extracts caused the ab-
sence or delay in germination of tested fungi, showing various inhibitory effects on the
growth rate reduction. Caraway was more efficient at lower concentrations, compared to
garlic and oregano. Moreover, it was the only extract to inhibit the growth of three spe-
cies (out of the four tested) during the whole period of incubation (7 days) at 25oC. Pre-
vious studies also reported strong inhibitory effect of caraway on Penicillium species.
Studies have shown that P. aurantiogriseum, P. corylophilum, P. commune and P. griseo-
fulvum were completely inhibited at 1% dose (8, 9).
Antifungal properties of the tested extracts are due to their major constitutive com-
ponents, carvacrol (from caraway and oregano), limonene (from caraway), thymol (from
13
APTEFF, 40, 1-220 (2009) UDC: 664.5:66.061.3:582.28
DOI: 10.2298/APT0940009D BIBLID: 1450-7188 (2009) 40, 9-16
oregano) and sulfur compounds (from garlic) (10, 11). Phenol compounds such as car-
vacrol, thymol, eugenol, vanillin, geraniol and cinnamaldehyd are known antimicrobial
agents (11-17). Phenolic OH-group is very reactive and easily forms hydrogen bonds
with active sites in enzymes (6). According to Soliman and Badea (18), caraway showed
inhibitory effect on Aspergillus flavus and A. parasiticus at 2000 ppm dose and on A.
ochraceus and Fusarium moniliforme at 3000 ppm. Nielsen and Rios (4) showed that
esential oils of mustard, garlic and clove are effective in preventing the growth of moulds
usually present in bread. Garlic show antifungal activity against certain Aspergillus, Pe-
nicillium and Fusarium species (19, 20). Guyenot et al. (3) investigated the protective
effect of essential oils of 16 spices against xerophilic fungi from genera Eurotium, Asper-
gillus and Penicillium, common spoilage organisms in bakery products. It was also re-
ported that oregano extract is capable to completely inhibit Aspergillus parasiticus at the
level of 2% in agar medium (6), which is in compliance with our results. Essential oils of
cinnamon and oregano proved to be very effective in growth inhibition of Fusarium
proliferatum (21).
CONCLUSION
This study proved that the tested spice extracts can be potentially protective agents
against filamentous fungi, frequent contaminants of food. Caraway extract exhibited high
efficacy already at 0.5% dose. Garlic was the most effective against E. nidulans and P.
implicatum. Oregano exhibited strong inhibitory effect although was unable to comple-
tely inhibit the fungal growth.
REFERENCES
1. van Egmond, P.H.: Mycotoxins: detection, reference materials and regulation, in

Introduction to food- and airborne fungi. Eds. R.A. Samsin, E.S. Hoekstra and J.C.
Frisvad, Ponson & Looyen, Wageningen, Utrech, Netherlands (2005) pp.332-338.
2. Silliker, J.H., R.P. Elliott, A.C. Bairol-Barker, F.L. Bryan, J.H.B. Christian, D.S.
Clark, J.C. Olson and Jr.T.A. Roberts: Microbial Ecology of Foods. Vol.II, Academic
Press, New York, London (1980).
3. Guynot, E.M., J. A. Ramos, L. Seto, V. Purroy, V. Sanchis and S. Marin: Antifungal
activity of volatile compounds generated by essential oils against fungi commonly
causing deterioration of bakery products. J. Appl. Microbiol. 94 (2003) 893-899.
4. Nielsen, V.P. and R. Rios: Inhibition of fungal growth on bread, by volatile com-
ponents from spices and herbs, and the possible application in active packaging, with
special emphasis on mustard essential oil. Int. J. of Food Microbiology 60 (2000)
219-229.
5. Mabrouk, S.S. and M.A.N. El-Shayeb: Inhibition of aflatoxin formation by some
spices. Z. Lebensm. Unters. Forsch. 171 (1980) 344-347.
6. Özcan, M.: Inhibitory effect of spices extracts on the growth of Aspergillus para-
siticus NRRL2999 strain. Z. Lebensm. Unters Forsch A 207 (1998) 253-255.
7. Boyraz, N. and M. Özcan: Antifungal effect of some spice hydrosols. Fitoterapia 76
(2005) 661-665.
14
APTEFF, 40, 1-220 (2009) UDC: 664.5:66.061.3:582.28
DOI: 10.2298/APT0940009D BIBLID: 1450-7188 (2009) 40, 9-16
8. Dimić, G. and S. Kocić-Tanackov: Antifungal activity of spices extracts on the

growth of Penicillium commune and Penicillium griseofulvum, 6th Congress of Medi-
cal Microbiology, Beograd, 11-14 June 2008, Book of Abstracts, pp. 335-336.
9. Dimić, G., S. Kocić-Tanackov and D. Karalić: Inhibitory Effect of Spice Extracts on
Growth of Penicillium aurantiogriseum and Penicillium corylophilum, 54th Interna-
tional Meat Industry Conference, Vrnjačka Banja, 18-20 June 2007, Book of Abs-
tracts, p. 89.
10. Baratta, M.T., H.J.D. Dorman, S.G. Deas, D.M. Biondi and G. Ruberto: Chemical
composition, antimicrobial and antioxidative activity of laurel, sage, rosmary, orega-
no and coriander essential oils. J. Essent. Oil Res. 10, 6 (1998) 618-627.
11. Ceylon, E. and C.Y. D. Fung: Antimicrobial activity of spices. J. Rapid. Meth. Au-
tom. Microbiol. 12 (2004) 1-55.
12. Bullerman, L.B., F.Y. Lien and S.A. Seir: Inhibition of growth and aflatoxin pro-
duction by cinnamon and clove oils, cinnamic aldehyde and eugenol. J. of Food Sci-
ence 42 4 (1977) 11007-1116.
13. Hitokoto, H., S. Morozumi, T. Wauke, S. Sakai and H. Kurata: Inhibitory effects of
spice on growth and toxin production of toxigenic fungi. Appl. Environ. Microbiol.
39 4 (1980) 818-822.
14. Moleyar, A.V. and P. Narasimham: Antifungal activity of some essential oil com-
ponents. Food Microbiol. 3 (1986) 331-336.
15. Mahmound, A.L.E.: Antifungal action and antiaflatoxigenic properties of some essen-
tial oil constituents. Lett. Appl. Microbiol. 19 (1994) 110-113.
16. Matamoros-Leon, B., A. Argaiz and A. Lopez-Malo: Individual and combined effects
of vanillin and potassium sorbate on Penicillium digitatum, Penicillium glabrum, and
Penicillium italicum growth. J. Food Protect. 62 5 (1999) 540-542.
17. Moriera, M.R., A.G. Ponce, C.E. del Valle and S.I. Rouza: Inhibitory parameters of
essential oils to reduce a foodborne pathogen. LWT 38 (2005) 565-579.
18. Soliman,K.M. and R.I. Badeaa: Effect of oil extrated from some medicinal plants on
different mycotoxigenic fungi. Food Chem. Toxicol. 40 (2002) 1669-1675.
19. Mei-chin, Y. and T. Shih-ming: Inhibitory effect of seven Allium plants upon three
Aspergillus species. Int. J. Food Microbiol. 49 (1999) 49-56.
20. Benkeblia, N.: Antimicrobial activity of essential oil extracts of various onions (Alli-
um cepa) and garlic (Allium sativum). LWT 37 (2004) 263-268.
21. Velluty, A., V. Sanchis, A.J. Ramos, J. Egido and S. Marin: Inhibitory effect of cin-
namon, clove, lemongrass, oregano and palmarose essential oils on growth and fumo-
nisin B1 production by Fusarium proliferatum. Int. J. Food Microbiol 89 (2003) 145-
154.
15
APTEFF, 40, 1-220 (2009) UDC: 664.5:66.061.3:582.28
DOI: 10.2298/APT0940009D BIBLID: 1450-7188 (2009) 40, 9-16
АНТИМИКРОБНА АКТИВНОСТ ЕКСТРАКАТА КИМА, БЕЛОГ ЛУКА И

ОРИГАНА НA ФИЛАМЕНТОЗНЕ ПЛЕСНИ
Гордана Р. Димић, Сунчицa Д. Коцић-Танацков, Душанка Ј. Пејин, Јелена Д. Пејин,

Илија Ј. Танацков и Данијела Туцо
Инхибиторнa активност екстраката кима, белог лука и оригана (0,07, 0,1, 0,5, 1
и 2%) је испитивана против четири врсте плесни. Екстракт кима је имао најјачи
инхибиторни ефекат спречавајући герминацију Emericella nidulans, Penicillium com-
mune и P. implicatum при концентрацији од 0,1% и Aspergillus tamarii при концен-
трацији од 0,5% током седам дана инкубирања на 25°C. Екстракт белог лука је само
парцијално инхибирао раст А. tamarii и P. commune и потпуно P. implicatum и Е.
nidulans при концентрацијама 0,5 и 1%. Оригано је парцијално инхибирао све че-
тири врсте плесни, али је раст колонија био значајно смањен, нарочито код Е.
nidulans (93,3%).
Received 24 July 2009

Accepted 14 September 2009
16
APTEFF, 40, 1-220 (2009) UDC:664.64.016.71.8:664.641.15+664.641.2:664.696
DOI: 10.2298/APT0940017D BIBLID: 1450-7188 (2009) 40, 17-24
PROPERTIES OF EXTRUDED SNACKS SUPPLEMENTED WITH AMARANTH

GRAIN GRITS
Ljubica P. Dokić, Marija I. Bodroža-Solarov, Miroslav S. Hadnađev

and Ivana R. Nikolić
Extruded amaranth grain products have specific aroma and can be used as snack
food, supplement in breakfast cereals, or as raw material for further processing. Extru-
ded products of corn-amaranth grits blends, containing 20% or 50% amaranth grain
grits, were produced by extrusion-cooking using a laboratory Brabender single screw
extruder 20 DN. Extrudates with various texture were obtained.
During extrusion process starch granules are partially degraded, hence rheological
properties were examined. All samples exhibited thixotropic flow behavior. Those sam-
ples in which part of the corn grits was replaced with amaranth one had lower viscosity
and exhibited lower level of structuration during storage.
KEY WORDS‫ ׃‬Amaranth, extrudate, proteins, starch, rheological, properties
INTRODUCTION
Amaranth grain has significant nutritional value. Its protein, mineral meters, fat and
cellulose percentage are higher compared to cereals (1, 2). Since this plant has similar
application as cereals, it is classified as pseudocereal (2). The origin of Amaranthus sp. is
from middle and south America, but for the last few decades this cultivar was introduced
in European countries as Austria, Poland, Hungary, Serbia and Montenegro (2, 3).
Extruded amaranth grain products have specific aroma and can be used as snack food,
supplement in breakfast cereals, or as raw material for further processing (4). Sanches–
Marroquim et al. investigated extruded blends of Amaranth with wheat and outs flour.
The optimal combination of these components, to their opinion is 50‫׃‬50, or 60‫׃‬40, res-
pectively (5).
Starch is an important industrial raw material for food products as well as for techni-
cal products. Corn starch is relatively easily isolated by wet-milling procedure. Amaranth
Dr. Ljubica Dokić, Assoc. Prof, ldokic@uns.ac.rs, Faculty of Tehnology, Bulevar Cara Lazara 1, 21000 Novi
Sad, Serbia; Dr. Marija Bodroža-Solarov, Miroslav Hadnađev, M. Sc. Institute for Food Technology, Bulevar
Cara Lazara 1, 21000 Novi Sad, Serbia; Ivana Nikolić, B. Sc., Faculty of Tehnology, Bulevar Cara Lazara 1,
21000 Novi Sad, Serbia
17
APTEFF, 40, 1-220 (2009) UDC:664.64.016.71.8:664.641.15+664.641.2:664.696
DOI: 10.2298/APT0940017D BIBLID: 1450-7188 (2009) 40, 17-24
starch is interesting because of its small granular size, 1–2 μm, which provides specific
functional properties, as good freeze–thaw stability and resistance to mechanical shear.
Amaranth starch content is approximately 63%, while protein content is 15%. The starch
has „waxy” characteristics and specific molecular composition, while proteins have high
quality and major content of S–amino acids and lysine (6).
There is no effective and easy method to isolate the starch from amaranth seed be-
cause of its small granules and high protein content. Considering the fact that starch and
protein are of high quality, some other methods were developed, like dry milling and
separation or extrusion process with the aim to provide products useful for food or non-
food purposes. During extrusion process, properties of native starch granules are modi-
fied and pregelatinized starch with changed rheological characteristics is obtained (7).
Numerous products with starch undergo thermal treatments during production and
application, when starch gels with specific rheological characteristics are formed. Mo-
dification of rheological properties can be achieved by mixing different starches. The aim
of this work was to investigate rheological characteristics of gels of amaranth and corn
grits blends, before and after extrusion process.
EXPERIMENTAL
Amaranth and corn were acquired from local commercial sources. Grits was obtained
by milling whole grain in laboratory mill (Bühler 202, Germany). Blends were prepared
in the following ratios: amaranth:corn, 20:80, 50:50 and 0:100, respectively.
Particle size distribution of grits is presented in Table 1.
Table 1. Particle size distribution of corn and amaranth grits
Fractions remained on sieves of particular size, (%)

Sample
850 μm 650 μm 450 μm 350 μm 250 μm
Corn grits 1.2 44.1 50.3 3.7 0.7
Amaranth grits 0.8 43.0 41.2 9.0 6.0
Grits blends were extruded using a laboratory Brabender single screw extruder 20
DN. Before extrusion, the moisture content of the grits was adjusted to 16%. Such mois-
ture content enables optimal extrusion conditions.
It was fed to the extruder under the following conditions‫׃‬
Compression ratio 4‫׃‬1
Temperature of the first zone 120°C
Temperature of the second zone 130°C
Temperature of the third zone 160°C
Die diameter 3 mm
Screw speed 120 rpm
The extruded product (flips) was analyzed for moisture (AOAC,1984, 8), bulk density
and expansion ratio.
18
APTEFF, 40, 1-220 (2009) UDC:664.64.016.71.8:664.641.15+664.641.2:664.696
DOI: 10.2298/APT0940017D BIBLID: 1450-7188 (2009) 40, 17-24
Extrudate density was calculated. Six products were randomly selected, weighted (m)
and measured (L, D), and the density was calculated as follows:
4× m
Density = 2
π×D ×L
where m is the mass, L – length of cooled extrudate with the diameter D (9).
Expansion ratio was calculated as the ratio of the diameter of extruded simple to that
of the extruded die.
Density and expansion ratio are given as average value of six calculations.
Statistical Analysis. Analysis of variance (ANOVA) and least significant differen-
ces (LSD) were performed by the Statistical Analysis System (Statistical, Tulsa, Oklaho-
ma, USA).
Strength, hardness and plasticity were determined on a universal instrument for tex-
ture analysis INSTRON 4301. Determination of hardness was performed with head with
1.6 mm diameter running at speed of 60 mm/min. Softness of sample by cutting was
examined using knife with 1 mm blade and at head speed of 60 mm/min. Probe for com-
pression with 30 mm diameter was used to examine the force needed to compress extru-
date from 1 cm length to 0.5 cm at the Instron head speed of 48 mm/min. The measu-
rements were performed in the six replicates and average value for strength, hardness and
plasticity were calculated.
The gels for rheological measurements were obtained by heating 6.25% suspension of
grits, in a Brabender viscoamylograph up to 95°C and then held at 95°C for 30 min. Gels
were made from grits blends as well as from extruded products milled to grits on a
laboratory mill. The obtained paste was cooled to room temperature and rheological mea-
surements of the gels were carried out at 20°C. Measurements were carried out right after
the cooling (0h) and 24 hours later (24h). Rheometer RheoStress 600HP (Haake, Ger-
many) with measurement set plate–plate (PP60Ti), with 60 mm diameter and 1 mm dis-
tance, was used. Flow curves were obtained by hysteresis loop method by the following
cycle:
ramp up 0-500 1/s in 4 min
constant γ 500 1/s in 2 min
ramp down 500-0 1/s in 4 min.
Micrographs were taken on a scanning electron microscope (JEOL–JSM–6460LV).
Results of the determination of expansion index and density of extrudates are presen-
ted in Table 2.
The initial moisture level drop from 16% to the final average level of 7.5% for extru-
dates, was within the expected range of 4-8%, reported by Koeppe (1987) as characte-
ristic for an extruded snack products. The difference in moisture content of extrudates is
not statistically important (Table 2).
19
APTEFF, 40, 1-220 (2009) UDC:664.64.016.71.8:664.641.15+664.641.2:664.696
DOI: 10.2298/APT0940017D BIBLID: 1450-7188 (2009) 40, 17-24
Table 2. Extrudate properties
Moisture content
Sample Expansion index Density (g/cm3)
(%) (db)
100% corn grits 7.6 ± 0.09a 4.03 ± 0.21a 0.095 ± 0.02a
80% corn : 20% amaranth grits 7.4 ± 0.03a 2.83 ± 0.11b 0.132 ± 0.01b
50% corn : 50% amaranth grits 7.6 ± 0.05a 1.83± 0.18c 0.346± 0.04c
Results are mean ± SD of six measurements
a
Means with different letters in the same column for each sample are significantly different at the 5% level
Addition of amaranth grits to extrusion blend proportionally reduced extrusion index

and increased density of extrudates, due to its reduced expansion properties compared to
corn. Reduced expansion resulted in a denser product with smaller air cell diameters as
shown in Figure 1.
Fig. 1. Scanning electron micrographs of amaranth:corn 50:50 (A) and corn extrudate (B)
It is difficult to produce expanded products by extrusion cooking of amaranth grain
alone, due to its high fat content (6-8% in whole grain). Fat provides a powerful lubricant
effect in extrusion cooking and reduced product expansion (11). Flips of 100% corn grits
have the highest expansion (4.03) and lowest density (0.095 g/cm3), which provides
demanded crispy structure during eating.
Blend of 50% corn and 50% amaranth grits resulted in a decreased expansion index
(1.83) compared to control 100% corn grits by 2.2 times, but in increased density (0.346
g/cm3 ) by 3.6 times.
Table 3. Texture of extrudates measured by Instron 4301

Sample Compression force Penetration force Cutting force
(N) (N) (N)
100% corn grits 13.2 ± 0.23a 3.2 ± 0.09a 19.2 ± 0.20a
80% corn : 20% amaranth grits 20.8 ± 0.32b 5.1 ± 0.11b 18.8 ± 0.18a
50% corn : 50% amaranth grits 24.5 ± 0.15b 11.6 ± 0.08c 18.4 ± 0.14a
a
Means with different letters in the same column for each samples are significantly different at the 5% level
Table 3 represents the differences in compression, penetration and cutting force (N)
for the extrudate with amaranth grits supplement compared to the control with 100% corn
20
APTEFF, 40, 1-220 (2009) UDC:664.64.016.71.8:664.641.15+664.641.2:664.696
DOI: 10.2298/APT0940017D BIBLID: 1450-7188 (2009) 40, 17-24
grits. Increasing the amount of amaranth grits in flips increased the resistance of extru-
dates, thus the difference in force for compression and penetration compared to corn grits
flips is significant. Cutting force which imitates biting, is not statistically much different
among treatments.
Hardness of flips with 50% amaranth grits represented by penetration force (11.6 N)
is by 3.6 times higher than the penetration force for sample with 100% corn grits (3.2N),
which is a major difference.
Generally, the addition of grits from whole amaranth grain, as a fiber–rich raw ma-
terial in extruded product formulations, resulted in increased density and hardness of the
product (12).
Figure 2 presents flow curves of gels obtained from corn grits and blends of corn and
amaranth grits in the ratios 80:20 and 50:50, determined immediately after the prepara-
tion and after 24h.
All systems exhibited thixotropic flow behavior. Gels prepared from corn grits had
the greatest value of shear stress, i.e. viscosity, and largest thixotropic loop area. The
structure is weak and it was broken–down by low share rates. During storage for 24h the
system with 100% corn grits is additionally structured and three–dimensional gel structu-
re is formed as a result of retrogradation of amylase fractions. Since retrogradation is a
process of binding by weak hydrogen bonds that are easily broken, the flow curve for
100% corn grits measured after 24 hours shows a peak characteristic for breaking such
structures down at very low shear rates. Gels of examined grits blends had lower viscosi-
ty than corn grits and during storage they did not build additional structure. This is a re-
sult of replacing part of the corn grits, whose starch contains amylose and amylopectin,
with amaranth grits whose starch includes small amount of amylose and high amount of
amylopectin fractions (13).
B ###
B ###
B ###
120 140
100 120
100
τ (Pa)
80
τ (Pa)
80
60
60
40
40
20
20
0 0
0 100 200 300 400 500 0 100 200 300 400 500
γ (1/s) γ (1/s)
Fig. 2. Flow curves of the gels from corn grits and corn–amaranth grits in the ratios 80:20
and 50:50 determined in 0 h and 24 h
21
APTEFF, 40, 1-220 (2009) UDC:664.64.016.71.8:664.641.15+664.641.2:664.696
DOI: 10.2298/APT0940017D BIBLID: 1450-7188 (2009) 40, 17-24
In such a way fraction which structurate during time (amylose) is partially replaced
by non gelling fraction. Increased amount of amaranth grits decreased viscosity and sur-
face of thixotropic flow curves.
From the flow curves, in Figure 2, fitting data in Herschel–Bulkley equation (14):
•
τ = τ0 + k γn
the values for yield stress τ0 and coefficients k and n were calculated. Yield stress
decreased with addition of amaranth grits, as part of the corn was replaced. Also, values
of yield stress τ0 increased for gel samples after 24 hours of storage due to the struc-
turation. Values of viscosity behavior index n for all samples are smaller than 1, which is
characteristic for thixotropic behavior.
Table 4. Values of yield stress, consistency index and viscosity behavior index
Yield stress τo Consistency Viscosity

Sample (Pa) index k behavior index n
0h 24h 0h 24h 0h 24h
100% corn grits 3.97 4.59 2.78 10.46 0.59 0.38
80% corn : 20% amaranth grits 1.96 2.75 2.49 6.75 0.57 0.41
50% corn : 50% amaranth grits 0.72 0.92 2.89 3.93 0.52 0.44
100% corn extrudate 0 0.57 0.30 12.59. 0.72 0.21
50% corn : 50% amar. extrudate 0 0 0.05 0.05 0.06 0.06
100% corn extrudate

corn/amaranth 50:50% extrudate
100% corn grits
120
corn/amarnth 50:50%
100
80
τ (Pa)
60
40
20
0
0 100 200 300 400 500
γ (1/s)
Fig. 3. Flow curves of the gels made from the corn grits and extruded snack, and from
50:50% corn:amaranth grits and extruded snack
Fig 3 presents flow curves of corn grits and grits blends of corn and amaranth in a
ratio 50:50, as well as flow curves of obtained extrudated products. Flow curves of gels
made from extrudates had lower viscosity than the gels made from corresponding grits.
22
APTEFF, 40, 1-220 (2009) UDC:664.64.016.71.8:664.641.15+664.641.2:664.696
DOI: 10.2298/APT0940017D BIBLID: 1450-7188 (2009) 40, 17-24
Starch granules in extrudates after thermal treatment were partly damaged during
extrusion process and they built weak gels. The obtained pastes had a lower viscosity,
and they can be classified to the category of pregelatined starch granules.
CONCLUSION
Increasing amount of amaranth grits in the extrusion blend causes increase of density
and hardness of the extrudated products and decrease in expansion index. When part of
the corn grits is replaced with amaranth grits viscosity of gels decreases compared to pure
corn grits. Also, extrusion process partially damages starch granules, thus obtained gels
of extrudated products have lower viscosity than the initial grits.
REFERENCES
1. Saunders, R.M., and R. Becker: Amaranthus: A Potential Food and Feed Recourse in:
Advances in Cereal Science (1984) 357-396.
2. Bodroža–Solarov, M.: Effects of Genotype and Sowing Date on Yield and Yield
Components of the Genus Amaranthus L, Ph.D.Thesis, Faculty of Agriculture, Uni-
versity of Novi Sad (2001) 1-113.
3. Kovacs, E.T., Maraz–Szabo, L., J. Varga: Influence of Variety and Type of Emulsi-
fier for Functional Food Quality and Amaranth Basis, Proceedings of XIV Internatio-
nal Congress „Cereal Bread 2000” (2001) 223-226.
4. Breene, W.M.: Food Uses of Grain Amaranth, Cereal Food World 36, 5 (1991) 426-
429.
5. Sanchez–Marroquan, A., Del Valle, F.R., Escobedo, M., Avita, R., Maya, S., and M.
Vega: Evaluation of whole amaranth (Amaranthus cruentus) flour, its air classified
fractions, and blends of these with wheat and outs as possible components for infant
formulas, J. Food Sci. 51, 5 (1986) 1231-1234.
6. Radosavljević, M., Jane J., L.A. Johanson: Isolation of Amaranth Starch by Diluted
Alkaline–Protease Treatment, Cereal Chem. 75, 2 (1998) 212-216.
7. Gonzales, R. J., Torres R. L., De Greef, D. M., Tosi E., E. Re: Brazilian Journal of
Chemical Engineering 19, 4 (2002) 391-395.
8. AOAC : Solids (total) and moisture in flour. Air oven method. In Official Methods of
Analytic Chemists, 14th edn. (1984) pp. 249.
9. Ding Q., P. Ainsworth, A. Plunkett, G. Tucker, H. Marson: The Effects of Conditions
on the Functional and Physical Properties of Wheat–Based Expanded Snacks, Journal
of Food Engineering 73 (2006) 142-148.
10. Koeppe, S. J., Harris, M.A., Hanna, J. H., Rupnow, C. E. Walker, and S.L. Cupett:
Physical Properties and Some Nutritional Characteristics of an Extrusion Products
with Defatted Amaranth Seeds and Defatted Maize Gluten Meal (80:20 Ratio), Cereal
Chemistry 64, 4 (1987) 332-336.
11. Ilo, S., Liu Y., E. Berghofer: Extrusion Cooking of Rice Flour and Amaranth Blends,
Lebensm.–Wiss.u.–Technol. 32 (1999) 79-88.
12. Onwulata, C. I., Konstance R. P., Strange E. D., Smith P. W., V.H. Holsinger: High
Fiber Snack Extruded from Triticale and Wheat Formulations, Cereal Food World 45,
10 (2000) 470-473.
23
APTEFF, 40, 1-220 (2009) UDC:664.64.016.71.8:664.641.15+664.641.2:664.696
DOI: 10.2298/APT0940017D BIBLID: 1450-7188 (2009) 40, 17-24
13. Bello–Perez, L. A., Colona, P., Roger, P., O. Peredes–Lopez: Macromolecular Fea-
tures of Amaranth Starch, Cereal Chem. 75, 4 (1998) 395–402.
14. Mezger T. : The Rheology Handbook, Vinsentz Verlag, Hannover (2002) pp.46.
СВОЈСТВА ЕКСТРУДАТА СА ДОДАТКОМ КРУПИЦЕ ОД СЕМЕНА

АМАРАНТУСА
Љубица П. Докић, Марија И. Бодрожа-Соларов, Мирослав С. Хаднађев

и Ивана Р. Николић
Поступком екструдирања на лабораторијском Брабендеровом 20 ДН екструдеру

добијени су екструдати мешавине кукурузног гриза са додатком 20% и 50% кру-
пице од семена амарантуса. Добијени су екструдати различите текстуре. Пове-
ћањем удела гриза од амарантуса у смеши за екструдирање долази до повећања
густине, смањења степена експанзије, повећања тврдоће. У процесу екструдирања
долази до делимичног оштећења скробне грануле те су испитане и реолошке карак-
теристике. Сви системи су по типу протицања били тиксотропни. Узорци код којих
је део кукурузне крупице замењен амарантусовом имали су ниже вискозитете и по-
казивали низак степен просторног структурирања током стајања.
Received 15 June 2009

24
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
REDUCTION OF WHEAT MIDDLINGS USING A CONVENTIONAL AND

EIGHT-ROLLER MILLING SYSTEMS
Aleksandar Z. Fišteš and Đuro M. Vukmirović1
Possibilities for the rationalization of the wheat flour milling process using the eight-
roller mill on the 1M and 2M passages of the reduction system have been investigated. At
the same roll gaps and under the same sieving conditions, the lower flour yield has been
obtained using an eight-roller mill compared to the conventional milling system (5-8 %)
followed by a higher energy requirements for grinding. By decreasing the roll gap setting
and increasing the upper size limit of flour in the process with the eight-roller mill it is
possible to increase flour yield and therefore decrease milling energy consumption per
unit mass of flour produced without deterioration of flour quality as determined by ash
content. With appropriate adjustments of the processing parameters in the eight-roller
milling system it is possible to achieve similar milling results to those in the conventional
system, while the overall investment, energy and maintenance costs are significantly
lower.
KEY WORDS: Wheat flour milling, process rationalization, eight-roller mill
INTRODUCTION
The objective of the wheat flour milling process is to separate the branny cover and
germ of the wheat kernel from the endosperm and achieve as high as possible flour
extraction with the lowest contamination of bran and germ that increase the ash content.
Breaking the wheat kernel is affected by corrugated cast steel rolls that gradually separate
the endosperm, bran and germ. Reduction of relatively pure endosperm to flour is achie-
ved by using smooth rolls. Segregation between the kernel parts occurs in plansifters,
where sieves separate particles of different size, and in purifires, where sieves and air-
flow separate particles of different size, specific gravity and shape (1).
Ever since the grinding of grain has been known to the mankind, possibilities and
solutions have been sought out in order to simplify the grinding process and make it more
efficient. New concepts and ideas only have chance of being successful if the yield as
well as the quality of the finished products are not affected and requirements such as
reduction of investment, operating and maintenance cost are met (2). The traditional
Aleksandar Z. Fišteš, M.Sc., fistes@uns.ac.rs, Faculty of Technology, Bulevar Cara Lazara 1, 21000 Novi Sad,
Serbia; Đuro M. Vukmirović, B.Sc., djuro.vukmirovic@gmail.com, Institute for Food Technology, Bulevar
Cara Lazara 1, 21000 Novi Sad, Serbia
25
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
wisdom in flour milling is that after every grinding step the ground material should be
sieved and the undersize material removed before regrinding (3). Over the years the main
equipment used in the grinding system has been redesigned to such an extent that it has
been possible to multiply the throughputs of these machines but flour process technology
has not changed fundamentally since the introduction of the roller mill, the purifier and
the plansifter (4). This is the reason why the double grinding of intermediate streams
before sieving has been one of the most notable process developments in flour milling
(3). Eight-roller mill (a total of 8 rolls in one housing) provides two grinding passages
without any intermediate sifting. Compared to a conventional process with the four-roller
mill, the introduction of the eight-roller mill into the milling flowsheet offers the
following advantages (2, 4, 5-8):
• fewer pneumatic suction lifts (conveying the stock from the roll to the sifter) –
resulting in lower material and installation costs
• lower pneumatic system air requirements – resulting in lower energy costs and
lower filter surface requirements for cleaning the pneumatic conveying air
• reduction of sifter surface
• smaller number of roll stands
• less spouting and auxiliary components
• lower space requirements for equipment installation – resulting in lower building
costs for the new milling plants or increase in grinding capacity within existing
limited building space
• more compact building design makes it easier to keep clean with less area to
fumigate
• with less equipment there is less cleaning and maintenance.
On the other hand, twin stage grinding ignores the milling principle that, after grin-
ding, coarse material is separated from the fines and therefore the conditions for con-
trolled milling are less favorable. The decision as to how many double grinding passages
can be applied in the flowsheet depends directly on the finished products to be made. It is
not always possible to equip installations exclusively with eight-roller mills (4, 8). This is
the reason why it is very important to define the passages and optimal roll parameters that
allow the introduction of the eight-roller mill into the milling process without deterio-
ration of the yield and quality of finished products.
Even though the eight-roller mill has found its place in the modern flour milling
process relatively little research has been performed on various factors affecting the
milling results using this technique mainly focusing on the front passages of the break
system (8-11) rather than the passages of the reduction system. It is evident that when the
first and second breaks are combined into a twin passage, the particle size distribution of
the stock will not be the same as with conventional single break system (2, 8, 11). There
are several disadvantages of using this break system. First, it grinds fine material, coarse
middlings, that should not pass to the next break roll, whose function is to separate
endosperm from bran. Second, it produces more break flour and fine middlings and less
coarse middlings and sizings that can be purified to produce clean middlings and low-ash
flour. Third, the capacity of the lower roll is limited because the ground material is lower
in density, which increases the volume to the roll (5). Tegeler (8) and Wanzenried (7)
26
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
stated that the granulation from flour produced in the eight-roller mill is finer while the
flour color is slightly whiter compared to the ones in the conventional process. The
research of Zwingleberg (9) and Zwingleberg and Artz (10) showed that appropriate
adjustment of the roll parameters is needed regardless of whether the eight-roller mill is
used on breakage or reduction passages.
Under industrial conditions, only the roll gap settings and feed rate (to a limited
degree) can be adjusted during the milling process. The aim of this work was to examine
and compare the effect that roll gap changes have on the milling results (degree of
particle size reduction, flour release, flour ash and milling energy requirements) obtained
using the conventional process and process with an eight-roller mill employed on the 1st
and 2nd midds (1M and 2M passages) of the reduction system.
EXPERIMENTAL
The samples were obtained from an industrial mill (120 t/day) intercepting the stream
(middlings) that would have been sent to the 1M. In this particular mill (having five
break, four sizing and six reduction passages) the eight-roller mills are not employed. The
samples (50 kg) were separated using the automatic sampler divider (Gompper-Maschi-
nen KG) into 0.5 kg batches and milled on a Variostuhl (model C Ex 2) – laboratory roll
stand (Miag). Smooth rolls 0.1 m in length and 0.25 m in diameter were used. Table 1
summarizes the experimental range of variables tested.
The experiments were designed to compare:
a.) conventional milling system – the entire stock following 1M was sieved for 3
min on a Bühler laboratory sifter (model MLU-300) and the part of the stock
held on the sieve fitted with 150 μm bolting cloth was milled on 2M
b.) eight-roller milling system – the entire stock following 1M was milled on the
2M without intermediate sifting
Table 1. Summary of experimental range of variables tested
Roll gap
Milling Roll Feed rate Fast roll
combinations Differential
system surface [kg/cm/min] speed [m/s]
[mm]
Conventional 1M-0.10; 2M-0.08 1M-0.20; 2M-0.15*

Smooth 1M-0.10; 2M-0.05 1.25 5
1M-0.08; 2M-0.05 1M-0.20; 2M-0.20
Eight-roller
*The slower feed rate on 2M corresponds to the amount of flour removed by intermediate sifting of the stock
leaving 1M
Sieve analysis of the entire stock following 2M was performed on the above Bühler
laboratory sifter for 3 min. For the conventional milling system the sieve openings were
250, 200 and 150 µm, along with the bottom collecting pan. For the eight-roller milling
system three different stacks of sieves were used. The first stack was the same as that
mentioned above. In the second, the sieve with the 150 µm bolting cloth was replaced
27
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
with a sieve having 180 µm bolting cloth, and for the third stack the sieve openings were
250 and 200 µm.
Two samples were milled and sieved under the same conditions. The weight distri-
bution among the streams was highly reproducible. Based on the 3σ rule, the 99.7% esti-
mated confidence interval for the data (weight percentages) presented in the paper is
±0.37%.
Flour yield F (%) in the eight-roller and conventional milling systems was calculated
from Equations (Eqs.) [1] and [2], respectively.
m2M
F (%) = * 100(%) [1]
M
m1M + m2 M
F (%) = *100(%) [2]
M
The symbols m and M stand for the weights of the flour and native feed, respectively.
The subscripts indicate flour following 1M or 2M. The ash content of the total quantity of
flour leaving the 2M was determined according to ICC standard method No.104/1 (12).
The milling energy consumption during all grinding runs was determined from the
wattmeter fitted as an integral part of the Variostuhl laboratory roll stand. Two different
power readings were recorded corresponding to operation with (P, kW) and without (P*,
kW) the material flow. The milling energy consumption, E kJ/kg, in the conventional and
eight-roller milling systems was calculated by Eqs. [3] and [4], respectively.
P1M − P1*M P − P2*M
E= t1M + 2 M t2M [3]
m1M m2M
( P1M − P1*M ) + ( P2 M − P2*M )
E= (t1M + t 2 M ) [4]
m2 M
Here m (kg) is the mass of flour obtained and t (s) is the time of the grinding run
determined by the chronometer. The significance of the differences between milling
results obtained using investigated milling systems have been tested by the paired Stu-
dent’s t-test.
Changes in the particle size distribution of the stocks leaving 2M, brought about by
the decrease of the roll gap, followed the same trends in both milling systems. Con-
sidering that the 2M feeds were different for the two milling systems, yields of the size
fractions of the milling output are not to be compared because the cumulative size distri-
butions (Fig. 1a and 1b) were given relative to the mass of the material milled on the 2M
and only serve to show the general trends. By decreasing the roll gap, the quantity of ma-
terial >200 μm (size fractions >250 μm and 250/200 μm) tends to decrease while the
flour yield (<150 μm) increased.
28
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
Fig. 1. Cumulative size distributions of the stocks following 2M milled through different
roll gaps in the a) conventional milling system and b) eight-roller mill system
The factors affecting particle size reduction can be classified into those arising from
the physicochemical properties of the material and those related to the design and
operation of the milling equipment (13). In a roller mill particles are subjected to shear
and compressive forces. The nature of deformation depends not only on the applied
stresses but also on the particle components upon which the stresses act. Compressive
stresses are more effective in causing the disintegration of the brittle endosperm material
while bran (tough and fibrous) is more prone to the ductile fracture imparted by shear
forces. The roll parameters such as roll gap, uniformity and the feed rate of stocks to
rolls, roll velocities, differential and the type and condition of roll surface influence the
magnitude of the stress and the relative contributions of compressive and shearing forces
(14). Middlings are composed primarily of endosperm (they also contain adhering bran
and germ) exhibiting viscoelasticity when fracturing. With roll differential closer to 1 the
compressive forces dominate in the grinding zone. As the roll differential increases,
greater shear stresses are imposed. Under the present grinding conditions, using the
smooth rolls and relatively small differential – 1.25 (usual for this stage of grinding pro-
cess), as the roll gap decreases greater compressive forces are imposed, thereby increa-
sing the number of endosperm fractures creating more flour. Simultaneously, the tougher
branny particles are flattened and remain in the coarsest size fractions of the milling
output (>200 μm). In addition to that, decreasing the roll gap increases the grinding zone
size (14) so the grinding action, under predominant compressive forces, is prolonged and
also contributes to the greater flour release as a result of increased number of endosperm
fractures. Ash is concentrated in the bran, with over half the total in the pericarp, testa
and aleurone and the ash content increases from the inner (endosperm) to the outer (bran)
29
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
part of the wheat kernel (15). The increase of the ash content of the coarsest fraction of
the stock following 2M, while roll gap had no influence on flour ash content (Table 2),
proves that bran particles remain intact. Scanlon and Dexter (16) and Scanlon, Dexter and
Biliaderis (17), in their studies of the effect of smooth roll grinding conditions on
reduction of wheat farina, also reported similar findings.
Table 2. Ash content in flour (<150 μm ) and the coarsest fraction (>250 μm) of the
stock following 2M in the conventional and eight-roller milling systems
Ash content (%)dm

Roll gap [mm] Conventional system Eight-roller system
>250 μm <150 μm >250 μm <150 μm
1M-0.10; 2M-0.08 1.16 0.38 1.12 0.37
1M-0.10; 2M-0.05 1.22 0.35 1.15 0.37
1M-0.08; 2M-0.05 1.27 0.35 1.17 0.38
At the same roll gap setting and under the same sieving conditions, the flour release
was lower in the process involving the eight-roller mill compared to the conventional
milling system (Fig. 2) and the difference is statistically significant (p<0.05). In the con-
ventional system, flour particles are removed from the stock before regrinding on 2M by
intermediate sifting, while in the absence of intermediate sifting remain in the material
feeding the lower pair of rolls of the eight-roller mill (stock following 1M).
Fig. 2. Flour release following 2M in the conventional and eight-roller milling system
30
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
These particles take on some of the stresses in the grinding zone which otherwise
would be used to reduce the remaining middlings to flour, thereby explaining the lower
flour release in the process with the eight-roller mill as well as the finer flour granulation
as a result of further grinding of flour particles. Only one sifting operation in the process
with eight-roller mill, compared to two in the conventional system, also contributed to
lower flour yield. The amount of the flour in the stock following 2M is considerably
higher in the eight-roller mill process, so the other probable cause could be inefficient
sifting of the flour. Sifting efficiency depends on a number of different factors such as
disposable sifter area, cloth tension, number of gyrations per minute, feed rate to the
sifter, etc. (18,19). However, under industrial conditions, replacement of the sieves in the
plansifter (changing the sieve aperture) is probably the easiest way to change the sieving
conditions and therefore influence the sifting efficiency. At the same, roll gap setting in
both milling systems, by replacing the 150 μm bolting cloth with sieves having 180 and
200 μm bolting cloths in the process with eight-roller mill, similar (p>0.05) or signifi-
cantly higher (p<0.01) flour yield has been achieved compared to the one in the con-
ventional system (Fig. 2). It needs to be pointed out that increasing the sieve size con-
tributes to more efficient sifting but at the same time increases the upper size limit of
flour, both causing the increase of the flour release.
By decreasing the roll gap setting in the process with the eight-roller mill compared to
the gap in the conventional system, without changing the sieving conditions, it is possible
to increase flour yield and it comes as a result of greater compressive forces imposed on
the particles of the stock. Neither the roll gap adjustments nor the sieving conditions
changes resulted in deterioration of flour quality since the flour ash remained constant
(p>0.05) (Table 3). It shows that previously mentioned adjustments in the process were
not followed by increased grinding of the bran which would otherwise pass into flour and
increase flour ash.
Table 3. Ash content in the total amount of flour following 2M in the conventional
and eight-roller milling systems
Ash content (%)dm

Conventional
Roll gap [mm] Eight-roller system
system
<150 [μm] <150 [μm] <180 [μm] <200 [μm]
1M-0.10; 2M-0.08 0.39 0.37 0.38 0.36
1M-0.10; 2M-0.05 0.36 0.37 0.36 0.35
1M-0.08; 2M-0.05 0.36 0.38 0.37 0.37
Eqs. [3] and [4] define milling energy consumption relative to the mass of flour ob-
tained. Under the same roll and sifting conditions, the energy required for grinding tends
to be higher in the eight-roller mill process compared to that in the conventional milling
system (Fig. 3) (p<0.01). It is mainly due to the lower flour yield in the eight-roller mill
process. The heavier load in the 2M grinding zone in the process with eight-roller mill
increases the power requirements in the operation with the material flow (P) and also
31
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
contributes to higher energy consumption. The higher feed rate to the lower pair of rolls
of the eight-roller mill, compared to feed rate on appropriate milling passage in conven-
tional milling system, is unavoidable because there is no intermediate sifting and there-
fore no removal of the undersized material before regrinding. By increasing the flour
release in the process with the eight-roller mill, by appropriate adjustment of the sieving
conditions and/or roll gap (following the data presented on Fig. 2), it is possible to reduce
the energy required for grinding (Fig. 3).
Fig. 3. Energy consumption in conventional and eight-roller milling systems
CONCLUSION
Compared to conventional milling system, the introduction of the eight-roller mill
into the milling flow sheet offers numerous advantages which significantly contribute to
the reduction of the production costs. With the appropriate adjustments of the sieving
conditions or/and roll gap setting in the process with the eight-roller mill it is possible to
achieve similar milling results to those in conventional milling system. This justifies the
use of the eight-roller mill on the 1st and 2nd midds (1M and 2M passages) of the wheat
flour milling process.
REFERENCES
1. Posner, E.S.: Wheat. In Handbook of Cereal Science and Technology. Eds. Kulp, K.
and J.G. Ponte, CRC Press, Taylor and Francis Group, New York (2000) pp.1-30.
2. Baltensperger, W.: New Development in the Mill Flow Charts Grinding Process
Using Eight-Roller Mills. Association of Operative Millers-Bulletin, December
(1993) 6327-6332.
32
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
3. Owens, W.G.: Wheat, corn and coarse grains milling. In Cereals Processing Tech-
nology. Ed. W.G. Owens, Woodhead Publishing Ltd., Cambridge (2001) pp. 27-52.
4. Baltensperger, W.: State-of-the-Art Grain Milling Technology. Association of Ope-
rative Millers-Bulletin, January (2001) 7583-7591.
5. Posner, E.S., and A.N. Hibbs: Wheat Flour Milling, AACC, St.Paul, Minnesota
(2005) p.212-213.
6. Eugster, W.: Advances in Process Technology. Association of Operative Millers –
Bulletin, October (2001) 7706-7707.
7. Wanzenried, H.: Benefits and results with 8-roller mill, model MDDL. Association
of Operative Millers-Bulletin, December (1991) 5977-5981.
8. Tegeler, V.C.: Eight-High Roller Mills vs. Four-High Roller Mills The Pros and
Cons. Association of Operative Millers-Bulletin, February (1999) 7229-7230.
9. Zwingelberg, H.: Verschiedene Walzenstuhlbeschüttungen und deren Auswirkungen
auf Produktanfall und Mineralstoffgehalt Teil 2: Versuche an nicht geputztem Grieß
als Einzel- und Doppelvermahlung. Die Mühle + Mischfuttertechnik, Heft 20 (1998)
649-654.
10. Zwingelberg, H., and B. Arzt: Verschiedene Walzenstuhlbeschüttungen und deren
Auswirkungen auf Produktanfall und Mineralstoffgehalt Teil 1: Versuche beim I.
Schrot und als Doppelvermahlung beim I. Und II. Schrot. Die Mühle + Mischfut-
tertechnik, Heft 18 (1998) 593-595.
11. Handreck, B., L. Pötschke and Ch. Senge: Intensives Aufschroten von Weizen im
Achtwalzenstuhl. Die Mühle + Mischfuttertechnik, Heft 26 (1999) 818-826.
12. ICC Standard No.104/1. Determination of ash in cereals and cereal products.
13. Campbell, G.M., P.J. Bunn, C. Webb and S.C.W. Hook: On predicting roller milling
performance Part II:The breakage function. Powder Technology 115 (2001) 243-255.
14. Haque, E.: Application of size reduction theory to roller mill design and operation.
Cereal Foods World 36 (1991) 368-374.
15. Kent, N.L.: Technology of cereals, Pergamon Press, Oxford (1975).
16. Scanlon, M.G. and J.E. Dexter: Effect of smooth roll grinding conditions on reduc-
tion of hard red spring wheat farina. Cereal Chemistry 63 (1986) 431-435.
17. Scanlon, M.G., J.E. Dexter and C.G. Biliaderis: Particle-size related physical pro-
perties of flour produced by smooth roll reduction of hard red spring wheat Farina.
Cereal Chemistry 65 (1988) 486-492.
18. Curran, S., W. Eustace and J. Gwirtz: The effect of cloth tension on sifting
performance. Association of Operative Millers-Bulletin, May (1994) 6379-6381.
19. Wingfield, J. and A. Ferrer: Multiple sieve sifter performance using various combi-
nations of feed rates, circles and speeds. Journal of Food Process Engineering 7
(1984) 91-110.
33
APTEFF, 40, 1-220 (2009) UDC:664.71-11:664.762:66.012
DOI: 10.2298/APT0940025F BIBLID: 1450-7188 (2009) 40, 25-34
ЕФЕКТИ МЛЕВЕЊА ПШЕНИЧНОГ ГРИЗА У КЛАСИЧНОМ И

ПОСТУПКУ СА ОСМОВАЉНОМ СТОЛИЦОМ
Александар З. Фиштеш и Ђуро М. Вукмировић
Испитивана је могућност рационализације технолошког поступка млевења пше-

нице укључивањем осмоваљне столице на пролазиштима млевења гриза 1М и 2М.
При истом вођењу ваљака и при употреби истог слога сита за просејавање млива, у
поступку са осмоваљном столицом остварује се мањи принос брашна него у кла-
сичном поступку (5-8%), док је специфични утрошак енергије по јединици масе
брашна већи. Нижим вођењем ваљака у поступку са осмоваљном столицом разлика
у оствареном приносу брашна у поменутим поступцима се смањује без промене
садржаја пепела у брашну. Корекцијом слога сита у поступку са осмоваљном сто-
лицом (повећањем величине отвора сејног ткива за одсејавање брашна) значајно се
повећава принос брашна у поменутом поступку, што доприноси смањењу специ-
фичног утрошка енергије за уситњавање. Такође, при томе није констатована про-
мена садржаја пепела у брашну. Одговарајућим прилагођавањем процесних пара-
метара могу се у поступку са осмоваљном столицом остварити ефекти уситњавања
блиски ефектима у класичном поступку, а истовремено се остварују значајне ин-
вестиционе и енергетске уштеде, што доприноси рационализацији технолошког
поступка млевења пшенице.

34
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
CHANGES IN NUTRITIVE AND TEXTURAL QUALITY OF APPLE

OSMODEHYDRATED IN SUGAR BEET MOLASSES AND SACCHAROSE
SOLUTIONS
Gordana B. Koprivica, Nevena M. Mišljenović, Ljubinko B. Lević and Vjera S. Pribiš
The paper describes texture and mineral content of apple, osmotically dehydrated in
sugar beet molasses as compared to apples treated in saccharose solution. Osmotic de-
hydration was conducted at constant temperature of 55°C and atmospheric pressure.
During the experiment, the concentration of sugar beet molasses was varied 40 to 80%,
the concentration of saccharose solutions was varied in the range of 30 to 70%, and the
most important kinetic parametars of the osmotic dehydration, after 1, 3 and 5 hours of
immersion were observed. During osmotic dehydration, in the samples which were trea-
ted in sugar beet molasses, the content of minerals was increased to a great extent that
enhanced their nutritive value. Textural quality parameter was evaluated from the maxi-
mum cut force, tested at Instron testing machine. It was found that the samples dehydra-
ted in saccharose solutions had a softer and more gentle texture – the maximum force
load decreased threefold as compared to the other samples.
KEY WORDS: Osmotical dehydration, apple, saccharose, sugar beet molasses
INTRODUCTION
Osmotic dehydration is used as pretreatment in the process of extension of the sus-

tainability of fruit and vegetable drying. The aim of osmotic dehydration is a partial re-
moval of water from the material with a simultaneous reduction of solid gain, in order to
obtain a better quality final product. In other words, osmotic dehydration is a process for
concentrating fruit and vegetables (1).
In the first phase of the process, fruits and vegetables are dipped in a hypertonic
aqueous solution whereupon part of moisture flows from fruits and vegetables as a con-
sequence of the difference in osmotic pressure of water in the plant tissue and hypertonic
aqueous solution, through the cell walls and surface tissue which act as a semi-permeable
membrane. However, as these structures are only partially permeable, at the same time
there is a diffusion of solute from osmotic solution into fruits and vegetables (2).
Gordana B. Koprivica, B. Sc., Res. Assist., gordanak@uns.ac.rs, Nevena M. Mišljenović, B. Sc, Res. Assist.,
nevenam@uns.ac.rs; Dr. Ljubinko B. Lević, Prof., megamum@uns.ac.rs, Dr. Vjera S. Pribiš, Prof., University
of Novi Sad, Faculty of Technology, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia
35
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
In addition to these two dominant processes a diffusion of cell juices from the plant
tissue into osmotic solution occurs, but to a much smaller extent, which is considered to
be minor but also affects the nutritive value of osmo-dehydrated fruits and vegetables (3,
4).
Partially, dehydrated fruits and vegetables can be used directly in human nutrition or
as a material for further drying to the appropriate moisture content (5).
The rate of diffusion of water from plant tissue depends on several factors such as:
temperature and concentration of osmotic solution, the size and geometry of plant tissue,
weight ratio of solution and plant tissue and the degree of solution mixing (6, 7).
Dehydration rate is the highest at the beginning of the process of osmotic dehydration
because of the large differences between osmotic pressure in the solution and the plant
tissue, as well as a small resistance to mass transfer (8).
The increase of the mass transfer rate during osmotic dehydration can be achieved in
several ways - by applying vacuum, ultrasound, high pressure, centrifugal force, etc. (1).
The choice of optimal hypertonic aqueous solution appears to be the key problem in
osmotic dehydration. So far, pure saccharose or its combinations with other sugars and
sodium-chloride has been proposed as the best solution for hypertonic aqueous solutions
(9, 10).
Besides saccharose, sugar beet molasses emerges as a suitable raw-material for the
preparation of hypertonic solutions. Molasses is the last syrup, from is not possible to
get the crystal sugar by usual procedures of crystallization. Molasses has a high content
of solids (around 80%) and contains, in average, 51% saccharose, 1% rafinose, 0.25%
glucose and fructose, 5% proteins, 6% betaine, 1.5% nucleosides, purine and pyramidine
bases, organic acids and bases (11).
In Serbia, sugar beet molasses has not yet been used as an ingredient in food industry.
Hence, the extensive research activities have been going on with the aim of introducing
molasses as a valuable ingredient in bakery, confectionery and meat processing industry
(1, 5). Sugar beet molasses is a cheap material available in large quantities and could be
used as a replacement for saccharose.
The aim of this paper is to examine the influence of hypertonic solution (sugar beet
molasses and sucrose solutions), used in the process, on the mechanical properties and
mineral composition and nutritional quality of treated apples.
EXPERIMENTAL
Apples for the experiment, were purchased on the local market. Prior to the treatment,
the apples were stored at 4ºC and then thoroughly washed and cut into cylindrical shapes,
20 mm in height and diameter with sharp apple corer.
Osmotic solutions were saccharose aqueous solutions (solid content: 30%, 50% and
70%) that were prepared by mixing commercial sugar with heated distilled water (30ºC)
to complete dissolution, and sugar beet molasses (from sugar factory in Bač, Serbia) with
40, 60 and 80% solid content. Solutions were made by mixing pure molasses with distil-
led water.
36
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
Osmotic dehydration was carried out at 55ºC under atmospheric pressure. After mea-
suring the initial mass, samples of apple were dipped into different concentrations of the
hypertonic saccharose solutions and sugar beet molasses.
Saccharose solutions had a concentration of 30% (in the study indicated as S1), 50%
(S2) and 70% (S3), while the concentrations of sugar beet molasses solutions were 40%
(in the further study indicated as M1), 60% (M2) and 80% solid content (M3).
Temperature monitoring
Osmotic solution
T
Samples
Thermostatic bath Pumps
Fig. 1. Setup for osmotic dehydration
The material to hypertonic solution ratio was 1:4. The immersion lasted for 1, 3 and 5
hours. After osmotic dehydration, the samples were washed with water and gently blotted
to remove excessive water from the surface. The next step was to measure the mass of the
samples. After measuring the mass, dry matter content and content of some mineral
matter (K, Ca, Na and Mg) were determined.
The samples were kept in an oven (Instrumentaria Sutjeska, Serbia) at 105°C for 24h,
until constant weight was attained.
The solid content of osmotic solutions was determined refractometrically (12).
The content of mineral matter were determined by atomic absorption spectrophoto-
meter AAS 30 – Carl Zeis (12-Cacak) and the texture of apple was determined by Instron
M4301 (13).
All analytical measurements were carried out in accordance to AOAC methods (14).
During the process the values of water content were followed as well as the changes
in weight and in the content of dry matter. These values allowed calculation of the fol-
lowing parameters: water loss (WL), weight reduction (WR), dry matter growth (SG)
(15).
⎡g⎤ w − w
WR ⎢ ⎥ = o [1]
⎣g⎦ wo
37
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
⎡ g ⎤ u − uo
SG ⎢ ⎥ = [2]
⎣g⎦ wo
⎡g⎤
WL ⎢ ⎥ = WR + SG [3]
⎣g⎦
where: Wo - initial weight of the sample (g), W - weight of the sample after osmotic
dehydration (g), uo - weight of dry matter in the fresh sample (g), u - weight of dry matter
in the sample after osmotic dehydration (g).
On the basis of these parameters, the rate of weight reduction (RWR ), the rate of
growth of dry matter (RSG) and the rate of water loss (RWL) during osmotic dehydration
are calculated.
⎡ g ⎤ WR
RWR ⎢ ⎥= t [4]
⎣ g ⋅ min ⎦
⎡ g ⎤ SG
RSG ⎢ ⎥= t [5]
⎣ g ⋅ min ⎦
⎡ g ⎤ WL
RWL ⎢ ⎥= t [6]
⎣ g ⋅ min ⎦
where t is the duration time of osmotic dehydration.
Kinetics of osmotic dehydration
Table 1. Kinetic parametars of osmotic dehydration of apple in sugar beet molasses
Concentration of Time, WR x 102, SG x 102, WL x 102,

molasses, % d.m. h g/g initial sample g/g initial sample g/g initial sample
weight weight weight
1 15.477 2.216 17.693
40 % 3 22.483 2.951 25.433
5 32.155 4.410 36.566
1 17.760 1.539 19.299
60 % 3 32.690 2.471 35.161
5 42.298 2.905 45.203
1 24.851 1.182 26.034
80 % 3 37.506 2.036 39.540
5 52.634 2.624 55.258
38
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
Tables 1 and 2 are show the changes of kinetic parameters during osmotic dehydra-
tion of apples depending on the concentration of sugar beet molasses and sucrose so-
lutions as well as duration of osmotic dehydration process.
Table 2. Kinetic parametars of osmotic dehydration of apple in saccharose solutions
Concentration of Time, WR x 102, SG x 102, WL x 102,

saccharose solution, h g/g initial sample g/g initial sample g/g initial sample
% d.m. weight weight weight
1 12.805 0.438 13.244
30 % 3 14.001 1.201 15.202
5 19.531 1.300 20.832
1 22.178 1.973 24.151
50 % 3 29.820 2.827 32.647
5 36.263 4.181 40.445
1 25.349 2.216 27.565
70 % 3 38.446 3.481 41.927
5 49.203 4.716 53.919
The process of osmotic dehydration decreased mass of the samples, weight reduction
(WR) increasing with time and increase in concentration of sugar beet molasses.
The apple dehydrated in the 80% sugar beat molasses shows the greatest mass loss of
52.634 x 10-2g/g of the initial sample mass.
The SG value indicates the degree of penetration of solids from the osmotic solution
to the apple sample. It shows a tendency to increase with increasing the immersion time,
while increasing the concentration of osmotic solution reduces the increase of dry matter.
Water loss is defined as the sum of weight reduction and solid gain. In all samples
WL was increased with increasing dehydration time. Higher concentration of sugar beet
molasses caused greater loss of water from the samples.
Mass transfer rate during osmotic dehydration
Tables 3 and 4 show the rate of mass transfer during osmotic dehydration as a func-
tion of time and concentration of sugar beet molasses and sucrose solution. The results
show that the osmotic dehydration was most intensive at the beginning of the process.
The rate of weight reduction is the highest in the first hour and 3 hours, after the process
has a tendency of stabilization. The rate of weight reduction is greater when 80% mo-
lasses and 70% sucrose solution were used as the osmotic solution in that case the value
of mass reduction being almost equal.
Loss of water is fast at the beginning of the process of osmotic dehydration. The cau-
se of this is a greater driving force in the beginning of the process of osmotic dehydra-
tion, i.e. greater difference in the osmotic pressures between surrounding hypertonic
solutions and the plant tissue.
39
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
Table 3. Rate of mass transfer in osmotic dehydration of apple in sugar beet molasses
Concentration of Time, Rate of WR x 104, Rate of SG x 105, Rate of WL x 104,

molasses, % d.m. h g/(g i.s.w.· min) g/(g i.s.w. · min) g/(g i.s.w. · min)
1 25.8 36.9 29.49
40 % 3 12.49 16.4 14.13
5 10.72 14.7 12.19
1 29.6 25.6 32.16
60% 3 18.16 13.7 19.53
5 14.1 9.68 15.06
1 41.42 19.7 43.39
80% 3 20.84 11.3 21.97
5 17.54 8,75 18.42
Table 4. Rate of mass transfer in osmotic dehydration of apple in saccharose solutions
Concentration of
Time, Rate of WR x 104, Rate of SG x 105, Rate of WL x 104,
saccharose solution,
h g/(g i.s.w.· min) g/(g i.s.w. · min) g/(g i.s.w. · min)
% d.m.
1 21.34 7.31 22.07
30 % 3 7.78 6,67 8.45
5 6.51 4,33 6.94
1 36.96 32.88 40.25
50% 3 16.57 15.71 18.14
5 12.09 13.93 13.48
1 42.25 36.93 45.94
70% 3 21.36 19.34 23.29
5 16.4 15.72 17.97
Mineral composition of apple before and after osmotic dehydration
It is well known that the minerals in solution have irreplaceable importance for nor-
mal functioning and revitalization of certain cell elements, organs and organisms of
plants and animals (10).
Figures 2-5, show the changes of the contents of the analyzed mineral components
(K, Na, Ca and Mg) in the apples dehydrated in saccharose solutions and sugar beet
molasses.
Since saccharose solution does not contain mineral component, a decrease in their
contents was observed during osmotic dehydration. On the other hand, because sugar beet
molasses is rich in minerals and other biogenic substances, and due to the diffusion of
these substances from the hypertonic solution into the sample, an increase in the content
of these components takes place in the treated apple.
40
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
Time, h
3
-40 -20 0 20 40 60
loss/increase in the content of K, %
saccharose-30% saccharose-50%
saccharose-70% molasses-40%
molasses-60% molasses-80%
Fig. 2. Loss/increase in the content of K in the apples during osmotic dehydration
5
Time, h
-50 0 50 100
loss/increase in the content of Na, %
Fig. 3. Loss/increase in the content of Na in the apples during osmotic dehydration
From the results we can conclude that the content of K decreased with extending
immersion time. Osmotic dehydration in solution S2, after 5 hours, led to the largest loss
in K content in apple (34.77%), while the smallest losses of K occured in the sample S1,
41
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
also after 5 hours (19.68%). After 3 hours of immersion, content of K was the highest in
the apple treated in the M3 solution and it was 159.06 mg/100g, and after 5 h of the
process the increase was the highest in apple dehydrated in M2 solution (48.24%).
The content of Na was the lowest in the apple dehydrated in S2 solution (50%) and
the highest in the apple treated in the saccharose solution with 30% of dry matter (S1).
Expressed in percentage, after 5 h immersion, the minimum loss was 26.91% and the
maximum was 39.01%.
After the first hour of immersion the increase of Na content was the highest in the M1
solution and it was 24.16%. At the end of the process of osmotic dehydration, after 5 h,
the maximum increase was 83.99%, and it was in the apple dehydrated in 80% molasses
(M3), but very similar results were achieved in the application of molasses with 60% of
dry matter as osmotic agent.
As in the previous two cases, the loss of Ca, during the process of osmotic dehydra-
tion was the smallest in the hypertonic solution with the lowest concentration of dry
matter. The loss was minimal in the S1 solution, and after 5 h it was 18.34% and the
highest in the sample S3 (37.4%). The increase of the content of Ca was the most
expressed in the sample dehydrated in 80% molasses, and it was 142.43%, which means
that the amount of Ca in that sample was about 2.5 times higher in comparison to the
amount of minerals in the starting, non-dehydrated sample.
5
Time, h
-50 0 50 100 150 200

loss/increase in the content of Ca, %
saccharose-30% saccharose-50% saccharose-70%

molasses-40% molasses-60% molasses-80%
Fig. 4. Loss / increase in the content of Ca in the apples during osmotic dehydration
42
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
Time, h
-50 0 50 100 150

loss/increase in the content of Mg, %
Fig. 5. Loss / increase in the content of Mg in the apples during osmotic dehydration
A very similar conclusion can be drawn on the basis of the results, shown in Figure 3
and related to the content of Mg in the apple sample after 1, 3 and 5 h of immersion. The
loss of this mineral was quite large and after 5 h was 41.14% in the sample dehydrated in
S3 solution. In comparison with the losses of other minerals (K, Na and Ca), the decrease
in Mg content was the highest.
The results of analysis of apple samples obtained by osmotic dehydration in sugar
beet molasses after 1, 3 and 5 h of immersion, indicate a significant increase in Mg con-
tent in the final product. The best results were achieved by applying 80% molasses as os-
motic medium, i.e. in the sample dehydrated in M3 solution. The quantity of the mineral
was, about 2.2 times higher in comparison to that before the osmotic dehydration.
On the basis of presented results, a general conclusion that the use of a sugar beet
molasses as hypertonic solution has a significant advantage in comparison to the sucrose
solution at all concentration and all immersion time.
Table 5 presents the results of texture determination in nondehydrated and osmo-
tically dehydrated apples. It is shown as an example of dehydrated apples in the sugar
beet molasses and sucrose solution with the highest concentrations of dry matter.
Based on the measured cutting force (mechanical properties of texture) it can be con-
cluded that the samples osmodehydrated in sucrose solutions were softer and gentler
because the cutting force was three times smaller in comparison to other samples of ap-
ples. There is no important difference in the strength between nondehydrated apples and
osmotically dehydrated in sugar beet molasses.
43
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
Table 5. Cutting force for different samples of apple
Cutting force, N
Sample Apple dehydrated in Apple dehydrated
Nondehydrated
saccharose solution in molasses
apple
(70% solid content) (80% solid content)
1. 9.4 2.7 10.3
2. 8.6 2.7 10.1
3. 9.9 3.5 7.0
4. 8.6 3.1 7.1
5. 10.3 2.6 8.7
6. 10.6 2.4 6.5
7. 9.3 2.2 8.1
8. 9.5 3.0 9.0
9. 7.3 2.6 10.2
10. 9.0 1.8 7.7
average 9.25 2.66 8.47
Osmotically dehydrated apple, in sugar beet molasses, is enriched with the Ca2+ ions
from molasses which bind to the R-COO– groups of pectic matter in apple. Therefore, the
final product has approximately the same as raw material.
CONCLUSION
The process of osmotic dehydration is the most intensive in the first hour. Water loss
and weight reduction after 5 h of dehydration are higher in comparison to the dehydration
over a shorter period of time. During osmotic dehydration of apple at 55° C water loss,
depending on the applied concentration of osmotic solution, is 6-22 times faster than the
increase of dry matter, which is very desirable and the candying effect of fruit is avoided.
By analyzing the content of mineral components (K, Na, Ca and Mg) in the samples
osmotically dehydrated in saccharose solutions and sugar beet molasses, some advanta-
ges of applying molasses as hypertonic solution, were observed.
Osmotic dehydration of apple with saccharose solutions decreases the content of mi-
neral substances in the fruit tissue, while the osmotic solutions of sugar beet molasses
increase the amount of mineral substances in the apple, and therefore increase its nutritive
properties.
Apples osmotically dehydrated in pure saccharose solution had a softer and more
gentle texture in comparison to the apple dehydrated in sugar beet molasses with 80% of
dry matter.
ACKNOWLEDGEMENT
This research is part of the project supported by the Ministry of Science and Tech-
nological Development of the Republic of Serbia, TR – 20112, 2008-2010.
44
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
REFERENCES
1. Filipčev, B., Lević, Lj., Pribiš, V., D. Kabić: Melasa šećerne repe kao pogodan hiper-
tonični rastvor za osmotski predtretman jabuke, XIII Savetovanje o biotehnologiji,
Čačak, 28-29. Zbornik radova 13 (2008) 323-331.
2. Koprivica, G., N. Mišljenović, Lj. Lević, T. Kuljanin: Influence of nutrients present
in sugar beet molasses and saccharose solutions on the quality of osmodehydrated
carrot, PTEP 13, 2 (2009) 184 -187.
3. Mizrahi, S., S. Eichler, J.L. Cuq: Osmotic dehydration in gel systems, Journal of
Food Engineering 49 (2001) 87-96.
4. Ramallo, L.A., R.H. Mascheroni: Rate of water loss and sugar uptake during the os-
motic dehydration of pineapple, Brazilian Archives of Biology and Technology 48, 5
(2005) 761-770.
5. Filipčev, B., Lj. Lević, O. Šimurina, T. Kuljanin: Priprema voća postupkom osmotske
dehidratacije u melasi šećerne repe sa ciljem njegove primene u pekarskoj industriji,
PTEP- Journal on Processing and Energy in Agriculture 10, 5 (2006) 154-157.
6. Shi, J., M. Le Maguer: Osmotic dehydration of foods: mass transfer and modeling
aspects, Food reviews International 18, 4 (2002) 305-335.
7. Sachetti, G., A. Gianotti, M. Della Rosa: Succrose – salt combined effects on mass
transfer kinetics and product acceptability. Study on apple osmotic treatments, Jour-
nal of Food Engineering 49 (2001) 163-173.
8. Mišljenović, N., G. Koprivica, Lj. Lević, M. Petkova, T. Kuljanin: Prenos mase to-
kom osmotske dehidratacije jabuke i mrkve u melasi šećerne repe, PTEP 12, 4 (2008)
211-214.
9. Filipović, N., V. Pribiš, Lj. Lević, D. Simović-Šoronja, B. Filipčev: Karakteristike
hleba sa jabukom osmotski dehidriranom u melasi šećerne repe, PTEP 11, 3 (2007)
120-123.
10. Lević, LJ., V. Filipović, T. Kuljanin: Osmotski tretman oblikovanog korena mrkve u
saharozi i melasi, PTEP 11, 3 (2007) 132-135.
11. Šušić, S., V. Sinobad: Istraživanja u cilju unapređenja industrije šećera Jugoslavije,
University in Belgrade Hem. Ind. 43, 1-2 (1989) 10 -21.
12. Milić, M., V. Karadžić, S. Obradović i koautori: Metode za laboratorijsku kontrolu
procesa proizvodnje fabrika šećera, Tehnološki fakultet, Novi Sad (1992).
13. Brennan, J. G.: Food Texture Measurement, Development in Food Analysis Techni-
ques-2, Applied Science Publishers LTD, London (1980).
14. AOAC, Official Methods of Analysis, Washington (2000), USA.
15. Le Maguer, M.: Osmotic dehydration: review and future directions, Proceedings of
the symposium in food preservation process, Brussels (1988): CERFCI vol. 1, 283-
309.
45
APTEFF, 40, 1-220 (2009) UDC:634.11:664.854:664.126+664.15:543.92
DOI: 10.2298/APT0940035K BIBLID: 1450-7188 (2009) 40, 35-46
ПРОМЕНА НУТРИТИВНОГ И ТЕКСТУРАЛНОГ КВАЛИТЕТА ЈАБУКЕ

ПРИ ОСМОТСКОЈ ДЕХИДРАТАЦИЈИ У МЕЛАСИ ШЕЋЕРНЕ РЕПЕ И
РАСТВОРИМА САХАРОЗЕ
Гордана Б. Копривица, Невена М. Мишљеновић, Љубинко Б. Левић

и Вјера С. Прибиш
У раду су испитивани текстура и минерални састав јабуке осмотски дехидри-

ране у хипертоничним растворима меласе шећерне репе у поређењу са својствима
јабуке третиране у растворима сахарозе. Осмотска дехидратација је извођена на ат-
мосферском притиску и константној температури осмотског раствора од 55ºЦ. То-
ком експеримента мењана је концентрација меласе шећерне репе од 40 дo 80% и
раствора сахарозе у опсегу 30-70% а праћене су и промене најважнијих кинетичких
параметара осмотске дехидратације након 1, 3 и 5 сати трајања имерзије. Утврђено
је да је осмотска дехидратација прихватљив поступак делимичног уклањања воде
из свеже јабуке - ниво влаге је 4-5 пута мањи у односу на ниво влаге у свежем узор-
ку. Током осмотске дехидратације, у узорцима третираним у меласи шећерне репе
дошло је до значајног повећања садржаја минералних материја па се самим тим по-
већала и њена нутритивна вредност. На основу измерених сила сечења (механичке
особине текстуре) закључено је да су узорци након осмотске дехидратације у
растворима сахарозе изразито мекши, сочнији и нежнији - сила сечења је три пута
мања у односу на остале узорке јабуке.

Accepted 28 August 2009
46
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.047
DOI: 10.2298/APT0940047M BIBLID: 1450-7188 (2009) 40, 47-52
USE OF MILK-BASED KOMBUCHA INOCULUM FOR MILK

FERMENTATION
Radomir V. Malbaša., Eva S. Lončar, Spasenija D. Milanović and Ljiljana A. Kolarov
In this investigation fermented milk beverages with 0.9% of milk fat were produced
using 10 and 15% (v/v) of traditional and milk-based kombucha inoculum by application
of appropriate technological process.
Milk fermentation using two types and concentrations of kombucha inoculum were
stopped when the pH reached 4.5. Sigmoidal fermentation profiles were noticed with tra-
ditional kombucha inoculums and linear with milk-based kombucha inoculums.
Chemical content and physico-chemical characteristics of kombucha fermented milk
beverages were typical and yoghurt-like for all obtained products.
The best textural and sensory characteristics possesed beverage obtained in fermen-
tation of milk using 10% (v/v) of milk-based kombucha inoculum.
KEY WORDS: Kombucha, milk fermentation, milk-based kombucha inoculum, product
quality
INTRODUCTION
Kombucha is well-known symbiotic association of several yeast species and acetic
bacteria, whose metabolic activity on tea sweetened with sucrose produces a pleasant
refreshing beverage, containing a number of useful compounds (1). In addition to sucro-
se, application of other sugars (glucose, fructose, lactose) is possible. This may have
major effect on the formation of lactic acid as product (2).
As far as lactose fermentation by kombucha is concerned, only a few investigations
have been reported; Lončar et al. (2001) reported findings of metabolic activity of kom-
bucha on milk (3), whilst, Belloso-Morales and Hernández-Sánchez (2003) described the
production of beverage from cheese whey (4). Malbaša et al. (2009) investigated milk
fermentation using kombucha inoculums with black, green tea and Jerusalem artichoke
tubers extract (5). Investigations of Lončar et al. (2001) and Malbaša et al. (2009) shows
significantly longer durations of kombucha fermentations on milk comparing to probiotic
yoghurt fermentation (1, 5).
This paper describes the manufacture of fermented milk beverages with 0.9% of milk
fat using 10 and 15% (v/v) of traditional kombucha inoculum and also milk-based kom-
Dr. Radomir V. Malbaša, Assist. Prof., Dr. Eva S. Lončar, Prof., Dr. Spasenija D. Milanović, Prof., Dr. Ljiljana
A. Kolarov, Assoc. Prof., University of Novi Sad, Faculty of Technology, Bulevar Cara Lazara 1, 21000 Novi
Sad, Serbia
47
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.047
DOI: 10.2298/APT0940047M BIBLID: 1450-7188 (2009) 40, 47-52
bucha inoculum by application of appropriate technological process. It investigates qua-

lity of obtained products comparing chemical composition, physico-chemical, textural
and sensory characteristics.
EXPERIMENTAL
Milk
Homogenised and pasteurized cow`s milk with 0.9% fat, 3.3% protein and 4.8% lac-
tose, produced in Novosadska mlekara, Novi Sad, Serbia, was used for the fermentation.
Kombucha culture
Local kombucha culture containing Acetobacter xylinum and at least five yeast strains
(Saccharomycodes ludwigii, Saccharomyces cerevisiae, Saccharomyces bisporus, Toru-
lopsis sp. and Zygosaccharomyces sp.) was applied for production of fermented milk
beverages (6).
Production of beverages
Fermented milk-based kombucha beverages were produced from milk inoculated at

43oC. Two amounts of traditional kombucha inoculum (5) obtained on sweetened black
tea of 10 and 15% (v/v) were used. The fermentations lasted up to reach pH 4.5. The ob-
tained gels were cooled to 8oC, homogenized by mixing, and packed in plastic glasses.
Two beverages marked T10% and T15% were obtained. Further, T10% and T15% were
used as inoculums for the next fermentation on milk following the procedure described
above. Inoculum amounts were 10% (v/v) of T10% and 15% (v/v) of T15%. Beverages
marked MB10% and MB15% were obtained.
Methods of analysis
The chemical content of obtained products was analyzed using standard methods (7).
Syneresis of whey was analyzed by method of Atamer et al. (8).
Water holding capacity (WHC) (%) represents the amount of whey drained after cen-
trifugation of 20 g of sample during 30 minutes at room temperature.
Textural characteristics were investigated using Texture analyser TA.Xplus (Micro
Stable System, England) at 4oC.
Sensory characteristics of obtained products were examined by qualified evaluators,
who assessed each particular element of quality as follows: appearance, colour, consis-
tency, odour and taste. Possible marks for each separate characteristic were in a range
from 1 to 5.
In Fig. 1 are presented fermentation curves for both inoculums and their concentra-
tions.
48
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.047
DOI: 10.2298/APT0940047M BIBLID: 1450-7188 (2009) 40, 47-52
T10%
6.5 T15%
MB10%
MB15%
6.0
5.5
pH
5.0
4.5
0 2 4 6 8 10 12 14
Time [hours]
Fig. 1. pH changes during kombucha fermentation on milk
Fermentations with traditional kombucha inoculum were stopped when pH reached

4.5. It needed about 12.5 hours for both inoculum concentrations. During first 6 to 7
hours were not noticed pH changes which caused sigmoidal fermentation curves.
Fermentations with milk-based kombucha inoculums were for about two times shor-
ter. pH changes were almost linear. The reason for such behaviour was absence of period
for adaptation to substrate which was necessary for microorganisms in traditional kom-
bucha inoculum. The very important fact is reasonably shorter fermentation with milk-ba-
sed kombucha inoculum primarily for the economic reasons.
Obtained results with milk-based inoculums are in accordance to investigations of
Lončar et al. (2001) and Milanović et al. (2002), where is proved that higher inoculum
concentration affects shorter fermentation time (3, 9).
Chemical content of obtained beverages is presented in table 1.
Table 1. Chemical content of fermented milk beverages
Beverage
Component(%) T10% T15% MB10% MB15%
Dry matter 9.52 9.55 10.09 11.01
Fat 0.86 0.86 0.89 0.88
Total proteins 3.13 3.06 3.61 3.44
Ash 0.72 0.66 0.84 0.82
Acidity (SHo) 28.6 31.6 27.6 31.2
Dry matter content is approximatly the same in T10% and T15%, and in MB10% and
MB15% is slightly higher. In MB15% is measured the highest dry matter content which
49
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.047
DOI: 10.2298/APT0940047M BIBLID: 1450-7188 (2009) 40, 47-52
is logical because of applied inoculum that itself contains the higher amount of dry mat-
ter. The same explanation could be used for the ash and protein content which is higher in
MB10% and MB15%.
Values of acidity expressed in SHo were higher in samples T15% and MB15% which
means that inoculum concentration affects acidity of product.
Average chemical content of obtained fermented beverages is in a range of results ob-
tained in earlier investigations of possibility of kombucha application in production of
fermented milk beverages (5).
In table 2 are presented some physico-chemical characteristics of fermented milk
beverage.
Table 2. Physico-chemical characteristics of fermented milk beverages
Beverage
Parameter
T10% T15% MB10% MB15%
Syneresis (mL) 35.0 34.5 29.0 30.0
Water holding capacity (WHC) (%) 34.5 36.3 32.0 35.3
Values of syneresis were in narrow interval as well as water holding capacity (WHC).
Higher WHC indicates better quality of products but it was not possible to point any of
produced beverages in a view of this parameter. Similar results obtained Duraković et al.
(2008) who investigated fermentation of milk with 0.9% milk fat with concentrated kom-
bucha inoculums (10).
Textural characteristics of obtained products are presented in table 3.
Table 3. Textural characteristics of fermented milk beverages
Beverage
Characteristic
T10% T15% MB10% MB15%
Firmness (g) 15.616 15.559 29.924 15.525
Consistency (gs) 444.158 433.774 701.611 408.725
Cohesivity (g) -5.340 -5.678 -24.359 -11.098
Index of viscosity (gs) -1.106 -1.235 -32.140 -6.777
Results presented in table 3 show that the highest firmness had MB10% and in the
other samples firmness was very similar. The same pattern was for the consistency. Con-
sistency value indicates higher density of the product.
Higher negative value represents more cohesive sample. The maximum value of co-
hesiveness was noticed also for MB10%. The same manner was found for the viscosity
index.
Values of textural parameters of quality of MB10% were significantly higher in com-
parison to the other produced fermented milk beverages and that characteristics recom-
mend this sample for further investigation and production.
Sensory characteristics of fermented milk products are the most important for consu-
mers. In table 4 are presented the results of sensory analysis.
50
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.047
DOI: 10.2298/APT0940047M BIBLID: 1450-7188 (2009) 40, 47-52
Table 4. Sensory characteristics of fermented milk beverages
Characteristic Beverage
T10% T15% MB10% MB15%
Appearance 5 5 5 5
Colour 5 5 5 5
Consistency 4 4 5 5
Odour 4.5 4.5 5 5
Taste 4 4 5 4.5
Scores of sensory analysis indicate that usage of milk-based kombucha inoculum af-
fect better characteristics of fermented milk beverage (Table 4), and taste of MB10% and
MB15% was characterized as mild and refreshing, but at all it was better composed in
MB10%. Appearance, colour and consistency of products were characteristic for the yog-
hurt-like products. Odour also was more intensive and typical in products obtained using
milk-based kombucha inoculums for fermentation.
The best sensory total mark is for MB10% which has the best textural characteristics
too (Table 3).
CONCLUSION
Milk-based kombucha inoculum is better than traditional one because of economic

reasons. It shortened milk fermentation for almost two times.
Chemical content and physico-chemical characteristics of kombucha fermented milk
beverages were typical for all obtained products.
The best textural and sensory characteristics had beverage obtained in fermentation of
milk using 10% (v/v) of milk-based kombucha inoculum which is potentially very inte-
resting for further investigations.
ACKNOWLEDGEMENT
Financial support from the Ministry of Science and Technological Development of
Serbia is highly acknowledged (Grant TR-20008).
REFERENCES
1. Dufresne, C. and E. Farnworth: Tea, kombucha and health: A Review. Food Research
International 33 (2000) 409-421.
2. Reiss, J.: Influence of different sugars on the metabolism of the tea fungus. Zeitschrift
fur Lebensmittel Untersuchung und Forschung 198 (1994) 258-261.
3. Lončar, E., Milanović, S., Carić, M., Malbaša, R. and M. Panić: Metabolička aktiv-
nost čajne gljive u mleku. Prehrambena industrija – Mleko i mlečni proizvodi 12, 1-2
(2001) 13-17.
4. Belloso-Morales, G. and H. Hernandez–Sanches: Manufacture of a beverage from
cheese whey using a tea fungus fermentation. Revista Latinoamericana de Microbio-
logia 45, 1-2 (2003) 5-11.
51
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.047
DOI: 10.2298/APT0940047M BIBLID: 1450-7188 (2009) 40, 47-52
5. Malbaša, R., Milanović, S., Lončar, E., Djurić, M., Carić, M., Iličić, M. and Lj. Ko-
larov: Milk – based beverages obtained by Kombucha application. Food Chemistry
112 (2009) 178-184.
6. Markov, S.L., Malbaša, R.V., Hauk, M.J. and D.D. Cvetković: Investigation of tea
fungus microbe associations. I. The yeasts. Acta Periodica Technologica 32 (2001)
133-138.
7. Carić, M., Milanović, S. and D. Vucelja: Standard Methods of Analysing Milk and
Dairy Products (in Serbian). Faculty of Technology, Novi Sad (2000).
8. Atamer, M., Carić, M., Milanović, S. i D. Gavarić: Kvalitat jogurta proizvedenog iz
UF mleka, Zbornik Matice srpske za prirodne nauke, Matica srpska Novi Sad 91
(1996) 19-26.
9. Milanović, S., Carić, M., Lončar, E., Malbaša, R., Panić, M. i D. Dobrić: Primena
koncentrata čajne gljive u proizvodnji fermentisanih mlečnih napitaka. Prehrambena
industrija - Mleko i mlečni proizvodi 13, 1-2 (2002) 8-13.
10. Duraković, K., Milanović, S., Carić M., Iličić, M., Đurrić, M., Tekić, M. i J. Lenđel:
Funkcionalni niskoenergetski fermentisani mlečni napitak proizveden uz primenu
kombuhe, Prehrambena industrija- mleko i mlečni proizvodi 19, 1-2 (2008) 66-73.
ПРИМЕНА МЛЕЧНО-ФЕРМЕНТИСАНОГ ИНОКУЛУМА КОМБУХЕ ЗА

ФЕРМЕНТАЦИЈУ МЛЕКА
Радомир В. Малбаша, Ева С. Лончар, Спасенија Д. Милановић, Љиљана А. Коларов
У овом истраживању су произведени ферментисани млечни производи од млека

са 0,9% млечне масти користећи 10 и 15% (v/v) традиционалног и млечно фермен-
тисаног инокулума комбухе применом одговарајућег технолошког процеса.
У свим ферментацијама је процес заустављан када је вредност pH достигла 4,5.
Уочене су сигмоидалне ферментационе криве где је коришћен традиционални ино-
кулум комбухе и линеарне за млечно ферментисани инокулум комбухе.
Хемијски састав и физичко-хемијске карактеристике комбуха ферментисаних
млечних производа су били типични и слични јогурту код свих производа.
Најбоље текстуралне и сензорне карактеристике је имао напитак произведен
ферментацијом млека са 10% (v/v) млечно ферментисаног инокулума.

52
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
ANTIOXIDANT ACTIVITY OF WHITE GRAPE SEED EXTRACTS ON DPPH

RADICALS
Anamarija I. Mandić, Sonja M. Djilas, Jasna M. Čanadanović-Brunet,

Gordana S. Ćetković and Jelena J. Vulić
Composition and antioxidant activity of grape seed extract (GSE) obtained from red
grape varietes are very well documented, in contrast to the white varietes. This paper
presents the results of polyphenols content of ethyl acetate extract of grape seeds, obtai-
ned from two white grape varieties, Italian Riesling and Župljanka, and their antioxidant
activity on the stable DPPH radical. The influence of the addition of GSE to raspberry
juice on the DPPH radical was also examined. Content of total polyphenols in GSEs
ranged between 81.6 and 82.8% (w/w), and the contetn of flavan-3-ols between 66.2 and
91.0% (w/w). HPLC results showed that the most abundant components in the extract
were (+)-catechin and (-)-epicatechin for both grape varieties. All tested GSEs exhibited
good antioxidant activity. IC50 values for the GSEs of Italian Riesling and Župljanka
were 0.79 and 0.95 mg sample/mg DPPH radical, respectivelly. Since the GSE of Italian
Riesling possesed stronger antioxidant activity, it was used for further experiments. The
IC50 value for raspberry juice was 4.18 mg raspberry juice/mg DPPH. The raspberry
juice with addition of 0.60 µg/mL of GSE showed antioxidant activity of 39.2%. The same
juice with the threefold concentration of vitamin C (1.81 µg/ml) exhibited similar antioxi-
dant activity (33.9%). Antioxidant activity of the same amount of juice without added an-
tioxidants was lower (15.7%). The results showed that the GSE of white varietes could be
considered as a good functional food ingredient.
KEY WORDS: Grape seed extract, white grape variety, antioxidant activity,
polyphenolics
INTRODUCTION
The use of synthetic antioxidants in the food has been under scrutiny for toxicological
reasons, and therefore the interest in the natural antioxidants has steadily been increasing
(1-3). The antioxidant and radical scavenging activities of a large number of polyphenolic
compounds isolated from plants have been studied (4-8). Green tea, grape seeds and skin
Dr. Anamarija I. Mandić, Institute for Food Technology, University of Novi Sad, Bulevar Cara Lazara 1, 21000
Novi Sad, Serbia; Dr. Sonja M. Đilas, Prof., Dr. Jasna M. Čanadanović-Brunet, Prof., Dr. Gordana S. Ćetković,
Prof., Jelena J. Vulić, B.Sc., Faculty of Technology, University of Novi Sad, Bulevar Cara Lazara 1, 21000
Novi Sad, Serbia
53
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
are considered as the remarkable rich sources of polyphenolics (6). Composition and anti-
oxidant activity of grape seed extract (GSE) obtained from red grape varietes are very
well documented, in contrast to the white varietes (9, 10). The GSE rich in polyphenols,
is a mixture of proanthocyanidines: monomers, olygomers and polymers of flavan-3-ols
((+)-catechin, (-)-epicatechin, (-)-epicatechin-O-gallate, and (-)-epigallocatechin) linked
with C4-C8 or C4-C6 bonds. The biological, pharmacological and medicinal properties
of the bioflavonoids and proanthocyanidins have been extensively reviewed. Besides the
free radical scavenging and antioxidant activity, proanthocyanidins exhibit vasodilatory,
anticarcinogenic, anti-allergic, antiinflammatory, antibacterial, cardioprotective, immune-
stimulating, anti-viral and estrogenic activities, as well as they act as inhibitors of the
enzymes phospholipase A2, cyclooxygenase and lipooxygenase (9).
The aim of this work was to determine the polyphenolic composition and antioxidant
activity on DPPH radical of GSEs obtained from two white grape varieties, Italian
Riesling and Župljanka. Also, the influence of the addition of GSE to a raspberry juice on
the DPPH radical was examined.
EXPERIMENTAL
Chemicals and Samples

ll solvents used for the extraction and spectrophotometric determination were of
analytic grade. Methanol and ethanol were purchased from Lachema (Neratovice, Czech
Republic) and ethyl acetate from Kemika (Zagreb, Croatia). Acetonitrile and methanol,
LiChrosolv, gradient grade for chromatography and vanillin were obtained from Merck
(Darmstadt, Germany). 2,2-Diphenyl-1-picrylhydrazil stable radical (DPPH), gallic acids,
(+)-catechin and (±)-epicatechin were purchased from Sigma-Aldrich Co. (St. Louis,
MO). Folin-Ciocalteau reagent was obtained from Fluka (St. Gallen, Switzerland).
he sample, pomace of Vitis vinifera cultivars, Italian Riesling and Župljanka, were
collected at the “Navip” winery. The grapes used for processing were harvested at optium
technological maturity, as judged by the indices of sugar and acid contents, established
by the laboratory in the “Navip” winery. The average sugar content in grapes was 17.0%,
and the acid content 6.4%. After pressing, the pomace was sampled for experiments. The
seeds were separated from other pomace components (skin and stems), dried at room
temperature, and stored at 24°C prior to extraction.
aspberry sample was purchased from a local market. Juice was separated from the
pulp, and the freshly obtained juice was two times diluted with water and used in the
antioxidant assay procedure.
Extraction procedure
The extraction was carried out according to the method described by Pekić et al. (11).
100 g of grape seeds was extracted with 400 mL of 90% ethyl acetate in a sealed bottle at
room temperature for 48 h, with occasional mixing. The obtained extract was filtrated,
and the solvent was removed by evaporation under the reduced pressure to the volume of
approximately 20 mL, at maximum temperature 40ºC. The concentrated ethyl acetate
solution was mixed with a five-fold volume of chloroform and the formed precipitate was
54
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
separated by filtration through a nutsch filter B-4. The precipitate was washed with
chloroform, dried in a vacuum desiccator and the yield of obtained GSE was calculated.
Total phenolics and flavan-3-ols in GSE
The amount of the total soluble polyphenols in the GSE was determined spectro-
photometrically according to the Folin-Ciocalteau method (12). Gallic acid was emplo-
yed as a calibration standard and the results were expressed as gallic acid equivalents in
grams per 1 kg of dry seed weight. The amount of total flavan-3-ols was assayed spectro-
photometrically by the vanillin method using (+)-catechin as a standard (13,14). The
values were expressed as catechin equivalents in grams per 1 kg of dry seed weight. All
analyses were performed in triplicate and the results were averaged.
HPLC Analysis of GSE
HPLC analyses were conducted on a Hewlett-Packard Liquid Chromatograph HP
1090 equipped with Diode Array Detector (DAD). A reversed-phase column (Zorbax SB-
C18, 5 µm, 3.0 x 250 mm i.d.), protected by guard column (Zorbax SB-C18, 5 µm, 4.6 x
12.5 mm i.d., Agilent, USA) was used throughout this research. The detection was
performed at 277 nm and the absorption spectra of the compounds were recorded
between 210 and 400 nm. The solvent gradient was formed by varying the proportion of
the solvent A (1% acetic acid in water, v/v) to the solvent B (acetonitrile) (15). The
solvent linear gradient elution programme was as follows: 0-20 min, 95-87% A; 20-30
min, 87% A; 30-46 min, 87-78% A; 46-55 min, 78-10% A; 55-65 min, 10% A. The co-
lumn was equilibrated to the initial conditions, 95% A, 10 min. The flow rate was set at
0.300 mL/min. The column was operated at room temperature (22°C). The sample
injection volume was 10 µL, and the injection was performed manually. The GSE was
dissolved in 10% (v/v) methanol in water and the obtained concentration was 1.0 mg/mL.
All solutions were filtered through 0.45 µm pore size nylon filters (Rotilabo–Spritzen-
filter 13 mm, Roth, Karlsruhe, Germany) before injecting them into the HPLC system.
Phenolic components in a sample extract were identified by matching the retention time
and their spectral characteristics against those of the standards. The purity of the peaks
was determined to ensure the identification. The external standard method was a tech-
nique used for quantification. For each component, (gallic acid, (+)-catechin, and (±)-epi-
catechin) a stock solution was made from the commercial standards dissolved in 10%
(v/v) methanol in water to obtain the concentration of 1.0 mg/mL.
The diluted stock solutions were used for calibration. The final concentrations were in
the range of 0.005-0.200 mg/mL. The peak areas from the chromatograms were plotted
against the known concentrations of the standards. The equations generated via linear
regression were used to establish the concentrations of the phenolic compounds in the
extracts. When reference compounds were not available, the calibration of (+)-catechin
was used. The results were presented as mean values with the standard deviations (SD).
Spectrophotometric assay for DPPH radical
In a test tube containing 3 mL of methanolic GSE solution, prepared at five different
concentrations ranged between 0.1 and 3.5 μg/mL, 1 mL of c(DPPH radical) = 90 μmol/L
55
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
(dissolved in 95% methanol in water, v/v) was added (test sample). The blank was
prepared by adding 1 mL of DPPH radical solution to 3 mL of 95% methanol. The
reaction mixture was allowed to stay in the dark at room temperature for 60 minutes.
The absorbance was read at 515 nm using 95% methanolic solution as a reference
solution (16). The percentage of the remaining DPPH radical (%DPPHrem) and antioxi-
dant activity (%AA) were calculated for each concentration of the GSE from the obtained
absorbance for the sample (As) and the blank (A0) using the following equations:
%DPPHrem = (As / A0)·100
%AA = 100 - %DPPHrem
The values of the calculated IC50 were expressed as mg GSE/mg of DPPH radical.
The IC50 value is the concentration of antioxidant, required for 50% scavenging of DPPH
radical in the specified time period.
Determination of DPPH scavenging activity of raspberry juice, vitamin C, and the
juice with the addition of 0.60 µg/mL GSE or 1.81 µg/mL vitamin C, differed from the
procedure given for GSE only in concentration range. Procedure for the scavenging
activity of rasperry juice was as follows: 2 mL of raspberry juice, two times diluted with
water, was further diluted to 50 mL with methanol, in order to prepare the stock solution.
The final concentrations used for the assay were in the range of 0.05-0.50 mg/mL, which
corresponds to 0.5-5.0 mL of fresh raspberry juice. Solution of vitamin C was prepared at
five different concentrations ranging from 0.20 to 2.00 μg/mL.
Proanthocyanidins from the grape seeds are extracted with the methanol, ethanol,
acetone and ethyl acetate, or with their mixtures for analytical and preparative purposes
(17). The differences in the GSE yields could be observed due to the extraction method
used. The yield of the GSE extract obtained with ethyl acetate for the Italian Riesling was
6.03 ± 0.53 g/kg of dry seeds, and it was higher than the yields (3.6-4.7 g/kg of dry seeds)
reported by Pekić et al. (11) for the same grape variety. These differences could be attri-
buted to the variations due to the seasons within the grape variety. Additionally, genetic
potential of the individual species for the polyphenol biosynthesis and maturation stage
may affect polyphenol content in seeds, along with the variations from season to season
(18-20). The yield of the GSE extract obtained with ethyl acetate for the Župljanka was
7.01 ± 0.75 g/kg of dry seeds.
In this paper, the total soluble polyphenols of the obtained GSE is expressed as gallic
acid equivalents and for the Italian Riesling it was 4.92±0.30 g/kg of dry seeds, as regards
81.6% (w/w GSE). Content of total soluble polyphenols for the Župljanka was 5.80±0.28
g/kg of dry seeds, as regards 82.8% (w/w GSE). Comparing the yields obtained from the
methanolic extracts, 1.84-4.07 g/kg of dry seeds, reported by Revilla et al. (21) for the
white grape varieties, the yields were similar to or higher than those obtained in this
research. The content of flavan-3-ols in GSE was 5.83±0.20 g/kg of dry seeds, as regards
91.0% (w/w GSE) for the Italian Riesling, and 4.64±0.20 g/kg of dry seeds, as regards
56
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
66.2% (w/w GSE) for the Župljanka. It may be concluded that the GSE of Župljanka
contains higher amounts of taninns than the GSE of Italian Riesling, due to the fact that
despite of the higher yield and total polyphenolics content of Župljanka the content of
flavan-3-ols was lower. The obtained results are in accordance with the results of Na-
kamura et al. (17). According to their results, the differences in the sensitivity to the
flavan-3-ols assay were observed, higher reactivity was observed in procyanidins which
are highly polymerized and lower reactivity was observed in catechins which are highly
esterified by gallic acid. In addition, epicatechin was more reactive than catechin.
Fig. 1. HPLC chromatogram of a GSE recorded at 277 nm. Peak assignment (tR, min):
1. gallic acid (9.3); 2. procyanidin-B3 (20.6); 3. procyanidin-B1 (22.4);
4. (+)-catechin (25.0); 5. procyanidin-B4 (27.5); 6. procyanidin-B2 (29.1);
7. (−)-epicatechin (33.4); 8. procyanidin-C1 (39.5); 9. dimer gallate (42.1);
and 10. Procyanidin-B5 (49.9)
A HPLC chromatogram of GSE is shown in Fig 1. The separation of the phenolic

compounds in GSE was achieved within 60 min. Under the described chromatographic
conditions, the components were eluted in the following order: gallic acid, procyanidin-
B3, procyanidin-B1, (+)-catechin, procyanidin-B4, procyanidin-B2, (−)-epicatechin, pro-
cyanidin-C1, dimer gallate, and procyanidin-B5. The content of the total phenolic com-
pounds determined by HPLC in GSE was 150 ± 2.10 mg/g GSE and 114 ± 2.90 mg/g
GSE, for the Italian Riesling and Župljanka, respectivelly. A dominant compound was
(+)-Catechin, with amount of 32.2 ± 1.30%, followed by (−)-epicatechin, 22.3 ± 0.72%
for the Italian Riesling, and 30.6 ± 1.13%, followed by (−)-epicatechin, 26.8 ± 0.93% for
the Župljanka. The content of phenolic compounds in the ethyl acetate extracts reported
by Guendez et al. (18) was on average three times lower determined by HPLC than the
one obtained by the Folin-Ciocalteau assay. These results are in agreement with the re-
sults obtained in this research. It could be assumed that lower results of polyphenolics
content obtained by HPLC determination are due to the fact that only monomers, dimers
57
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
and trimers of procyanidins could be determined, while the obtained higher results for
spectrophotometric assay are owing to the presence of highly polymerized procyanidins
and other substances in extract.
Antioxidant activity of GSE is assessed by scavenging of the stable DPPH radical.
With increasing concentrations of GSE, antioxidant activity on investigated free radical
increased. The IC50 for the GSEs of Italian Riesling and Župljanka were 0.79 and 0.95
mg sample/mg DPPH radical, respectivelly, and it is in the same range as the values ob-
tained by Bakkalbaşi et al (22) for the GSEs of the white grape varieties using acetone for
the extraction (0.52–0.82 mg sample/mg DPPH radical). The antioxidant activities of
different concentrations of GSEs of Italian Riesling and Župljanka are presented on Fig.
2. Since the better antioxidant activity was obtained for the Italian Rieslings GSE, the
extract of Italian Riesling was used for the further investigation.
100
Italian Riesling Župljanka
90
80
70
AA (%)
60
50
40
30
20
10
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Concentration (µg/mL)
Fig. 2. The antioxidant activities of different concentrations of GSEs of Italian Riesling

and Župljanka on DPPH radical
The IC50 values for the raspberry juice and vitamin C solution were 4.18 and 1.88
mg/mg DPPH, respectively. The raspberry juice with addition of 0.60 µg/mL of GSE
showed the antioxidant activity of 39.2%. The same juice with the threefold concen-
tration of vitamin C (1.81 µg/ml) exhibited similar activity (33.9%) on DPPH radical.
Antioxidant activity of the same amount of juice without any addition of antioxidants was
15.7%.
CONCLUSION
The high flavan-3-ols content and the antioxidant activity on stable DPPH radical of
the GSEs, obtained from white grape varieties, Italian Riesling and Župljanka, were de-
termined. GSE exerts better antioxidant activity on DPPH radical in comparison with jui-
ce containing threefold concentration of vitamin C and juice without the addition of anti-
oxidant. Antioxidant activity of GSE is further confirmed through the addition of GSE to
58
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
fresh raspberry juice, which resulted in stronger scavinging activity of DPPH radical than
the activity of pure juice. Tested white grape varieties could be used as a source for pro-
ducing GSE, like red varieties, which are commercially used for industrial production of
GSE as a dietary supplement and a natural food additive.
ACKNOWLEDGEMENT
These results are part of a research on the project No. TR 23011, financed by the Mi-
nistry of Science and Technological Development of the Republic of Serbia.
REFERENCES
1. Arvanitoyannis, I.S., D. Ladas and A. Mavromatis: Potential Uses and Applications

of Treated Wine Waste: A Review. Int. J. Food Sci. Tech. 41 (2006) 475-487.
2. Pokorny, J.: Introduction, in Antioxidants in Food. Eds. J. Pokorny, N.Yanishlieva,
and M. Gordon, CRC Press and Woodhead Publishing Limited, Cambridge, England
(2001) pp. 1-3.
3. Bajaj, S., A. Urooj and P. Prabhasankar: Effect of Incorporation of Mint on Texture,
Colour and Sensory Parameters of Biscuits. Int. J. Food Prop. 9 (2006) 691-700.
4. Bravo, L.: Polyphenols: Chemistry, Dietary Sources, Metabolism and Nutritional Sig-
nificance. Nutr. Rev. 56 (1998) 317-333.
5. Robards, K.: Strategies for the Determination of Bioactive Phenols in Plants, Fruit
and Vegetables. J. Chromatogr. A. 1000 (2003) 657-691.
6. Moure, A., J.M. Cruz, D. Franco, J.M. Domínguez, J.Sineiro, H. Domínguez, M.J.
Núñez and J.C. Parajó: Natural Antioxidants from Residual Sources. Food Chem. 72
(2001) 145-171.
7. Zin, Z.M., A.A. Hamid, A. Osman, N. Saari and A. Misran: Isolation and Identifica-
tion of Antioxidative Compound from Fruit of Mengkudu (Morinda citrifolia L.). Int.
J. Food Prop. 10 (2007) 363-373.
8. Cano, A. and M.B. Arnao: Hydrophilic and Lipophilic Antioxidant Activity in Diffe-
rent Leaves of Three Lettuce Varieties. Int. J. Food Prop. 8 (2005) 521-528.
9. Bagchi, D., M. Bagchi, S.J. Stohs, D.K. Das, S.D. Ray, C.A. Kuszynski, S.S. Joshi
and H.G. Pruess: Free Radicals and Grape Seed Proanthocyanidin Extract: Impor-
tance in Human Health and Disease Prevention. Toxicology. 148 (2000) 187-197.
10. Gabetta, B., N. Fuzzati, A. Griffini, E. Lolla, R. Pace, T. Ruffilli and F. Peterlongo:
Characterization of Proanthocyanidins from Grape Seeds. Fitoterapia. 71 (2000) 162-
175.
11. Pekić, B., V. Kovač, E. Alonso and E. Revilla: Study of the Extraction of Proantho-
cyanidins from Grape Seeds. Food Chem. 61 (1998) 201-206.
12. Singleton, V.L., R. Orthofer and R.M. Lamuela-Raventos: Analysis of Total Phenols
and other Oxidation Substrates and Antioxidants by Means of Folin-Ciocalteu Rea-
gent. Met. Enzym. 299 (1999) 152-178.
13. Sun, B., J.M. Ricardo-da-Silva and I. Spranger: Critical Factors of Vanillin Assay for
Catechins and Proanthocyanidins. J. Agric. Food Chem. 46 (1998) 4267-4274.
59
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
14. Nakamura, Y., S. Tsuji and Y. Tonogai: Analysis of Proanthocyanidins in Grape Seed
Extracts, Health Foods and Grape Seed Oils. J. Health Sci. 49 (2003) 45-54.
15. Li, W., H. Fong, K. Singletary and J. Fitzloff: Determination of Catechins in Com-
mercial Grape Seed Extract. J. Liq. Chromatogr. R. T. 25 (2002) 397-407.
16. Espin, J.C., C. Soler-Rivas and H.J. Wichers: Characterization of the Total Free
Radical Scavenger Capacity of Vegetable Oils and Oil Fractions Using 2,2-diphenyl-
1-picrylhydrazyl Radical. J. Agric. Food Chem. 48 (2000) 648-656.
17. Escribano-Bailon, M. and C. Santos-Buelga: Polyphenol Extraction from Foods, in
Methods in Polyphenol Analysis. Eds. C. Santos-Buelga and G. Williamson, RSC Pu-
blishing,Cambridge, U.K. (2003) pp. 1-16.
18. Guendez, R., S. Kallithraka, D.P. Makris and P. Kefalas: Determination of Low Mo-
lecular Weight Polyphenolic Constituents in Grape (Vitis vinifera sp.) seed extracts:
Correlation with Antiradical Activity. Food Chem. 89 (2005) 1-9.
19. Kennedy, J.A., G.J. Troup, J.R. Pilbrow, D.R. Hutton, D. Hewitt, C.R. Hunter, R.
Ristić, P.G. Iland and G.P. Jones: Development of Seed Polyphenols in Berries from
Vitis vinifera L. cv. Shiraz. Austral. J. Grape Wine Res. 6 (2000) 244-254.
20. De Freitas, V.A.P. and Y. Glories: Concentration and Compositional Changes of Pro-
cyanidins in Grape Seeds and Skin of White Vitis vinifera Varieties. J. Sci. Food
Agric. 79 (1999) 1601-1606.
21. Revilla, E., J.M. Escalona, E. Alonso and V. Kovač: The phenolic Composition of
Table Grapes, in Food Flavors: Generation, Analysis and Process Influence. Ed. G.
Charalambous, Elsevier Science, Amsterdam (1995) pp. 1579-1596.
22. Bakkalbasi, E., O.Yemi, D. Aslanova and N. Art: Major Flavan-3-ol Composition and
Antioxidant Activity of Seeds from Different Grape Cultivars Grown in Turkey. Eur.
Food Res. Technol. 221 (2005) 792-797.
АНТИОКСИДАТИВНА АКТИВНОСТ ЕКСТРАКАТА СЕМЕНА

БЕЛОГ ГРОЖЂА НА DPPH РАДИКАЛE
Анамарија И. Мандић, Соња М. Ђилас, Јасна М. Чанадановић-Брунет,

Гордана С. Ћетковић и Јелена Ј. Вулић
Састав и антоксидативне особине екстраката семена из сорти црног грожђа су

детаљно описане у литератури, док су екстракти семена сорти белог грожђа мање
испитиване. У раду су приказани резултати одређивања садржаја полифенолних је-
дињења и антиоксидативна активност етилацетатног екстракта семена две сорте
белог грожђа, Италијански ризлинг и Жупљанка, на DPPH радикалe. Утврђен је
утицај додатка екстракта, као антиоксиданта, у сок од малине на исту радикалску
врсту. Садржај укупних полифенолних једињења у екстрактима износио је од 81,6
до 82,8 % (w/w), а садржај флаван-3-ола је био између 66,2 и 91,0 % (w/w). Са-
држаји најзаступљенијих компонената, (+)-катехина (32,17 ± 1,30%) и (-)-епикате-
хина (22,30 ± 0,72%), утврђени су HPLC методом. Сви испитивани екстракти пока-
зали су добру антиоксидативну активност. Вредност IC50 за екстракт семена грожђа
сорте Италијански ризлинг износила је 0,79, а за екстракт сорте Жупљанка 0,95 mg
узорка/mg DPPH. Пошто је екстракт семена грожђа сорте Италијански ризлинг по-
60
APTEFF, 40, 1-220 (2009) UDC: 634.8.076:631.53.01:66.061.34:543.645
DOI: 10.2298/APT0940053M BIBLID: 1450-7188 (2009) 40, 53-61
казао бољу антиоксидативну активност, он је коришћен за даља испитивања. IC50

вредност за сок од малине износила је 4,18 mg узорка/mg DPPH. Сок од малине са
додатком екстракта од 0,60 µg/ml показао је антиоксидативну активност од 39,2%.
Сличну антиоксидативну активност (33,9%) имао је и сок од малине са додатком
витамина Ц у три пута већој концентрацији (1,81 µg/ml). Антиоксидативна актив-
ност исте количине сока без додатка антиоксиданата је била знатно нижа и износи-
ла је 15,7%. Наведени резултати испитивања указују на могућност коришћења екс-
тракта семена сорти белог грожђа као доброг функционалног додатка.

61
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.05
DOI: 10.2298/APT0940063M BIBLID: 1450-7188 (2009) 40, 63-69
TEXTURAL CHARACTERISTICS ОF FERMENTED MILK BEVERAGES

PRODUCED BY KOMBUCHA
Spasenija D. Milanović, Mirela D. Iličić, Katarina G. Duraković and Vladimir R. Vukić
Rheological properties of fermented dairy products are very important parameters of

the product quality. The behaviour of gel formed during fermentation of milk is influen-
ced by a great number of factors, such as: milk composition, starter culture, flavourings
addition, etc. The aim of this research was to examine the influence of fat content, and
kombucha inoculum concentration on textural characteristics of fermented milk bevera-
ges: firmness, consistency, cohesiveness and viscosity index after production and during
10 days of storage. Higher fat content of beverage affects the firmness, consistency, cohe-
siveness and viscosity index, while higher amount of inoculum in beverages has an oppo-
site effect on textural characteristics of samples during storage.
KEY WORDS: Milk, fermentation, kombucha, textural characteristics
INTRODUCTION
Fermented milk beverages are very important in human nutrition due to their high nu-
tritive value and high content of valuable components. The dairy beverages included in
this group of food products differ by kind of milk, fermentation type, consistency, milk
fat content, additives, etc. (1).
Kombucha is a symbiosis of yeast and acetic acid bacteria, which is traditionally cul-
tivated on black tea with sucrose addition. Product of this cultivation is a pleasant, slight-
ly sour and slightly carbonated refreshment beverage. The previous findings showed that
kombucha can be cultivated on different substrates such as black and green tea, beer,
coca-cola, wine, molasses, topinambure extract, herbs and whey (2, 3).
Besides refreshment effect, due to products of metabolitic activity, kombucha beve-
rage has a wide range of prophylactic and therapeutic properties. Kombucha is used to
treat headache, arteriosclerosis, reuma, problems with metabolism and immune system,
burns, skin injuries, etc.
Antibiotic activity of kombucha towards Helicobacter pylori, Escherichia coli, Sta-
phylococcus aureus and Agrobacterium tumefaciens has been proved, mostly due to the
production of acetic acid during fermentation (4-6). Organic acids produced during
Dr. Spasenija D. Milanović, Prof., senadm@uns.ac.rs, Mirela D. Iličić, M.Sc., Katarina G. Duraković B.Sc.,
Vladimir R. Vukić, M.Sc., University of Novi Sad, Faculty of Technology, Bulevar Cara Lazara 1, 21000 Novi
Sad, Serbia
63
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.05
DOI: 10.2298/APT0940063M BIBLID: 1450-7188 (2009) 40, 63-69
fermentation are responsible for most of the characteristics of kombucha. The role of
kombucha in detoxication is probably connected with the ability of glucuronic acid to
bind toxins.
Kombucha fermented beverage is suitable for milk fermentation (7, 8). The physico-
chemical and sensory characteristics of obtained beverage pointed to the posibilities and
necesity of further technological investigations of this group of high-nutritive functional
milk beverages.
Generally, rheological properties of fermented dairy products affect significantly the
quality of the product. The viscosity and gel structure are influenced by a great number of
factors, including milk composition, especially contents of fat and proteins. In the case of
low fat products, behaviour of proteins during the gelation process is of particular im-
portance (9).
The aim of this research was to examine the influence of different milk fat content
and kombucha inoculum concentration on textural characteristics of fermented milk be-
verages: firmness, consistency, cohesiveness and viscosity index after production and du-
ring 10 days of storage.
EXPERIMENTAL
Pasteurized, homogenized milk of 1.0% and 2.2% of fat content („AD IMLEK Beo-
grad - Novosadska mlekara division“ Novi Sad) was used for dietary fermented milk
beverages production in the laboratory conditions.
The following materials were used for fermentation:
1) probiotic starter culture – Delvo-Yog MY-721, „DSM Food Specialites“ Nether-
lands, 0.005%;
2) inoculum (I) – tea fungus cultivated in black tea with addition of sucrose (subs-
trate), as C-atom source was concentrated by microfiltration (using ceramic mem-
brane - pores diameter 200 nm; temperature 25°C, pressure 40 kPa and fluid flow
5 L/min).
Fermented milk beverages were produced from milk with 1.0% and 2.2% fat content
according to the technological process previously described in the literature (10).
Plan of experiment is presented in Table 1.
Table 1. Plan of experiment

No Sample Fat content (%) Inoculum (%)
1 1.0% fat 10% I 1.0 10
2 1.0% fat 15% I 1.0 15
3 2.2% fat 10% I 2.2 10
4 2.2% fat 15% I 2.2 15
Fermented milk beverage samples were analyzed by Texture Analyser TA XP

(Stable Micro System, Godalming, England) through a single compression test, using a
back extrusion cell (A/BE) disc (diameter 35 mm; distance 30 mm; speed 10 mm s-1) and
an extension bar, using 5 kg load cell. Firmness, consistency, cohesiveness and index of
viscosity were measured at 5°C during 10 days of storage .
64
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.05
DOI: 10.2298/APT0940063M BIBLID: 1450-7188 (2009) 40, 63-69
Firmness
The change of firmness in samples obtained from milk of 1.0% and 2.2% of fat by
adding 10% and 15% of kombucha inoculum, after production and during storage is
presented in Fig. 1.
18.00
17.00
16.00
firmness (g)
1.0% fat 10% I

15.00 1.0% fat 15% I
2.2% fat 10% I
2.2% fat 15% I
14.00
13.00
12.00
0 5 10
time (day)
Fig. 1. Firmness of samples after production and during storage

The firmness of sample 1 (1.0% fat 15% I) was the highest (14.85 g). A decrease of
firmness after storage was found in sample 2 (1.0% fat 15% I), and the lowest value after
10 days (13.58 g). The highest increase of firmness during storage was determined in
sample 3 (2.2% fat 10% I). The obtained results show that higher fat content affects the
increase of beverage firmness, while the higher amount of inoculum in beverage decrea-
ses the firmness of samples during storage. It is known that the increase of milk fat
increases the firmness of yoghurt samples (11), while the increase of inoculum amount
results in a decrease of firmness due to dilution, i.e. lower dry matter content in the
sample. The biggest difference in firmness was noticed on the fifth day of storage and it
was by 22.6% higher in sample 3 (2.2% fat 10% I) than in the other samples.
Consistency
Fig. 2 presents the change of consistency of samples produced from milk of 1.0% and
2.2% milk fat by adding 10% and 15% of kombucha inoculum, after production and
during storage for 5 and 10 days.
65
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.05
DOI: 10.2298/APT0940063M BIBLID: 1450-7188 (2009) 40, 63-69
490.00
470.00
450.00
consistency (gs)
430.00 1.0% fat 10% I

1.0% fat 15% I
410.00 2.2% fat 10% I
2.2% fat 15% I
390.00
370.00
350.00
0 5 10
time (day)
Fig. 2. Consistency of samples after production and during storage
The consistency value indicates the density of the product. It is evident that the higher
the value of the consistency, means the higher the product density. The results show the
same trend as in the firmness analysis of fermented milk beverages. The highest con-
sistency value after production (426.34 gs) was found with sample 2 (1.0% fat 15% I).
However, the consistency of this sample decreased during storage, and after 10 days the
consistency value was the lowest (367.02 gs). The highest increase of consistency during
storage was found for sample 3 (2.2% fat 10% I). It can be seen that the obtained value of
consistency for the sample produced from milk with 2.2% of fat and 15% of kombucha
inoculum, increases by 29.0% between 5 and 10 days. Consistency of fermented milk
beverage with 1.0% of fat shows an opposite trend compared to milk with 2.2% of fat. It
is obvious that the increase of milk fat content results in an increase of the consistency
level, while the increase of inoculum amount leads to a decrease of the consistency value
of samples during storage.
Cohesiveness
Fig. 3 shows the change of cohesiveness of fermented milk beverage samples during
10 days storage.
Samples produced from milk of 2.2% fat have greater cohesiveness after production.
Fermented milk beverage sample made with 10% of inoculum showed the highest co-
hesiveness during storage, and the measured values were in the range from -7.72 g after
production to -12.60 g after 10 days of storage. The lowest cohesiveness after production
(-6.07 g) and after 10 days of storage (-9.53 g) was found for sample 2 (1.0% fat 15% I).
The difference of cohesiveness values during storage between samples produced from
milk of 1.0% fat is 23% on average during the storage, and it is significantly lower than
for samples made from milk of 2.2% fat (63%). The obtained results show that higher
66
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.05
DOI: 10.2298/APT0940063M BIBLID: 1450-7188 (2009) 40, 63-69
milk fat content affects the increase of cohesiveness, while the effect of inoculum amount
increase is opposite, i.e. the cohesiveness of produced fermented milk beverage decrea-
ses.
-5.00
-6.00
-7.00
cohesiveness (g)
-8.00 1.0% fat 10% I

1.0% fat 15% I
-9.00 2.2% fat 10% I
2.2% fat 15% I
-10.00
-11.00
-12.00
-13.00
0 5 10
time (day)
Fig. 3. Cohesiveness of samples after production and during storage
Viscosity index
-1.00
-2.00
-3.00
-4.00
index of viscosity (gs)
-5.00 1.0% fat 10% I

1.0% fat 15% I
-6.00
2.2% fat 10% I
-7.00 2.2% fat 15% I
-8.00
-9.00
-10.00
-11.00
0 5 10
time (day)
Fig. 4. Viscosity index of samples after production and during storage
Similarly to cohesiveness, the lower measured value of the viscosity index, means the
higher viscosity index (Fig. 4). The viscosity index after production and after 10 days of
67
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.05
DOI: 10.2298/APT0940063M BIBLID: 1450-7188 (2009) 40, 63-69
storage was measured in sample 3 (2.2% fat 10% I), -2.70 gs and -10.68 gs, respectively.
Samples made from milk of 2.2% milk fat had higher viscosity index after production.
The lowest viscosity index after production (-1.48 gs) and after 10 days storage (-3.96 gs)
was determined in sample 2 (1.0% fat 15% I).
The obtained values are similar to the results of fermented milk beverages obtained
from milk of 0.9% fat using concentrated tea fungus inoculum, 1.5% and 3.0%, regarding
the decrease of textural characteristic values at increased concentrations of inoculum
(10).
CONCLUSION
Samples that contained higher level of fat had much better textural characteristics
from those of fermented milk beverages produced from milk of 1.0% fat content. Signi-
ficant change of textural characteristics was noticed during the first 5 days of storage.
Generally, the sample produced from milk of 2.2% fat with addition of 10% kombucha
inoculum has the best textural characteristics during storage.
ACKNOWLEDGEMENT
This research is part of the project TR-20008 which is financially suported by the
Ministry of Science and Technological Development of the Republic of Serbia.
REFERENCES
1. Tamime, A.Y. and R.K. Robinson: Yoghurt Science and Technology, Woodhead
publishing limited, Cambridge, England, (2004) p. 619.
2. Malbaša, R., S. Milanović, E. Lončar, M. Djurić, M. Carić, M. Iličić and Lj. Kolarov:
Milk-based beverages obtained by Kombucha application. Food Chemistry 112
(2009) 178-184.
3. Malbaša, R., E. Lončar, M. Djurić, and I. Došenović: Effect of sucrose concentration
on kombucha fermentation on molasses. Food Chemistry 108 (2008) 926-932.
4. Sreeramulu, G., Y. Zhu, and W. Knol: Kombucha Fermentation and Its Antimicrobial
Activity. Journal of Agricultural and Food Chemistry 48, 6 (2000) 2589-2594.
5. Greenwalt, C.J., K.H. Steinkraus and R. Ledford: Determination and characterization
of the antimicrobial activity of the fermented tea Kombucha. Lebensmittel
Wissenschaft und Technologie 31 (1998) 291-296.
6. Hartman, A. M., L. E. Burleson, A.K. Holmes and C. Geist: Effects of Chronic Kom-
bucha Ingestion on Open-field Behaviors,Longevity, Appetitive Behaviors, and Or-
gans in C57-BL/6 Mice: A Pilot Study. Nutrition 16 (2000) 755-761.
7. Lončar, E., S. Milanović, M. Carić, R. Malbaša i M. Panić: Metabolička aktivnost
čajne gljive u mleku. Prehrambena industrija – Mleko i mlečni proizvodi 12 (2001)
13-17.
8. Milanović, S., M. Carić, E. Lončar, M. Panić, R. Malbaša and D. Dobrić: Primena
koncentrata čajne gljive u proizvodnji fermentisanih mlečnih napitaka. Prehrambena
industrija – Mleko i mlečni proizvodi 13 (2002) 8-13.
68
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:663.88:637.05
DOI: 10.2298/APT0940063M BIBLID: 1450-7188 (2009) 40, 63-69
9. Iličić, M.: Improvement of probiotic yoghurt characteristics by transglutaminase ap-

plication. ''Second European Workshop on Food Engineering and Technology''.
Massy, 26-27 May 2008, Book of Abstracts p. 19.
10. Duraković. K., S. Milanović, M. Carić, M. Iličić, M. Đurić, M. Tekić i J. Lenđel:
Funkcionalni niskoenergetski fermentisani mlečni napitak proizveden uz primenu
kombuhe, Prehrambena industrija – Mleko i mlečni proizvodi 19 (2008) 66-73.
11. Iličić, M., S. Milanović, M. Carić, M. Đurić, M. Tekić, D. Šašić: Teksturalne karak-
teristike fermentisanih mlečnih proizvoda. Prehrambena industrija – Mleko i mlečni
prozvodi 19 (2008) 79-83.
ТЕКСТУРАЛНЕ ОСОБИНЕ ФЕРМЕНТИСАНИХ МЛЕЧНИХ НАПИТАКА

ДОБИЈЕНИХ ПРИМЕНОМ КОМБУХЕ
Спасенија Д. Милановић, Мирела Д. Иличић, Катарина Г. Дураковић и

Владимир Р. Вукић
Реолошке особине ферментисаних млечних производа представљају веома ва-

жан фактор за квалитет производа. Својства гела добијеног ферментацијом млека
зависе од различитих фактора, као што су: састав млека, стартер култура, аромe,
итд. У раду је испитан утицај садржаја млечне масти (1,0% и 2,2%), и различитих
концентрација инокулума комбухе (10% и 15%) на текстуралне особине фермен-
тисаних млечних производа: чврстоћу, конзистенцију, кохезивност и индекс виско-
зитета, након производње и током 10 дана складиштења. Добијени резултати по-
казују да повећање садржаја млечне масти доводи до повећања чврстоће, конзи-
стенције, кохезивности и индекса вискозитета узорака. Ферментисани млечни на-
пици произведени са већом концентрацијом инокулума комбухе имају ниже вред-
ности текстуралних карактеристика. Највеће промене текстуралних карактеристика
у узорцима уочавају се током првих 5 дана складиштења.

Accepted 1 October 2009
69
APTEFF, 40, 1-220 (2009) UDC:633.34:66.040.2:543.645
DOI: 10.2298/APT0940071P BIBLID: 1450-7188 (2009) 40, 71-77
PROTOCOL FOR USING PROTEIN SOLUBILITY AS AN INDICATOR OF

FULL-FAT SOYBEAN HEAT TREATMENT
Dragan V. Palić and Sophia E. Coetzee
When the degree of full-fat soybean (FFSB) processing is determined using protein
solubility as an indicator of heat treatment extent, a problem represents the lack of a
standard with known value of protein solubility, against which the protein solubility of
heat treated FFSB would be determined. Also, a special practical problem imposes the
fact that universal ranges of units for describing the degree of FFSB processing are used
globally, without taking into consideration specific regional differences. In this paper, a
protocol was proposed for establishing unit ranges for defining under-, adequately- and
over-processed FFSB when protein solubility is used as an indicator of the extent of heat
treatment.
KEY WORDS: Full-fat soybeans, degree of processing, extent of heat treatment, protein
solubility
INTRODUCTION
The use of full-fat soybeans (FFSB) in animal feeds has been limited because of the
uncertainty of the exact availability of the amino acids. This arises due to both the
presence of biologically active substances with an anti-nutrient action, which are con-
tained in raw soybean, as well as the effect that processing has on the availability of the
amino acids contained therein. Processing of the raw FFSB by means of heat destroys the
anti-nutrients, thus making them fit for use in monogastric diets. The problem relating to
the availability of the amino acids in the heat-treated soybeans arises due to the fact that
only an optimum level of heat treatment will produce maximal availability of the amino
acids to the animal. Both under- and over-processing result in decreased availability of
amino acids (4).
Amongst other authors, Holmes (1), Ruiz et al. (2) and Zarkadas et al. (3) showed that
moderate heating is necessary to increase the digestibility of soybean protein for non-
ruminants. A comparatively mild heating leads to partial protein degradation (denatura-
tion of tertiary and quaternary structures), allowing more effective penetration by diges-
tion enzymes.
One of the major concerns is: what happens when FFSB is under- or over-processed?
Is the one more detrimental than the other? To define under- and over-processing is easy
Dr. Dragan Palić, dragan.palic@fins.uns.ac.rs, Institute for Food Technology, Bulevar Cara Lazara 1, 21000
Novi Sad, Serbia; Sophia E. Coetzee, M. Sc., Agricultural Research Council, ARC-Animal Production Institute,
Private Bag X2, Irene 0062, South Africa
71
APTEFF, 40, 1-220 (2009) UDC:633.34:66.040.2:543.645
DOI: 10.2298/APT0940071P BIBLID: 1450-7188 (2009) 40, 71-77
in theoretical terms, but is it easy to define it in practice? The following mechanisms are
involved in under- and over-processing:
Under-processing: Residual trypsin inhibitor mediates its effects via the digestive
processes, affecting both endogenous and exogenous amino acid losses. It also binds and
inactivates the pancreatic enzyme trypsin (4). The result is that protein digestibility is re-
duced and swelling of the pancreas occurs, caused by the production of additional trypsin
and chymotrypsin.
Over-processing: In this process, proteins are more than partially denaturated and
amino acid availability is reduced. This is because the Maillard reaction takes place, i.e.
reducing sugars react with the epsilon-amino group of lysine (4).
As a consequence, the objective of heating processes for full-fat soybeans, intended
for inclusion in diets for poultry and pigs, is to maintain optimum balance between de-
gradation of anti-nutrients on the one hand and maintenance of protein digestibility on the
other.
Commonly used methods for assessing the processed FFSB quality are those for the
determination of:
1. Urease Activity Index (UAI)

2. Trypsin Inhibitor Activity (TIA)
3. Protein Solubility in KOH (PSKOH)
4. Nitrogen Solubility Index (NSI)
5. Protein Dispersibility Index (PDI)
6. Lysine availability
In a critical assessments of methods, Palic et al. (5, 6), established that some of com-
monly used methods, e.g. UAI, have limitations, as they can be used only to determine
the under-processed FFSB, that some of the methods are very complicated to perform,
such as TIA and Lysine availability, and concluded that protein solubility is the best
indicator for FFSB quality control and that therefore PSKOH, NSI and PDI methods
would be the prefferd choice. A problem represents the fact that there is a lack of a stan-
dard with known protein solubility, against which the protein solubility of processed
FFSB would be determined.
Special practical problem imposes the fact that universal ranges of units for PSKOH,
PDI and NSI methods, for describing the under-, adequately- and over-processed FFSB,
are used globally, without taking into consideration specific regional differences (7).
The aim of this study was to develop a protocol for establishing unit ranges for defi-
ning under-, well- and over processed FFSB, when protein solubility is used as an indica-
tor of the extent of heat treatment.
EXPERIMENTAL
In the absence of a standard with known value of protein solubility, the solution is to
use an “indirect” standard, which can be obtained through in vivo trial with animals.
Therefore, the proposed protocol for establishing the degree of FFSB processing,
when protein solubility is used as an indicator of heat treatment extent, consists of the
following steps.
72
APTEFF, 40, 1-220 (2009) UDC:633.34:66.040.2:543.645
DOI: 10.2298/APT0940071P BIBLID: 1450-7188 (2009) 40, 71-77
Step 1. Raw soybeans processing
A number, but not less than five, samples of FFSB processed at different temperatu-
res, i.e. exposed to different extents of heat, is produced. In this study, raw soybeans, with
moisture of 10-11%, were processed by dry extrusion, using industrial „Insta-Pro 2000R“
single screw extruder at 8 temperatures: 110, 120, 127, 136, 140, 145, 151 and 165°C,
with the processing time ranging between 30 and 40 seconds.
Step 2. In vivo trial

Samples of FFSB produced in Step 1 are fed to animals and their performance is mo-
nitored. In the work presented, a total of 384 male Ross broilers were randomly allocated
to 48 pens, each containing 8 birds. On arrival, all broilers were sorted into equal weight
groups, and assigned at random to the different treatment pens, such that initial average
weight and weight distribution were similar for all pens. They were allocated to one of
eight dietary treatments containing the processed FFSB. The average body weight gain
(ADWG), in the period from 0 to 14 days of age, and feed conversion ratio (FCR), on day
14, were monitored as production parameters.
Step 3. Choice of laboratory method

A laboratory method for determining protein solubility is chosen. In this study, as a
model-method for determining protein solubility as an indicator of the extent of soybean
processing, the Protein solubility in potassium hydroxide (PSKOH), as described by Palic
(8), has been chosen. Eight samples of FFSB processed in Step 1 were analysed for
PSKOH in duplicates by five laboratories.
Step 4. Establishing ranges of units for chosen method for describing the degree
of FFSB processing
The ranges of units for chosen method, for describing under-, adequately- and over-
processed FFSB, are established. An illustration of the relation between in vivo animal
performance, measured by average daily weight gain (g), and the PSKOH values (in %)
is shown in Figure 1. Assuming that the FFSB samples at 135oC and 145oC are adequa-
tely processed, the values X and Y of PSKOH (%) for those two samples, represent the
border-points of the range of adequately-processed FFSB. Consequently, unknown FFSB
sample, whose PSKOH protein solubility value falls between X and Y points, would be
assessed as adequately-processed.
73
APTEFF, 40, 1-220 (2009) UDC:633.34:66.040.2:543.645
DOI: 10.2298/APT0940071P BIBLID: 1450-7188 (2009) 40, 71-77
120 100
115
90
Average daily weight gain (g)
110
80 X
105
Y
100 70
KOH (%)
95
60
90
50
85
80 40
115 125 135 145 165
Temperatures (°C)
AWG KOH
Fig 1. Illustration of the relation between in vivo animal performance, measured by

average daily weight gain (g), and the KOH protein solubility (%) for FFSB samples
processed at different temperatures
Statistical analysis. Data were analyzed using the statistical program SAS/STAT (9).
The experiment was designed as a randomized complete block with five replicates per
treatment. Analysis of variance (ANOVA) was used to test for differences between treat-
ments. Treatment means were separated using Fishers' protected t-test least significant
difference (LSD) at the 5 % level of significance.
The results of the in vivo trial are shown in Table 1 and Figure 2.
Table 1. Average body weight gain (ABWG) and feed conversion ratio (FCR) for
chickens fed FFSB processed at different temperatures in the period
from 0 to 14 days of age
Temperature (oC) ABWG (g) FCR (kg/kg)

110°C 87.8bc 2.081d
120°C 96.0bc 1.893cd
127°C 108.0bc 1.768c
136°C 138.3a 1.382a
140°C 132.0a 1.466a
145°C 123.0a 1.529a
151°C 97.2b 1.679c
164°C 79.8c 1.891cd
SEM1 7.94 0.081
LSD2 22.81 0.232
CV%3 19.1 11.5
a,b,c,d
Values in the same column with different superscript differ significantly (P<0.05);
1
SEM = Standard error of the means; 2LSD = Least significant difference; 3CV% = Coefficient of variation
74
APTEFF, 40, 1-220 (2009) UDC:633.34:66.040.2:543.645
DOI: 10.2298/APT0940071P BIBLID: 1450-7188 (2009) 40, 71-77
150 2.2
140
2.0
130
Feed conversion ratio (kg/kg)

120
1.8
Body weight gain (g)
110
100 1.6
90
1.4
80
Well prosessed FFSB range
70
1.2
60
50 1.0
110 115 120 125 130 135 140 145 150 155 160 165 170
Temperature (ºC)
In-vivo body weight gain In-vivo FCR
Fig. 2. Average daily body weight gain in the period from 0 to 14 days of age, and feed
conversion ratio on day 14, of broiler chickens fed FFSB processed at different
temperatures
Statistical analysis of the results showed that the best performance was achieved by
chickens that were fed the FFSB processed at 136oC, 140oC and 145oC and that there was
no significant difference between them (P>0.05). However, the difference between the
groups that received the FFSB processed at 1270C and 136oC, as well as at 145oC and
151oC, were significant (P<0.05). Based on these parameters, a relation between the tem-
perature of extruding and the in vivo assessment of the degree of FFSB processing has
been derived and is shown in Table 2.
Table 2. Relation between the temperature of extruding and the in vivo assessment of the
degree of FFSB processing
Degree of FFSB processing Temperature of extrusion (oC)

Under-processed < 136
Adequately - processed 136 – 145
Over-processed > 145
The results of the protein solubility in potassium hydroxide (PSKOH) are shown in
Table 3.
75
APTEFF, 40, 1-220 (2009) UDC:633.34:66.040.2:543.645
DOI: 10.2298/APT0940071P BIBLID: 1450-7188 (2009) 40, 71-77
Table 3. Results of the determination of protein solubility in potassium hydroxide
(PSKOH) in FFSB samples processed by dry extrusion at different temperatures
Temperature (oC) PSKOH (%)1

110°C 90.45
120°C 89.21
127°C 86.87
136°C 76.51
140°C 73.87
145°C 67.14
151°C 67.99
164°C 61.05
1
Mean values of the results obtained in five laboratories
FFBS samples processed at temperatures between 136oC and 145oC, represented ade-
quately-processed FFSB (Table 1). The mean values for PSKOH for these two samples,
obtained at five laboratories (Table 3), were 76.51 % and 67.14%, or for the practical ap-
plication, 77 % and 67% respectively. Therefore, the following ranges, shown in Table 4,
for describing the degree of FFSB processing when PSKOH method is used, have been
established.
Table 4. Ranges for describing the degree of FFSB processing using PSKOH method
Degree of FFSB processing PSKOH (%)

Under-processed >77%
Adequately- processed 67% - 77%
Over-processed <67%
CONCLUSION
Using the protocol described in this study, the ranges for any laboratory method
which uses protein solubility as an indicator of the extent of FFSB heat treatment, can be
established. The numerical value of the units for the method(s) established by using the
proposed protocol, take into consideration regional differences such as soybean quality
and may be safely applied for FFSB quality control, regardless of what the globally ac-
cepted unit ranges for the specific method(s) are.
ACKNOWLEGMENT
This study has been supported by the Protein Research Foundation of South Africa.
REFERENCES
1. Holmes, B: Quality control of raw materials and final products in fullfat soybean
production. Proc. of Fullfat Soybean Regional Conference, Milan, 14-15 April 1987,
Book of Abstracts, p. 246.
76
APTEFF, 40, 1-220 (2009) UDC:633.34:66.040.2:543.645
DOI: 10.2298/APT0940071P BIBLID: 1450-7188 (2009) 40, 71-77
2. Ruiz, N., F. de Belalcazar and G. J. Diaz: Quality Control Parameters for Commercial
Full-Fat Soybeans Processed by Two Different Methods and Fed to Broilers. J. Appl.
Poult. Res. 13 (2004) 443-450.
3. Zarkadas, L. N. and J. Wiseman: Influence of processing of full fat soya beans inclu-
ded in diet for piglets. 1. Performance. Animal Feed Science and Technology 118
(2005) 109-119.
4. Monary, S.: Fullfat Soya Handbook, American Soybean Association, Brussels (1989)
p.6.
5. Palic, D., K. Moloto, E. S. Coetzee and O. Djuragic, O: Critical assessment of labora-
tory methods for full-fat soybean quality control. 1st International Congress on Food
Technology, Quality and Safety, Novi Sad, 13-15 November 2007, Proceedings p. 197.
6. Palic, D., J. Levic, S. Sredanovic, O. Djuragic: Quality control of full-fat soybeans
using urease activity: critical assessment of the method. Acta Periodica Technologica,
39 (2008) 47-53.
7. Palic, D: Quality control of processed full-fat soybeans: Choice of method. XI Inter-
national Feed Technology Symposium, Vrnjacka Banja, 30 March–3 June 2005, Pro-
ceedings p. 96.
8. Palic, D: Quality control of processed full-fat soybeans using protein solubility in
KOH: Critical review and modification of the method. 11th International Feed Tech-
nology Symposium, Vrnjacka Banja, 30 March – 3 June 2005, Proceedings p.106.
9. SAS/STAT User's Guide, Version 8, SAS Institute Inc., Cary, NC:SAS Institute
(1999).
ПОСТУПАК ЗА КОРИШЋЕЊЕ РАСТВОРЉИВОСТИ ПРОТЕИНА КАО

ИНДИКАТОРА ТЕРМИЧКОГ ТРЕТМАНА ПУНОМАСНЕ СОЈЕ
Драган В. Палић и Sophia E. Coetzee
Када се растворљивост протеина користи као индикатор степена термичке об-
раде пуномасне соје, проблем представља недостатак стандарда у односу на који би
се одређивала растворљивост протеина обрађеног сојиног зрна. Посебан практичан
проблем представља чињеница да се за изражавање различитих степена термичког
третмана, примењује опсег јединица које су глобално прихваћене, не водећи при
томе рачуна о регионалним разликама у квалитету сирове соје. У овом раду је
предложен поступак за утврђивање опсега јединица за дефинисање недовољно,
адекватно и сувише термички обрађене пуномасне соје, када се растворљивост
протеина користи као индикатор степена термичког третмана.

77
APTEFF, 40, 1-220 (2009) UDC:636.085.1/.3:636.085.55:636.2+636.3
DOI: 10.2298/APT0940079P BIBLID: 1450-7188 (2009) 40, 79-86
COMPARISON OF THREE IN VITRO METHODS FOR DETERMINING AND

PREDICTING THE ORGANIC MATTER DIGESTIBILITY OF COMPLETE
DIETS FOR RUMINANTS
Dragan V. Palić and Klaas-Jan Leeuw
In this study, the organic matter digestibility (OMD) of six complete diets for rumi-
nants has been determined in-vivo in trials with sheep and in-vitro using two-stage Tilley
and Terry (T&T) method, gas production (GP) technique and multi-enzyme incubation
(EDOM) procedures. The mean OMD values obtained in vivo and using T&T, GP and
EDOM techniques were 684, 716, 685 and 710 g OM/kgDM respectively and did not
differ significantly (P>0.05). The obtained in vitro results were regressed against deter-
mined in-vivo values to derive prediction equations. Using the T&T technique, the pre-
diction equation OMD (in_vivo) = -17.36 + 0.98 x OMD (in_vitro_T&T), (R2 = 0.75;
RMSE = 37.59) has been obtained. The equation OMD (in_vivo) = 198.98 + 0.71 x OMD
(in_vitro_GP), (R2 = 0.21; RMSE = 66.36) has been derived for Gas production procedure,
while the equation OMD (in_vivo) = 102 + 0.82 x OMD (in_vitro_EDOM), (R2 = 0.86;
RMSE = 27.30) has been generated for multi-enzyme incubation technique. The results of
this study showed that the OMD of complete diets for ruminants can be successfully
determined, and in-vivo values predicted, using multi-enzyme incubation procedure,
which is important because of the fact that rumen liquor, needed for the in-vitro two-
stage T&T and GP techniques is not always available to analytical laboratories.
KEY WORDS: Ruminants, complete diets, organic matter digestibility, in vitro

techniques, prediction
INTRODUCTION
Energy value of ruminant feeds and its bio-availability is of great importance for
animal feed manufactures and end users. The amount of available energy in feeds for ru-
minants is described either by its metabolisable energy or by organic matter digestibility
(1), since the value of organic matter digestibility is very close to the corresponding di-
gestibility of energy (2). The most accurate way of obtaining information on digestibility
of organic matter of feeds for ruminants is by conducting in vivo digestibility experi-
Dr. Dragan Palić, dragan.palic@fins.uns.ac.rs, Institute for Food Technology, Bulevar Cara Lazara 1, 21000
Novi Sad, Serbia; Klaas-Jan Leeuw, M. Sc., Agricultural Research Council, ARC-Animal Production Institute,
Private Bag X2, Irene 0062, South Africa
79
APTEFF, 40, 1-220 (2009) UDC:636.085.1/.3:636.085.55:636.2+636.3
DOI: 10.2298/APT0940079P BIBLID: 1450-7188 (2009) 40, 79-86
ments. Since this method is expensive and time consuming, laboratory methods for
routine prediction of the in vivo organic matter digestibility of ruminant feeds are an
irreplaceable tool for routine quality control in the feed industry.
In vitro methods for determining organic matter digestibility (OMD) by the use of ru-
men liquor fermentation techniques have become well established despite the limitations
on the use of rumen liquor for digestibility studies and the request for fistulated animals
(which are not available to all laboratories) for the collection of fresh rumen liquor. Wi-
dely accepted in vitro rumen liquor fermentation procedures for determining the organic
matter digestibility of ruminant feeds are the two-stage Tilley and Terry (T&T) method
(3) and Gas Production (GP) technique developed by Menke and Steingass (4).
The alternative to rumen liquor is the use of incubation of feeds with exogenous
enzymes, which has the aim to mimic the digestive processes in the animal. Enzymes can
break down different parts of the plant constituents, which can be divided into those that
make up the structure of the plant (cell-wall constituents) and the material within the cells
(cell-content constituents). Cell content is essentially completely digestible in vivo, whe-
reas cell-wall constituents vary in digestibility (5). Enzymes therefore need to remove the
cell contents and to solubilise unlignified and moderately lignified cell-wall to a signifi-
cant extent. Most enzymatic methods for organic matter digestibility determination have
been developed for forage feedstuffs, with a few used for compound feeds (6). Hvelplund
et al. (7) used a multi-enzymatic incubation method for estimating the enzymatic digesti-
bility of organic matter (EDOM) of straws. Palic and Muller (8) demonstrated the ability
of this method to determine the OMD of a variety of other feedstuffs for ruminants. Apart
from the above-mentioned two references, more citations of this method in the literature
have not been found.
The aim of this study was to compare the two rumen-fluid-based methods, i.e. the two-
stage in vitro method (3) and gas production (4) with the under-utilized, multi-enzyme in-
cubation procedure of Hvelplund et al., (7) for determining the OMD of compound feeds for
ruminants and to develop equations for predicting the in vivo OMD of complete diets for
ruminants using results obtained by in vitro T&T, GP and EDOM techniques.
EXPERIMENTAL
Six complete diets for ruminants were used in this study. The chemical composition of
the diets was determined by the Official Methods of AOAC International (9).
The OMD of investigated diets was determined in the in vivo trial, as well as by three in
vitro methods. In vitro procedures used for OMD determination were:
Two-stage method (3). This method has two main stages. In the first, 0.5 g of dried
sample is incubated anaerobically with rumen liquor inoculum in a buffered sollution, for
48 hours at 38oC in the dark. The second stage involves digestion with pepsin-HCl for 48
hours at 38oC. The OMD is calculated as the difference between the organic matter in the
original sample and in the residue.
Gas Production technique (4). An amount of 200 mg of dried sample is introduced
into a special piston-syringe, after which rumen liquor is added. The production of gas is
measured during a 24-hour incubation at 39oC and the OMD is calculated from the fol-
lowing equation:
80
APTEFF, 40, 1-220 (2009) UDC:636.085.1/.3:636.085.55:636.2+636.3
DOI: 10.2298/APT0940079P BIBLID: 1450-7188 (2009) 40, 79-86
OMD (%) = 14.88 + 0.889 x gp + 0.45 x CP + 0.65 x CA

where:
gp = Gas produced (ml)
CP = Crude protein (%)
CA = Crude ash (%)
EDOM procedure (7). About 0.5 g of sample is incubated with pepsin-HCl solution
for 24 hours at 40oC to dissolve protein, followed by incubation with a buffered enzyme
solution (consisting of cellulase, cellobiase, hemicellulase and amyloglucosidase) first at
40oC for 19 hours and then at 60oC for 19 hours. The residue was dried and ashed, and
the insoluble organic matter in the sample was determined as the difference.
In vivo organic matter digestibility was determined according to Steg et al. (10) using
four castrated adult male sheep per diet. Water was freely available at all times, while the
feed was standardized at 1000 g of dry matter daily for each animal. An 11-day adap-
tation period during which the animals were fed the trial ration was followed by a 10-day
collection period during which the exact amounts of feed, residues and the faecal pro-
duction were recorded. The organic matter contents of both feed and faeces were measu-
red and their difference represened the digestible organic matter.
Data was analysed using the statistical programme GenStat (11).
The chemical composition of studied diets is shown in Table 1.
Table 1. Chemical composition (%) of complete diets used in the study
Diet DM (%) CP (%) CF (%) CA (%)

1 89.39 9.44 1.45 6.37
2 85.35 13.99 1.09 7.00
3 86.24 13.45 1.29 8.69
4 87.45 14.52 2.69 6.72
5 86.07 21.33 1.07 7.20
6 87.04 9.40 1.34 8.08
DM = Dry matter
CP = Crude protein
CF = Crude fibre
CA = Crude ash
The organic matter digestibility of complete diets for ruminants determined in in vivo
trial and by three in vitro procedures is shown in Table 2.
81
APTEFF, 40, 1-220 (2009) UDC:636.085.1/.3:636.085.55:636.2+636.3
DOI: 10.2298/APT0940079P BIBLID: 1450-7188 (2009) 40, 79-86
Table 2. The OMD of complete diets for ruminants (g OM/kg DM) determined in vivo
and by three in vitro methods
Diet Organic matter digestibility (g OM/kg DM)1

In vivo T & T2 GP3 EDOM4
1 595 625 665 622
2 716 756 686 781
3 674 724 618 726
4 710 776 708 708
5 781 749 609 801
6 628 664 599 624
Mean 684a 716a 685a 710a
SD5 66.7 58.9 43.1 75.8
1
Means of three replicates
2
Two-stage in vitro method (3)
3
Gas production method (4)
4
Multi-enzyme incubation method (7)
5
SD = Standard deviation
a
Means with same subscript in a row do not differ significantly (P>0.05)
The mean values obtained by in-vivo, T&T, GP and EDOM procedures were 684, 716,
685 and 710 g OM/kg DM respectively, and did not differ significantly (P>0.05).
The values obtained by laboratory procedures were then regressed against determined
in vivo values and the functions for each in vitro procedure for predicting the in vivo
OMD of complete diets have been derived.
The following equation, shown also in Figure 1, to predict the in vivo OMD from the
results of in vitro T&T method has been obtained:
OMD (in_vivo) = -17.36 + 0.98 x OMD (in_vitro_T&T),

R2 = 0.75; RMSE = 37.59
The regression of the OMD results obtained by Gas production method against the in
vivo values, resulted in the following equation, shown also on Figure 2.
OMD (in_vivo) = 198.98 + 0.71 x OMD (in_vitro_GP)

R2 = 0.21; RMSE = 66.36
The equation, shown also on Figure 2, for predicting the in vivo OMD using Multi-
enzyme incubation (EDOM) method, was as follows:
OMD (in_vivo) = 102 + 0.82 x OMD (in_vitro_EDOM)

R2 = 0.86; RMSE = 27.30
82
APTEFF, 40, 1-220 (2009) UDC:636.085.1/.3:636.085.55:636.2+636.3
DOI: 10.2298/APT0940079P BIBLID: 1450-7188 (2009) 40, 79-86
Fig. 1. Relationship between the OMD values of compound feeds for ruminants
determined in vivo and by in vitro T&T method
Fig. 2. Relationship between the OMD values of complete diets for ruminants determined
in vivo and by in vitro GP method
83
APTEFF, 40, 1-220 (2009) UDC:636.085.1/.3:636.085.55:636.2+636.3
DOI: 10.2298/APT0940079P BIBLID: 1450-7188 (2009) 40, 79-86
Fig. 3. Relationship between the OMD values of complete diets for ruminants determined in
vivo and by in vitro EDOM method
The organic matter digestibility of complete diets for ruminants in this study was pre-
dicted best by EDOM method, closely followed by the two-stage T&T procedure.
A low correlation between Gas production and in vivo results (R2 = 0.21) was some-
how unexpected, since Palic and Muller (8), investigating the OMD of feedstuffs for ru-
minants, established an R2 = 0.81 for the relationship between the OMD values determined
in vivo and by in vitro Gas production method.
The prediction of OMD of compound feeds for ruminants by the use of enzymes have
been up to date applied mostly to forages. Enzymatic methods have been much less
studied with energy and protein feeds (6), and in those seldom cases, the authors used
single-enzyme incubations. Dowman and Collins (12), using incubation with pepsin-HCl,
followed by the treatment of 12 complete diets with cellulase, reported a correlation of R2
= 0.87 between in vivo and in vitro values, whereas Aufrere and Michalet-Doreau (6)
found an R2 = 0.90 using 24 energy-rich feeds. The procedure of Hvelplund et al. (7)
used in this study, might have an advantage over the above-mentioned, as it uses multi-
enzyme incubation and therefore may better mimic the digestion in the animal gastro-
intestinal tract.
CONCLUSION
Although conducted on a small number of samples, the results of this study yielded a
clear reference point and showed that the organic matter digestibility (OMD) of complete
diets for ruminants can be successfully determined, and the in vivo OMD successfully
predicted, using a multi-enzyme incubation procedure. This is important because of the fact
that rumen liquor, needed for the in vitro Tilley and Terry and Gas Production techni-
ques, is not always available to analytical laboratories. Further work, with inclusion of
more samples of complete diets, is needed to confirm the results of this study.
84
APTEFF, 40, 1-220 (2009) UDC:636.085.1/.3:636.085.55:636.2+636.3
DOI: 10.2298/APT0940079P BIBLID: 1450-7188 (2009) 40, 79-86
REFERENCES
1. Barber, G.D., D.I. Givens, M.S. Kridis, N.W. Offer and I. Murray: Prediction of the
organic matter digestibility of grass silage. Anim. Feed Sci. Tech. 28 (1990) 115-128.
2. Thomas, P.C. Predicting the nutritive value of compound feeds for ruminants. In:
Feedstuffs Evaluation. Eds. W. Haresign and D.J.A. Cole, Butterworths, London
(1990) p. 301
3. Tilley, J.M.A. and R.A. Terry: A two stage method for the in vitro digestion of forage
crops. J. Brit. Grassl. Soc. 18 (1963) 104-111.
4. Menke, K.H.and H. Steingass: Estimation of the energetic feed value obtained from
chemical analysis and in vitro gas production using rumen fluid. Anim. Res. Dev. 28
(1988) 7-55.
5. Jones, D.I.H.and M. K. Theodorou: Enzyme techniques for estimating digestibility.
In: Forage Evaluation in Ruminant Nutrition. Eds. D. I. Givens, E. Owen, R. F. E.
Axford and H. M. Omed, CABI publishing, New York (2000) p. 155.
6. Aufrere, J. and B. Michalet-Doreau, B: Comparison of methods for predicting
digestibility of feeds. Anim. Feed Sci. Tech. 20 (1988) 203-218.
7. Hvelplund, T., M. R. Weisberg and K. Soegaard: Use of in vitro digestibility methods
to estimate in vivo digestibility of straws. TSAP Science Conference, Arusha, 3-4
Aug 1990, Proceedings vol. 26, p. 70.
8. Palic, D., H. Muller: Prediction of the in vivo organic matter digestibility of feed-
stuffs for ruminants using in vitro techniques. Savremena Poljoprivreda 55, 1–2
(2006) 127-132.
9. AOAC INTERNATIONAL, Official Methods of Analysis, 17th Edition, AOAC
INTERNATIONAL, Gaithersburg, MD 20877-2417, USA (2000).
10. Steg, A., J. M. van der Merwe, B. Smits and V. A. Hindle: Prediction of the digesti-
bility of feedstuffs: recent developments. Annual report, ID-DLO, The Netherlands
(1987).
11. Payne, R.W., D.A. Murray, S.A. Harding, D.B. Baird and D. M. Soutar: GenStat for
Windows (10th Edition), Introduction. VSN International, Hemel Hempstead, UK
(2007).
12. Dowman, M. G., F. C. Collins: The use of enzymes to predict the digestibility of
animal feeds. J. Sci. Food Agric 33 (1982) 689-696.
ПОРЕЂЕЊЕ ТРИ IN VITRO МЕТОДЕ ЗА ОДРЕЂИВАЊЕ И ПРОЦЕНУ

СВАРЉИВОСТИ ОРГАНСКЕ МАТЕРИЈЕ У ПОТПУНИМ СМЕШАМА ЗА
ПРЕЖИВАРЕ
Драган Палић и Klaas-Jan Leeuw
У овом раду је одређена сварљивост органске материје (OMD) у шест потпуних

смеша за преживаре и то in vivo, у огледима на овцама, и in vitro, коришћењем
двостепене Tilley и Terry (Т&Т) (3) методе, технике мерења продукције гаса (GP)
(4) и поcтупка мулти-ензимске инкубације (EDOM) (7). Средње вредности сварљи-
вости органске материје добијене in vivo и коришћењем Т&Т, GP и EDOM in vitro
85
APTEFF, 40, 1-220 (2009) UDC:636.085.1/.3:636.085.55:636.2+636.3
DOI: 10.2298/APT0940079P BIBLID: 1450-7188 (2009) 40, 79-86
метода износиле су 684, 716, 685 и 710 g органске материје/кг суве материје и нису се
значајно разликовале (P>0,05). Регресионом анализом добијене су једначине за пред-
виђање in vivo OMD на основу резултата добијених in vitro методама. За Т&Т ме-
тоду добијена је једначина OMD (in_vivo) = -17,36 + 0,98 x OMD (in_vitro_T&T), (R2=
0,75; RMSE=37,59, за GP технику OMD (in_vivo)=198,98 + 0,71 x OMD (in_vitro_GP),
(R2=0,21; RMSE=66,36), док је за EDOM методу изведена једначина OMD (in_vivo)=
102 + 0,82 x OMD (in_vitro_EDOM), (R2=0,86; RMSE = 27,30). Резултати овога рада
показују да се сварљивист органске материје потпуних смешa за преживаре може
успешно одредити, и њене in vivo вредности успешно предвидети, коришћењем
мулти-ензимске инкубационе методе, што је веома важно с обзиром на чињеницу
да буражни сок, неопходан за Т&Т и GP методе, није увек доступан аналитичким
лабораторијама.

Accepted 28 Septembar 2009
86
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:637.04/.05
DOI: 10.2298/APT0940087P BIBLID: 1450-7188 (2009) 40, 87-94
COMPOSITIONAL CHARACTERISTICS OF COMMERCIAL YOGHURT

BASED ON QUANTITATIVE DETERMINATION OF VIABLE LACTIC ACID
BACTERIA
Dragana Pešić-Mikulec and Gordana B. Niketić
Yoghurt quality is particularly difficult to standardize because of the many forms,

varieties, manufacturing methods, ingredients and consumer preferences that exist. Since
these factors will always play an important role, it is unlikely that a uniform yoghurt qua-
lity concept will ever emerge, such as has been developed for other dairy products. There
are a number of common denominators, however that have bearing on yoghurt quality.
Since a number of producers are recognized within the broad category entitled yoghurt.
This situation makes yoghurt an interesting, challenging, but also a confusing area to
work in. The present investigation was undertaken to isolate from commercial yoghurt
the strains involved in its manufacture and determine the characteristics of Streptococcus
thermophilus and Lactobacillus delbrueckii subsp.bulgaricus. This study is concerned
with the lactic acid bacteria (L.delbrueckii subsp. bulgaricus and S. thermophilus)
growth in yoghurt from involving different procedures and with the determination of the
number of lactic acid bacteria in dependence of the temperature and acidity in the
period of storage. Predominant samples of yoghurt were with 11-107/ml lactic acid
lactococci (44.28%).
KEY WORDS: LAB, yoghurt, viable lactic acid bacteria, probiotics
INTRODUCTION
One of the first records of yoghurt consumption comes from the Middle East during
the times of the Conqueror Genghis Khan in the 13th century, whose armies were
sustained by this healthful food. Yoghurt and other fermented dairy products have long
been a staple in the diets of cultures of the Middle East, Asia, Russia and Eastern
European countries, such as Bulgaria. Yet, the recognition of yoghurts special health be-
nefits did not become apparent in Western Europe and North America until the 20th
century, as a result of research done by Dr. Elie Metchnikoff. Dr. Metchnikoff (12, 13) is
the first researcher who proposed fermented dairy products with beneficial properties.
Conducted research on the health benefits of lactic acid-producing bacteria and postu-
Dr. Dragana Pešić Mikulec, dpesic@sbb.rs, Senior research fellow, Department of Food Microbiology, Institute
of Veterinary Research, Autoput 3, 11050 Novi Beograd, Serbia; Dr. Gordana B. Niketić, Research fellow, JPS
Dairy Institute, Autoput 3, 11050 Novi Beograd, Serbia
87
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:637.04/.05
DOI: 10.2298/APT0940087P BIBLID: 1450-7188 (2009) 40, 87-94
lated that the longevity of peoples of certain cultures, such as the Bulgarians, was related
to their high consumption of yoghurt and fermented dairy products.
The benefits of yoghurt depends for „live active cultures“ or „living yoghurt cultu-
res“. Yoghurt is made by fermenting milk with friendly bacteria, mainly Lactobacillus
delbrueckii subsp.bulgaricus and Streptococcus thermophilus. Yoghurt is a traditional
food and beverage in many countries and especially in Serbia. Yoghurt consumption in
Serbia has increased during the last decade. Product quality and satisfaction of consumer
expectation are discussed since they are essential for the continued successful growth of
the yoghurt market. Much emphasis is placed on yoghurt flavor, body, and texture. The
specific objectives of the study were: a) to determine the effect of cell lactis acid bacteria
L.delbrueckii subsp. bulgaricus and S. thermophilus on growth in yoghurt from a dif-
ferent producers and b) to determine the number of lactic acid bacteria of the temperature
and acidity in the period of storage (1,2). Initial counts of Lactobacillus and Strepto-
coccus in the samples of yoghurts were in the range from 8 to 1x106 /g resp. Ratio of
Lactobacillus : Streptococcus at the start of the test varied from 1:1 to 1: 2.7.
EXPERIMENTAL
Yoghurt production
Yoghurt is made by fermenting milk with friendly bacteria, in Serbia mainly with
Lactobacillus delbrueckii subsp.bulgaricus and Streptococcus thermophilus. The milk
sugar or lactose is fermented by these bacteria to lactic acid which causes the charac-
teristic curd to form. This process gives yoghurt its refreshingly tart flavor and unique
pudding like texture. The yoghurt qualities were judged to be satisfactory without defec-
tive taste. The baccilli/cocci ratio in the pre-fermented milk, unstable with free cells, was
stabilized when the strains were enterapped. The yoghurt starter cultures play an impor-
tant role during the production of yoghurt.
High cell numbers to about 5-107 C.F.U. ml-1 with a steady bacilli/cocci ratio were
present in the effluent milk. The starter culture for most yoghurt production is a symbio-
tic blend of Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus
in relation 50:50. Although they can grow independently, the rate of acid production is
much higher when used together than either of the two organisms grown individually.
Streptococcus thermophilus grows faster and produces both acid and carbon dioxide. The
format and carbon dioxide produced stimulates growth of lactobacilli. On the other hand,
the proteolytic activity of lactobacilli produces stimulatory peptides and amino acids for
use by streptococci. These microorganisms are ultimately responsible for the formation of
typical yoghurt flavor and texture. Yoghurt that contains live bacterial cultures may help
to live longer and may fortify immune system. The yoghurt mixture coagulates during
fermentation due to the drop in pH. The streptococci are responsible for the initial pH
drop of the yoghurt mix to approximately 5.0. The lactobacilli are responsible for a fur-
ther decrease to pH 4.0. The following fermentation products contribute to flavor:
• lactic acid
• acetaldehyde
88
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:637.04/.05
DOI: 10.2298/APT0940087P BIBLID: 1450-7188 (2009) 40, 87-94
• acetic acid
• diacetyl
The acid also restricts the growth of food poisoning bacteria. During the yoghurt
fermentation some flavors are produced, which give yoghurt its characteristic flavor.
Cultured media for the enumeration of Streptococcus thermophilus

and Lactobacillus delbrueckii subsp. bulgaricus mixtures
Yoghurt was sampled 1 day after manufacture and the samples were taken in the mar-
ket. Samples from 2 replicate experiments were processed. There are at present several
culture media for the differential enumeration of mixtures of Streptococcus thermophilus
and Lactobacillus delbrueckii subsp.bulgaricus. The use of these media is important for
the control of the relation between these two microorganisms in the yoghurt production
starter medium, as well as in the follow-up of both populations during the production and
later ripening of yoghurt. The number of lactobacilli was determined on the MRS agar (6)
and the number of streptococci was determined on M17 agar medium after 48 h of incu-
bation was investigated, making use of the different morphologies of the colonies as a
means of different enumeration. Total bacterial count was determined by direct counting
on a microscope glass. Material for the research was 70 samples of yoghurt from 9 dif-
ferent dairy producers. The samples were from Belgrade trade market. As a laboratory
control we used the mixed cultures of L.delbrueckii subsp. bulgaricus and S.thermo-
philus. Cultures enumerated in this study were lactic acid starter cultured used for manu-
factured of yoghurt. Lactic acid bacteria were enumerated using Elliker (7, 8) and MRS
(5, 6) solid agar plates, used for the isolation lactobacilli (3, 4). Reconstituted milk pow-
der was used for the storage lactic acid bacteria in the refrigerator. Strains of lactic acid
lactococci were an aerobically transferred three times at 370C for 48 h (BBLGas Pak
System). The species designation of isolated was confirmed by Gram stain colonial ap-
pearance. MRS agar plate was incubated at 37oC without further adjustment. Plates or
tube prepared with MRS agar were placed in plastic bags in anaerobe conditions and then
incubated. M17 agar plates were used for the detection of streptococci. Each plate was
overplayed with the same solid medium and then incubated at 30-37oC. Duplicate plates
were prepared for each medium for the required dilutions (7, 9). Plates were examined
after 48 and 72 h. To evaluate the factors that might be responsible for excessive acid
development during yoghurt storage, 9 brands of plain commercial yoghurt were pur-
chased from local retail markets and stored at 8, 12, 20oC, and analyzed weekly for 288
hours to monitor changes in acidity (oSH), total viable Lactobacillus and Streptococcus,
coliform, yeasts and moulds.
Microorganisms
Strains of Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgari-

cus isolated from commercial products were used. Commercial yoghurt samples were
used with no modifications, mixing 1 g of yoghurt in 10 ml saline solution and effective
successive dilutions from 10 -2 to 10-6.
89
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:637.04/.05
DOI: 10.2298/APT0940087P BIBLID: 1450-7188 (2009) 40, 87-94
Microorganisms: Streptococcus thermophilus and Lactobacillus delbrueckii subsp.

bulgaricus isolated from commercial yoghurt were grown at 43oC in M.R.S. and M17
agar.
Isolation and identification: The lactis acid bacteria were isolated by API system for
lactobacili and streptococi identification. The API system was incubated at 37oC for 72 h.
The various colonies formed were identified. Criteria examined included: Gram reaction,
cell morphology, catalasa reaction, growth at 50oC, acid production from lactose, sucrose,
trehalose, maltose and manitol, thermoresistence to 63oC, growth in NaCl 2%, hydrolysis
of arginine, and growth at pH 9.6.
Culture medium and growth conditions
Enumeration of microorganisms: Streptococcus thermophilus and Lactobacillus del-

brueckii subsp.bulgaricus were enumerated by surface spreading 0.1 ml samples on Elli-
ker agar medium and by direct microscopic count on the microscope slade.
RESULT AND DISCUSSION
The optimum growth temperature for Lactobacillus delbrueckii subsp. bulgaricus oc-
curred at 45oC and Streptococcus thermophilus has its optimum between 40oC and 45oC.
The optimum temperature of the mixed culture was 45oC. The growth of these cultures at
different initial pH in Elliker medium shows that the optimum pH for pure and mixed
cultures is from 6.5 to 8.0. In Table 1 the results of the chemical analyses and total bac-
terial count (on the microscope slide) in yoghurt from different producers are shown.
Table 1. Relationship between Streptococcus and Lactobacillus levels and average

acidity
Producers Acidity (oSH) Lactobacillus Streptococcus Relation

(L) (S) S/L
(I) 40.00 65 45 1 : 1.4
(II) 43.00 70 30 1 : 2.33
(III) 41.00 73 27 1 : 2.7
(IV) 39.00 55 54 1 : 1.2
(V) 37.00 50 50 1:1
(VI) 42.00 65 45 1 : 1.44
(VII) 38.00 50 50 1:1
(VIII) 42.00 67 43 1 : 1.55
(IX) 40.00 71 29 1 : 2.44
Control 32.00 50 50 1:1
90
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:637.04/.05
DOI: 10.2298/APT0940087P BIBLID: 1450-7188 (2009) 40, 87-94
80
70
60
50
Lactobacillus (L)
40
Streptococcus (S)
30
20
10
0
1 2 3 4 5 6 7 8 9 10 11
Fig. 1. Relationship between viable count of Streptococci/ Lactobacilli in the different

commercial yoghurt samples
The results of a brief microbiological evaluation of yoghurt and of a consumer pre-

ference survey are presented also in Fig.1. Lactic acid bacteria are broadly used as a star-
ter cultures for industrial production of fermented food. The requirement of Serbian
Regulation is a minimum number od 106 viable yoghurt organisms (L.delbrueckii subsp.
bulgaricus and S. thermophilus) must be present The ratio of the yoghurt organisms is
also important in determining the quality of yoghurt. In this study we determined the ef-
fect of cell for lactic acid bacteria (L.delbrueckii subsp. bulgaricus and S. thermophilus)
on growth in yoghurt from a different procedures and made a quantitative estimation of
the number of lactic acid bacteria the temperature and acidity in the period of storage.
Material for the research were 70 samples of yoghurt from 9 different dairy producers.
The commercial samples were from the Belgrade trade market. Lactic acid bacteria
were enumerate using Elliker solid medium agar plates, MRS solid agar plates for iso-
lation of lactobacilli and M17 agar plates for the isolation of streptococci. There was a
wide range in the total number of LAB. Predominant samples of yoghurt were with 11-
107/ml LAB (44, 28%) of examined samples.
Yoghurt taste is most satisfactory after 72 hours of storage at 8oC when acidity was
44.14o SH. Coliforms, yeasts or moulds were not detected at any time during storage.
There were consistent differences in oSH among brands with means ranging from 54.1 to
40.28 at 8oC, 63.1 to 47.74 at 12oC, 64.46 to 45.96 at 20oC indicating that it was possible
for the pH of yoghurt to remain ≥ 8.0. All brands showed initial counts of Lactobacillus
and Streptococcus in the range 8 to 1x106/g resp. Ratio of Lactobacillus: Streptococcus at
the start of the test varied from 1 : 1 to 1 : 2.7. The decrease in counts of Streptococcus
and Lactobacillus in most of brand stored at 20oC. It is apparent that strain of bother
Streptococcus and Lactobacillus vary in their survival and in their ability to maintain a
higher oSH in yoghurt during storage.
91
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:637.04/.05
DOI: 10.2298/APT0940087P BIBLID: 1450-7188 (2009) 40, 87-94
Table 2. Temperature storage of yoghurt and the number of viable lactobacilli and
streptococci at the corresponding pH
Time of Number of viable lactobacilli and streptococci in

Temperature Acidity
storage the samples of yoghurt (in 000000)
storage (oC) (oSH)
(hours) 1-10 11-100 101-200 201-300 301-400
8 72 1 3 2 4 0 44.14
8 120 4 2 3 1 0 40.28
8 192 5 4 1 0 0 51.24
8 288 8 2 0 0 0 54.10
12 72 2 2 2 3 1 47.74
12 120 5 2 1 1 1 53.84
12 192 4 3 3 0 0 52.00
12 288 7 3 0 0 0 63.10
20 72 3 1 5 1 0 45.96
20 120 3 2 4 0 0 51.52
20 192 7 2 1 0 0 64.46
20 288 8 2 0 0 0 59.21
The product is resulting from milk by fermentation with a mixed starter culture con-
sisting only Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus
and these two organisms have a symbiotic relationship during the manufacture of yoghurt
with the ratio of Streptococcus thermophilus to Lactobacillus delbrueckii subsp. bulga-
ricus constantly changing. Both organisms produce lactic acid as the main fermentation
products. For a proper flavor development, the ratio of Streptococcus thermophilus to
Lactobacillus delbrueckii subsp.bulgaricus should be in the range of 1:1 to 3:1. Con-
sistency is an important attribute in evaluating the quality of yoghurt. There are numerous
factors which affect consistency including milk composition, heat treatment, homogeni-
zation, use of stabilizer, yoghurt culture, and mechanical handling of the yoghurt.
CONCLUSION
During the past two decades, there has been renewed interest in the study of the nu-
tritional and therapeutic aspects of dairy products (12, 13, 14). A majority of reviewed
papers suggested potential therapeutic benefit following the consumption of fermented
dairy products containing viable lactic acid bacteria (LAB) count and decreased coliform
count in the intestine as observed in the analysis. The beneficial effect of normal gut flora
temporarily brings yoghurt live microorganisms which have probiotics effects (3, 7).
While numerous researchers have suggested that lactic cultures and cultured dairy pro-
ducts provide several nutritional and therapeutic benefits to the consumer there exist a
few reports in which some of the benefits have been questioned. Lactic acid bacteria are
broadly used as starter cultures for industrial production of fermented food. It has been
known for many years that lactic acid bacteria may positively influence the gastroin-
testinal tract of human and other mammals. The beneficial effects include the inhibition
of undesirable microorganisms, reduction in cholesterol level and reduction of the risk of
colon cancer (10, 11, 14, 7).
92
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:637.04/.05
DOI: 10.2298/APT0940087P BIBLID: 1450-7188 (2009) 40, 87-94
REFERENCES
1. American Public Health Association: Standard methods for the examination of dairy
products, 13 th ed American Public Health Association, Washington D.C., USA
(1972).
2. Amoroso M.J., Manca M.C. de Nadra, G. Oliver: Growth and sugars utilization by
mixed cultures of Lactobacillus delbrueckii subsp.bulgaricus and Streptococcus Sali-
varius subsp, thermophilus isolated from Argentina, World Journal of Microbiology
and Biotechnology 8 (1997) 50-54.
3. Bautista E.A., R.S. Dahya, M.L. Speck: Identification of compounds causing symbio-
tic growth of Streptococcus thermophilus and Lactobacillus bulgaricus in milk. Jour-
nal of Dairy Research 33 (1966) 299- 307.
4. Barbour Anne E., F.G. Priest: The preservation of lactobacilli: A comparison of three
methods. Letters in Applied Microbiology 2, 4 (1986) 57-95.
5. Clunies- Parnell E., Y. Kakuda, A.K. Smith: Microstructure of yoghurt as affected by
heat treatment of milk. Milchwissenchaft 42, 7 (1987) 413-417.
6. De Man, J.D., M. Rogosa, A. Sharpe.: A Medium for the Cultivation of Lactobacilli.
J. Appl. Bact. 23 (1960) 130-135.
7. Davis J.G., McLachlan. Yoghurt in the United Kingdom: Chemical and microbiolo-
gical analyses. Dairy Ind. 39, 5 (1974) p. 149,150, 152, 154,157,177.
8. Datta, N., M.G. Hayes, H.C. Deeth, A.L. Kelly: Significance of frictional heating for
effects of high pressure homogenization on milk. Journal of Dairy Research 72, 4
(2005) 1−7.
9. Galesloot, T.E., F. Hassing, H.A. Veringa: Symbiosis in yoghurt. Neth. Milk Dairy J.
22 (1968) 50-63.
10. Metchnikoff. E., The prolongation of life: Optimistic Studies. Revised Edition, Trans-
lated by Mitchell,C., Heineman Ltd.London, UK (1907).
11. E. Metchnikoff, Optimistic studies New York: Putman’s Sons, (1908), 161-183.
12. Peral M.C de Portillo, M.J. Amoroso, G. Oliver: Culture medimu for differential enu-
meration of lactic acid bacteria in yoghurt. Milchwissenchaft 43, 8 (1988) 490-491.
13. Sinha R.P., H.W. Modler , D.B. Emmons: Changes in acidity and starter bacteria in
commercial yoghurts during storage. Cultured Dairy Products Journal 24, 2 (1989)
12-14.
14. Tamime A.Y., R.K. Robinson: Yoghurt science and technology, Printed in Great Bri-
tain by Wheaton and Co.Ltd., Exeter, UK (1985) 23-210.
15. Boyanova L., M. Stephanova-Kondratenko, I. Mitov: Anti-Helicobacter pylori acti-
vity of Lactobacillus delbrueckii subsp. bulgaricus strains: preliminary report Letters
in Applied Microbiology 48 (2009) 579-584.
93
APTEFF, 40, 1-220 (2009) UDC: 637.146.3:637.04/.05
DOI: 10.2298/APT0940087P BIBLID: 1450-7188 (2009) 40, 87-94
КВАНТИТАТИВНО ОДРЕЂИВАЊЕ БАКТЕРИЈА МЛЕЧНЕ КИСЕЛИНЕ

КОМЕРЦИЈАЛНИХ УЗОРАКА ЈОГУРТА
Драгана Пешић-Микулец и Гордана Б. Никетић
Квалитет јогурта је тешко стандардизовати због великог броја и врста бактерија

млечне киселине у процесу производње. Због свих ових особина уједначеност ква-
литета јогурта тешко се постиже. Испитивања у овом раду обухватила су изолаци-
ју, однос и број бактерија млечне киселине (L. delbrueckii subsp. bulgaricus и S. ther-
mophilus) које се користе за производњу одређених варијетета јогурта карактерис-
тичних за поднебље Србије. Одређиван је број бактерија млечне киселине у раз-
личитим комерцијалним узорцима јогурта. Квалитет узорака јогурта одређиван је у
зависности од температуре и времена чувања. Код већег броја узорака јогурта на-
ђено је 11-107/мл живих бактерија млечне киселине. Укус и арома јогурта били су
задовољавајући након 72 часа при температури складиштења од 8оC. Однос бак-
терија млечне киселине Streptococcus и Lactobacillus износио је од 1:1 до 1: 2,7 код
свежих узорака јогурта, да би даљим складиштењем на неповољним температурама
дошло до поремећаја односа до 3:1 стартер култура јогурта.
Received 5 May 2009

94
APTEFF, 40, 1-220 (2009) UDC: 664.11:663.81:543.645
DOI: 10.2298/APT0940095S BIBLID: 1450-7188 (2009) 40, 95-102
ANTIOXIDANT ACTIVITY OF POLYPHENOL-ENRICHED APPLE JUICE
Slađana M. Savatović, Aleksandra N. Tepić, Zdravko M. Šumić and Milan S. Nikolić
This paper shows that it is possible to improve antioxidant activity of apple juice by
extraction of polyphenolic compounds from apple pomace, as waste, and their addition to
the apple juice. Raw apple juice was prepared by pressing of apple mash. After thermal
treatment of raw apple juice, depectinisation, additional clarification and filtration, the
clarified juice was obtained. In raw and clarified apple juice soluble solids, acidity,
reducing sugar, total sugars and brown component content were determined, as well as
total dry matter, ash, acidity, reducing sugar, total sugars, total pectins, cellulose and
starch content in apple mash and pomace. The total cotent of phenolics in clarified apple
juice and apple pomace extract, determined spectrophotometrically using the Folin-
Ciocalteu reagent, was 0.496 mg/ml and 6.505 mg/g, respectively. The antioxidant acti-
vity of clarified and polyphenol-enriched clarified juice (with addition of apple pomace
extract in the concentrations 0.05 g, 0.1 g, 0.5 g and 1 g of phenolic compounds per liter
of clarified apple juice) was examined on stable 1,1-diphenyl-2-picrylhydrazyl (DPPH)
free radicals. Based on the obtained results it can be concluded that polyphenol-enriched
clarified juice was more effective on DPPH radicals than the clarified apple juice.
KEY WORDS: Apple, apple juice, apple pomace, phenolic compounds, antioxidant
activity
INRODUCTION
Different studies have shown that free radicals present in the human organism cause
oxidative damage to various biomolecules, such as lipids, proteins and nucleic acids, and
thus are involved in the initiation phase of degenerative disease. Phenolic and other phy-
tochemical antioxidants found in fruits and vegetables are bioactive compounds capable
of neutralizing free radicals and may play a major role in the prevention of certain disea-
ses (1). Also, dietary supplements and food fortification may be an alternative route to the
consumption of minor plant components that may have health benefits.
Apple (Malus domestica) has been the leading fruit variety according to its world pro-
duction. The most important industrial utilization of apple is the juice production. Apples
contain 85% of water, 12-14% of carbohydrates, about 0.3% of proteins, minor quantity
Slađana M. Savatović, M.Sc., Aleksandra N. Tepić, M.Sc., Assist., Zdravko M. Šumić, B.Sc., Milan S. Nikolić,
B.Sc., University of Novi Sad, Faculty of Technology, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia
95
APTEFF, 40, 1-220 (2009) UDC: 664.11:663.81:543.645
DOI: 10.2298/APT0940095S BIBLID: 1450-7188 (2009) 40, 95-102
of lipids (<0.1%), minerals and vitamins (2). The differences in chemical composition of
apples are due to the region of growth, variety, ripening state during harvesting, agrono-
mic and environmental conditions. About 80% of apple carbohydrates are soluble sugars:
sucrose (~ 2.1%), glucose (2.4%) and fructose (5.9%). Apples contain about 2.4% of total
dietary fibers, and it is proved that they contain sorbitol (2). Malic acid is the predomi-
nant organic acid in apples (80-90% of total acids), and its content varies depending on
the variety, ripeness, and environmental conditions during growing and storage. The phe-
nolic compounds content in apples is 36 ± 19 mg/kg fresh weight (3). The most important
groups of phenolic antioxidants present are flavonols (with quercetin glycosides as the
main representative), monomeric and oligomeric flavanols, dihydrochalcones (e.g., phlo-
ridzin), anthocyanins, p-hydroxycinnamic and p-hydroxybenzoic acids (3, 4). Apples are
not rich in vitamin C (its content in fresh fruit can be only a few miligrams) (5).
The technological process of apple juice production at industrial scale include raw
material preparation for processing, mashing, pressing, thermal treatment, depectiniza-
tion, clarification, filtration, pasteurization and packing. The raw material preparation for
further processing consists of washing and inspection. During processing, losses of anti-
oxidant components occur, especially due to the exposure of raw material to oxygen and
thermal treatment of raw material and juice. The conventional apple juice production
(direct pressing of apple mash or pressing after mash depectinization) results in a juice
poor in phenolics and with only 3-10% of the antioxidant activity of fresh apples. The
fact that in conventional apple juice production techniques most of the phenolics do not
get deteriorated or lost during juice manufacture, but remain in the pomace, suggests that
pomace can be considered as a source of phenolic antioxidants.
Several possibilities exist in the juice production chain to enhance the phenolic con-
tent of apple juice, by the choice of cultivation methods, raw material, production
methods, processing, storage and distribution conditions (3). Also, the juice with a high
content of phenolic bioactive components can be produced by enrichment with phenolics
from other sources. For this reasons, the possibility of utilization of apple pomace antio-
xidant compounds by their extraction, and enrichment of clarified juice, was examined in
this work. The objectives of this study were: (I) to obtain raw and clarified apple juice;
(II) to determine soluble solids, acidity, sugar and brown component content in obtained
juices; (III) to determine chemical compositions of apple mash and apple pomace; (IV) to
perform an extraction of pomace obtained after direct pressing of apple mash; (V) to
determine phenolic content of apple pomace and clarified apple juice spectrophotome-
trically using the Folin-Ciocalteu reagent; (VI) to obtain polyphenol-enriched apple juice,
by adding pomace extract to the previously produced clarified juice, and (VII) to compare
the antioxidant activity of clarified and polyphenol-enriched clarified juice on stable 1,1-
diphenyl-2-picrylhydrazyl (DPPH) free radicals.
EXPERIMENTAL
Chemicals and samples
Methanol and sodium carbonate were obtained from „Zorka” Šabac (Serbia). 1,1-Di-
phenyl-2-picrylhydrazyl (DPPH), Folin-Ciocalteu reagent and gallic acid were purchased
96
APTEFF, 40, 1-220 (2009) UDC: 664.11:663.81:543.645
DOI: 10.2298/APT0940095S BIBLID: 1450-7188 (2009) 40, 95-102
from Sigma Chemical Co. (USA). These chemicals were of analytical reagent grade.
Other chemicals and solvents used were of the highest analytical grade.
For apple juice production in laboratory conditions Red Delicious variety (from a lo-
cal market) was used.
Apple juice production
Apples were washed and mashed, and the mash was pressed using laboratory cider
press. The obtained raw apple juice was heated to 85-90°C and treated with 0.04 g/l Ra-
pidase Pro (DSM) at 45-50°C, during 1.5 hours. After depectinization, enzyme was in-
activated by heating the raw juice up to 85-90°C/2-3 min., cooled and treated with gela-
tine and bentonite solutions (0.005% and 0.05%, respectively). The clarified juice was fil-
tered through Büchner funnel using vacuum, filled into PET bottles and stored at -18°C.
Apple pomace that remained after juice production was stored in PE bags at -18°C.
Determination of soluble solids, acidity, sugar and brown component content

in raw and clarified apple juice
In raw and clarified apple juice soluble solids, acidity and sugar content were deter-
mined according to valid Regulations (6). Brown component was assayed by measuring
absorbance of the sample extract at 380 nm (7).
Determination of chemical compositions of apple mash and apple pomace
In apple mash and pomace total dry matter, ash, acidity, reducing sugar, total sugars,
total pectins, cellulose and starch content were determined according to valid Regulations
(6). Cellulose content was determined according to Kirschner-Gannak (7).
Extraction procedure
Sample of apple pomace (20 g) was extracted at room temperature using an ultraso-
nic bath, Heidolph DIAX 900 (Heidolph Instruments GmbH, Kelheim, Germany). The
extraction was performed three times with different amounts of 80% methanol: 160 ml in
60 min, 80 ml in 60 min, 80 ml in 30 min. The total extraction time was 150 min. The ob-
tained three extracts were combined and evaporated to dryness under reduced pressure.
The yield of extract was 2.65 g.
Total phenolics
Total phenolics in apple pomace extract and clarified apple juice were determined
using the Folin-Ciocalteu reagent (8). The reaction mixture was prepared by mixing 0.1
ml of water solution of extract (concentration 50 mg/ml) or 0.1 ml juice, 7.9 ml of distil-
led water, 0.5 ml of the Folin-Ciocalteu’s reagent and 1.5 ml of 20% sodium carbonate.
After 2 h, the absorbance at 750 nm (UV-1800 spectrophotometer, Shimadzu, Kyoto, Ja-
pan) was obtained against blank prepared in a similar manner, by replacing the extract
97
APTEFF, 40, 1-220 (2009) UDC: 664.11:663.81:543.645
DOI: 10.2298/APT0940095S BIBLID: 1450-7188 (2009) 40, 95-102
with distilled water. The total phenolic content, expressed as mg of gallic acid equiva-
lents per g of apple pomace extract or per ml of apple juice, was determined using cali-
bration curve of gallic acid standard.
Apple juice enrichment
Apple pomace extract was added to the previously produced clarified juice in the con-
centrations 0.05 g, 0.1 g, 0.5 g and 1g of phenolic compounds per liter of apple juice.
DPPH radical-scavenging spectrophotometric assay
The potential antioxidant activity of apple juice was assessed on the basis of the sca-
venging activity of the stable DPPH free radicals according to the method of Yen and
Chen (9). The juice (1 ml) was diluted with distilled water. The range of the investigated
juice concentration was 5-50%. An aliquot (1 ml) of diluted juice was added to 3 ml of
absolute methanol and 1 ml of methanolic DPPH solution (concentration 0.3 mmol/l).
The mixture was shaken and left at room temperature for 10 min, then the absorbance
was measured at 517 nm using a UV-1800 spectrophotometer (Shimadzu, Kyoto, Japan).
The blank probe contained all components except the radicals. The antioxidant activity
on the basis of the capability to scavenge the DPPH radicals (AADPPH) was estimated
from the differences in absorbance of DPPH solution with or without juice (control) and
the inhibition percent was calculated using the following equation:
AADPPH (%) = (AControl - ASample)/AControl × 100
where AControl is the absorbance of the control reaction (containing all reagents except the
juice) and ASample is the absorbance in the presence of the juice. The values of antioxidant
activity were investigated for the various concentrations of the juice.

Chemical composition of raw and clarified apple juice is given in Table 1. As previo-
usly said, the chemical composition of apples depends on the number of parameters.
The yield of raw apple juice was 55.6%.
The differences in chemical composition of raw and clarified apple juice are due to
the influence of different treatments during juice production.
Table 1. Chemical composition of raw and clarified apple juice

Component Raw juice Clarified juice
Soluble solids (g/100g) 15.15 15.25
Acidity (g malic acid/100g) 0.21 0.22
Reducing sugars (g/100g) 11.41 14.14
Total sugars (g/100g) 13.67 14.42
Brown component (μg K2Cr2O7/ml) 117.5 98.0
98
APTEFF, 40, 1-220 (2009) UDC: 664.11:663.81:543.645
DOI: 10.2298/APT0940095S BIBLID: 1450-7188 (2009) 40, 95-102
As can be seen from Table 1, soluble solids of raw apple juice were somewhat lower
than in clarified juice (15.15% and 15.25%, respectively). Acidity of raw juice was also
lower than that of clarified juice (0.21 g/100g and 0.22 g/100g of dry matter, respec-
tively). Results of chemical composition of raw and clarified apple juice are in accor-
dance with the available literature (5). According to Hui et al. (5), dry matter content in
apple is somewhat lower than obtained in this research.
Chemical compositions of apple mash and apple pomace are given in Table 2.
Table 2. Chemical composition of apple mash and pomace

Component Apple mash Apple pomace
Total dry matter (%) 14.95 18.65
Ash (g/100g) 0.29 0.35
Acidity (% malic acid) 0.17 0.15
Reducing sugar (g/100g) 11.68 11.5
Total sugars (g/100g) 11.85 13.3
Total pectins (g/100g Ca-pectate) 0.54 0.19
Cellulose (g/100g) 0.71 1.58
Starch (g/100g) 0.87 2.11
The Folin-Ciocalteu method is a rapid and widely-used assay to investigate the total
phenolic content. The content of total soluble phenolics of clarified apple juice and apple
pomace extract was expressed as gallic acid equivalent and it was 0.496 mg/ml and 6.505
mg/g, respectively.
The potential antioxidant activity of apple juice was assessed on the basis of the
scavenging activity of the stable DPPH free radicals. Antioxidant activity of clarified
apple juice without and with addition of apple pomace extract in the concentrations 0.05
g, 0.1 g, 0.5 g and 1g of phenolic compounds per liter of apple juice is shown in Fig. 1.
100
juice
80 juice + 0,05 g polyphenolic compounds per 1l
juice + 0,1 g polyphenolic compounds per 1l
60 juice + 0,5 g polyphenolic compounds per 1l
AADPPH (%)
juice + 1 g polyphenolic compounds per 1l
40
20
0
5 10
25 50
Juice concentration (%)
Fig. 1. Antioxidant activity of clarified apple juice without and with addition of apple
pomace extract in the concentrations 0.05 g, 0.1 g, 0.5 g and 1g of phenolic compounds
per liter of apple juice
99
APTEFF, 40, 1-220 (2009) UDC: 664.11:663.81:543.645
DOI: 10.2298/APT0940095S BIBLID: 1450-7188 (2009) 40, 95-102
Based on the DPPH radical-scavenging spectrophotometric measurments, it can be

observed that apple juice showed dose-dependent antioxidant activity (AADPPH). Also, it
can be concluded that apple juice without addition of apple pomace extract was less ef-
fective on DPPH radicals than with addition of apple pomace extract. With increasing
concentrations of added phenolic compounds ranged from 0.05 g to 1 g per liter of apple
juice, AADPPH increased from 36.42% to 90.94%, at juice concentration of 5%. This
indicates that enriched apple juice, with addition of 0.5 g of phenolic compounds per liter
of apple juice, had an antioxidant activity that was 3.8 times higher than that of the cor-
responding clarified juice.
The IC50 value is a parameter used to measure antioxidative activity and it is defined
as the juice concentration required for 50% scavenging of DPPH radicals under experi-
mental condition employed. A smaller IC50 value corresponds to a higher antioxidant
activity. The IC50 values of apple juice without and with addition of apple pomace extract
in the concentrations 0.05 g and 0.1 g of phenolic compounds per liter of apple juice, de-
termined based on antioxidant activities, were 13.89%, 10.77% and 8.16%, respectively.
It was observed that with increasing concentrations of added phenolic compounds to the
juice antioxidant activity increased.
The fact that phenolic compounds that contribute to antioxidant activity preferentially
remain in the pomace offers a possibility for apple juice optimization with respect to phe-
nolic content and antioxidant activity. Thus, it is a challenging option to search for me-
thods in which the phenolics may be extracted from the pomace and later added to the
final apple juice.
CONCLUSION
• The differences in chemical composition of raw and clarified apple juice are due
to the influence of different treatments during juice production;
• Soluble solids of raw apple juice were somewhat lower than in clarified juice
(15.15% and 15.25%, respectively);
• Raw juice acidity was also lower than in clarified juice (0.21 g/100g and 0.22
g/100g of dry matter, respectively);
• The content of total soluble phenolics of clarified apple juice and apple pomace
extract was expressed as gallic acid equivalent and it was 0.496 mg/ml and
6.505 mg/g, respectively;
• Apple juice without addition of apple pomace extract was less effective on
DPPH radicals than with addition of apple pomace extract;
• With increasing concentrations of added phenolic compounds, ranging from
0.05 g to 1 g per liter of apple juice, AADPPH increased from 36.42% to 90.94%,
at juice concentration 5%;
• The enriched apple juice by adding apple pomace extract in the concentration of
0.5 g of phenolic compounds per liter of apple juice, had an antioxidant activity
that was 3.8 times higher than that of the corresponding clarified juice, at the
juice concentration of 5%.
100
APTEFF, 40, 1-220 (2009) UDC: 664.11:663.81:543.645
DOI: 10.2298/APT0940095S BIBLID: 1450-7188 (2009) 40, 95-102
ACKNOWLEDGEMENT
These results are part of the project No. 23011, which is financially supported by the
REFERENCES
1. Kaur, C. and H.C. Kapoor: Antioxidants in fruits and vegetables - the millennium's
Health. Int. J. Food Sci. Tech. 36 (2001) 703-725.
2. http://www.nal.usda.gov/fnic/foodcomp
3. van der Sluis, A.A., M. Dekker, G. Skrede and W.M.F. Jongen: Activity and con-
centration of polyphenolic antioxidants in apple juice. 1. Effect of existing production
methods. J. Agric. Food Chem. 50 (2002) 7211-7219.
4. Escarpa, A. and M. Gonzalez: High-performance liquid chromatography with diode-
array detection for the performance of phenolic compounds in peel and pulp from
different apple varieties. J. Chrom. 823 (1998) 331-337.
5. Hui, Y.H.: Nutritional Values of Fruits, in Handbook of fruits and fruit processing.
Eds. C.S. Moreno, S.P. Teresa, B. Ancos and M.P. Cano, Wiley-Blackwell Publi-
shing, Iowa, USA (2006) pp.30-31.
6. Pravilnik o metodama uzimanja uzoraka i vršenja hemijskih i fizičkih analiza radi
kontrole kvaliteta proizvoda od voća i povrća, Službeni list SFRJ 29/83.
7. Vračar, Lj.: Priručnik za kontrolu kvaliteta svežeg i prerađenog voća, povrća i pečurki
i osvežavajućih bezalkoholnih pića, Tehnološki fakultet, Novi Sad (2001) p.79.
8. Singleton,V.L., R. Orthofer and R.M. Lamuela-Raventos: Analysis of total phenols
and other oxidation substrates and antioxidants by means of Folin–Ciocalteu reagent.
Meth. Enzymo. 299 (1999) 152-178.
9. Yen, G.C and H.Y. Chen: Antioxidant activity of various tea extracts in relation to
their antimutagenicity. J. Agric. Food Chem. 43 (1995) 27-32.
АНТИОКСИДАТИВНА АКТИВНОСТ СОКА ОД ЈАБУКА ОБОГАЋЕНОГ

ПОЛИФЕНОЛНИМ ЈЕДИЊЕЊИМА
Слађана М. Саватовић, Александра Н. Тепић, Здравко М. Шумић

и Милан С. Николић
У овом раду испитана је могућност побољшања антиоксидативне активности

сока од јабука додатком полифенолних једињења екстрахованих из тропа, спо-
редног производа насталог приликом добијања сока од јабука. Матични сок је про-
изведен стандардним поступком производње – пресовањем каше од јабука. Након
термичке обраде матичног сока, депектинизације, бистрења и филтрирања добијен
је бистри сок. У матичном и бистром соку одређен је садржај киселина, суве
материје, шећера и смеђе компоненте. Такође, одређен је садржај суве материје,
киселина, шећера, скроба, целулозе, пектина и пепела у каши и тропу јабука. Са-
држај укупних полифенолних једињења у бистром соку од јабука и екстракту тро-
па, одређен спектрофотометријски, методом по Folin-Ciocalteu, износи 0.496 mg/ml
101
APTEFF, 40, 1-220 (2009) UDC: 664.11:663.81:543.645
DOI: 10.2298/APT0940095S BIBLID: 1450-7188 (2009) 40, 95-102
и 6.505 mg/g. Антиоксидативна активност бистрог сока, као и бистрог сока од ја-
бука обогаћеног полифенолним једињењима (са додатком екстракта тропа у кон-
центрацијама 0.05 g, 0.1 g, 0.5 g и 1 g полифенолних једињења po 1l бистрог сока)
испитана је на стабилне 1,1-дифенил-2-пикрилхидразил (DPPH) радикале. Обогаће-
ни сок од јабука показао је израженију антиоксидативну активност на DPPH ра-
дикале од сока без додатка екстракта тропа.

102
APTEFF, 40, 1-220 (2009) UDC: 664.844:635.652:543.641
DOI: 10.2298/APT0940103V BIBLID: 1450-7188 (2009) 40, 103-110
DIETARY FIBER CONTENT IN SOME DRY BEANS
Mirjana A. Vasić, Biserka L. Vujičić, Aleksandra N. Tepić,

Jelica M. Gvozdanović-Varga and Zdravko M. Šumić
Dietary fibers are one of the main nutritive components, along with proteins, fats and
oils, carbohydrates, minerals and vitamins. Also, they are one of the basic parameters of
dry beans technological quality and nutritive value. Physical characteristics and the
main chemical composition of sixteen dry bean varieties (Phaseolus vulgaris) had been
examined in this study. Using statistical analyses, correlation between certain parame-
ters of chemical composition was established.
KEY WORDS: Dry beans, physical characteristics, chemical composition, dietary fibers
INTRODUCTION
Dry beans (Phaseolus vulgaris) are very important in human diet (1, 2). They are one
of the most important sources of plant proteins, carbohydrates, soluble and insoluble
fibers, certain minerals and vitamins (3, 4, 5). Dietary fiber content is one of the most im-
portant parameters of technological quality and physiological value of dry beans and
other legumes (6).
The definition of dietary fibers is still controversial and several definitions have been
suggested. The most widely accepted definition is a physiological one, in which "dietary
fibers" correspond to the plant wall residues that are resistant to enzymatic hydrolysis in
the small intestine. A chemical definition describes dietary fibers as non-starch poly-
saccharides. The most commonly used definition of dietary fibers is the following: "die-
tary fibres are oligosaccharides, polysaccharides and the (hydrophilic) derivatives which
cannot be digested by the human digestive enzymes to absorbable components in the
upper alimentary tract" (7).
Dietary fibers do not constitute a defined chemical group, but are a combination of
chemically heterogeneous substances such as celluloses, hemicelluloses, pectins, lignins,
gums and polysaccharides from seaweeds or bacteria. Celluloses, hemicelluloses and pec-
tins, also referred to as structural components of cell walls are also classified as dietary
fibers: secreted gums (e.g. gum arabic), reserve gums (bean gums, guar gums) and poly-
Dr. Mirjana A. Vasić, senior research associate, vasicka@ifvcns.ns.ac.yu, Dr. Jelica M. Gvozdanović-Varga,
scientific associate, Institute of Field and Vegetable Crops, Maksima Gorkog 30, 21000 Novi Sad, Serbia;
Aleksandra N. Tepić, M.Sc., Assist.; Dr. Biserka L. Vujičić, Prof., Zdravko M. Šumić, B.Sc., Faculty of
Technology, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia
103
APTEFF, 40, 1-220 (2009) UDC: 664.844:635.652:543.641
DOI: 10.2298/APT0940103V BIBLID: 1450-7188 (2009) 40, 103-110
saccharides from seaweeds (carrageenans, agar, alginates); some workers also include
resistant starch (fractions of starch that are not digested by small intestinal enzymes).
Since then, intensive research of the role and importance of these compounds has been
done (8). Dietary fibers show a number of health benefits, like prevention of cardio-
vascular diseases, decrease in blood cholesterol and glucose level, prevention of digestive
system carcinogenic diseases, constipation prevention, etc (9).
The aim of this paper is to compare dietary fiber content in Serbian and bean varieties
from other countries. The hierarhical cluster method of the multivariate analysis was used
to classify the tested varieties according to the chemical composition of all dietary fiber.
Their relative relationship regarding the total dietary fiber content was also examined.
EXPERIMENTAL
Sixteen dry beans varieties - domestic (Levač, Panonski tetovac, Balkan, Dvadesetica,
Aster, Belko, Sremac, Galeb, Zlatko, Jovandeka, Slavonski Zeleni) and foreign (Spinel,
Naya Nayahit, C-20, Igman and Prelom) were examined in this work. All samples were
grown at the Institute of Field and Vegetable Crops, Novi Sad, Rimski Šančevi, in 2006.
The physical analyses of dry bean seeds were done in sample of 50 seeds. Physical
measurements were done in whole seeds. Classification according to color of seed coat
was performed visually. Determination of shape of bean seeds was conducted by the
method of Dekaprelevic (3). According to the seed length-to-width and thickness-to-
width ratio, the examined genotypes were classified into five botanical forms or groups
(10).
Dry beans were milled and stored in hermetically closed jars. Chemical analyses
included total dry matter (11); total dietary fibers (12); cellulose according to Kirschner-
Gannak method (13), and pectin compounds (14, 15).
Pearson coefficient of correlation among traits were calculated. Hierarchical clus-
tering of varieties (Single linkage method or nearest neighbour by Euclidean distance for
Distance metric) was done using a computer statistical package STATISTICA.
As has been said above, eleven domestic and five foreign varieties were chosen for
the study. Domestic varieties have been currently used in the industry (2), while foreign
varieties were chosen due to their origin and wide dispersion (16). All beans are from
Phaseolus vulgaris species, except for Igman, which is from Phaseolus coccineus. In
Serbia, varietis with determinate growth (I type of habitus) have been grown, and these
were the samples of this research. Five varieties were indeterminate, while three of them
upright (II type of habitus), and Spinel was type III of growth. Variety Levač is typical
tetovac, internationally recognized market class from the Balkan Peninsula with IV type
of growth (17). The origin, status, type of habitus, seed color and shape, and 1000 seed
mass are given in Tables 1 and 2.
104
APTEFF, 40, 1-220 (2009) UDC: 664.844:635.652:543.641
DOI: 10.2298/APT0940103V BIBLID: 1450-7188 (2009) 40, 103-110
Table 1. The main characteristics of dry bean samples
No. Genotype Origin of genotype Status Type of habitus

1 Levač Serbia IFVCNS Variety IV
2 Aster Serbia IFVCNS Variety I
3 Spinel USA Variety III
4 Panonski tetovac Serbia IVCSP Variety I
5 Balkan Serbia IFVCNS Variety I
6 Naya nayahit USA Variety II
7 Dvadesetica Serbia IFVCNS Variety I
8 Sremac Serbia IFVCNS Variety I
9 Jovandeka Serbia Landraces I
10 Galeb Serbia IVCSP Variety I
11 Prelom Bulgaria Variety II
12 Belko Serbia IFVCNS Variety I
13 C-20 USA Variety II
14 Igman Bosnia and Herzegovina Variety I
15 Zlatko Serbia IFVCNS Variety I
16 Slavonski zeleni Serbia Landraces I
Seed color, seed shape and seed size according to mass of 1000 seeds are quality
traits, important market characteristics and a stable cultivar trait (4, 18). Examined spe-
cies could be distinguished according to seed color, and most had white seeds. The seeds
had one of four shape forms (Table 2). The variation in 1000-seed mass from 161.8 g to
648.5 g measured in this study indicates that the tested genotypes differed significantly in
their seed size.
Table 2. Seed color and shape and 1000 seed mass of dry bean samples
Seed colour Seed shape 1000

No. Genotype seed
English Form English Form mass
1 Levač white albus kidney compressus 592.7
2 Aster white albus cylindr. oblongus 412.0
3 Spinel white albus semi-flat subcompr. 329.9
4 Panonski tetovac white albus semi-flat subcompr. 394.8
5 Balkan white albus ellipsoid ellipticus 317.7
6 Naya nayahit black niger ellipsoid ellipticus 168.3
7 Dvadesetica white albus kidney compressus 339.8
8 Sremac greenish-yellow griseus cylindr. oblongus 349.2
9 Jovandeka seed coat patterns versicolor cylindr. oblongus 405.0
10 Galeb white albus ellipsoid ellipticus 353.7
11 Prelom white albus ellipsoid ellipticus 223.0
12 Belko white albus ellipsoid ellipticus 294.1
105
APTEFF, 40, 1-220 (2009) UDC: 664.844:635.652:543.641
DOI: 10.2298/APT0940103V BIBLID: 1450-7188 (2009) 40, 103-110
Table 2. Continuation
Seed colour Seed shape 1000

No. Genotype seed
English Form English Form mass
13 C-20 white albus ellipsoid ellipticus 179.2
14 Igman white albus ellipsoid ellipticus 648.5
15 Zlatko gold-yellow aureus cylindr. oblongus 380.0
16 Slavonski zeleni greenish-yellow griseus cylindr. oblongus 342.5
Dry matter symbolizes the content of chemical compounds out of water. It consists of
soluble (sugars, acids, etc.) and insoluble compounds (starch, cellulose, hemicellulose,
protopectin, etc.). In our research, the lowest dry matter was measured in the variety C-
20, and the highest in Igman (Table 3), which are higher values than those given by
Kojnov (3), Costa et al. (6) and Todorović et al. (2).
Table 3. Total dietary fiber (TDF), cellulose and pectin content (g/100g dry matter) in
dry bean samples
Dry Pectins
Genotype TDF Cellulose
matter pectin pectic acid protopectin total
Levač 90.57 21.22 4.23 0.61 0.23 1.16 2.00
Aster 90.67 26.97 4.84 0.66 0.27 1.78 2.71
Spinel 92.00 30.62 4.36 0.45 0.25 1.46 2.16
Panonski tetovac 91.32 21.74 4.84 0.54 0.19 2.46 3.19
Balkan 90.65 17.98 3.88 0.38 0.24 2.00 2.62
Naya nayahit 90.61 30.96 3.68 0.28 0.14 1.51 1.93
Dvadesetica 90.46 31.69 5.33 0.41 0.15 1.47 2.04
Sremac 90.85 19.20 3.93 0.34 0.19 1.20 1.73
Jovandeka 91.69 27.58 3.47 0.38 0.15 1.24 1.77
Galeb 91.11 30.93 4.29 0.35 0.17 1.59 2.12
Prelom 90.83 19.12 4.52 0.38 0.19 1.51 2.08
Belko 90.80 33.76 4.15 0.59 0.22 1.23 2.04
C-20 90.45 26.50 4.05 0.42 0.17 1.28 1.87
Igman 92.46 25.14 5.45 0.31 0.12 0.87 1.30
Zlatko 91.27 23.82 3.65 0.47 0.19 1.32 1.99
Slavonski zeleni 91.05 27.55 4.20 0.35 0.19 1.48 2.02
Mean 91.05 25.92 4.30 0.43 0.19 1.47 2.10
As dietary fibers are compounds from edible parts of plants, resistant to digestion and
absorption in human intestine, and prone to complete or partial fermentation in human
colon, they have a growing importance in the diet of modern people, being exposed to
stress and environmental pollution. In this research, cellulose, pectic acid and protopectin
are characterized as insoluble, and pectin as soluble dietary fiber.
106
APTEFF, 40, 1-220 (2009) UDC: 664.844:635.652:543.641
DOI: 10.2298/APT0940103V BIBLID: 1450-7188 (2009) 40, 103-110
According to the literature, dry beans contain 15-25% of TDF (19, 20, 21, 6), which
was also confirmed by this research (Table 3). Examined dry bean varieties contained
17.98-33.76 % (in dry matter), with mean value of 25.92%. According to the results of
this work, dry beans can be characterized as rich in dietary fibers.
Dry beans belong to cellulose-rich foods. Vasić et al. (22) reported cellulose content
in dry beans of 3.83-5.43%, Granito et al. (21) 4.65-5.61%, while Tepić et al. (5) reported
3.47-3.89%. In this research, cellulose content in dry beans was in the range from 3.18%
for Jovandeka to 5.04% for Igman. The mean value of 4.30% agrees well with the
literature data.
Pectin compounds are of polysaccharide origin, and are considered as soluble fibers.
They can be found in all fruits and vegetables. Pectin compounds have beneficial effects
in human organism, as lowering fats and cholesterol absorption, influencing the main-
tenance of glucose level in blood, increase the feces mass, thus preventing cardiovascular
and digestion system carcinogenic diseases, etc. (9, 8). Because of the importance of
pectins, dry beans should be included in the diet.
Among examined dry bean varieties, Igman contained the least pectin compounds
(1.21%) (Table 3). Panonski Tetovac was the variety richest in pectin compounds, with
2.91%. In average, the examined dry beans had 2.10% of total pectin compounds. Pectin
content was the lowest for Igman, and highest for Aster (0.30 and 0.60%, respectively);
lowest and highest pectic acid content was for Igman (0.11%) and Aster (0.24%), res-
pectively; the poorest in protopectin was Igman (0.80%) and richest Panonski Tetovac
(2.25%).
The mutual correspondence between different pectin compounds is more obvious
after the analyses of correlation between all features (Table 4). All pectin compounds are
in significant correlation with total pectin content, with protopectin being in almost com-
plete correlation (r = 0.95). Pectic acid and pectin are also in high correlation. However,
they are not in correlation with protopectin.
Table 4. Pearson's coefficients of correlation between examined dry bean features

Dry matter
Seed shape
Pectic acid
Seed color
1000 seed
Cellulose
Habitus
pectins
Pectin
Total
mass
TDF
Origin 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25
Habitus 1.00
Seed color -0.29 1.00
Seed shape 0.44 -0.22 1.00
1000 seed
- - 0.36 1.00
mass
Dry matter - - - 0.51* 1.0
TDF - - - -0.21 - 1.00
Cellulose - -0.60* 0.35 0.42 0.21 - 1.00
Total
- -0.29 0.24 - -0.28 - - 1.00
pectins
107
APTEFF, 40, 1-220 (2009) UDC: 664.844:635.652:543.641
DOI: 10.2298/APT0940103V BIBLID: 1450-7188 (2009) 40, 103-110
Table 4. Continuation
Total pectins
Dry matter
Seed shape
Pectic acid
Seed color
1000 seed
Cellulose
Habitus
Pectin
mass
TDF
Pectin 0.23 -0.23 0.43 0.25 -0.23 - - 0.52* 1.00
Pectic acid 0.28 -0.23 0.21 - -0.24 - - 0.57* 0.70* 1.00
Protopectin -0.22 -0.25 - -0.27 -0.23 -0.21 - 0.95* 0.23 0.35
*p = 0,05, r = 0.05
Among mutual correlation between pectin compounds, only two more signifficant
correlations were observed – (dry matter content : 1000 seed mass) and (seed color :
cellulose content) (Table 3). The dependence between seed color and cellulose content
was also observed in previous investigations (3, 18, 16), especially when they were more
detailed and connected with edible and technological quality of seeds (4, 17).
In the research aiming at examining a larger number of genotypes, the most effective
way of perceiveing the whole set of data is the use of some methods of multivariate
analyses. The hierarhical cluster method (Single linkage method or nearest neighbor) of
multivariate analysis was used to classify the tested varieties according to the chemical
composition of all dietary fiber, out of origin, type of habitus and seed color and shape.
The dendogram or Cluster Tree (Figure 1.) was constructed using the Euclidien distance.
Fig. 1. Dendogram of connection of examined dry bean varieties depending on their dry
matter and dietary fiber content
The distances are not high (1.5), but from cluster tree, there are clearly distinguishe-
able three groups, with four members and four genotypes, which make a separate group.
In each separate group, one genotype with colored seed, of domestic and foreign variety,
of different type of growth is placed, which points out that TDF content was not in
correlation with the main morphological features of dry beans. The dendogram starts with
108
APTEFF, 40, 1-220 (2009) UDC: 664.844:635.652:543.641
DOI: 10.2298/APT0940103V BIBLID: 1450-7188 (2009) 40, 103-110
the group of varieties with lowest TDF content, and ends up with the group of varieties
containing a maximum of TDF. The most distant, i.e. the most different from other
varieties, was Belko, with the highest TDF, and Igman with highest dry matter.
CONCLUSION
According to the correlation between pectin compounds content in dry beans, their
mutual correspondence was observed. Two more significant correlations, between dry
matter-to-1000 seed mass, and seed colour-to-cellulose content, were also noticed. The
hierarchical cluster method of multivariate analyses showed that TDF content was not in
correlation with the main morphological features of eleven examined dry beans varieties,
which should be a subject of further research in dietary fiber distribution in dry bean
seeds. However, on the basis of total dietary fiber content, dry beans can be characterized
as dietary fiber-rich food.
ACKNOWLEDGEMENTS
This research is part of the Project No. 20077, supported by the Ministry of Science
and Technological Development of the Republic of Serbia.
REFERENCES
1. Abreu, J.M.F.A., and M. Bruno-Soares: Chemical composition, organic matter di-
gestibility and gas production of nine legume grains. Animal feed science technology.
70 (1998) 49-57.
2. Todorović, J., Vasić, M., and V. Todorović: Pasulj i boranija. Grafomark, Laktaši,
Republika Srpska (2008) p. 333.
3. Kojnov, G.: Fasulat v Balgarija, Sofija (1973) p. 263.
4. Kelly, J. D., Kolkman, J. M., and K. Schneider: Breeding for yield in dry bean (Pha-
seolus vulgaris L). Euphytica. 102 (1998) 343-356.
5. Tepić A., Vujičić B., Vasić M. and A. Lučić: Amino acids and phytic acid in some
Serbian varieties of dry beans (Phaseolus vulgaris). 2nd International Congress on
Food and Nutrition, Istanbul, Turkey, 24-26 October 2007, Book of Abstracts, p 170.
6. Costa G., Quiroz-Monici K., Reis S., and A. de Oliveira: Chemical composition, die-
tary fibre and resistant starch contents of raw and cooked pea, common bean, chick-
pea and lentil legumes. Food Chem. 94 (2006), 327 – 330.
7. Thebaudin, J.Y., Lefebvre, A.C., Harrington, M., and C.M. Bourgeois: Dietary fibres:
nutritional and technological interest. Trends in Food Sci. & Technol. 8 (1997) 41-48.
8. McCleary, B.V., and L. Prosky: Advanced Dietary Fibre Tehnology, Blackwell Sci-
ence (2001) pp 63-76.
9. Anderson J., Smith B., and C. Washnock: Cardiovascular and renal benefits of dry bean
and soybean intake, Am. J. Clin. Nutr. 70 (1999) 464S – 474S.
10. EC/GR, Europen Phaseolus Database, 1998: Descriptor for the European Phaseolus
Database data, CD-rom.
11. Pravilnik o metodama uzimanja uzoraka i vršenja hemijskih i fizičkih analiza radi
kontrole kvaliteta proizvoda od voća i povrća, Službeni list SFRJ 29/83.
109
APTEFF, 40, 1-220 (2009) UDC: 664.844:635.652:543.641
DOI: 10.2298/APT0940103V BIBLID: 1450-7188 (2009) 40, 103-110
12. AOAC Official Method 985.29. Total Dietary Fibre in Foods, Enzymatic-Gravimetric
method.
13. Vračar, Lj.: Priručnik za kontrolu kvaliteta svežeg i prerađenog voća, povrća i pečurki
i osvežavajućih bezalkoholnih pića. Tehnološki Fakultet, Novi Sad (2001) pp 79-82.
14. Dische, Z.: Modification of the carbazole reaction of hexaronic acids for the study of
polyuronides. J. Biol. Chem. 183 (1950) p. 489.
15. International Federation of Fruit Juice Producers. I.F.J.U. - Analyses 26 (1964) 1-6.
16. Vasić Mirjana: Genetička divergentnost pasulja. Genetic divergence in a bean collec-
tion. Zadužbina Andrejević, Beograd (2004) p. 94.
17. Vasić, M., Mihailovic, V., Mikic, A., and J. Gvozdanović-Varga: Serbian bean mar-
ket classes. 6th European Conference on Grain Legumes "Integrating legume biology
for sustainable agriculture", November 2007, Lisbon, Portugal, Book of Abstracts, p.
117.
18. De La Cuadra, C., De Ron, A.M., ans R. Schachl (editors): Handbook on evaluation of
Phaseolus germplasm, PHASELIEU-FAIR-PL97-3463, Misión Biológica de Galicia,
Spania (2001) 109.
19. www.usaid.gov
20. Garcia O., Infante R., and C. Rivera: Determination of total, soluble and insoluble
dietary fibre in two new varieties of Phaseolus vulgaris L. using chemical and enzy-
matic gravimetric methods. Food Chem. 59 (1997) 171-174.
21. Granito, M., Michel, C., Frías, J., Champ, M., and M. Guerra: Fermented Phaseolus
vulgaris: acceptability and intestinal effects. Eur. Food Res. Technol. 220 (2005) 182-
186.
22. Vasić, M., Gvozdanović-Varga, J., and J. Navalušić: Determining chemical composi-
tion of bean seed by multivariate analysis. Proc. XXXIV ESNA annual meeting. Novi
Sad, Serbia and Montenegro (2004) pp. 300-304.
САДРЖАЈ ДИЈЕТЕТСКИХ ВЛАКАНА У НЕКИМ СОРТАМА ПАСУЉА
Александра Н. Тепић, Мирјана А. Васић, Бисерка Л. Вујичић,

Јелица М. Гвоздановић-Варга и Здравко М. Шумић
Дијететска влакна се сматрају основним хранљивим компонентама, заједно са

протеинима, мастима, угљеним хидратима, минералима и витаминима. Један од
основних параметара технолошког квалитета и нутритивне вредности пасуља је и
садржај дијететских влакана. У раду су испитане физичке карактеристике и садржај
основних компоненти хемијског састава шеснаест сорти пасуља селекције Научног
института за ратарство и повртарство, са посебним освртом на садржај дијететских
влакана. Статистичком анализом утврђена је корелација између појединих пара-
метара хемијског састава.

110
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
THE EVOLUTION OF LACTIC ACID BACTERIA COMMUNITY DURING

THE DEVELOPMENT OF MATURE SOURDOUGH
Tanja D. Žugić-Petrović, Nataša M. Joković and Dragiša S. Savić
In order to follow the composition and changes in lactic acid bacteria (LAB) po-
pulation of rye flour sourdough that was continuously propagated by a repeated inocu-
lation, sixty-two strains of LAB were isolated and characterized. The LAB were the only
bacteria detected, both at the end of the second propagation step and in the stage of
mature sourdough (after two weeks of continuous daily refreshment). The stable ecolo-
gical system in rye sourdough could be established from the second propagation step
onward. The predominant genera of LAB during the development of sourdough were lac-
tobacilli, which were grouped in eight clusters. Heterofermentative lactobacilli were in
majority in both propagation step two and a mature sourdough participating 56% and
70% of total bacterial count, respectively. The identification based on a phenotypic cha-
racterization that was carried out by using a set of 36 tests, showed that the lactobacilli
contained in the two sourdough steps did not clearly belong to any known species of the
genus Lactobacillus. In addition, the structure of the bacterial population were monito-
red by two statistical techniques (Hierachical Cluster Analysis and Principal Component
Analysis), being applied to phenotypical characteristics of the isolates.
KEY WORDS: Lactic acid bacteria, sourdough, Hierachical Cluster Analysis, Principal
Component Analysis
INTRODUCTION
Sourdough fermentation is a process in which a mixture of flour and water is fermen-

ted with lactic acid bacteria (LAB) and yeast. It is an ancient way to improve flavor, tex-
ture and microbiological shelflife of bread and has a natural, additive-free image. Both
nutritional and technological quality could be considerably enhanced in cereal foods rich
in dietary fibers by utilizing sourdough.
The LAB included in sourdough fermentation may originate from flour (spontaneous
fermentation), preceding sourdough or a commercial starter culture. Spontaneous sour-
dough fermentation begins with aerobic fermentation immediately upon mixing flour and
Tanja D. Žugić-Petrović, B.Sc., Nataša M. Joković, M.Sc., Dr. Dragiša S. Savić, Prof., savic@junis.ni.ac.rs,
Laboratory for Food Science and Biotechnology, Faculty of Technology, Bulevar oslobodjenja 124, 16000
Leskovac, Serbia
111
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
water. Once oxygen is depleted, anaerobic fermentation begins with the growth of LAB
and yeasts. The production of acids by LAB enables their rapid growth when the pH
value has dropped to low for other microorganisms to grow. So, the LAB become the
most abound microflora in the sourdough, and they are therefore responsible for the final
stages of sourdough (1, 2). This type of processing is still the basic practice for preparing
so-named predoughs. When the fermented dough is used as an inoculum for a succeeding
fermentation run, the adaptation of the microbial assotiation to the process becomes
greater (2).
The microflora of spontaneously fermented dough depends on the microflora of raw
materials, and is variable in terms of kind, origin and storage conditions of the flour, as
well as the technological parameters of the fermentation process applied. The impact of
these parameters during a continuous propagation of sourdough causes the selection of a
characteristic microbiota (3). The wholemeal rye flour may contain 104-106 cfu of unspe-
cified bacteria per gram (1, 4, 5), in which 102 - 103 cfu g-1 belong to LAB (5). According
to our previous research (1), LAB participated in one third of rye flour bacterial
populations were detected on MRS agar. In mature sourdough, LAB ranged between 107
and 108 cfu g-1 (6).
Microbiological studies have revealed that more than 50 species of LAB occur in
sourdough (2). Sourdough LAB originate generally from the genera Lactobacillus,
Leuconostoc, Pediococcus or Weissella and the majority of the strains belong to the ge-
nus Lactobacillus (3). Since sourdough is a suitable substrate for most lactobacilli, more
than half of the species in this genus occur in sourdoughs or in related cereal fermen-
tations (7). Homofermentative (Lb. acidophylus, Lb. delbrueckii subsp. delbrueckii, Lb.
amilovorus), as well as heterofermentative (Lb. sanfranciscensis, Lb. panis, Lb. pontis)
lactobacilli, were found in rye spontaneously fermented sourdoughs (8, 9).
The isolation of the strains from microbial communities and their identification is still
a necessary approach when the characterization of physiological and technological pro-
perties of microbial community members is of interest, as in the case of sourdoughs.
Sourdoughs are microbial systems that have not been thoroughly investigated and new
species of LAB may occur in this ecosystem due to the continuous propagation in the
sourdough system (9). In this paper, the composition and changes in LAB microflora of a
rye flour sourdough that was continuously propagated by a repeated inoculation, were
examined. The results represent the resumption of those previously obtained (1). Statis-
tical procedures applied to physiological characteristics of the isolates were used to re-
present the composition of LAB and detect physiological factors affecting their differen-
tiation.
MATERIALS AND METHODS
Sourdough production and propagation
Doughs were made by mixing 100 g rye flour ("Žitopromet" Zaječar, Serbia with
11.7 % and 0.99 % water and ash content, respectively) and 60 cm3 of sterile tap water in
aseptic conditions (prior to work a mixer dish and blades were moistened with ethanol
and flamed). Sourdough formation started by a spontanous fermentation followed by se-
112
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
veral propagation steps using 10% of previously fermented dough (calculated on a total
dough weight) as inoculum for the next fermentation. Each of the sourdoughs were
fermenting for 24 h at 30oC.
Enumeration and isolation of lactic acid bacteria
For isolation end enumeration of the bacteria, 10 g of dough were homogenized with
90 cm3 of sterile 0.85% saline. Serial dilution was spread plated on MRS (Torlak, Bel-
grade, Serbia) and incubated under the anaerobic incubation (GasPak, BBL, Cockeys-
ville) at 30oC for 48 h.
The isolation of bacteria was done both at the end of the second propagation step and
after two weeks of a daily refreshment (the mature sourdough stage). After enumeration,
the colonies were randomly isolated from MRS plates, transferred to MRS broth and,
after the incubation (48 h, 30oC), purified three times by streaking on the MRS agar and
then checked for morphology, Gram stain and catalase test (determined by transferring
fresh colonies from agar medium to a glass slide and adding 5% H2O2). Each of Gram
positive and catalase negative cultures were separated for further examination.
Physiological characterisation of lactic acid bacteria

A set of 36 tests (including morphology, Gram staining characteristic and a catalase
test) was used to clasify the isolates. The tests used to determine catalase activity, gas
production, arginine and esculine hydrolisis, the ability to acidify, cloth and reduce lac-
mus milk (1%), growth at different temperatures (15oC and 45oC) and the concentration
of NaCl (4, 6.5 and 8%) were performed by using previously described methods (6, 10).
Other tests were observing the growth on entero and citrate agar (HIMEDIA, Mombai,
India) and the ability to form diacetyl (11) and exopolysaccharides (formation of slimy
colonies on MRS agar with sucrose as carbon source, 20 g/dm3).
Acid production from carbohydrates (L-arabinose, D-xylose, galactose, mannitol, tre-
halose, mannose, raffinose, lactose, maltose, sucrose, glucose, fructose, rhamnose, sorbo-
se, ribose, salicin, cellobiose, melibiose, inuline, sorbitol, Sigma) was evaluated by the
procedure as follows: filter sterilized solution of sugar was added to basal MRS medium
(without glucose) and with 0.16 g/dm3 bromcresol-purple to a final concentration of 10
g/dm3. The filter sterillized (0.22 μm, Millipore, Saint-Quentin, France) sugar solution
was dispensed (0.9 cm3) in the microtube (1.5 cm3). The cells suspension (0.1 cm3) obtai-
ned by centrifuging 5 cm3 of 16-h old MRS broth culture and resuspension of the sedi-
ment in 5 cm3 sterile saline was used to inoculate microtube. The apperance of yellow
color in medium after the incubation (48 h at 30oC) was considered as a positive result.
Determination of TTA and pH
Total titrable acidity (TTA) and pH of sourdough were determined on an aliquot of 10
g sourdough, blended with 90 cm3 destilled water. The pH of this suspension was deter-
mined by using HANNA HI 9025 meter. For TTA determination, the same aliquot was
titrated against 0.1M NaOH to final pH 8.5. TTA was expressed as the amount (cm3) of
NaOH used.
113
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
Statystical analysis
The relationship among the isolated strains was determined by Hierachical Cluster
Analysis (HCA) and Principal Component Analysis (PCA) using Statistica 7.0 (StatSoft
Inc. USA) for Windows. The results of phenotypical tests were coded as + positive, - ne-
gative or +- weak positive or delayed (positive after 7 days of incubation) reaction. For
morphology, three codes were used – cocci, + long bacillus cells and +- common bacillus
cells. HCA was carried out using the algorithm Unweighed Pair-Group „Average Link-
age Analysis“. Distances between the clusters were assessed using „Percent of dissagre-
ement« and its translation in »similarity level“ assuming that 0% disagreement=100% si-
milarity. PCA was implemented using the state-of-the-art algorithm known as NIPALS
(None Linear Iterative Partial Least Squares).
The total cell count of the continuous rye flour fermentations was determined over 3
propagation steps (PS) and in the stage of mature sourdough (MD) after two weeks of
daily propagations. The additional characterization was achieved by the pH and TTA
measurement at the end of the first three propagation steps, as well as in mature sour-
dough (Figure 1). The significant increase of the number of bacteria was observed only at
the end of the first 24 h fermentation after flour and water mixing, reaching the level of
ca 8 log cfu g-1 which varied slightly during the propagation period and the formation of
the MD (Figure 1). The pH values that reached 3.7 at the end of PS2 were maintained till
the end of the study, and TTA were ca 13 from the second refreshment onward (Figure
1), thus demonstrating the acid production in the sourdough.
Fig. 1. The kinetics of pH (■), TTA (▲) and the number of lactic acid bacteria (•) in start
dough (SD), sourdough et the end of propagation steps (PS1, PS2 and PS3) and mature
sourdough (MD) after two weeks of continuous daily propagations
The acidity, the viable count, and the composition of LAB microflora during repeated
succeeding spontaneus fermentations were related to the first 2-propagation step. The lac-
114
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
tic acid bacteria did not dominate microflora in the initial dough (as a consequence of not
dominating in rye flour, as well), making one-third of bacterial population that can be
detected on MRS agar (Figure 2A), but their acid resistance enabled them to continue
growing during a natural sourdough fermentation. The LAB dominated the entire sour-
dough microflora (ca 80%, Figure 2A) at the end of the first propagation step, but maxi-
mum values of acidification were not achieved yet. Maximum acidification was achieved
and varied slightly after the second propagation steps (Figure 1) when LAB were the only
bacteria found in sourdough (Figure 4A). The number of bacteria found in sourdough
(8.0-8.7 log cfu g-1) during all propagation steps was in accordance with previously pu-
blished results where counts of soudoughs LAB determined by enumeration on MRS ran-
ged between 107-108 (6, 9). It can be assumed that the stable ecosystem of mature sour-
dough could be established from the second propagation step onward, when the pH and
total titrable acidity of sourdoughs were ca 3.5 and ca 13, respectively as the consequence
of the action of LAB that were at the level of 8.0-8.7 log cfu g-1.
Fig. 2. Participation of LAB in microflora determined on MRS (A), and particule genera
in lactic acid bacteria population (B) in rye flour - RF, propagation steps 1 - PS1 (1) and
2 - PS2 and mature sourdough - MD
On the basis of the tests applied, sourdough anaerobe and/or facultative anaerobe iso-
lates fall within well-recognised LAB genera. The strains of genera Enterococcus found
in rye flour remained in the first two propagation steps, and the participation of entero-
cocci in LAB count were significantly reduced from 70% in rye flour to 3% in PS2, and
were not detected in MD (Figure 2B). Some enterococi were reported as common inhabi-
tants of vegetables (12) and were isolated from rye flour (E. faecium, E. avium, E. casse-
liflavus and E. durans - 4) and sourdoughs (4, 6).
Greater LAB diversity was observed at the end of the first 24 h fermentation and 6
genera were detected: Enterococcus, Streptococcus, Leuconostoc, Weisella, Pediococcus
and Lactobacillus (1). It was shown (16) that in the sourdough obtained by continuous
daily refreshments at 30oC, subdominant LAB such as E. faecium and P. pentosaceus are
stronger acidifiers than Lb. sanfranciscensis at the beginning of the sourdough produc-
tion. Those species inhibiting indigenous microorganisms other than LAB by lowering
115
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
the pH, might prepare the environment for the establishment of typical species (e.g.
Lactobacillus spp.) of mature sourdoughs. Their most crucial role was played at the be-
ginning of the sourdough production (13). Nevertheless, once lactobacilli colonized
dough this determined a rapid decrement of LAB cocci which remained at subdominant
concentrations or at the concentration that enabled their detection (13). There is no work
reporting on the prevalence of enterococci strains in the sourdough environments at the
stage of mature sourdough.
After the first propagation step, lactobacilli were detected totaling up to one-third of
the entire LAB community and all of these were homofermentative. After the second
refreshment, lactobacilli comprised the greatest group of LAB (97%), while MD micro-
flora consisted only of the strains belonging to the genus Lactobacillus (Figure 2B). This
confirmed the already published reports (9, 6, 14) that rye and wheat sourdoughs micro-
bial populations were dominated mainly by bacteria of the genus Lactobacillus. The he-
terofermentative lactobacilli were first detected at the end of PS2 (56%) and in MD par-
ticipate with 70 % of bacterial count. With the increasing fermentation time, a shift
towards the predominance of hetrofermentative lactobacilli was reported in rye sour-
dough (14). A rather stable association of lactobacilli was a result of the selective pres-
sure exerted by the environmental conditions through a continuous propagation, adding
flour and water at regular intervals (14).
The identification of lactobacilli strains was not possible because they did not con-
form to any species description. A high percentage of disagreement was accounted bet-
ween our isolates and the obligately homofermentative (Lb. amilovorus and Lb. acido-
phylus), facultatively heterofermentative (Lb. plantarum) and obligately heterofermen-
tative (Lb. sanfransciscensis, Lb. brevis, Lb. fermentum, Lb. fructivorans, Lb. pontis, Lb.
panis) lactobacilli commonly isolated from sourdoughs. The main goal of this paper was
not to fully identify the isolates, since it is well known that it is very difficult to distin-
guish lactobacilli by their physiological properties (9, 16), especially in fermentations of
non-sterilized substrates. Environmental parameters lead to a well-adapted, stable flora
within this dough, and physiological properties could not be assorted to reference orga-
nisms from other habitats as a result of adaptation to different environments.
Statistical procedures based on phenotypic properties have been commonly used for
the analysis of sourdough microbial communities (6, 16). The techniques appeared to be
very effective in representation of the composition and a relative position of LAB com-
munities. In fact, in most papers statistical analysis was based on clustering of isolates
and on the presentation of distribution of isolates in graphical formats. This kind of re-
presentation had significant advantages, allowing a direct comparison between the groups
and detecting physiological factors affecting the differentiation.
Thirty-two and thirty strains were isolated at the end of PS2 and MD, respectively.
All of these were Gram-positive and catalase-negative, and all, but one, were rod-shaped.
The one remaining LAB isolate was coccus-shaped.
The similarity among strains of LAB isolated from PS2 and MD, as well as from
propagation step 1 (our previous work, 1), is summarized in the dendrogram shown in Fi-
gure 3, while the phenotypic characteristics of the clusters are summarized in Table 1.
Nine diferent LAB clusters were identified at the 80% similarity level. Within the
clusters, the profiles were not identical but very similar.
116
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
Fig. 3. Dendrogram of similarities based on phenotypic tests of 80 LAB strains isolated at

the end of propagation steps 1 (1) and 2, and from mature sourdough. Note: Grouping
was performed by Hierachical Cluster Analysis using „Unweighed Pair-Group Average
Analysis“ algorithm
All isolated lactobacilli grown at 45oC produced acid on glucose, maltose and sucrose
and could not be grown on entero agar (Table 1). The heterofermentative lactobacilli of
PS2 and MD dominated LAB microflora with 56% and 70% of total bacterial count,
respectively. Only the homofermentative lactobacilli of cluster VII showed the imposi-
billity of growing both in milk and fermenting lactose (Table 1). Eight strains of cluster
VIII were the only lactobacilli that could grow in the presence of 6.5 % and 8 % NaCl. It
was not possible to identify lactobacilli isolates on the species level scoring the results of
the tests applied with the species literature description (6, 8, 17).
One of two lactobacilli types of isolates from PS1 clustered together with isolates
from PS2 and MD with ca 80% interspecies similarity level (cluster VII on Figure 3 and
Table 1). The lactobacilli of this cluster from PS2 and MD explicitly differed from those
from PS1 (1) in producing acid from mannitol, as well as in disability to grow in the pre-
sence of 8% NaCl (Table 1). Another lactobacilli cluster from PS1, containing only two
isolates, was separated in cluster I and showed to be most related to cluster II (ca 70 %
similarity level) consisting of five isolates from PS2.
The coccoid isolate from PS2 deemed to belong to genus Enterococcus according to
its ability to hydrolise esculine and to grow at 15oC and 45oC, as well as to grow in entero
agar and in the presence of 6.5% NaCl (12, 18, 19). The identification of enterococcus
was not possible because they could not be related to known species description. Ente-
rococci isolated from PS1 and PS2 were comprised in one cluster with ca 78% interspe-
cies similarity level (Figure 3). When compared to those from PS1 (1), the enterococci
isolated from PS2 showed to be different in fermentation mannose and raffinose.
117
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
Table 1. Physiological and biochemical characteristics of the clusters of the lactic acid
bacteria isolates from sourdoughs after propagation step 1 (1) and 2 and from mature
sourdough
Cluster No I II III IV V VI VII VIII IX
Number of LAB1 2+0+0 0+5+0 0+4+6 0+5+8 0+6+5 0+3+2 8+5+4 0+3+5 8+1+0
Morphology2 B B B B B BL B BS C
CO2 from glucose - - + w w w - - -
Growth at 45oC + + + + + + w* w +
15oC - - -* - - + + + +
Arginine hydrolysis - - - + + + - - +
Esculin hydrolysis - - -* - w* - - + +
Growth with
+ + +* -* + + + + +
4%NaCl
6.5%NaCl - - - - - - w* + +
8%NaCl - - - - - - -* + -
Growth in milk3 +a +ac -a* +acr* +acr* - -* +acr +acr
Entero agar - - - - - - - - +
Citrate agar - - - - - + -* + -*
Diacetile - - - - - - -* -* +*
EPS production - - - - - r - r -
Acid from
L-arabinose + + + + +* + - - w
D-xilose - w - -* w + - - -
galactose + + + +* + w + + +
mannitol - w - - + - +* + +*
trehalose + - - - +* - -* - +*
mannose +* - - - - - + + +*
rafinose + + + + + + - - +*
lactose + + + + + + - w +*
fructose + + - - + - + + +
rhamnose - - - - - - - - -
sorbose - - - - - - - - -
ribose - - -* w* w + -* +* -
salicine -* - - - - - - - +
celobiose w +* - - + - - + +
melibiose w w + +* + w - - +
inuline w w* - - +* - -* + +
sorbitol - - - - - - - + -
+ positive, - negative, w - weakly positive, All strains produce acid from maltose, sucrose and glucose.
1
the first number represents number of isolates from propagation step 1, the second from propagation step 2, and the third
from mature sourdough
2
C: cocci, B: rods, BL: long rods
3
acidification (a), clothing (c) and reducing (r) of lacmus milk, 1%
* - properties differ among strains of the same type
In addition, PCA was performed in order to evaluate the similarity among lactobacilli
isolates and detect physiological characteristics affecting their differentiation. The phy-
siological characteristics for isolated strains showed that four principal components (PC)
were able to explain 50 % of the variance (data not shown). The plot of the first two
components (Figure 4) made it possible to separate the 71 lactobacilli strains (10 from
PS1, 31 from PS2 and 30 from MD) into distinct groups (Figure 4A). Enterococus cluster
118
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
was explicitly separated from all lactobacilli (data not shown) and was not included in
further statistical analysis.
Figure 4. Scores plot (A) and loadings plot (B - the loadings less than 0.5 not presented)
from PCA of the physiological properties of lactobacilli isolated at the end of propagation
steps 1 (1) and 2, and from mature sourdough Note: The explained variance by PC1 and
PC2 were 21 % and 12 %, respectively. Numbers represent clusters from Figure 2
The clusters formed by HCA analysis (Figure 3) can be observed on plots from the
PCA (Figure 4A). Each cluster of strains was represented by the prediction interval
ellipse for its samples, as a measure of dispersion. The coordinates for the ellipse were
computed from the data and a number of observations given, and showed the prediction
interval for a single new observation (13). The prediction interval ellipse desribes the area
in which a single new observation can be expected to fall with the probability of 90 % (a
coefficient that controlled the ellipse was chosen to be 0.9). As shown in Figure 4A,
lactobacilli from clusters I and II formed by HCA could not be clearly separated by PCA
and made a single ellipse (I-II) representing related isolates.
The Principal component 1 (PC1) that explained 21 % of total variance in the data,
clearly separated clusters VII and VIII from other lactobacilli types, while PC2 (account
12 % of total variance) separated clusters I-II, III, VI and VII from clusters V and VIII
(Figure 4A). According to Figure 4B, PC1 distinguished the strains according to CO2
production, growth at 15 oC and 45 oC, growth in medium containing 6.5 % NaCl, as well
as the acid production from mannose, raffinose, arabinose, fructose and mannose. PC2
differentiated the strains according to their growth in milk and fermentation of celobiose.
Only PC3 which explained 15% of the variation (data not shown) was defined by the
cell shape (long or common rods).
The growth in medium supplemented with 4% NaCl and acid formation from galac-
tose, trehalose, sucrose, galactose, salicin and ribose, as well as growth on citrate agar,
had low loadings on PC1 and PC2 (close to zero). Therefore, it could be concluded that
these properties had no significant influence on the attained classification of isolates. Ne-
119
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
vertheless, the ribose and citrate fermentation had a significant influence on PC3 and
PC4, respectively (data not shown) and they could not be omitted from further studies of
the classification of isolates.
The lactobacilli of cluster VII were assembled in the upper left part of the graph (Fi-
gure 4A) that displayed homofermentetative bacilli with the ability to ferment fructose
and mannose but not lactose, raffinose, melibiose, xilose and arabinose (Figure 4B, Table
1). The strains grouped in the lower left part of the graph represented cluster VIII. The
isolates from this cluster showed growth in 8% NaCl, with the ability to hydrolyze escu-
lin, ferment inuline, and some strains had the potential to produce diacetile (Figure 4B,
Table 1).
Six strains isolated from PS2 and five strains from MD were located in the lower
right part of the graph (cluster V). They displayed a high activity in milk and arginine
hydrolysis. Two thermophilic isolates from PS1 and the majority of strains isolated after
PS2 (clusters I-II, III, IV and VI) which were located in the upper right part of the graph,
were characterized by disability to grow in 8% NaCl, to ferment sorbitol and to form
diacetyl.
CONCLUSION
In this study, the statistical analyses provided indices for the evaluation of the distan-
ces among the population members and their differentiation. Principal component 1 from
PCA succeeded in differentiating homo- from heterofermentative isolates with the excep-
tion of only cluster I-II. This cluster comprised homofermentative lactobacilli and was
located among other heterofermentative clusters (III, IV, V and VI). PCA differentiated
strains on the base of, mainly, pentose fermentation (Figure 4B) and the bias was intro-
duced because the strains from cluster I-II ferment arabinose (Table 1). Although the
strains may differ, pentoses (arabinose, xylose or ribose) were usually fermented by obli-
gately heterofermentative and seldom by homofermentative lactic acid bacteria. Further
investigation of the isolates (for example DNA-DNA hybridization, rRNK sequencing
and GC spectrum) is necessary to reveal whether the isolates belonged to some hitherto
unknown species of the genus Lactobacillus.
ACKNOWLEDGEMENT
This research was supported under the project PTR 2042 by the Ministry of Science
and Environmental Protection of the Republic of Serbia.
REFERENCES
1. Savić, D., T. Savić, M. Škrinjar and N. Joković: Profile of lactic acid bacteria in rye
flour and sourdough. J. Culture Collect. 5 (2007) 38-45.
2. Hammes, W.P., M.J. Brandt, K.L. Francis, J. Rosenheim, M.F.H. Seitter and S.A.
Vogelmann: Microbial ecology of cereal fermentations. Trends Food Sci. Technol. 16
(2005) 4-11.
3. Arendt, E. K., L. Ryan, and F. Dal Belo: Impact of sourdough on the texture of bread.
Food Microbiol. 24 (2007) 165-174.
120
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
4. Rocha, M. and X. Malcata: On the microbiological profile of traditional portoguese

sourdough, J. Food Protect. 62 (1999) 1416-1429.
5. Salovaara, H.: Lactic acid bacteria in cereal-based products, in Lactic acid bacteria
Microbiological and Functional Aspects, Eds. S.Salminen, A. von Wright and A.
Ouwehand. New York. USA: Marcel Dekker. (2004) pp. 431-451.
6. Ricciardi, A., E. Parente, P. Piraino, M. Paraggio and P. Romano: Phenotypic charac-
terisation of lactic acid bacteria from sourdoughs for Altamura brad produced in Apu-
lia (southern Italy). Int. J. Food Microbiol. 98 (2005) 63-72.
7. Gänzle, M.G., N. Vermeulenb, and R.F. Vogel: Carbohydrate, peptide and lipid meta-
bolism of lactic acid bacteria in sourdough. Food Microbiol. 24 (2007) 128-138.
8. Müller, M., M. Ehrmann, and R.F. Vogel: Lactobacillus frumenti sp. nov., a new lac-
tic acid bacterium isolated from rye-bran fermentations with a long fermentation pe-
riod. Int. J. Syst. Evolut. Microbiol. 50 (2000) 2127-2133.
9. Rosemquist, H. and A. Hansen: The microbial stability of two bakery sourdoughs ma-
de from conventionally and organically grown rye Food Microbiol. 17 (2000) 241-
250.
10. Strohmar, W. and H. Diekmann: Die Mikroflora eines Langzeit-Sauerteiges. Zeit.
Lebens. Untersuch. Forscg. 194 (1992) 536-540.
11. Arnau, J., F. Jorgensen, S. Madsen, A. Vrang and H. Israelsen: Cloning, expression,
and characterization of the Lactococcus lactis pfl gene, encoding pyruvate formate-
lyase. J. Bacteriology. 179, 18 (1997) 5884–5891.
12. Anonymous: Nomenclature et principales sources d'isolement des espèces du genre
Enterococcus (http://www.bacterio.cict.fr/bacdico/ee/tenterococcusisolement.html)
13. StatSoft, Inc. (2006). Electronic Statistics Textbook. Tulsa, OK: StatSoft. WEB:
http://www.statsoft.com/textbook/stathome.html
14. Corsetti, A., L. Settanni, S. Valmorri, M. Mastrangelo and G. Suzzi: Identification of
subdominant sourdough lactic acid bacteria and their evolution during laboratory-
scale fermentations. Food Microbiol. 24 (2007) 592–600.
15. Müller, M., G. Wolfrum, P. Stolz, M. Ehrmann, and R.F. Vogel: Monitoring the
growth of Lactobacillus species during a rye flour fermentation. Food Microbiol. 18
(2001) 217-227.
16. Parente, E. and A. Ricciardi: A statistical procedure for the analysis of microbial
communities based on phenotypic properties of isolates. J. Microbiol. Meth. 49
(2002) 121–134.
17. Hammes, W. and C. Hertel: The genera Lactobacillus and Carnobacterium, in The
Procaryotes (vol 4), Ed. M. Dworkin, Springer, London (2007) pp. 320-403
18. Manero, A. and A. Blanch: Identification of Enterococcus spp. with a Biochemical
Key. App. Environ. Microbiol. 65, 10 (1999) 4425-4430.
19. Leblanc, D.: Enterococcus, in The Procaryotes (vol 4), Ed. M. Dworkin, Springer,
London (2007) pp. 175-204.
121
APTEFF,40, 1-220 (2009) UDC: 664.654.1:664.64.016.3/.7
DOI: 10.2298/APT0940111Z BIBLID: 1450-7188 (2009) 40, 111-122
РАЗВОЈ ПОПУЛАЦИЈЕ БАКТЕРИЈА МЛЕЧНЕ КИСЕЛИНЕ У ТОКУ

ФОРМИРАЊА ЗРЕЛОГ КИСЕЛОГ ТЕСТА
Тања Д. Жугић-Петровић, Наташа М. Јоковић и Драгиша С. Савић
У циљу праћења састава и промена популације бактерија млечне киселине

(БМК) у киселом тесту припремљеног сукцесивним свакодневним премешавањем
теста од ражаног брашна, 62 сојева БМК је изоловано и окарактерисано. БМК су
једини микроорганизми изоловани на крају другог премешавања, као и у фази зре-
лог киселог теста (након 2 седмице свакодневног сукцесивног премешавања). Ста-
билни еколошки систем у киселом тесту припремљеном од ражаног брашна успос-
тавља се након другог премешавања. Род БМК који доминира у току развоја ки-
селог теста су лактобацили и који су груписани у 7 група. Хетероферментативни
лактобацили доминирају од другог премешавања, као и у фази зрелог киселог тес-
та, при чему учествују са 56%, односно 70% у укупном броју БМК. Идентифика-
ција на основу фенотипских и физиолошких својстава (применом 36 теста) пока-
зала је да се лактобацили иззоловани из киселих теста не могу јасно сврстати у до
сада познате врсте рода Lactobacillus. Поред тога, праћена је структура бактеријске
популације у киселом тесту применом две статистиче технике (хијерархијска
кластер анализа и анализа главних компоненти) на фенотипска својства изолата.
Received 28 August 2009

122
CHEMICAL TECHNOLOGY AND PROCESS ENGINEERING
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
DETERMINATION OF THE PHOTOCATALYTIC ACTIVITY OF TiO2

COATINGS ON CLAY ROOFING TILE SUBSTRATES - METHYLENE BLUE
AS MODEL POLLUTANT
Eva S. Lončar, Miroslava M. Radeka, Snežana B. Petrović, Andrea S. Skapin,

Ognjen Lj. Rudić and Jonjaua G. Ranogajec
The photocatalytically active mesoporous coatings, based on titanium dioxide sols

(Degussa), of the fired clay roofing tiles substrate were prepared by using poly(ethylene
glycol) (PEG) M-600 and M-4000, as the structure directing agents. The coatings were
deposited using spray technique followed by thermal treatment. Photocatalytic activity of
the TiO2 coatings was evaluated by aqueous solution of methylene blue as model dye,
deposited on the top of the coatings, after irradiation with UV light. The results were
compared with the photocatalytic efficiency of some commercial self-cleaning products
(clay roofing tiles, glass). The newly design coatings showed an interesting decolourisa-
tion performance (over 30 % after 24 h). It appeared that the procedure of photocatalytic
activity determination, in the case of porous substrates, should be renewed by a pre-
adsorption process.
KEY WORDS: ТiO2 coatings, clay roofing tile, photocatalytic activity
INTRODUCTION
Semiconductor photocatalysis (SPC) has attracted a great deal of attention over the
last 30 years. The application of semiconductor in heterogeneous photocatalysis in order
to eliminate various pollutants including many pesticides, surfactants and carcinogens in
aqueous systems and in the air (1-7), has been extensively investigated. Titanium dioxide
(TiO2 - titania) appeared to be the most suitable photocatalytic semiconducting material
due to its high stability toward photocorrosion and relatively favorable band-gap energy.
Titania occurs in three crystalline forms: brookite, anatase and rutile. Rutile is the stable
phase while anatase and brookite are metastable. Anatase TiO2 has become the foremost
semiconductor material for SPC application since it is biologically inert, mechanically
robust, relatively inexpensive and highly reactive.
Dr. Eva S. Lončar, Prof., Snezana B. Petrovic, B. Sc., Ognjen Rudic, B. Sc., Dr. Jonjaua G. Ranogajec, Prof.,
University of Novi Sad, Faculty of Technology, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia; Miroslava M.
Radeka, Assoc. Prof., University of Novi Sad, Faculty of Technical Sciences, Dept. of Civil Engineering, Trg
Dositeja Obradovica 6, 21000 Novi Sad, Serbia; Dr. Andrea S. Skapin, Ph. D., Slovenian National Building and
Civil Engineering Institute, 1000 Ljubljana, Dimičeva 12, Slovenia
125
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
Photocatalytic chemical reactions occur on the surface of an SPC material. The pho-
tocatalyst generates electron/hole pairs which are capable to initiate a series of chemical
reactions when it is illuminated by appropriate light source: UV light (λ≤387 nm) (4, 5,
8-11) or visible light (artificial solar and sunlight) (1, 2, 12-17). Pairs of photo-generated
hole (h+) and electron (e-) induce the formation of aggressive species such as hydroxyl or
superoxide radicals from the moisture and atmospheric oxygen. These species are strong
enough to oxidize and decompose organic materials or smelling gas and kill bacteria (5,
11, 13, 15-17). The process is predominantly determined by the fundamental physical
properties of the SPC material surface. Mesoporous coatings are predicted to have great
potential to increase the photocatalytic activity by enlarging the specific surface area
(18). High surface area and interconnectivity in the porous network of mesoporous coa-
tings are desirable in order to optimize the activity of the surface and provide diffusion,
charge or light transfer or reactant access into the cavities. The coatings with a greater
surface area can improve the degradation rate of organic pollutants by the adsorption and
the concentration of the reactants, but still allowing the latter to diffuse from the adsorp-
tion sites to the TiO2 surface.
To assess the photocatalytic efficiency of a TiO2 coating a broad range of pollutants,
both of organic and inorganic nature, can be used. They can be classified into three cate-
gories:
• Dyestuffs (19, 20),
• Organic compounds (21),
• Inorganic gases (22).
Dyes are degraded by TiO2 under the influence of UV or solar light. The decom-
position is assessed by decoloration measurements (color removal ratio), as well as
chromatographic investigations. The widespread application of, for example, methylene
blue (MB) originates from the fact that it is mainly nontoxic and convenient for the use as
a dye. MB exibits strong absorption in the visible light (λmax =664 nm; ε664 = 7.4×104 M-1
cm-1) but not in the UVA region. This fact presents the perceived effectiveness of the MB
test, which is considered by the International Organization of Standardization (ISO) as a
standard test for photocatalytic surfaces (16, 23). However, the assessment of decompo-
sition processes of dyes by decoloration measurements is still a subject of discussion.
The present research was focused on obtaining mesoporous coatings using TiO2 and
poly(ethylene glycol) (PEG) of different molecular weight, as a structure directing agent
applied on ceramic roofing tile surface. The functionality of these coatings have been
tested by the photocatalytic decomposition of MB in aqueous solution. These values were
compared with the activity of the commercial products (clay roofing tiles and glass
samples). Major attention has been devoted to the selection of a suitable sample pre-
paration. Namely, ceramic roofing tiles as a porous material, without photocalytic coa-
ting, also adsorb a certain quantity of MB solution decreasing its concentration. In order
to assess the contribution of the photocatalytic activity of the prepared mesoporous titania
coating, the photocatalytic tests were modified compared to the tests where a non-porous
substrate was used. The precursory adsorption of methylene blue was established until
the adsorption process of the dye was complete. After that the decomposition rate of the
dye under the UVA light irradiation (photocatalytic activity) was determined by recor-
ding its absorption spectrum.
126
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
EXPERIMENTAL
Photocatalytic dispersion. The photocatlytic dispersion, sol, was prepared by mixing

2.5 mass % TiO2 aqueous (deionised water) colloidal dispersion (VP Disp.W 2730 X,
Degussa, Frankfurt, Germany) and PEG (HOCH2(CH2OCH2)nCH2OH) with molecular
weight 600 (M-600) or 4000 (M-4000), (Baker, Germany). Titania particle size in colloi-
dal dispersion was reported to be (d-50= 50-100 nm), dry matter content 30.0 +/-1.0
mass%, and pH value 5-7.
Tile substrate. The clay roofing tiles produced in the industrial conditions (a.d. Polet,
member of Nexe group, Novi Bečej, Serbia ) were cut in the form of square-shaped slabs
(dimensions 3.5 x 3.5 x1.5 cm) and used as substrates for photocatalytic coating pre-
paration. The used raw material for the tile production was based on illite-kaolinite clay
material and carbonates (dolomite, calcite).
Preparation of TiO2 layer on clay roofing tile substrate. The photocatlyic sol was
frequently stirred in order to obtain stable coating solution. The stability of the suspen-
sion is essential to achieve the necessary consistency.
Three layers of the photocatalytic sol were deposited by spray technique on the top of
the clay roofing tiles. Drying of the photocatalytic sol layers lasted 30 min at 25oC and
50% air RH. They were subsequently heated in an oven at: 290oC for 30 minutes, case of
the sol based on 2.5 mass% TiO2 and PEG 600 (marked as NM 600), and at 400oC for 30
minutes, the sol based on 2.5 mass% TiO2 and PEG 4000 (marked as NM 4000).
Photocatalytic activity of TiO2 coatings. The photocatalytic activity of the TiO2
coatings was evaluated by examining the discoloration of the methylene blue, in the pro-
cedure adjusted to porous substrate, as follows.
As first, the degree of adsorption of the dissolved MB molecules, by the surface of
the mesoporous titania coatings and tile substrate, was measured by a pre-adsorption test.
A cylindrically shaped glass cell with the inner diameter of 3 cm and a height of 6 cm
was attached to the substrate using silicon glue. Both the test cell and substrate are
marked as a test sample. The concentration of MB for the pre-adsorption test and for the
photocatalytic test was 20 and 10 μmol/L, respectively. Twelve milliliters of the MB for
the pre-adsorption test were poured into the test cell. The part of the tile was drowned
into the MB solution of the same concentration. The adsorbtion of the MB (20 μmol/L)
by the tile sample proceeded in the dark for 12 h. The procedure was continued in the
dark with the test solutions of MB (10 μmol/L) for 24 h up to 36 h (until the adsorption of
the dye was completed). The adsorption was considered complete if the differences in the
concentration of MB measured after 30, 60, 120 and 150 min were less than 5%.
In the above procedure, the adsorption of the dye was complete and the test samples
were irradiated with UVA light (Osram Eversun lamp / I= 0,67 mW/cm2/ distance
between the UV lamp and the reactor 18 cm) for 1.5, 2.5, 3.5 and 24 h, Fig. 1a and 1b.
The emission spectrum of the lamp light and absorption spectrum of MB are shown in
Fig. 2. The maximum irradiation of the lamp is in the range of 320-380 nm. It is evident
that the absorption spectrum of MB is not in the range of the emission spectrum of the
lamp.
127
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
a) b)
Fig. 1. Measurement of the photocatalytic activity of TiO2 coating: a) in the protective
box, b) schematic view of the measurements
Fig. 2. Absorption spectrum of MB and the emission spectrum of the light

source
The photocatalytic activity of the materials was monitored on a UV/VIS spectro-
photometer (Evolution 600 / Thermoscientific, England / water as the reference sample)
by measuring absorption spectra of MB (λ=664 nm) as a function of the irradiation time.
Photocatalytic activity of the TiO2 coatings was calculated using the relation 1:
TiO2 activity = [(co – c)/co] x [c1/co] x 100 [1]
where co is the concentration of the test solution of MB before irradiation: c is the

concentration of MB after UV irradiation, and c1 is the concentration of MB after the pre-
adsorption test.
The concentrations c and c1 were determined from the calibration curves showing the
dependence of the absorbance at λmax (664 nm) of MB solutions as a function of the
concentration of MB solutions.
The above procedure was applied on the clay roofing tile samples with mesoporous
coatings (NM 600, NM 4000) and clay roofing tiles without catalyst-mesoporous coating
(marked as N), as well as on the commercial self-cleaning products (clay roofing tile-E /
Germany; glass-SG / France). For each experiment four measurements were performed.
128
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
The clay roofing tiles (a.d. Polet, member of the Nexe group) are complex materials
with total porosity of about 13%. During the firing procedure stable secondary mineral
phases as gehlenite, anortite or diopside, are formed. Besides stable mineral phases, a
certain amount of ″active centers″ are present on the porous tile surfaces. These centers
can influence different adsorption processes. Evidentlly, the porosity of the tiles increases
the consumption of the applied titania dispersion prepared for self-cleaning and decreases
the photocatalytic activity of the formed titania coating. The mesoporous TiO2 coatings
based on PEG 600 (NM 600) and PEG 4000 (NM 4000) have similar surface area (≈100
cm2/g). Both coatings have an average pore size diameters about 10 nm. In the case of the
NM 4000 samples the pore size distribution was wider, with a higher amount of small
pores (24). The thickness of the coatings NM 600 and NM 4000 was about 3 μm. Fig. 3
shows the SEM micrograph of the sample NM 4000 (thickness about 3 μm).
Fig. 3. SEM micrograph of the mesoporous TiO2 coating NM 4000 (x5000)
The photocatalytic activities of the TiO2 coatings were evaluated by using the aqueo-
us solution of the MB, according to DIN 52980:2007-11 (25) and ISO/DIS 10678 (23).
The photocatalytic activity of the mesoporous coatings can be defined as a two step pro-
cess: pre-adsorption (elimination of the influence of MB adsorption) (20, 26, 27), and
MB irradiation test (determination of the photocatalytic activity). Maximum absorption
appeares at λ=664 nm and gradually decreases during the irradiation time. Like many
thiazine dyes, MB has a tendency to dimerise. The dimer of MB, (MB)2, has an absorp-
tion maximum at 614 nm (28). As a consequence of dimerization the size of the MB
molecules increases, which can be an important step in the catalysis since diffusion
towards and from catalyst surface is an important step in the decolorization of MB so-
lution. The decrease of the absorption maximum at 614 nm indicates that the photo-
catalytic coating possesses a photonic efficiency for degradation, Fig. 4. The decrease of
the concentration of MB solution at 614 nm and 664 nm, during irradiation time, clearly
shows that the sample NM 600 is active.
129
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
Fig. 4. Dependance of the absorbance spectra of MB solution (10 μmol/L) in contact with
NM 600 on UVA irradiation time
The photocatalytic activities of the samples were calculated from equation 1. The
activities of the samples NM 600, NM 4000, were compared with the values for the
commercial self-cleaning products: glass- SG, clay roofing tiles-E and clay roofing tiles
without mesoporous TiO2 coating (N), Fig. 5.
100
N NM 600 NM 4000
SG E
80
60
Activity (%)
40
20
0 5 10 15 20 25
Time (h)
Fig. 5. Photocatalytic activity of tested samples during the irradiation time
Several conclusions can be drawn from Fig. 5. Firstly, no significant differences in
the activity were obtained after 24h irradiation time of the samples NM 600, NM 4000
and E. The activity of the commercial glass-SG was lower in comparison with the
samples NM 600, NM 4000 and E. Secondly, these differences are greater up to 3.5 h of
irradiation. These results demonstrate the fact that samples NM 600 and NM 4000 are the
most active ones. The MB molecules were probably more adsorbed (24) in the case of
these samples due to the fact that the coatings have a higher surface area and, con-
sequently, the decrease of the MB concentration in aqueous solution was more pro-
nounced. Later (after 24h irradiation), a deactivation phenomenon was observed. The
active catalyst sites were probably blocked by the formation of the intermediate products
and, consequently, the photocatalytic activity of these systems was decreased. This con-
clusion is in concordance with our previous investigation (29). Namely, the samples NM
130
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
600 and NM 4000 can be partially regenerated by washing with water (rainfall simula-
tion).
A control photocatalytic experiment, to evaluate the dye decolorization, was carried
out with the irradiated samples without catalyst (N). The obtained results indicate that
decolorisation of the MB was not negligible, Fig. 5. It seems that MB is not an ideal
model pollutant, probably the surface of the clay roofing tiles has also some redox
potentional. More reliable values of the photocatalytic activity could be the difference
between the activity values of NM 600 and N samples / NM 4000 and N samples.
Unfortunately, this procedure was not possible to apply in the case of the commercial
products (the reference samples for these experiments were missing).
CONCLUSION
The photocatalytic activity of the samples based on 2.5 mass% TiO2 and PEG 600
(NM 600), 2.5 mass% TiO2 and PEG 4000 (NM 4000), clay roofing tiles without catalyst
- mesoporous coating (N), commercial roofing tiles (E) and commercial glass (SG) were
determined as a two - step process: pre-adsorption, and photocatalytic test. Their efficien-
cy in regard to the MB degradation was found to vary significantly in the laboratory
conditions. The activities of the mesoporous coatings NM 600 and NM 4000 were higher
up to 3.5 h of irradiation time than the activities of the commercial samples. After 24h of
irradiation the efficiency of the prepared mesoporous coating NM 600, NM 4000 was
equal to the values of the commercial self-cleaning ceramic roofing tiles (E), while the
activity of the commercial self-cleaning glass (SG) was significantly lower.
This study suggested that porous substrate without titania coating, due to an adsorp-
tion process, also undergoes a decolorization phenomenon of the MB aqueous solution.
The MB test should include the pre-adsorption test, especially in the case of porous ma-
terials such as ceramic roofing tiles.
ACKNOWLEDGMENT
This study is a part of the Project Eureka, E!3969, financially supported by the
REFERENCES
1. Ma, Y. and J. Yao: Photodegradation of rhodamine B by TiO2 thin film. J. Pho-
tochem. Photobiol. A: Chem. 116 (1998) 167-170.
2. Ruan, S., F. Wu, T. Zhang, W. Gao, B. Xu and M. Zhao: Surface state of TiO2
nanoparticles and photocatalytic degradation of methyl orange in aqueous TiO2
dispersions. Mater. Chem. Phys. 69 (2001) 7-9.
3. Devipriya, S.and S. Yesodharan: Photocatalytic degradation of pesticide contami-
nants in water, Sol. Energy Mater. Sol. Cells 86 (2005) 309-348.
4. Mills, A., J. Wang and M. Mcgrady: Method of Rapid Assessment of Photocatalytic
Activities of Self-Cleaning Films. J. Phys. Chem. B 110 (2006) 18324-18331.
5. Schmidt, H., M. Naumann, T.S. Müller and M. Akarsu: Application of spray tech-
niques for new photocatalytic gradient films on plastics. Thin Solid Films 502 (2006)
132-137.
131
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
6. Balasubramanian,G., D.D.Dionysiou, M.T.Suidan, I. Baudin and J.M. Laine: Evalua-

ting the activities of immobilized TiO2 powder films for the photocatalytic degra-
dation of organic contaminants in water. App. Catal. B Environ. 47 (2004) 73-84.
7. Arsac, F., D Bianchi,., J.M. Chovelon, P.Conchon, C. Ferronato, A. Lair and M.
Sleiman: Photocatalytic degradation of organic pollutants in water and in air. An
analytical approach. Mater. Sci. Eng. C 28, 5-6 (2008) 722-725.
8. Vohra, M.S. and K. Tanaka: Photocatalytic degradation of aqueous pollutants using
silica-modified TiO2. Water Res. 37, 16 (2003) 3992-3996.
9. Wark, M., J. Tschirch, O. Bartles, D. Bahnemann and J. Rathouskỳ: Photocatalytic
activity of hydrophobized mesoporous thin films of TiO2. Microporous Mesoporous
Mater. 84 (2005) 247-253.
10. Evans, P., S. Mantke, A. Mills, A. Robinson and D.W. Sheel: A comparative study
of three techniques for determining photocatalytic activity. J. Photochem. Photobiol.,
A: Chem. 188 (2007) 387-391.
11. Mills, A. and M. McGrady: A study of new photocatalyst indicator inks. J. Photo-
chem. Photobiol., A: Chem. 193 (2008) 228-236.
12. Zhang, F., J. Zhao, L. Zang, T. Shen, H. Hidaka, E. Pelizzetti and N. Serpone: Pho-
toassisted degradation of dye pollutants in aqueous TiO2 dispersion under irradiation
by visible light. J. Mol. Cat. A: Chem. 120 (1997) 173-178.
13. Naman, S.A., Z:A.-A. Khammas and F.M. Hussein: Photo-oxidative degradation of
insecticide dichlorovos by a combined semiinductors and organic sensitizers in
aqueous media. J. Photochem. Photobiol. A: Chem. 153 (2002) 229-236.
14. Yu, J., J.C. Yu, B. Cheng and X. Zhao: Photocatalytic activity and characterization
of the sol-gel derived Pb-doped TiO2 thin films. J. Sol-Gel Sci. Technol. 24 (2002)
39-48.
15. Wilhelm, P. and D. Stephan: Photodegradation of rhodamine B in aqueous solution
via SiO2@TiO2 nano-spheres. J. Photochem. Photobiol. A: Chem. 185 (2007) 19-25.
16. Yang, J., H. Bai, Q. Jiang and J. Lian: Visible-light photocatalysis in nitrogen-car-
bon-doped TiO2 films obtained by heating TiO2 gel-film in an ionized N2 gas. Thin
Solid Films 516 (2008) 1736-1742.
17. São Marcos, P., J. Marto, T. Trindade and J.A. Labrincha: Screen-printing of TiO2
photocatalytic layers on glazed ceramic tiles. J. Photochem. Photobiol. A: Chem. 197
(2008) 125-131.
18. Machida, M., K. Norimoto, T. Watanabe, K. Hashimoto and A. Fujishima: The ef-
fect of SiO2 addition in super-hydrophilic proprety of TiO2 photocatalyst. J. Mater.
Sci. 34 (1999) 2569-2574.
19. Kanta, A., R. Sedev and J. Ralston: Fabrication of silica-on-titania and titania-on-
silica nanoparticle assemblies. Colloids Surface: Physicochem. Eng. Aspects 292
(2007) 1-7.
20. Tschirch, J., D. Bahnemann, M. Wark and J. Rathouskỳ: A comparative study into
photocatalytic properties of thin mesoporous layers of TiO2 with controlled meso-
porosity. J. Photochem. Photobiol., A: Chem. (2008), www.sciencedirect.com.
21. Tobaldi, D.M., A. Tucci, , G. Camera-Roda, G. Baldi and L. Esposito: Photocatalytic
activity for exposed building materials. J. Eur. Ceram. Soc. 28 (2008) 2645-2652.
132
APTEFF, 40, 1-220 (2009) UDC: 666.74:666.3.017:666.3.052:549.514
DOI: 10.2298/APT0940125L BIBLID: 1450-7188 (2009) 40, 125-133
22. Lu, M.C., G.D. Roam, J.N. Chen and C.P. Huang: Adsorption characteristics of di-
chlorovos onto hydrous titanium dioxide surface. Wat. Res. 30, 7 (1996) 1670-1676.
23. ISO/DIS 10678: Fine ceramics (advanced ceramics, advanced technical ceramics):
Determination of photocatalytic activity of surfaces in an aqueous medium by de-
gradation of methylene blue, standard under development.
24. M. Hadnađev, D. Zorić, R. Marinković-Nedučin, E. Lončar, J. Ranogajec: Characte-
ristics of self-cleaning TiO2 coating on clay roofing tiles, 20th Congress of Society of
Chemists and Тechnologists of Macedonia, Ohrid, 17-20 September 2008, Book of
abstracts, p.254.
25. DIN 52980/2008: Photocatalytic activity of surfaces: Determination of photocata-
lytic activity by degradation of methylene blue.
26. Strataki, N., V. Bekiari and P. Lianos: Study of the conditions affecting dye adsorp-
tion on titania films and of their effect on dye photodegadation rates, J. Haz. Ma-
terials 146 (2007) 514-519.
27. Mills, A., M. Sheik, C. O′Rourke and M. McFarlane: Adsorption and photocatalysed
destruction of cationic and anionic dyes on mesoporous titania films: Reactions at
the air-solid interface, App. Cat. B: Enviromental 89 (2009) 189-195.
28. Mills, A. and J. Wang: Photobleaching of methylene blue sensitized by TiO2: an am-
biguous system?, J. Photochem. And Photobiol. A: Chemistry 127 (1999) 123-134.
29. Ranogajec, J., R. Marinković-Nedučin, M. Radeka, V. Ducman, A. Skarpin, O. Ru-
dić and D. Zorić: Photocatalytic and Superhydrophilic Phenomena of TiO2 Coated
Clay Roofing Tiles, Ecers 11th International Conference and Exhibition of the Eu-
ropean Ceramic Society, Cracow, 21-25 June 2009, Programme and Book of Ab-
stracts, p.190-191.
ОДРЕЂИВАЊЕ ФОТОКАТАЛИТИЧКЕ АКТИВНОСТИ ТiO2 ПРЕВЛАКА

НА ЦРЕПУ КОРИШЋЕЊЕМ МЕТИЛЕН ПЛАВОГ КАО МОДЕЛ
ПОЛУТАНТА
Ева С. Лончар, Мирослава М. Радека, Снежана Б.Петровић,
Андреа С. Скапин, Огњен Љ. Рудић и Јоњауа Г. Раногајец
Фотокаталитички активне мезопорозне превлаке нанете на површину глиненог
црепа направљене су од сола титанијум диоксида (Degussa, Germany) и поли(ети-
лен гликола-PEG) М-600 и М-4000 који је носилац мезопорозне структуре. Прев-
лака је нанета коришћењем спреј технике и термичког третмана. Фотокаталитичка
активност TiO2 превлака је процењена коришћењем метлен плавог као модeл боје,
постављене на превлаку, након зрачења UV зрацима. Резултати су поређени са ре-
зултатима фотокаталитичке активности неких комерцијалних производа (глинени
цреп, стакло). Новоформирана превлака је показала значајан ниво обезбојавања
раствора (преко 30% после 24h). Установљено је да поступак одређивања фотока-
талитичке активности за случај порозног црепа би требало да буде иновиран уво-
ђењем процеса предадсорпције.

133
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
MATHEMATICAL MODELLING OF FLUX RECOVERY DURING CHEMICAL

CLEANING OF TUBULAR MEMBRANE FOULED WITH WHEY PROTEINS
Nataša Lj. Lukić, Svetlana S. Popović and Jelena Dj. Marković
Membrane process efficiency in the dairy industry is impaired by the formation of

deposits during filtration processes. This work describes cleaning procedures for ceramic
tubular membrane (50 nm) fouled with whey proteins. Also, mathematical modelling was
performed to obtain models which allow deeper insight into the mechanisms involved
during cleaning procedures. The caustic solutions (0.2%w/w, 0.4%w/w and 1.0%w/w
NaOH) and the mixture of two commercial detergents (0.8%w/w P3-ultrasil 69+0.5%
w/w P3-ultrasil 67 and 1.2% P3-ultrasil 69+0.75 P3-ultrasil 67) were used as chemical
cleaning agents. The results showed that the best flux recovery was achieved with
0.4%w/w NaOH solution. After analyzing the experimental data, five parameter and six
parameter kinetic models were suggested for alkali and detergent cleaning, respectively.
The changes of total and specific resistances, as well as the change of the effective pore
diameter and deposit thickness during cleaning are estimated by applying these models.
KEY WORDS: Ceramic membrane, whey proteins, kinetic models, alkali cleaning,
detergent cleaning
INTRODUCTION
Membrane separation processes are commonly used in the dairy industry as they
alone provide possibility for achieving both the fractionation and concentration without
phase change while preserving physical and chemical characteristics of the main dairy
components. Recently, the application of ultrafiltration and microfiltration has become
increasingly widespread for the production of whey protein concentrate, a product with
high nutritional value.
However, membrane application is often restricted by membrane fouling, which ine-
vitably leads to flux decrease throughout the membrane. During ultra and microfiltration
of whey, fouling is mainly governed by pore plugging and gradual adsorption of whey
proteins at the membrane surface (1). Consequently, membrane cleaning is an essential
step in maintaining the permeability and selectivity of membrane processes.
Nataša Lukić, B.Sc., Junior Res., nlukic@tf.uns.ac.rs, Svetlana Popović, M.Sc., Assist., popovics@tf.uns.ac.rs,
Jelena Marković, B.Sc., Assist., jmarkovic@tf.uns.ac.rs, University of Novi Sad, Faculty of Technology,
Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia
135
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
Chemical methods are mainly used and most efficient cleaning agents are: alkali solu-
tions (2-4) and formulated detergents (5, 6). Mathematical models of cleaning give the
possibility of getting a deeper insight into the present cleaning mechanisms. Some
authors (2, 7-9) have proposed models which describe mechanisms of membrane clea-
ning by applying certain chemical agents.
The main objective of this work was to propose kinetic models for alkali and de-
tergent cleaning of tubular ceramic membrane that was fouled with whey proteins.
EXPERIMENTAL
Experimental apparatus and materials
The schematic diagram of the experimental apparatus, made of stainless steel, is

shown in Fig. 1. The feed solution was circulated by a rotary vane pump PO511 (Cmf,
Italy). During an experimental run, the permeate and the retentate were recycled back to
the feed reservoir to avoid feed concentrating. Flow rate and transmembrane pressure
(TMP) across the membrane module were simultaneously adjusted by a bypass valve and
the main flow valve. The TMP was monitored with manometers and the flow rate was
measured using a rotameter. The feed solution temperature was kept constant and mea-
sured by a digital thermometer mounted inside the feed reservoir. The permeate was col-
lected and weighted continuously on a digital balance (EW 1500-2M, Kern, Germany)
and the data were transmitted to a personal computer (PC).
Fig. 1. Experimental apparatus
The experiments were carried out using Membralox™ monotubular ceramic mem-
brane, 250 mm long, with 7 mm ID and 10 mm OD (Pall Exekia, France). The total filtra-
tion area of the membrane was 46.2 cm2. The membrane, with mean pore size of 50 nm
made of ZrO2 layer on an α-alumina support, was investigated.
136
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
A reconstituted whey solution from a whey powder (donated by Novosadska mlekara,

Serbia) of the following composition: 11.8% (w/w) protein, 75.0% (w/w) lactose, 3.3%
(w/w) fat, 9.5% (w/w) ash and 2.3% (w/w) water, was employed for all fouling expe-
riments. The feed solution was prepared by dissolving the whey powder in deionised wa-
ter to obtain a concentration of 10 g/L. The pH value of the prepared feed solution was
6.0 for all experiments.
Deionized water was used for rinsing steps which were performed before and after
each fouling experiment occurred. Two alkaline cleaning agents were used during mem-
brane cleaning: sodium hydroxide and a mixture of commercially available detergents
P3-Ultrasil 67 and P3-Ultrasil 69 (Henkel, Germany). Studied concentrations of alkali
solution were 0.2%w/w, 0.4%w/w and 1.0%w/w, while detergent solutions were pre-
pared using the following concentrations: 0.8%w/w P3-Ultrasil 69+0.5%w/w P3-Ultrasil
67 and 1.2% P3-Ultrasil 69+0.75 P3-Ultrasil 67. According to the manufacturer, P3-Ul-
trasil 67 is a neutral detergent which consists of alkylaminoxide (15-30%) and proteolytic
enzyme (<5%) while P3-Ultrasil 69 is a mild alkaline detergent which consists of phos-
phonates (5-15%) and salts of organic acids (5-15%). Cleaning solutions viscosity are
supposed to be equal to the viscosity of water (5.47·10-4Pa·s, at 50°C).
Operating conditions
Each experiment consisted of the following steps: pure water flux measurement,
fouling, rinsing, chemical cleaning, rinsing, and pure water flux measurement. Since the
enzymes in detergents show the highest activity at 50°C, all cleaning steps were per-
formed at this temperature. The cross flow velocity during fouling was low to intensify
formation of the fouling layer. However, the cross flow velocity during rinsing and clea-
ning steps was higher to enhance removal of the fouling layer from the membrane
surface. The operating conditions are briefly outlined in Table 1.
Table 1. Operating conditions for the experimental procedure
Step v (m/s) TMP (kPa) t (min) T (°C) Feed stream

Pure water flux measurement 1.73 30 30 25 Water
Fouling 0.43 30 60 25 Whey (10g/L)
Rinsing 1.73 30 30 25 Water
Cleaning / / / 50 NaOH/ Ultrasil P3
Rinsing 1.73 30 30 25 Water
Pure water flux measurement 1.73 30 30 25 Water
Kinetic model
The total resistance (Rtf), due to the deposition of fouling material at surface and
inside pores of the membrane, was chosen as a value that represents fouling intensity in
the most suitable way. Kinetic model which estimates time change of total resistance
(Rtf), which should be reduced during chemical cleaning, was introduced. Furthermore,
137
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
the model predicts the change of the effective pore diameter and deposit thickness during
cleaning.
The total resistance (Eq. [1]) can be presented as the sum of the individual resistan-
ces: the resistance of membrane (Rm), the resistance due to concentration polarization
(Rcp), the hydraulic resistance of deposits at the membrane surface (Rc) and the resistance
due to in-pore fouling (Rin).
Rtf = Rm + R f = Rm + Rcp + Rc + Rin [1]
The model involves Darcy’s equation:
TMP
J cs =
μ cs Rtf
[2]
also used to calculate the initial membrane resistance:
TMP TMP
Jw = ⇒ Rm =
μ w Rm μw J w [3]
The resistance due to concentration polarization was omitted since the membrane was
rinsed with deionized water. The cake resistance (Rc) was calculated assuming a first-
order change of time derivate of the cake deposit (7):
dRc
= − p1 Rc ⇒ Rc = exp(− p1t + p 2 ) [4]
dt
The resistance due to in-pore fouling can be determined as a function of second order
(alkali cleaning Eq. [5a]) and as a function of third order (detergent cleaning Eq. [5b]):
alkali cleaning Rin = p 3 a t 2 + p 4 a t + p 5 a [5a]
detergent cleaning Rin = p3 d t 3 + p4 d t 2 + p5 d t + p6 d [5b]
After replacing all specific resistances into Eq. [1], the kinetic model attains the form:
alkali cleaning Rtf = Rm + exp( − p1t + p 2 ) + p3 a t 2 + p4 a t + p5 a [6a]
detergent cleaning Rtf = Rm + exp( − p1t + p2 ) + p3 d t 3 + p4 d t 2 + p5 d t + p6 d [6b]
Modified Karmen-Kozeny equation shows the connection between effective pore
diameter (de) and the in-pore resistance:
36hk (1 − ε )2 l [7]
Rin = = k ⋅ d e−2
ε 3 d e2
Combining Eq. [7] with Eqs. [5a] and [5b], parameters ka (in the case of alkali
cleaning) and kd (in the case of detergent cleaning) can be calculated using conditions at
the end of the cleaning process:
alkali cleaning (
k a = d e2 (t end ) p 3 a t end
2
+ p 4 a t end + p 5 a )[8a]
detergent cleaning kd = d 2
e (t )( p
end t3
3 d end +p t 2
4 d end + p 5 d t end + p6 d ) [8b]
Detailed description of the applied calculation procedure can be found in other papers
(10, 11).
Mathematical models for effective pore diameter result from equations [5a] and [5b]
using the determined k values:
138
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
0 .5
⎛ ka ⎞
alkali cleaning d e ( t ) = ⎜⎜ ⎟⎟ [9a]
⎝ p3a t + p4 a t + p5 a ⎠
2
0.5
⎛ kd ⎞
detergent cleaning d e ( t ) = ⎜⎜ ⎟⎟ [9b]
⎝ p 3 d t + p 4 d t + p 5 d t + p6 d ⎠
3 2
Deposit thickness during cleaning (δ(t)) can be estimated as follows:

d − d e (t )
d e (t ) = d o − 2δ(t ) ⇒ δ(t ) = o [10]
2
Fouling of membrane
Since the fouling steps were carried out under the same operating conditions, the
same extent of fouling was achieved in each experiment, as shown in Fig. 2. As expected,
significant permeate flux decline occurred during filtration of reconstituted whey solution
until the pseudo-steady state, approximately 8.1% of the initial value, was reached. It can
be observed that the flux decline is sharp within the first few minutes due to the for-
mation of the concentration polarization layer. Further flux decline, yet at a smaller rate,
can be associated with the gradual adsorption of protein deposits on and inside the
membrane surface.
200 0.2% NaOH

180 0.4% NaOH
1.0% NaOH
160
0.8% P3-Ultrasil 69+0.5% P3-Ultrasil 67
-2 1
140 1.2% P3-Ultrasil 69+0.75% P3-Ultrasil 67

Permeate flux Lm h
120
100
80
60
40
20
0
0 10 20 30 40 50 60
t, min
Fig. 2. Permeate flux during fouling
Cleaning of membrane
The permeate fluxes achieved during the steps that followed the fouling experiments
(rinsing, chemical cleaning and final rinsing) are given in Fig. 3. Rinsing with deionized
water (a-series in Fig. 3) was implemented so that loosely bound deposits are removed
139
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
from the membrane surface. Since the achieved rinsing efficiency was close to 20%,
which is less than required, chemical cleaning was needed.
0.2% NaOH
0.4% NaOH 0.8% P3-Ultrasil 69+0.5% P3-Ultrasil 67
1.0% NaOH 1.2% P3-Ultrasil 69+0.75% P3-Ultrasil 67
350 350
a) b) c) a) b) c)
Pre-rinsing Chemical cleaning Final rinsing Pre-rinsing Chemical cleaning Final rinsing
300 300
Flux recovery, dm3m-2h-1
Flux recovery, dm3m-2h-1

250 250
200 200
150 150
100 100
50 50
A B
0 0
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
t, min t, min
(a) (b)
Fig. 3. Flux recovery during alkali (a) and detergent cleaning (b)
Detailed information about rinsing efficiency can be found elsewhere (12).

Permeate flux curves for the NaOH solutions and for detergent solutions, during
cleaning, are given in Fig. 3 as b-series. It can be noticed that the permeate flux increased
within the first few minutes and more or less remained constant during the rest of
cleaning time. Considerably higher flux recovery was achieved during cleaning with
NaOH solutions compared to detergent solutions, especially with 1% w/w solution of
NaOH. Flux curves during final rinsing, that followed cleaning steps, are given in Fig. 3
as c-series. Considerable increase in permeate flux, compared to the flux obtained during
cleaning, can be observed. These results indicate significant influence of the final rinsing
on flux recovery. The highest flux recovery was achieved by cleaning with 0.4%w/w
alkali solution, which can save large amounts of chemicals compared to 1.0%w/w NaOH,
recommended by the membrane manufacturer.
Mathematical modelling - total and specific resistances
The total resistance, Rtf, during chemical cleaning was determined according to Eq.
[2]. The results (ten representative points in Fig. 4) show considerable decrease of the
total resistance at the beginning of cleaning. Further, as alkali cleaning progressed (Fig.
4a), the decrease weakened till the end of cleaning, while in the case of detergent
cleaning (Fig. 4b), even a slight increase was noticed.
Mathematical modelling, through applying Levenberg-Marquardt method (ORIGIN
6.1) on Eq. [6a] and [6b], was carried out using the Rtf data in order to estimate parameter
values for both alkali and detergent cleaning. The parameter values were than used to
calculate total and specific resistances (the cake resistance, Rc, from Eq. [4] and the in-
pore resistance, Rin, from Eq. [5a] and [5b]). The resistance of the membrane, Rm, was
determined applying Eq. [3] to the pure water flux measurements.
140
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
4.00E+012 6.00E+012
A Cleaning agent: 0.4% NaOH B
Rtf=4.94E11+exp(-P1*t+P2)+P3a*t^2+P4a*t+P5a
Total and particular resistances, m-1
3.50E+012
Goodness of fit: R^2 = 0.98675 Rtf
Total and particular resistances (m-1)

5.00E+012
3.00E+012 Rin
4.00E+012
2.50E+012
Cleaning agent: 0.8+0.5% w/w Ultrasil(69+67)
2.00E+012 3.00E+012 Rtf=Rm+exp(-P1*t+P2)+P3d*t^3+P4d*t^2+P5d*t+P6d
Goodness of fit: R^2 = 0.99754
1.50E+012 Rtf
2.00E+012
1.00E+012 Rin
Rm 1.00E+012
5.00E+011
Rm
Rc Rc
0.00E+000 0.00E+000
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
Cleaning time, min Cleaning time, min
(a) (b)
Fig. 4. Total and specific resistances during alkali (a) and detergent cleaning (b)
Since the resulting curves for total and specific resistances show similar trends during
cleaning, only two representative cases are presented: cleaning with 0.4% w/w caustic
solution and 0.8% w/w P3-ultrasil 69+0.5% w/w P3-ultrasil 67 (Fig. 4). By analyzing
specific resistances an exponential decrease of the cake resistance within a few minutes
was observed. The in-pore resistance decreased slowly during the alkali cleaning.
However, re-fouling was noticed during second half of detergent cleaning. Statistical
parameters obtained for all cases of cleaning solutions are presented in Table 2.
Table 2. Model parameters for alkali and detergent cleaning
NaOH
p1 P2 p3a p4a p5a
concentration
0.2%w/w 1.65±0 27.72±0.07 (4.58±2.59)·108 (3.49±8.10)·109 (1.12±4.60)·1010
0.4%w/w 2.85±0.58 30±0.38 (17.10±4.16)·108 (-7.54±1.35)·1010 (1.68±7.86)·1010
1%w/w 1±0.12 27.9±0.03 (6.49±1.5)·108 (-3.18±5.26)·1010 (111.98±3.83)·1010
Ultrasil
(69+67) p1 p2 p3d p4d p5d p6d
concentration
0.8+0.5%w/w (15.07±0) (29.15±0) (13.44±1.08)·107 (-81.47±5.21)·108 (145.6±6.99)·109 (38.44±2.21)·1011
1.2+0.75%w/w (4.63±0)·1010 26.26±0.36 (6.68±3.88)·107 (-3.86±1.88)·109 (5.70±2.51)·1010 (406.74±7.96)·1010
Mathematical modelling - effective pore diameter and thickness of the layer
Protein adsorption within membrane pores leads to pore narrowing, which manifests
as the in-pore resistance according to the Carmen-Kozeny Eq. [7]. Model parameters
141
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
were obtained by fitting experimental data (Eq. [8a] and [8b]) and the results (changes in
effective pore diameter and deposit thickness) are shown in Fig. 5a and Fig. 5b, res-
pectively. It can be noticed that the greatest effective pore diameter was achieved during
30 min cleaning period with 1%w/w alkali solution. However, a maximum pore diameter
after 20 min cleaning period was achieved with 0.4%w/w alkali solution, so it can be
concluded that higher concentration of alkali solution is unnecessary. By analysing
deposit thickness it can be concluded that some amount of proteins remains in the pores
throughout the entire cleaning process. Deposits are present inside pores regardless of the
applied cleaning solution and its concentration. It is worth mentioning that the lowest
deposit thickness is achieved for cleaning with 1%w/w alkali solution but after 20 min of
cleaning the deposit thickness is the same as for 0.4%w/w alkali solution. This confirms
previous statement that it is unnecessary to use higher concentrations of alkalis.
0.2% NaOH 0.2% NaOH
0.4% NaOH 2.00E-008
0.4% NaOH
5.00E-008 1.0% NaOH
1.0% NaOH A B
0.8% P3-Ultrasil 69+0.5% P3-Ultrasil 67 1.80E-008
0.8% P3-Ultrasil 69+0.5% P3-Ultrasil 67
4.50E-008 1.2% P3-Ultrasil 69+0.75% P3-Ultrasil 67 1.2% P3-Ultrasil 69+0.75% P3-Ultrasil 67
Effective pore diameter, m
1.60E-008
Deposit thickness, m
4.00E-008 1.40E-008
1.20E-008
3.50E-008
1.00E-008
3.00E-008
8.00E-009
2.50E-008 6.00E-009
4.00E-009
2.00E-008
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Cleaning time, min Cleaning time, min
Fig. 5. Effective pore diameter and deposit thickness during A) alkali and B) detergent
cleaning
CONCLUSION
Defining mathematical model for both alkali and detergent cleaning gives possibility
to predict changes in total and specific resistances with time as well as the changes of the
effective pore diameter and deposit thickness. A five parameter model for alkali cleaning
and a six parameter model for the detergent cleaning were suggested. The models con-
firm the decrease of total resistance within a first few minutes and almost complete elimi-
nation of the cake resistance. The in-pore resistance decreased during alkali cleaning
whereas during detergent cleaning even an increase was observed. It can be noticed that
even though deposits are present within pores after cleaning, further rinsing with deioni-
zed water contributes to the enhancement of cleaning. Alkali solution of 0.4%w/w NaOH
was proved to have the highest cleaning strength which guaranties almost full flux re-
covery.
142
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
ACKNOWLEGMENT
This research was financially supported by the Ministry of Science and Technological
Development of the Republic of Serbia (Project No. 142045).
REFERENCES
1. Ramachandra Rao, H. G.: Mechanisms of flux decline during ultrafiltration of dairy

products and influence of pH on flux rates of whey and buttermilk. Desalination 144
(2002) 319-324.
2. Bartlett, M., Bird, M. R. and J.A. Howell: An experimnetal study for the development
of a qualitative membrane cleaning model. Journal of Membrane Science 105 (1995)
147-157.
3. Bird, M. R. and M. Bartlett: Measuring and modelling flux recovery during the
chemical cleaning of MF membranes for the processing of whey protein concentrate.
J. Food Eng. 53 (2002) 143-152.
4. Nigam, M. O., Bansal, B. and X.D. Chen: Fouling and cleaning of whey protein
concentrate fouled ultrafiltration membranes. Desalination 218 (2008) 313-322.
5. Argüello, M. A., Alvarez, S., Riera, F. A. and R. Alvarez: Utilization of enzymatic
detergents to clean inorganic membranes fouled by whey proteins. Separation and
Purification Technology 41 (2005) 147-154.
6. Muñoz-Aguado, M. J., Wiley, D. E. and A.G. Fane: Enzymatic and detergent
cleaning of a polysulfone membrane fouled with BSA and whey. J. Membr. Sci. 117
(1996) 175-187.
7. Bird, M. R. and P.J. Fryer: An analytical model for the cleaning of food process plant.
In: IChemE Symposium Series No. 126 (1992) 325-330.
8. Matzinos, P. and R. Álvarez Effect of ionic strength on rinsing and alkaline cleaning
of ultrafiltration inorganic membranes fouled with whey proteins. Journal of Mem-
brane Science 208 (2002) 23-30.
9. Zondervan, E., Betlem, B. H. L. and B. Roffel: Development of a dynamic model for
cleaning ultrafiltration membranes fouled by surface water. Journal of Membrane Sci-
ence 289 (2007) 26-31.
10. Popović, S. S.: Flux Regeneration after Membrane Filtration of Whey Proteins. MSc
Thesis, University of Novi Sad, 2008.
11. Popović, S. S., Tekić, M. N. and M.S. Djurić: Kinetic models for alkali and detergent
cleaning of ceramic tubular membrane fould with whey proteins. J. Food Eng. 94
(2009) 307-315.
12. Popović, S. S., Milanović, S., Iličić, M., Lukić N. Lj. and I.M. Šijački.: Flux recovery
of ceramic tubular membranes fouled with whey proteins: Some aspects of membrane
cleaning. Acta Peroiodica Technologica 39 (2008) 101-109.
143
APTEFF, 40, 1-220 (2009) UDC: 637.02:542.816:66.011+66.013.8
DOI: 10.2298/APT0940135L BIBLID: 1450-7188 (2009) 40, 135-144
МАТЕМАТИЧКО МОДЕЛОВАЊЕ РЕГЕНЕРАЦИЈЕ ФЛУКСА ТОКОМ

ХЕМИЈСКОГ ЧИШЋЕЊА ТУБУЛАРНЕ МЕМБРАНЕ ЗАПРЉАНЕ
ПРОТЕИНИМА СУРУТКЕ
Наташа Љ. Лукић, Светлана С. Поповић и Јелена Ђ. Марковић
У индустрији млека мембране се чисте веома често због интензивног прљања

мембрана протеинима. У овом раду описан је поступак хемијског чишћења цевне
керамичке мембране запрљане протеинима сурутке. Такође су предложени модели
који описују чишћење мембрана алкалним растворима и растворима детерџента.
Резултати су показали да се најбоља регенерација флукса постиже 0.4% раствором
NaOH-а. Након анализе експерименталних података, модели са укупно пет и шест
параметара су предложени за алкално чишћење и чишћење детерџентом, респек-
тивно. Дефинисање модела омогућује процену промене укупног и појединачних
отпора као и промене ефективног пречника пора и дебљине слоја адсорбованог
унутар пора мембране током времена.

144
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
OSMOTIC DEHYDRATION OF RED CABBAGE IN SUGAR BEET

MOLASSES – MASS TRANSFER KINETICS
Nevena M. Mišljenović , Gordana B. Koprivica, Ljubinko B. Lević, Bojana V. Filipčev

and Tatjana A. Kuljanin
The paper describes a study of osmotic dehydration of red cabbage in sugar beet
molasses of different concentrations (40, 60 and 80%) at 50°C and under atmospheric
pressure. The best results were obtained at the sugar beet molasses of 80% as an osmotic
medium.
The most important kinetic parameters of the process were determined: water loss,
solid uptake, weight reduction, normalized solid content and normalized moisture con-
tent. The kinetic parameters were determined after 1, 3 and 5 hours.
Mass transfer coefficients were calculated using Hawkes and Flink’s model and the
results indicate that the diffusion of water and solids was the most intensive during the
first three hours of dehydration.
KEY WORDS: Osmotic dehydration, red cabbage, mass transfer kinetic, sugar beet
molasses
INTRODUCTION
Osmotic dehydration is an effective way to reduce the water content in plant and
animal tissue with minimal negative effect on nutritive and sensorial properties of the
final product. Osmotic dehydration, mainly of fruits and vegetables, is performed by
immersing them in various hypertonic solutions.
Concentrated saccharose solution, sodium chloride solutions and their combinations
are usually used as hypertonic solution (1). At the Faculty of Technology in Novi Sad, a
method has been developed for osmotic drying in sugar beet molasses as hypertonic
solution. Sugar beet molasses appears to be an excellent medium for osmotic dehydra-
tion, primarily due to the high content of dry matter (80%), which provides high osmotic
pressure in the solution (2), as well as the specific chemical composition, characterized
by high contents of vitamins, minerals, antioxidants and betain (3).
Nevena M. Mišljenović, B.Sc., Res. Assist, nevenam@uns.ac.rs, Gordana B. Koprivica, B.Sc., Res. Assist.,
gordanak@uns.ac.rs, Dr. Ljubinko B. Lević, Prof., megamum@uns.ac.rs, Bojana V. Filipčev, M.Sc. Res.,
bojana.filipcev@fins.ns.ac.rs, Dr. Tatjana A. Kuljanin, Assist. Prof., kuljanin@uns.ac.rs, University of Novi
Sad, Faculty of Technology, Bulеvar Cara Lazara 1, 21000 Novi Sad, Serbia
145
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
The complex cellular structure of plant tissue acts as a, not completely selective,
semi-permeable membrane, which allows two main countercurrent flows: water from the
plant tissue flows into the osmotic solution whereas osmotic solute diffuses from the
solution to the tissue (4, 5, 6).
During osmotic dehydration, the tendency is to increase the diffusion of water from
the sample into the surrounding solution and decrease penetration of solids from the
solution into the plant tissue, on the other hand (7). However, in the case when sugar beet
molasses is used as hypertonic solution, the penetration of mineral substances, vitamins,
etc. to the tissue can be considered as favorable because the nutritional value of thus
treated fruits and vegetables is higher (8).
The rate and dewatering degree of the material and changes in its chemical com-
position depend on the sort of the osmotic solution used, the kind and the size of raw
material, as well as the ratio of material to osmotic solution, temperature, dehydration
time, and type of apparatus. Rate of osmotic dehydration is the highest at the beginning
of the process. It results from the largest difference of osmotic pressure between the
osmotic solution and the cell sap of the material and small mass transfer resistance at this
stage of the process (9).
Red cabbage after osmotic dehydration in sugar beet molasses can be used in the
baker industry for the production of a nutritionally valuable food. Breads are darker and
with a very pleasant, caramel-specific, taste, which is derived from sugar beet molasses.
Antioxidative potential of breads were significantly increased (10).
The influence of different concentrations of sugar beet molasses and dehydration time
on the efficiency of osmotic dehydration process of red cabbage was examined in this
study. Kinetic parameters, rate of mass transfer and overall mass transfer coefficients for
water and solute were determined in this paper.
Mass transfer model
During the osmotic dehydration process, three main process variables are usually
measured: moisture content, change in weight and change in soluble solids. Of these,
water loss (WL), weight reduction (WR), solids gain (SG), normalized moisture content
(NMC) and normalized solid content (NSC) were calculated as follows:
⎡g ⎤ w − w
WR ⎢ ⎥ = o [1]
⎣g ⎦ wo
⎡ g ⎤ u − uo
SG ⎢ ⎥ = [2]
⎣g⎦ wo
WL
⎡ g ⎤ = WR + SG [3]
⎢g⎥
⎣ ⎦
146
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
NMC = X/Xo [4]
NSC = u/uo [5]
where: wo– initial sample weight (g), w – sample weight after osmotic dehydration (g), uo
– initial solid content in the fresh sample (g), u – solid content in the sample after osmotic
dehydration (g), Xo – initial moisture content of the fresh sample before osmotic
treatment (g), X – moisture content in the sample after osmotic dehydration (g).
A model was proposed by Hawkes and Flink (9) to describe the kinetics of moisture
loss and solid gain:
NMC = 1 - kwθ -0.5 [6]
NSC = 1 + ksθ -0.5 [7]
where kw (s-0.5) and ks (s-0.5) represent the overall mass transfer coefficients for water and
solute respectively, and θ (s) is the dehydration time. Under static conditions mass
transfer coefficients depend on the solution concentration and contact temperature.
Based on the above parameters, the rate of weight reduction (RWR), rate of solid
gain (RSG) and the rate of water loss (RWL) were calculated.
⎡ g ⎤ WR
RWR ⎢ ⎥ = [8]
⎣g ⋅ s⎦ θ
⎡ g ⎤ SG
RSG ⎢ ⎥ = [9]
⎣g ⋅s⎦ θ
⎡ g ⎤ WL
RWL ⎢ ⎥ = [10]
⎣g ⋅ s⎦ θ
EXPERIMENTAL
Red cabbage was purchased on a local market in Novi Sad, Serbia and stored at 4°C.
Prior to the treatment, the red cabbage was thoroughly washed and cut into cubes,
dimension 1x1 cm. Sugar beet molasses in different concentrations (40, 60 and 80% dry
matter) were used as osmotic solution. Sugar beet molasses was obtained from the sugar
factory Bač, Serbia. For dilution of sugar beet molasses distilled water were used.
Osmotic dehydration was conducted in an apparatus at 55°C under atmospheric pressure
(Fig. 1).
147
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
Temperature monitoring
Osmotic solution
T
Samples
Thermostatic bath Pumps
Fig. 1. Apparatus for osmotic dehydration
The material to solution ratio was 1:4 (w/w). Finally, the cabbage pieces were
removed from the osmotic solutions, washed with water and gently blotted to remove
excessive water. The samples were weighed. Kinetic parameters were determined after 1,
3 and 5 hours.
The samples were kept in an oven (Instrumentaria Sutjeska, Serbia) at 105°C for 24
hours until a constant weight was attained (11).
Moisture content of the samples was determined by the oven drying method ac-
cording to AOAC (12).
Table 1 shows changes in dry matter content in the samples of red cabbage during
osmotic dehydration as a function of the concentration and dehydration time. The
increase in concentration and immersion time during the osmotic dehydration resulted in
higher content of dry matter in the samples of red cabbage. The highest value of dry
matter content in red cabbage (29.35%) was achieved when 80% solid content sugar beet
molasses was used as the osmotic solution, and when the immersion time was 5 hours.
In addition to the changes of dry matter content, the changes in kinetic parameters
during the osmotic dehydration of red cabbage are also shown in the table 1.
During the dehydration, mass of the samples was reduced. Higher value of WR
parameter was found when the immersion time and concentration of sugar beet molasses
were higher. The red cabbage dehydrated in the 80% molasses for 5 hours, lost about
45% of the initial weight.
The SG value indicates the degree of penetration of solids from the osmotic solution
in the samples. Solid gain, during the osmotic dehydration of red cabbage, showed a
tendency to increase with increasing the immersion time and concentration of molasses.
148
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
Penetration of the solute from the osmotic solution into the sample can be limited by
applying starch edible coatings (13).
Table 1. Changes of dry matter content and the kinetic parameters during osmotic
dehydration of red cabbage in the sugar beet molasses
Concentration of Time, Dry matter, WR, g/g initial SG, g/g initial WL, g/g initial
molasses, % d. m. (h) (%) sample weight sample weight sample weight
1 13.63 0.163918 0.031934 0.195852

40 % 3 17.46 0.235882 0.051424 0.287306
5 19.17 0.262369 0.059420 0.321789
1 16.18 0.210605 0.042771 0.253377
60 % 3 23.44 0.349175 0.067584 0.416758
5 24.34 0.375188 0.067084 0.442271
1 16.48 0.214393 0.049435 0.263828
80% 3 24.62 0.368816 0.075422 0.444238
5 29.35 0.456772 0.079420 0.536192
Increasing the dehydration time caused greater loss of water from the sample. Higher
concentrations of molasses increased the osmotic pressure in the hypertonic solution, and
therefore the driving force for dehydration was higher. The highest water loss (WL)
(0.5362 g / g of initial sample weight) was observed in the sample which was dehydrated
in molasses with 80% solid content for 5 hours.
Table 2 shows the mass transfer rate during the osmotic dehydration as a function of
the immersion time and concentration of sugar beet molasses. The results show that the
osmotic dehydration was the most intensive at the beginning of the process.
The rate of mass reduction, the rate of water loss and the rate of solid gain were the
highest during the first hour of the process. Mass transfer rate decreased continuously
from the first to the third hour, and after the third hour showed a stabilization tendency.
The mass transfer rate was the most intensive when sugar beet molasses with 80% solid
content was used as osmotic solution, which can be explained by greater driving force
during the process of osmotic dehydration, i.e. by the greater difference between the
osmotic pressures of the hypertonic solution and the plant tissue.
The objective of osmotic dehydration is the removal of water from plant tissue and, at
the same time, minimizing the penetration of substances from the osmotic solution into
the vegetable tissue. The increase in the rate of solid gain is directly dependent on the
concentration of osmotic solution and inversely proportional to the size of sugar mole-
cules (4, 9).The results of this work indicated that water loss from the samples was faster
than the penetration of the solute into the samples (Fig. 2). Broken line in Fig. 2 is a
theoretical case which shows the situation when the rate of water loss is equal to the rate
of solid gain. The slope of the straight line indicates the ratio of water loss and solid gain
rates. During the dehydration of red cabbage in sugar beet molasses (40% of dry matter)
water loss was 6.6 times faster than the solid gain.
149
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
Table 2. Mass transfer rate during osmotic dehydration of red cabbage in sugar beet
molasses
Rate of weight
Concentration of Time, Rate of solids gain, Rate of water loss,
reduction,
molasses, % d. m. (h) g/(g i.s.w.·s)·106 g/(g i.s.w.·s)·105
g/(g i.s.w.·s)·105
40 % 1 4.55 8.87 5.44
3 2.32 5.05 2.82
5 1.46 3.3 1.79
60 % 1 5.85 11.9 7.04
3 3.43 6.64 4.09
5 2.08 3.37 2.46
80% 1 5.96 13.7 7.33
3 3.62 7.41 4.36
5 2.54 4.41 2.98
i.s.w. – initial sample weight
8 y = 0.6603x - 0.4399
R2 = 0.9988
7
y = 0.5606x + 0.3683
RWL*10 (g/g i.s.w.*s)
6 R2 = 1
5 y = 0.4688x + 0.9017
R2 = 1
4
5
3
2
1
0
0 2 4 6 8 10 12 14 16 18
6
RSG*10 (g/g i.s.w.*s)
Fig. 2. Comparison of the RWL with the RSG during the process of osmotic
dehydration of red cabbage in sugar beet molasses (● - 40% sugar beet molasses; ■ - 60%
sugar beet molasses, ▲ - 80% sugar beet molasses)
Typical results of the change of the dehydration parameters (NMC and NSC) for the
red cabbage immersed in sugar beet molasses (40%, 60% and 80%) at 55°C are shown in
Fig. 3. With increase of the solution concentration, an increase in the dehydration rate
was observed. Extensive dehydration took place within the first 3 h, during which the wa-
ter removal ranged between 30 and 50% of the initial moisture content in the plant tissue.
After the third hour, the water diffusion slowed down and for the next two hours the
water content was reduced by 3 - 10%. The results of this work indicated that the time of
dehydration can be limited to 3 hours. During 5 hours of dehydration there was the
diffusion of the solute from the osmotic solution into the plant tissue, although the most
150
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
intensive diffusion took place within the first 3 hours of the process. After the third hour
of the process, in the samples dehydrated in sugar beet molasses with 60% and 80% solid
content, the increase in solute diffusion was very small, whereas in the sample dehydra-
ted in 40% molasses, the diffusion continued at a greater rate. Lazarides et al. (14) indi-
cated that the diffusion of the solute into the plant tissue can be limited by lowering the
temperature to 45 °C, since the selectivity of the semi-permeable cell membrane of plant
tissue reduces at the temperatures above 45°C.
0,80
40 % molasses 2,0
0,75
1,9
0,70
1,8
0,65
1,7
NMC
0,60 60 % molasses NMC
NSC
NSC
0,55 80 % molasses 1,6
0,50 80 % molasses 1,5

60 % molasses
0,45
40 % molasses 1,4
0,40
1,3
50 100 150 200 250 300
Time (min)
Fig. 3. Influence of dehydration time and concentration of osmotic media on the NMC
and NSC parameters
The overall mass transfer coefficients for water and solute was determined on the ba-
sis of equations 6 and 7 (Fig. 4 and 5). Mass transfer coefficients depend on the tempera-
ture and concentration of hypertonic solution. At a constant temperature, the dependence
of these coefficients was studied as a function of the duration of immersion time and con-
centrations of sugar beet molasses.
A increase in the mass transfer coefficient was observed for water and solute with
increasing concentration of the osmotic solution of sugar beet molasses. The mass trans-
fer coefficient decreased with the immersion time.
Higher values of mass transfer coefficients for water in the samples obtained during
the first hour of osmotic dehydration can be explained by a higher driving force during
the process, i.e. by the higher concentration gradient of moisture in the initial stages of
osmotic dehydration, during the diffusion of free water.
151
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
0.006
0.005
40% sugar beet
molasses
0.004
60% sugar beet
Kw (s-0.5)
molasses
0.003
80% sugar beet
molasses
0.002
0.001
0
1 3 5
Time (h)
Fig. 4. Mass transfer coefficient for water during osmotic dehydration of red cabbage in
sugar beet molasses at 55°C
0.012
0.01
0.008 40% sugar beet

molasses
)
-0.5
60% sugar beet

0.006
Ks (s
molasses
80% sugar beet
0.004 molasses
0.002
0
1 3 5
Time (h)
Fig. 5. Mass transfer coefficient for solute during osmotic dehydration of red cabbage in
sugar beet molasses at 55°
152
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
CONCLUSION
Sugar beet molasses seems to be a osmotic solution suitable for the osmotic dehydra-
tion of red cabbage. Higher solution concentration caused higher water loss, higher solids
gain and higher mass reduction of red cabbage. Osmotic dehydration was the most inten-
sive in the first hour of dehydration. After 3 hours of dehydration the rate of mass transfer
decreased so that the processing time can be limited to 3 hours.
About seven times higher was the rate of water loss than the rate of solid gain, which
is very desirable in order to avoid the effect of sugar penetration into the sample.
ACKNOWLEDGEMENT
This research is part of the project supported by the Ministry of Science and Techno-
logical Development of the Republic of Serbia, TR – 20112, 2008-2010.
REFERENCES
1. Jokić, A., J. Gyura, Lj. Lević, and Z. Zavargo,: Osmotic dehydration of sugar beet in
combined aqueous solutions of sucrose and sodium chloride. Journal of Food
Engineering 78 (2007) 47-51.
2. Lević LJ., V. Filipović and T. Kuljanin: Osmotski tretman oblikovanog korena mrkve
u saharozi i melasi. Journal on Processing and Energy in Agriculture 11, 3 (2007),
132-135.
3. Šušić S. and V. Sinobad: Istraživanja u cilju unapređenja industrije šećera Jugo-
slavije. Hemijska Industrija 43, 1-2 (1989) 10 -21.
4. Kowalska, H. and A. Lenart: Mass transfer during osmotic dehydration of plant
tissue. In Proceedings of the IX seminar, Properties of Water in Food, Warsaw (1998)
131-142.
5. Rastogi, N. K. and K. S. M. S. Raghavarao: Mass transfer during osmotic dehydration
of carrot: comparison of different methods for estimation of effective diffusivities, in
Engineering and food (Part 2) (G73-G76). Eds. R. Jowitt,. London: Sheffield Aca-
demic Press (1997).
6. Salvatori, D., A. Andres, A. Chiralt and P. Fito: Concentration profiles in apple tissue
during osmotic dehydration, in Engineering and food (Part 2) (G77-G80). Eds. R.
Jowitt, London: Sheffield Academic Press (1997).
7. Matuska M., A. Lenart and H. N. Lazarides: On the use of edible coatings to monitor
osmotic dehydration kinetics for minimal solids uptake, Journal of Food Engineering
72 (2006) 85-91.
8. Koprivica G., N. Mišljenović, Lj. Lević, M. Petkova and V. Pribiš: Osmotsko sušenje
jabuke u rastvorima saharoze i melase šećerne repe: promena nutritivnih karakte-
ristika finalnog proizvoda. Journal on Processing and Energy in Agriculture 12, 4
(2008) 215-218.
9. Moreira, R. and A. M. Sereno: Evaluation of mass transfer coefficients and volu-
metric shrinkage during osmotic dehydration of apple using sucrose solutions in static
and non-static conditions. Journal of Food Engineering 57 (2003) 25–31.
153
APTEFF, 40, 1-220 (2009) UDC:635.34:635.078:664.15:66.012.5
DOI: 10.2298/APT0940145M BIBLID: 1450-7188 (2009) 40, 145-154
10. Lević, Lj., Pribiš, V., Filipčev, B., i T. Kuljanin: Promena antioksidativnog
potencijala finih kvasnih peciva pri dodavanju voća i povrća osmotski dehidriranom u
melasi šećerne repe. Journal on Processing and Energy in Agriculture 12, 3 (2008)
124-126.
11. Milić, M., V. Karadžić, S. Obradović: Metode za laboratorijsku kontrolu procesa
proizvodnje fabrika šećera, Tehnološki fakultet Novi Sad, Novi Sad (1992).
12. AOAC, Official Methods of Analysis (2000), Washington, USA.
13. Lević Lj., G. Koprivica, N. Mišljenović, B. Filipčev, O. Šimurina and T. Kuljanin:
Effect of starch as an edible coating material on the process of osmotic dehydration of
carrot in saccharose solution and sugar beet molasses. Acta Periodica Technologica
39 (2008) 29-36.
14. Lazarides, H. N. and N. E. Mavroudis: Kinetics of osmotic dehydration of a highly
shrinking vegetable tissue in a salt – free medium. Journal of Food Engineering 30
(1996) 61-74.
ОСМОТСКА ДЕХИДРАТАЦИЈА ЦРВЕНОГ КУПУСА У МЕЛАСИ

ШЕЋЕРНE РЕПE – КИНЕТИКА ПРЕНОСА МАСЕ
Невена М. Мишљеновић , Гордана Б. Копривица, Љубинко Б. Левић,

Бојана В. Филипчев и Татјана А. Куљанин
У раду је проучавана осмотска дехидратација црвеног купуса у раствору меласе

шећерне репе различитих концентрација (40, 60 и 80 %), на температури од 55oC и
атмосферском притиску. Најбољи резултати дехидратације су постигнути кад је
као осмотски раствор коришћена 80% меласа шећерне репе.
У раду су одређени најважнији кинетички параметри процеса: губитак влаге,
прираст суве материје, губитак масе, нормални садржај суве материје и нормални
садржај влаге. Кинетички параметри процеса су одређени након 1, 3 и 5 сати дехи-
дратације.
Коефицијенти преноса масе, одрeђени према моделу Hawkes-а и Flina–а, ука-
зују да је дифузија воде и растворка најинтензивнија у прва три сата дехидратације.

154
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
DETERMINATION OF TENSION STRENGTH IN THE LONGITUDINAL AND

CIRCUMFERENTIONAL DIRECTION IN GLASS-POLYESTER COMPOSITE
PIPES
Slaviša S. Putić, Marina R. Stamenović, Branislav B. Bajčeta and Dragana D. Vitković
Polymer composite pipes with glass fiber reinforcement have today a wide usage in
the chemical and process industries. The basic subject of this paper is the determination
and distribution of stresses and strains in longitudinal and circumferentional directions
of glass-polyester pipes under tension test. Also, the tension strengths in both directions
are determined out. Tension test was performed on an electro-mechanical test machine
on flat samples and rings obtained by cutting of pipes produced by the method “Filament
winding” with glass fibers reinforcement ±55°. Also, the micromechanical analysis on
fracture surfaces was done by SEM, which provided the knowledge about models and
mechanisms of fracture on applyed loading.
KEY WORDS: Glass-polyester composite pipe, tension test, ring test, micromechanical
analysis
INTRODUCTION
Traditional materials in the chemical and process industries are today successfully
replaced by composite materials. More and more pipes, tanks, reservoirs, pressure vessels
are made of these materials. The advantages are in relatively small mass with good
strentgh/stiffness ratio, good static and dynamic properties, as well as good resistance to
corrosion.
The basic subject of this paper is to determine distribution of streses and strains in the
longitudinal and circumferentional direction of glass-polyester composite pipe of defined
construction subjected to tension. The basic hypothesis is that the current standard of
choice, design and production of glass-polyester composite pipe does not give enough
information about the behavior of pipes during the internal pressure, and this problem is
still connected to the individual investigations which are conducted in individual world
laboratories.
Dr. Slaviša Putić, Assoc. Prof., slavisa@tmf.bg.ac.rs, Faculty of Technology and Metallurgy, Karnegijeva 4,
11000 Belgrade, Serbia; Marina Stamenović, M.Sc., Belgrade Polytechnic, Brankova 17, Belgrade, Serbia;
Branislav Bajčeta, B. Sc., Dragana Vitković, B. Sc., Faculty of Technology and Metallurgy, Karnegijeva 4,
11000 Belgrade, Serbia
155
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
An example of research is the paper (1) in which authors studied the dependence of
reinforcement angle in polymer composite pipes subjected to internal pressure. The rela-
tionship between stress in circumferentional and longitudinal direction (2:1) in the case of
different angle of reinforcement has been checked out. Hoffman’s criteria of crack was
used. The analysis showed that angle of reinforcement has a great influence on the pipe
strength and the optimal value for glass-epoxy, carbon-epoxy, and boron-epoxy composi-
te pipe is approximately 45°÷55° (1). Deviation of this values increase the possibility of
finall fracture. There is a resemblance between the paper (2) and presented experiment.
Mechanical properties of glass-polyester pipes with angle of reinforcement of ±54° and
90° were tested (2). Experiments were carried out with internal pressure and stresses and
strains were measured in the logitudinal and circumferential directions. In (3), the authors
also tested composite pipes subjected to internal pressure and found the values of stresses
and strains in both directions. Also, they investigated the influence of stress distribution
on creeep. The subject of paper (4) is also the determination of stresses and strains in the
longitudinal and circumferentional directions, conected with the investigation of crack
initiation and propagation. The final result was the knowledge about stress in both directi-
ons which does not cause cracks in the pipe and their propagation until the final fracture.
EXPERIMENTAL
Composite pipes have been fabricated in the lab conditions. The properties of used
glass-polyester pipes were given in the official certificates of the particular producers of
components. The producers of reinforced glass fibers are A.D. “OHIS” and “Vidoe Smi-
levski-Bato” from Gostivar (Macedonia) by certificate confirm “E” glass with 1% of al-
kali. Thermo-reactive resin was used as matrix by the producer “Color”-Medvode (Slo-
venia). Certificate was given for “COLPOLY 7510” for the type: UP/SOM- highly
reactive, with low viscose polyester on the basis of ortophthalic acid and standard glycol.
Fig. 1. Scheme of cutting of the test pipes; (left) flat samples R-BR; (right) rings P-BR
156
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
The pipes were made by the method “Filament Winding”, with ±55° angle of glass fi-
bers reinforcement. The specimens for tests, (flat specimens, R-BR, and rings, P-BR),
were cut from the samples of pipes according to the standard dimensions, Fig. 1., the flat
specimens 250x25(20 gage area)x3.5 mm, and the rings φ70x35x3.5 mm (average values
of all tested samples). The cut was done on machine type NC-2010 (Nr 95110, Ar 001)
by the tools with diamond top and the moving speed which lowers the heating of the
sample.
Testing of flat test specimens was done on a servo-hydraulic testing machine
SCHENCK TREBEL RM 100, and ring test on the servo-hydraulic testing machine
INSTRON 1332 with the controller INSTRON FAST TRACK 80800, using of hydraulic
jaws. The testing was defined by standard ASTM D 3039 (5, 6). Six flat test specimens
(P-BR) and six rings (R-BR) were tested. Loading was registered with measuring cell of
the capacity of 100 kN. Displacements were measured by double extensiometer HOT-
TINGER DD1.
During the test the diagrams stress-strains (σ-ε) were plotted, Fig. 2.
95
5
90
85
4
80
75 1 P-BR
2
70 6
65 3
60
Stress, σ (MPa)
55 6
50 5
2 R-BR
45 1
4
40 3
35
30
25
20
15
10
5
0
0,0 0,5 1,0 1,5 2,0 2,5 3,0
Strain, ε (%)
Fig. 2. Comparison of the stress-strain (σ–ε) diagrams from the two tests
Tensile strength was calculated according to the Equation [1] for flat test specimens
(longitudinal direction), and, according to the Equation [2] for the rings (circumferentio-
nal direction):
Pmax
R m ,l = [1]
b⋅d
157
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
Pmax [2]
R m ,c =
2 ⋅b ⋅d
where: Rm,l, in MPa, is the tensile strength in the longitudinal direction; Rm,c, in MPa, is
the tensile strength in the circumferentional direction; Pmax, kN, is the maximal applied
load force; b, mm, is the width of test specimen (flat specimens or rings); and d, mm, is
the thickness of test specimen (flat specimens or rings).
Module of elasticity E(l,c) (GPa) was calculated using Equation [3], and the relation-
ship ΔΡ/Δε was determined by linear regression of rectilinear parts of obtained stresses-
strains curves.
Δσ ΔP 1
E( l , c ) = = ⋅ [3]
Δε Δε 2 ⋅ b ⋅ d
Tension strentgh in longitudinal direction (flat samples, R-BR)
The calculated values of tensile strength and module of elasticity in the longitudinal
direction are shown in Table 1.
Table 1. Test results of flat samples

Sample Width of test Thickness of test Cross section Maximal Tensile Module of
specimen specimen A0 (mm2) load force strength elasticity
b (mm) d (mm) Pmax (kN) Rm,l (MPa) El (GPa)
R-BR-1 16.50 3.62 59.73 2.74 46 8.90
R-BR-2 15.72 3.41 53.61 2.57 48 8.90
R-BR-3 15.45 3.72 57.47 2.47 43 7.54
R-BR-4 13.55 3.48 47.15 2.08 44 10.90
R-BR-5 14.95 3.62 54.12 2.81 52 12.20
R-BR-6 15.15 3.48 52.72 2.85 54 9.75
The relative agreement of the obtained values of maximal load force Pmax can be
noticed, except for the test specimen R-BR-4, which is of smaller geometrical dimensions
area, and which crashed much earlier than the others during the test. That is the reason
why this test specimen has the smallest maximal load force.
According to the results of six tested specimens, the average value of tension strength
is 47.8 MPa, and the average value of module of elasticity 9.70 GPa. It can be concluded
that for this kind of mechanical testing, there is relatively small deviation of measured
and average values for tensile strength and module of elasticity. For tension strength
minimal deviation is 0.4% for test specimen R-BR-2, maximal 13.0% for test specimen
R-BR-6. As for module of elasticity, minimal deviation is 0.5% for test specimen R-BR-
6, maximal 25.8% for test specimen R-BR-5.
The explanation for higher deviation of result of the module of elasticity is the fact
that it was relatively more difficult to determine precisely the module elasticity because
of the relatively unstable linear part of diagram in Fig. 2 and relatively small starting
curve stress-strain (σ–ε). As for the tension strength, it is known that because of the angle
of winding fibers and different distribution of tension along the axis of fibers, all the
158
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
fibers are not loaded in the same way. The result of that is the different moment of
breaking fibers, that is, some fibers break under lower and some under higher loading
(Fig. 3). Fibers that break easier cause disturbance in the zone of breaking and local ten-
sions occur next to the broken fiber, so that it leads to different maximal loading force on
the fracture.
Important influence of shear components of tension can be seen in the dependence of
tension-strain (σ–ε), which is not linear (Fig. 2) for most of composites. Non-linear cha-
racteristics occurred on approximately 20÷25% of value of maximal stress. The increase
of stress brought to the debonding and cracks between fiber-matrix connection and mac-
ro-cracks which cause fracture. The result of that is the break of fibers and local
delamination, but pipe still carried out the loading. With further increase of load force,
the local deformations were spreading, the whole groups of fibers broke, resulting in
progressive delamination and final fracture. The fracture was followed by strong acoustic
effect which was a consequence of the break of a great number of fibers.
Delamination of layers is certainly a phenomenon of destruction of these test samples.
The delamination has the appearance (Fig. 4) which matches the interlaminar shear stress.
The confirmation of these conclusions is the SEM micrograph shown in Fig. 5, where
under the higher magnification we can see the previously mentioned phenomenon.
Fig. 3. Different time of fibers breaking shown by SEM
Fig. 4. Delamination of test specimen Fig. 5. Delamination under the

during the load interlaminar shear stress; SEM
micrograph
159
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
Tension strentgh in circumferentional direction (ring samples, P-BR)

The calculated values of tension strength in circumferentional direction and module
of elasticity are shown in Table 2.
Table 2. The results of testing of all ring samples
Sample Width Thickness Cross section Maximal load Tensile Module of
of ring of ring 2.b.s (mm2) force strentgh elasticity
b (mm) s (mm) Pmax (kN) Rm,c (MPa) Ec (GPa)
P-BR-1 35.0 3.35 234.5 19.13 82 13.5
P-BR-2 35.0 3.35 234.5 18.47 79 13.1
P-BR-3 35.0 3.25 227.5 15.86 70 13.0
P-BR-4 35.0 3.15 220.5 19.48 88 12.9
P-BR-5 35.0 3.3 231.0 21.47 93 13.6
P-BR-6 35.0 3.45 241.5 17.21 71 12.8
The relative agreement of the obtained values of maximal load forces Pmax can be
seen, except for the test ring specimen P-BR-3, where the smallest value was obtained
(Table 2). This ring cracked earlier than the others during the testing, so its calculated
tension strength in circumferentional direction is the smallest. It is assumed that because
of this there is slightly irregularly set tool inside the ring which caused irregular increase
of force, curved ring and earlier crack.
The average value of tension strength in the circumferentional direction is 80.5 MPa,
and the average module of elasticity 13.2 GPa according to results of six tested rings.
Minimal deviation is 1.9% for the ring P-BR-1, maximal 15.2% for the ring P-BR-5. As
for the module of elasticity, minimal deviation is 0.8% for the ring P-BR-2, maximal
3.0% for the rings P-BR-5 and P-BR-6.
The explanation of the presented results is relatively similar as with flat samples (R-
BR). For the module of elasticity the fact that it was relatively difficult to determine pre-
cisely module of elasticity, because of relatively unstable linear part on the diagram and
small starting curving of curve stress-strain (Fig. 2). It must be taken in consideration that
the test was done by modified and specific method of testing of rings for tensile strength,
so a specially made tools were used. Because of that, there were certain irregularities in
the measurements that affected the results.
As for the analysis of the break itself, it has to be said that it occurred with a great
stretch of ring samples (Fig. 6) and strong acoustic effect. It is characteristic that all
tested rings had evident break with cracking of fibers in two directions (conditionally,
crossed ±45°). Also, stretching and bursting of fibers is characteristic which can be a
consequence of the lack of matrix (Fig. 7). The fibers did not crack on exactly determined
surfaces but randomly in all directions.
Also, the stresses-strains (σ-ε) curves shown in Fig. 2., are not linear for this testing
either. The nonlinearity occurred on approximately 40% of the value of maximal load
force. That is also the confirmation of the important participation of shear components of
tension. The starting crack initiated in one of the layers by the breaking of connection
fiber-matrix or cracking of fiber, which caused increased concentration of tension, was
spreading further and caused the occurrence of macro-crack, leaving one group of fibers
still together.
160
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
Fig. 6. Straining of rings during the testing Fig. 7. Stretching and bursting of
fibers under the lack of matrix
With a further increase of the loading and tension of macro-crack spread in the
adjacent layers, causing delamination, but the ring still carried out the loading. Delami-
nation is a phenomenon of the destruction of all rings (Fig. 8).
Fig. 8. Delamination of rings
Comparison and analysis of results of the two tension tests
Now we compare and analyze the results obtained in the both tests. Fig. 2 shows
comparative stress-strain (σ–ε) diagrams, Fig. 9 the comparative calculated tension
strengths and Fig. 10 module of elasticity from both tests. This kind of presentation is
justified having in mind the condition of tension in pipes (cylindrical samples) and the
fact that in the circumferentional direction there are twice as high stresses compared to
the longitudinal direction. In our experiment this ratio is approximately 1.7. On the other
hand, it was expected because the structure of composite pipe is neither heterogeneous
nor homogenous, and composite materials have different and specific mechanisms of
damages and breaking.
Some major differences in the values of module of elasticity obtained in the two tests
do not exist. It was real to expect that higher values would be obtained for the module of
elasticity in the circumferentional direction than for those in the longitudinal direction. In
this case their relation is approximately 1.4.
161
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
100 15
14
90 13
elasticity
80 12
11
rength
70 10
)
E(l, (GPa
Pa)
60 9
8
Rm (M
Module of
Tension st
50 7
c)
6
(l, c)
40
5
30 4
20 3
2
10 1
0 0
6 6
BR
5
5 R P-
Sa
4 P-B Sa
4
m 3
3
BR
m pl
pl
BR e 2
R-
R-
e 2
1
1
Fig. 9. Comparative presentation of Fig. 10. Comparative presentation

the tension strength obtained in the two of the module
tests of elasticity obtained in the two
tests
CONCLUSION
The aim of the paper was the determination and distribution of streseses in glass-
polyester composite pipes in the circumferentional and longitudinal directions. Also, the
tension strentgh in both directions were calculated. In the previously standard tests,
performed on flat test specimens and rings, these properties were determined to get some
starting points for further tests.
Fig. 2 can be used to explain the mechanisms of damage and crack behavior during
the loading in both tension tests. The nonlinearity of diagram stress-strain was found
(σ–ε) in both tests. It occurred at approximately 20÷25% of value of the maximal load
force for samples R-BR and 40% for samples P-BR. This nonlinearity occurred in the
first part of the curve, but it is continued in the whole process. The decrease of the curve
slope occurred with increase of tension load in both tests. Thus, based on Fig. 2 the
following conclusion can be drawn:
1. Linear part at the beginning of the curve for samples R-BR is much lower than
for samples P-BR. Besides the determination of the curve defining the module
of elasticity is difficult. As a result, maximal deviations of the calculated
modules of elasticity for R-BR was much higher (25.8% compared to 3.0%).
The cause of this nonlinearity in the beginning shows the fact that because of
tension in the longitudinal direction sliping of fibers occurred, and this caused
first cracks between the fibers and matrix and creation of zones of increased
concentration of tension. On the other hand, during the streching of rings
strength of wound fibers was higher and they carried the loading;
2. We can explain the fact that the curve slope and values of tension strength of the
samples P-BR are higher than for samples R-BR. During the loading of samples
P-BR the matrix was streching until it cracked, which the wound fibers where
162
APTEFF, 40, 1-220 (2009) UDC: 678.7+666.189.21:539.41
DOI: 10.2298/APT0940155P BIBLID: 1450-7188 (2009) 40, 155-163
carrying the loading. The crack appeared on the thickness of the ring, but the
following layers took over the loading. The case of samples R-BR is different,
and with these curves in the second, growing part, slight curving was seen. It
can be concluded that even at a higher tension, there is seen almost linearity (to
approximately 90% of maximal tension). This means that after the first cracks
there was dominant shear stress which caused progressive delaminating and
gradually, but equally lead to the final crack;
3. It is obvious that samples P-BR have higher tension strentgh, but also withstand
higher strains.
REFERENCES
1. Ni, Ai-Qing; Zhu, Yi-Wen; Wang, Ji-Hui: Effects of winding angle on the strength of
composite pipe under internal pressure. Wuhan Ligong Daxue Xuebao/Journal of
Wuhan University of Technology 28, 3 (2006) 10-13.
2. Karpuz, P. : Mechanical Characterization Of Filament Wound Composite Tubes By
Internal Pressure Testing, Master of Science Thesis, Department of Metallurgical and
Materials Engineering, University of Ankara (2005) 104.
3. Guan, Z. W., Boot, J. C. : Creep Analysis of Polymeric Pipes Under Internal Pressure.
Polym. Eng. Sci. 41 6 (2004) 955-961.
4. Flueler, P., M. Farshad: Arrest of rapid crack propagation in polymer pipes. Mater.
Struct. 28, 2 (1995) 108-110.
5. Annual book of ASTM Standards, Vol.15.03, American Society for Testing and
Materials, Philadelphia, PA. (1999).
6. ASTM D 3039-76, Standard Test Method for Tensile Properties of Fiber-Resin
Composites, Annual Book of ASTM Standard, v 36 (1980) 734-739.
ОДРЕЂИВАЊЕ ЗАТЕЗНЕ ЧВРСТОЋЕ У УЗДУЖНОМ И ОБИМНОМ

ПРАВЦУ У СТАКЛО-ПОЛИЕСТЕР КОМПОЗИТНИМ ЦЕВИМА
Славиша С. Путић, Марина Р. Стаменовић, Бранислав Б. Бајчета

и Драгана Д. Витковић
Цеви од полимерних композита са стакленим ојачањем данас имају широку
примену у хемијској и процесној индустрији. Предмет овог рада је одређивање и
расподела напона и деформација у уздужном и обимном (тангенцијалном) правцу
стакло-полиестер цеви при испитивању затезањем. Такође, одређена је и затезна
чврстоћа у оба наведена правца. Испитивање затезањем је изведено на електроме-
ханичким кидалицама на равним узорцима и прстеновима сеченим од цеви произ-
ведених методом намотавања, са намотавањем стакленог ојачања под углом ±55°.
Микромеханичка анализа је изведена на површинама прелома коришћењем СЕМ,
на основу које су претпостављени модели и механизми настанка лома услед приме-
њених оптерећења.

163
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
THE USE OF L-ASCORBIC ACID IN SPECIATION OF ARSENIC

COMPOUNDS IN DRINKING WATER
Aleksandar R. Stanić, Saša I. Jovanić, Nikola J. Marjanović and Zvonimir J. Suturović1
Arsenic speciation, besides total arsenic content determination, is very important in

analysis of water, foodstuffs, and environmental samples, because of varying degrees of
toxicity of different species. For such purpose hydride generation atomic absorption
spectrometry can be used based on the generation of certain types of hydride, depending
on the pH value and pretreatment in different reaction media. In this study, we have
investigated the effect of L-ascorbic acid as the reaction medium as well as the pre-redu-
cing agent in speciation of arsenic by hydride generation-atomic absorption spectrometry
in order to determine monomethyl arsonic acid (MMA) in the presence of inorganic
forms of arsenic.
KEY WORDS: Total arsenic, speciation, L-ascorbic acid, atomic absorption

spectrometry, hydride generation
INTRODUCTION
Although content of arsenic in lithosphere (without atmosphere and hydrosphere) is

relatively low (5·10-4 mass %) (1), high toxicity of arsenic demands continuous determi-
nation of total arsenic as well as its different species in the environment. Reported con-
centrations of arsenic in soil are between 0.1 and 40 mg kg-1 (2), while in non-contami-
nated soil average concentration is 5- 6 mg kg-1 (3,4). By using arsenic based pesticides,
fertilizers, irrigation, fossil fuel burning, disposal of industrial and other waste products,
humans are exposed to the action of arsenic in the environment (5). Arsenic concentration
in the parts of plants used for human nutrition is usually low, even when grown on
contaminated ground (6), despite the fact that higher concentrations are found in plants
grown on contaminated soil near mines and smelteries (7-10). Monomethyl arsonic acid
(MMA) and dimethyl arsinic acid (DMA) are better absorbed and also better assimilated
(incorporated) by plants than arsenic inorganic compounds (11). Arsenic concentration in
marine water systems is in the range of 2 - 3 μg/l (12), while in marine organisms it is in
the range of 1-30 mg kg-1 of arsenic (13) as arsenobetaine (14). Rivers, lakes and ground-
Aleksandar R. Stanić, B.Sc., Saša I. Jovanić, M.Sc., Institute of Public Health, Zmaj Jovina 30, 24000 Subotica,
Serbia; Dr. Nikola J. Marjanović, Prof., Dr. Zvonimir J. Suturović, Prof., University of Novi Sad, Faculty of
Technology, Bulеvar Cara Lazara 1, 21000 Novi Sad, Serbia
165
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
water contain up to 100 μg/l of arsenic, drinking water of Bangladesh and West Bengal,
India contains more than 100 μg/l of arsenic (15).
Arsenic is one of the elements of public concern because of its highly toxic and
carcinogenic properties (I group) (16-19). Exposure to arsenic can cause a variety of ad-
verse health effects, including dermal, respiratory, cardiovascular and gastrointestinal
changes. Also, arsenic is genotoxic and mutagenic (20, 21). Inorganic forms are highly
toxic, and arsenite is apparently more toxic than arsenate. The methylated organic species
(MMA) and (DMA) are less toxic than the inorganic forms (20, 22, 23).
The Serbian Regulation on Hygienic Propriety of Drinking Water (25), European
Union regulation (Directive EU 98/83/EC), directive of WHO (World Health Organiza-
tion) and EPA (United States Environmental Protection Agency) prescribe maximum
contaminant level of arsenic in drinking water of 10 μg/l. However, these standards defi-
ne only total arsenic MRL (Maximum Residue Level), regardless of arsenic forms and
without request of arsenic speciation. Besides, U.S. Department of Health and Human
Services, Agency for Toxic Substances and Disease Registry in toxicological profile for
arsenic, specifies the level of 0.3 μg/kg of inorganic form per day, as a limit under which
there are no toxic effects in case of human oral consumption longer than one year (24).
Arsenic speciation can be very important in the analysis of water, foodstuffs, and
environmental samples, besides of total arsenic analysis. Possible method used in the spe-
ciation of arsenic compounds is atomic absorption spectrometry with hydride generation,
and it is based on the generation of certain types of hydride, depending on the pH value
and pretreatment in different reaction media. The aim of this work was to explore the
impact and the possible use of L-ascorbic acid as the reaction medium and the pre-re-
ducing agent for hydride generation of different arsenic compounds in the process of
speciation, in order to determine MMA in the presence of inorganic forms of arsenic.
Until now L-ascorbic acid was used as a chemical agent for arsenic preservation and
stabilization in samples with arsenic concentrations 1mg/ml (30, 45, 46). L-ascorbic acid
is also used as pre-reducing agent for the determination of total arsenic and total inor-
ganic forms [As(III) and As(V)], in the presence of thiourea (28, 31) or in combination
with KI (33, 34, 36, 38). In other cases, mixture of L-ascorbic acid and KI was used for
reduction of inorganic As(V) and organic compound arsenic MMA by using cryogenic or
HPLC separation (26, 27, 29, 32, 37). Anderson et al. (44) reported KI utilization in the
determinations of total inorganic arsenic forms. In this paper, results of the study on the
effect of L-ascorbic acid as the reaction medium and the pre-reducing agent are given. To
the best of our knowledge, there are no data about the L-ascorbic acid usage in the
determination of MMA in the presence of inorganic forms of arsenic by hydride genera-
tion-atomic absorption spectrometry.
EXPERIMENTAL
All chemicals were of analytical grade unless otherwise specified. Water used was
obtained from a purification system (Elga) and the conductivity of reagent water was 0.05
μS cm-1. The arsenic stock solutions were prepared as follows: 1 mg/ml As(V)
(As2O5·2H2O in H2O) (Carlo Erba Analytical), 6.6495 mg/ml NaAsO2 (Merck), 1mg/ml
MMA and 0.039376g original solid disodium methyl arsonate hexahydrate >99% (Chem
166
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
Service) were dissolved in 10 ml reagent water; 1mg/ml DMA and 0.028568g original
solid sodium cacodylat trihydrat >98% (Fluka) were dissolved in 10 ml reagent water.
Working standard solutions were prepared daily by diluting appropriate volumes of stock
solution in reagent water. Sodium hydroxide p.a. (Zorka Pharma), sodium tetrahydrobo-
rate p.a. (Fluka), hydrochloric acid 37% (Merck), potassium iodide p.a. (Fluka), L-as-
corbic acid p.a. (Zorka Pharma). Reagent water without detected presence of arsenic
compounds and drinking water with 3 and 4 μg/l of total arsenic were used as samples.
Analyses were performed on an Atomic Absorption Spectrometer Shimadzu AA-680
with Hydride Vapour Generation (HVG-1 3 channels, Shimadzu) connected to the ins-
trument. The sample solutions were pumped into a manifold where they reacted with
acid and sodium tetrahydroborate solution. Using nitrogen, the generated arsines were
swept to a gas-liquid separator and then to a heated T-shaped absorption cell.
The possibility of using L-ascorbic acid in speciation of As(V) and MMA in drinking
water was investigated on three channels system with peristaltic pump (HVG-1) for
hydride generation. Through the channel for the sample aspiration, water solutions of ar-
senic compounds or solutions of arsenic compounds with hydrochloric acid, L-ascorbic
acid or KI were introduced. Hydrochloric acid was introduced through the second chan-
nel and through the third channel reduction solution of sodium tetrahydroborate. After
appropriate investigation and statistical treatment (“t”-test), calibration curve was made
using standard solution of As(III). During the determination of MMA it was necessary to
wait longer for the hydride generation and extended washing of hydride system was also
necessary.
During investigation of the influence of L-ascorbic acid on hydride generation,

through channel for acid were introduced different concentrations of L-ascorbic acid. The
effect of 1-10% L-ascorbic acid on the absorbance signals, in the absence or presence of
HCl in sample is shown in Fig. 1.
The absorption signals of As(III) and DMA increased with the increase of L-ascorbic
acid concentration, while the response for As(V) and MMA was low, caused by the dif-
ferent degrees of protonation of arsenic compounds. Protonation and the formation of
arsines are pH dependent. The pK values of arsenite-As(III), arsenate-As(V), MMA and
DMA, are 9.23, 2.25, 2.60, 6.19, respectively (47). Under these conditions reduction pro-
ducts (arsines) appear, AsH3 from arsenite-As(III) and arsenate-As(V) (boiling point -
55ºC), CH3AsH2 from MMA (boiling point 2ºC) and (CH3)2AsH from DMA (boiling
point 35.6ºC) (47). All obtained signals were lower than in case when hydrochloric acid
was used in hydride generation as reported previously (27). The obtained results were
similar to those obtained by applying the other organic acids like oxalic acid, acetic acid,
tartaric and citric acid. They were also in accordance with the investigation of Anderson
et al. (44). In the case when HCl was present in the sample, higher values of analytical
signals (Fig. 1, dashed line) were obtained and maximum values of relative sensitivity
were obtained when lower concentrations of L-ascorbic acid was used (Fig. 1).
167
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
DMA MMA A s (III) A s (V )
D M A + 0 .1 m o l/l H C L M M A + 0 .1 m o l/l H C L A s (III)+ 0 .1 m o l/l H C L A s (V )+ 0 .1 m o l/l H C L
120
100
80
60
Abs
40
20
0
0 2 4 6 8 10 12
-2 0
L -a s c o r b i c a c id m a s s c o n c e n tr a tio n (% )
Fig. 1. Effect of L-ascorbic acid concentration on the absorption signals of As(III),

As(V), MMA and DMA. Three-channel system: the first channel with sample of
arsenic (10 μg/l As each) in water (dashed line, sample in 0.1 mol/l HCl), the
second channel with different concentrations of L-ascorbic acid and the third
channel with 0.4% NaBH4 in 0.5% NaOH
Similar effect was observed in the case when KI was added as pre-reducing agent in
the samples (Fig. 2). There was no reduction of arsenic(V) compounds and MMA,
because pH value was not sufficiently low. MMA was reduced partially when KI and 0.1
mol/dm3 HCl were added to the sample (Fig. 2, dashed line). Because of the relatively
high pH value, the reduction of As(V) practically was not observed.
D M A M M A A s (III) A s (V )
D M A + 0 . 1 m o l/ l H C l M M A + 0 . 1 m o l/l H C l A s ( I I I) + 0 . 1 m o l/l H C l A s ( V ) + 0 . 1 m o l/l H C l
100
80
60
Abs
40
20
0
0 2 4 6 8 10 12
-2 0
L - a s c o r b ic a c id m a s s c o n c e n tr a tio n (% )
Fig. 2. Effect of L-ascorbic acid concentration on the absorption signals of As(III),

As(V), MMA and DMA. Three-channel system: the first channel with sample of
arsenic (10 μg/l As each) + 1.5ml 40% KI in water (dashed line, sample in 0.1
mol/l HCl), the second channel with different concentrations of L-ascorbic acid
and the third channel with 0.4% NaBH4 in 0.5% NaOH
168
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
The effect of potassium iodide concentration on the degree of As(III), As(V), DMA
and MMA reduction, without or with L-ascorbic acid (Fig. 3, dashed line), was studied.
KI concentration varied within the range of 0.5-3% and results of this experiment are
shown in Fig. 3. As(III) is the reduction form of arsenic and the presence of KI or L-
ascorbic acid has no efect on reduction and generation of its hydride. In the presence of
L-ascorbic acid, DMA was not reduced, neither hydride was generated, because the pH
value of the solution was too low. In the absence of L-ascorbic acid in the sample, re-
duction of MMA was low or there was no reduction, while when L-ascorbic acid was
added, reduction of MMA was complete. Reduction of As(V) was complete in the
presence of L-ascorbic acid. Similar results were obtained by Anderson et al. for As(III)
and As(V) in the absence of L-ascorbic acid (44). These characteristics of As(V) and
MMA were used for the arsenic speciation, in order to determine MMA in the presence
of inorganic forms of arsenic. The results of this investigation has shown that very good
arsenic reduction was obtained when the KI concentration was about 1.5 %. Burguera et
al. in their investigation (47) reported similar observations.
M M A +KI M M A + K I+ L -a s c o rb ic a c id
A s (V )+ K I A s (V )+ K I+ L -a s c o rb ic a c id
A s (III )+ K I A s (III)+ K I+ L -a s c o rb ic a c id
280 D M A+KI D M A + K I+ L -a s c o rb ic a c id
230
180
Abs
130
80
30
-2 0 0 0 ,5 1 1 ,5 2 2 ,5 3 3 ,5
% KI
Fig. 3. Effect of KI mass concentration on the absorption signals of As(III), As(V), DMA
and MMA. Three-channel system: the first channel with sample arsenic (20 μg/l
As each) + different concentration KI in 2 mol/l HCl (dashed line, sample with L-
ascorbic acid), the second channel with 5 mol/l HCl and the third channel with
0.4% NaBH4 in 0.5% NaOH
The influence of the time needed for complete reduction of As (V) and MMA under
optimized experimental conditions is shown in Fig. 4.
Fig. 4 shows that at least 10 minutes were needed for the complete reduction of
MMA and about 40 minutes for arsenate-As(V). DMA was not reduced, neither hydride
was generated, because of the solution pH value (42, 44).
169
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
MMA A s(V ) A s(III) DM A

280
230
180
Abs
130
80
30
-2 0 0 10 20 30 40 50 60 70
tim e (m in )
Fig. 4. Effect of time needed for reduction with KI on the absorption signals of As(III),
As(V), MMA and DMA. Three-channel system: the first channel with sample
arsenic (20 μg/l As each) + 1ml 40% KI + 1ml 5% L-ascorbic acid in 2 mol/l HCl,
the second channel with 5 mol/l HCl and the third channel with 0.4% NaBH4 in
0.5% NaOH
MMA As(V) A-0,6%NaBH4+0,1mol/l HCl

As(III) DMA B-0,6%NaBH4+5 mol/l HCl
280 C-0,4%NaBH4+0,1mol/l HCl
D-0,4%NaBH4+5 mol/l HCl
230
180
Abs
130
80
30
-20 A B C D
combination NaBH4-HCl
Fig. 5. Effect of different reaction media on the generation of arsines of As(III), As(V),
MMA and DMA. Three-channel system: the first channel with sample of arsenic
(20 μg/l As each) + 0.125ml 40% KI in 5 mol/l HCl, the second and third chan-
nels are described as A, B, C and D
170
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
In an attempt to find optimal experimental conditions under which L-ascorbic acid

was used in process of speciation of As(V) and MMA, four experimental conditions we-
re applied (A, B, C and D) as shown in Fig. 5. Under condition of “D” (Fig. 5), MMA
signal was low. The condition “D” was used for the determination of total inorganic ar-
senic [As(III) and As(V)], despite of the presence of L-ascorbic acid. In the case when L-
ascorbic acid was added, MMA was determined too. Concentration of MMA was obtai-
ned as the difference between As(III)+As(V)+MMA in the presence of KI and L-ascorbic
acid and As(III)+As(V) in the presence of KI only. Namely, the aim was to obtain mi-
nimal value of the analytical signal of MMA, not the maximal difference between the
analytical signals of MMA and of inorganic arsenic compounds.
Using appropriate conditions developed in this study, two different samples were
analyzed (reagent and drinking water). Under investigated conditions As(III) was com-
pletely reduced and its hydryde was generated, while hydride was not generated in case
of DMA (because of low pH) and only the content of As(V) and MMA were determined.
As(III) and DMA behave equally both in the presence and in the absence of L-ascorbic
acid, so investigation under chosen conditions had no effect. The results are shown in
Table 1. After separate examination of all standard solutions (As(III), As(V) and MMA)
for calibration curves construction and after the conduction of “t”-test (ttab=2.10, degree
of freedom was 18, number of replicate was n=10, α=0.05), for calibration range of
0.004-0.018 mg/l t value of 0.21-2.07 was obtained. Values of “t”-test indicate that there
is no significant difference between the previously mentioned standard solutions, so
As(III) was chosen for the calibration curves definition.
Table 1. Results for water samples, % recovery and % RSD
Total arsenic Total arsenic

Added (ppb) Recoveryc RSD
Sample Analyzed concentration concentration
(%) (%)
typea speciesb (ppb) before (ppb) after
As(V) MMA (n=5) (n=5)
spiking spiking
RW As(V) 0.0 8.0 8.0 8.1 101.3 1.2
DW As(V) 3.0 8.0 8.0 10.5 88.8 2.4
RW As(V) 0.0 4.0 4.0 3.7 92.5 2.0
RW As(V)+MMA 0.0 8.0 8.0 14.0 87.6 1.3
DW As(V)+MMA 3.0 8.0 8.0 17.7 89.1 1.3
RW As(V)+MMA 0.0 4.0 4.0 7.3 90.7 2.1
DW As(V)+MMA 4.0 4.0 4.0 11.7 92.1 2.7
a- RW-reagent water and DW- drinking water,

b- reaction media is D (Fig. 5); for determination As(V) only KI is added, for determination As(V)+MMA
KI+L-ascorbic acid are added,
c-recovery of analyzed species
Results reported in this study showed that the recovery of added spike was 82-101%
and RSD (%) was 3.6, which are acceptable results for this level of content (ppb). It was
confirmed that the addition of L-ascorbic acid to the samples as the pre-reducing agent
enables the arsine generation from MMA and it was shown that these analyses can be
used in speciation of arsenic compounds. Ascorbic acid is frequently applied as a pre-
reductant for As(V) and MMA together with iodide in order to prevent the liberation of
171
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
iodine. The influence of ascorbic acid on the generation of hydrides of various arsenic
species has not been previously studied in detail. In reference (35) it was mentioned that
there is a possibility that this phenomenon is the result of better protonation in the pre-
sence of L-ascorbic acid or easier approach of the pre-reducing agent (KI) to methylated
arsenic compounds. It is evident that in the process of reduction agent iodine was made
and it was reduced with L-ascorbic acid into iodide, which entered into the process of
reduction of arsenic compounds again. The detailed conclusions about the L-ascorbic
acid influence could be obtained by the investigations of kinetics of these chemical reac-
tions.
Based on the results obtained by the determination of As(V) and MMA in water
samples it can be concluded that L-ascorbic acid can be used as the reaction medium and
the pre-reducing agent for hydride generation of different arsenic compounds in the pro-
cess of speciation, in order to determine MMA in the presence of inorganic forms of
arsenic.
REFERENCES
1. Arsenijević, S.: Opšta i neorganska hemija, Naučna knjiga, Beograd (1986) p. 82.
2. Kabata-Pendias, A.: Trace Elements in Soils and Plants, CRC Press, Boca Raton
(2000) p. 225.
3. Bowen H.J.M.: Environmental Geochemistry of The Elements, Academic Press, New
York (1979) p. 333.
4. Fergusson, J.E.: The Heavy Elements: Chemistry, Environmental Impact and Health
Efect, Pergamon Press, Oxford (1990) pp.7-30.
5. Sandberg, G.R., and I.K. Allen: Arsenical Pesticides, American Chemical Society,
E.A. Woolson Ed. Washington, DC (1975) p. 124.
6. O′Neill, P.O.: Heavy Metals in Soils., Wiley, New York (1990) pp. 105-121.
7. Temple, P.J., S.N.Linzon and B.L. Chai: Contamination of vegetation and soil by
arsenic emissions from secondary lead smelters. Environ. Pollut. 12 (1977) 311-320.
8. Xu, J., and I. Thornton: Arsenic in garden soils and vegetable crops in Cornwall, En-
gland: implications for human health. Environ. Geochem. Health. 7 (1985) 131-133.
9. Li, X., and I. Thornton: Arsenic, antimony and bizmuth in soils and pasture herbage
in some old metalliferous mining areas in England. Environ. Geochem. Health. 15
(1993) 135-144.
10. Loredo, J., A. Ordóñez, J.R. Gallego, C. Baldo and J.G. García-Iglesias: Geochemical
characterization of mercury mining spoil heaps in the area of Mieres (Asturias, nor-
thern Spain). J. Geochem. Explor. 67 (1999) 377-390.
11. Carbonell-Barrachina, A., F. Burló, D. Valero, E. Lopez, D. Martinez-Romero and F.
Martinez-Sanchez: Arsenic toxicity and accumulation in turnipas affected by arsenic
chemical speciation. J. Agric. Food Chem. 47 (1999) 2288-2294.
12. Hunt, L. E. and A.G. Howard: Arsenic speciation and distribution in the Carnum es-
tuary following the acute discharge of contaminated water from a difused mine. Mar.
Pollut. Bull. 28 (1994) 33-38.
13. Ballin, U., R. Kruse and H. A. Rüssel Fresenius’: Determination of total arsenic and
speciation of arseno-betaine in marine fish by means of reaction-headspace gas chro-
172
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
matography-utilizing flame-ionization detection and element specific spectrometric

detection. J. Anal. Chem. 350 (1994) 54-61.
14. Edmonds, J.S., K.A. Francesconi, J.R. Cannon, C.L. Raston, B.W. Skelton, and A.H.
White: Izolation, crystal structure and syntesis of arsenobetaine, the arsenical consti-
tuent of the western rock lobster Panulirus longipes cygnus George. Tetrahedron Lett.
18 (1977) 1543-1546.
15. Beard, J.: Safe to drink-for a few ponce a year, a life can be saved. New. Sci. 2127
(1998) 10-16.
16. IARC Monographs on Evaluation of Carcinogenic Risk to Humans, International
Agency for Research on Cancer, Lyon 23 (1980) 37-141.
17. IARC Monographs on Evaluation of Carcinogenic Risk to Humans Suppl, Internatio-
nal Agency for Research on Cancer, Lyon 7 (1987) 100-106.
18. Enterline, P.E., V.L. Henderson and G.M. Narsh: Exposure to arsenic and respiratory
cancer a reanalysis. Am. J. Epidemiol. 125 (1987) 929-938.
19. Ferreccio, C.R., C.W. Gonzalez, J.S. Solari and C.N. Noder: Bronchopulmonary can-
cer in workers exposed to arsenic; a case control study. Rev. Med. Chile. 124 (1996)
119-123.
20. Hopenhayn, C., M.L. Biggs, A.H. Smith, D.A. Kalman and L.E. Moore: Methylation
study of a population environ mentelly exposed to arsenicin drinking water. Environ.
Health Perspect. 104 (1996) 620-629.
21. Stohrer, G.: Arsenic: opportunity for risk assessment. Arch. Toxicol. 65 (1991) 525-
531.
22. Hung, T.C. and S.M. Liao: Arsenic species in the well water and sediments of the
black foot disease area in Taiwan. Toxicol. Environ. Chem. 56 (1996) 63-73.
23. Ng, J., D. Johnson, P., Imary, B.Chiswell and M. Moore: Speciation of arsenic me-
tabolites in the urin of occupational workers and experimental rats using an optimised
hydride cold-trapping method. Analyst. 123 (1998) 929-935.
24. Draft Toxicological Profile for Arsenic, U.S. Department of Health and Human Ser-
vices, Agency for Toxic Substances and Disease Registry (2005) 45.
25. Pravilnik o higijenskoj ispravnosti vode za piće (Sl. List SRJ 42/98).
26. Munoz, O., D. Velez, M.L. Cervera and R. Montoro: Rapid and quantitative release,
separation and determination of inorganic arsenic [As(III)+As(V)] in seafood pro-
ducts by microwaveassisted distillation and hydride generation atomic absorption
spectrometry. J. Anal. At. Spectrom. 14 (1999) 1607–1613.
27. Petrov, P.K., J.M. Serafimovska, S. Arpadjan, D.L. Tsalev and T. Stafilov: Influence
of EDTA, carboxylic acids, amino- and hydroxocarboxylic acids and monosaccha-
rides on the generation of arsines in hydride generation atomic absorption spectro-
metry. Central European Journal of Chemistry 6, 2 (2008) 216–221.
28. Niedzielski, P. and M. Siepak: Speciation analysis of inorganic form of arsenic in
ground water samples by hydride generation atomic absorption spectrometry with
insitu trapping in graphite tube. Central European Journal of Chemistry 3, 1 (2005)
82–94.
29. Reyes, M.N.M., M.L. Cervera, R.C. Campos and M. de la Guardia: Non-chromato-
graphic speciation of toxic arsenic in vegetables by hydride generation-atomic fluore-
scence spectrometry after ultrasound-assisted extraction. Talanta 75 (2008) 811–816.
173
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
30. Salgado, S.G., M.A.Q. Nieto and M.M.B. Simon.: Assessment of total arsenic and
arsenic species stability in alga samples and their aqueous extracts. Talanta 75 (2008)
897–903.
31. Wei, C. and J. Liu: A new hydride generation system applied in determination of ar-
senic species with ion hromatography–hydride generation-atomic fluorescence spec-
trometry (IC–HG-AFS). Talanta 73 (2007) 540–545.
32. Reyes, M.N.M., M.L. Cervera, R.C. Campos and M. de la Guardia.: Determination of
arsenite, arsenate, monomethylarsonic acid and dimethylarsinic acid in cereals by
hydride generation atomic fluorescence spectrometry. Spectrochimica Acta Part B 62
(2007) 1078–1082.
33. Musil, S. and T. Matoušek: On-line pre-reduction of pentavalent arsenicals by thio-
glycolic acid for speciation analysis by selective hydride generation–cryotrapping–
atomic absorption spectrometry. Spectrochimica Acta Part B 63 (2008) 685–691.
34. Nielsen, S. and E.H. Hansen: Determination of As(III) and As(V) by flow injection-
hydride generation-atomic absorption spectrometry via on-line reduction of As(V) by
ICI. Analytica Chimica Acta 343 (1997) 5-17.
35. Ioannou, P.V.: On the Direct Reduction of Arsonic Acids to Arsenoso Compounds:
Mechanisms a Preparations. Appl. Organometal. Chem. 14 (2000) 261–272.
36. Gonzálvez, A., A. Llorens, M.L. Cervera, S. Armenta and M. de la Guardia: Non-
chromatographic speciation of inorganic arsenic in mushrooms by hydride generation
atomic fluorescence spectrometry. Food Chemistry, in press.
37. Guo, T., J. Baasner and D.L. Tsalev: Fast automated determination of toxicologically
relevant arsenic in urine by flow injection-hydride generation atomic absorption
spectrometry. Analytica Chimica Acta 349 (1997) 313-3I8.
38. C. de Souzaa, S.V, C.T. Pintob and R. G. Junqueira: In-house method validation: Ap-
plication in arsenic analysis. Journal of Food Composition and Analysis 20 (2007)
241–247.
39. Shraim, A., B. Chiswell and H.Olszowy: Speciation of arsenic by hydride generation–
atomic absorption spectrometry (HG–AAS) in hydrochloric acid reaction medium.
Talanta 50 (1999) 1109–1127.
40. Shraim, A., B. Chiswell and H. Olszowy: Use of perchloric acid as a reaction me-
dium for speciation of arsenic by hydride generation–atomic absorption spectrometry.
Analyst 125 (2000) 949–953.
41. Cordos, E.A., T. Frentiu, M. Ponta, B. Abraham and I. Marginean: Optimisation of
analytical parameters in inorganic arsenic (III and V) speciation by hydride genera-
tion using L-cysteine as prereducing agent in diluted HCl medium. Chemical Spe-
ciation and Bioavailability 18 (2006) 1-9.
42. Carrero, P., A. Malavé, J.L. Burguera, M. Burguera and C. Rondón: Determination of
various arsenic species by flow injection hydride generation atomic absorption spec-
trometry: investigation of the effects of the acid concentration of different reaction
media on the generation of arsines. Analytica Chimica Acta 438 (2001) 195–204.
43. Bermejo-Barrera, P., J. Moreda-PinÄeiro, A. Moreda-PinÄeiro and A. Bermejo-Bar-
rera: Selective medium reactions for the àrsenic(III)', àrsenic(V)', dimethylarsonic
acid and monomethylarsonic acid determination in waters by hydride generation on-
174
APTEFF, 40, 1-220 (2009) UDC: 628.1.033:628.16:546.19:577.164.2
DOI: 10.2298/APT0940165S BIBLID: 1450-7188 (2009) 40, 165-175
line electrothermal atomic absorption spectrometry with in situ preconcentration on

Zr-coated graphite tubes. Analytica Chimica Acta 374 (1998) 231-240.
44. Anderson, R.K., M. Thompson and E. Culbard: Selective reduction of arsenic species
by continuous hydride generation. Analyst 111 (1986) 1143-1152.
45. Feldman, C.: Improvements in the arsine accumulation-helium glow detector proce-
dure for determining traces of arsenic. Anal. Chem. 51, 6 (1979) 664–669.
46. Edwards, M., S. Patel, L. McNeill, H.W. Chen, M. Frey, A.D.Eaton, R.C. Antweiler
and H.E. Taylor: Considerations in As analysis and speciation. J. Am. Water Works
Assoc. 90 (1998) 103-113.
47. Burguera, J.L., M. Burguera, C. Rivas and P. Carrero: On-line cryogenic trapping
with microwave heating for the determination and speciation of arsenic by flow injec-
tion-hydride generation atomic absorption spectrometry. Talanta 45 (1998) 531–542.
48. Chanthai, S., N. Suwamat, C. Ruangviriyachai and P. Danvirutai: Speciation Analysis
of Arsenic (III), Arsenic (V) and Total Arsenic by Continuous Flow HG-AAS in Thai
Fruit Wines and Distilled Spirits. ASEAN Food Journal 14 (2007) 181-196.
ПРИМЕНА Л-АСКОРБИНСКЕ КИСЕЛИНЕ ПРИ ОДРЕЂИВАЊУ

РАЗЛИЧИТИХ ОБЛИКА АРСЕНА У ВОДИ ЗА ПИЋЕ
Александар Р. Станић, Саша И. Јованић, Никола Ј. Марјановић и

Звонимир Ј. Сутуровић
Приликом анализе различитих врста узорака вода, животних намирница и узо-

рака из животне средине, указује се потреба да се осим одређивања садржаја укуп-
ног арсена изведе и одређивање различитих облика арсена. Атомска апсорпциона
спектрометрија са грађењем хидрида је метода којом је могуће одредити различите
облике арсена применом различитих услова стварања хидрида. Л-аскорбинска ки-
селина се може употребити као реакциони медијум, односно као прередукциони
агенс у стварању хидрида арсена током специјације. Циљ овог рада је да се испита
могућност коришћења Л-аскорбинске киселине при одређивању органског облика
арсена, монометил арсонске киселине (ММА), у води за пиће у присуству неорган-
ског арсена (Аs(III) и As(V)).
Received 18 May 2009

Accepted 28 July 2009
175
APTEFF, 40, 1-220 (2009) UDC: 628.3.034.2:664.1:628.345
DOI: 10.2298/APT0940177S BIBLID: 1450-7188 (2009) 40, 177-182
TREATMENT OF SUGAR BEET THICK JUICE SPENT WASH BY CHEMICAL

AND NATURAL COAGULANTS
Marina B. Šćiban, Mile T. Klašnja and Mirjana G. Antov
The possibility of treatment of wastewater from bioethanol production by aluminium

sulfate and natural coagulant extracted from common bean seed was studied. The highest
coagulation activity at pH 6.5 is reached with analum dose of 1 g/l, but only a little lower
coagulation activities were obtained by the dose of 0.05 and 0.10 g/l, which is more
favourable for economic and environmental reasons. When natural coagulant from
common bean was applied the highest coagulation activity, 14.3%, at pH 6.5 is reached
with a dose of 0.5 ml/l. However, when common bean natural coagulant was used
simultaneously with alum, the highest turbidity removal resulting in 24% coagulation
activity was achieved and this was more efficient than when alum or natural coagulant
were used.
KEY WORDS: Tick juice stillage; coagulation; alum; natural coagulants
INTRODUCTION
Sugar beet and intermediates from beet-processing are very good raw materials for
bioethanol production, due to their content of fermentable sugars which can be directly
used for fermentation without any pretreatment. Molasses is a traditional raw material for
distilleries, but it is not sufficient for bioethanol production as a fuel. Raw extraction
juice has the lowest price from beet-processing intermediates, but its disadvantage is low
storability. Thick juice is more expensive than extraction juice but its storability is
excellent, that is comparable with molasses (1).
Effluent originating after alcohol distillation, known as distillery wastewater, spent
wash, stillage, and so on, is highly loaded and causes extensive soil and water pollution.
The production and characteristics of spent wash are highly variable and dependent on
feedstock and various aspects of the ethanol production process. For example, an average
molasses based distillery generates 15 L of spent wash per 1 L of ethanol produced (2).
Elimination of pollutants from distillery effluent is becoming increasingly important from
the environmetal point of view. Due to the large volume of these effluent and presence of
certain hardly biodegradable compounds, the treatment of this stream is rather chal-
Dr. Marina B. Šćiban, Assoc. Prof., msciban@uns.ac.rs, Dr. Mile T. Klašnja, Prof., Dr. Mirjana G. Antov,
Assoc. Prof., University of Novi Sad, Faculty of Technology, 21000 Novi Sad, Bulevar Cara Lazara 1, Serbia
177
APTEFF, 40, 1-220 (2009) UDC: 628.3.034.2:664.1:628.345
DOI: 10.2298/APT0940177S BIBLID: 1450-7188 (2009) 40, 177-182
lenging by conventional methods. Therefore, to improve the existing treatments, a num-

ber of novel phisico-chemical and biological treatments have been conducted (3).
The main objective of the present study is preliminary investigation of the usage of
aluminium sulfate and/or natural coagulants from common bean, in the treatment of thick
juice spent wash. Aluminium sulfate (alum) an inorganic salt is the most widely used
coagulant in water and wastewater treatment. The disadvantage of this process is the pre-
sence of aluminium in the obtained sludge, which causes further environmental problems.
Moreover, the obtained sludge can not be used as a feed. Our previous studies indicated
the ability of extract from common bean (Phaseolus vulgaris) seed to act as a natural
coagulant (4). In this study, common bean natural coagulant will be investigated for its
suitability to substitute alum in thick juice spent wash treatment.
EXPERIMENTAL
Wastewater
Wastewater was collected from the Laboratory for ethanol and yeast, Faculty of
Technology, Novi Sad, after bioethanol production experiments (1). The characteristics
of wastewaters are shown in Table 1.
Table 1. Typical characteristic of thick juice spent wash
Parameter Range Average value
pH 4.68 – 4.85 4.74
Settleable matter (ml/l) 1.8 – 3.5 2.5
Suspended solids - SS (mg/l) 8540 - 9815 9322
Total solids - TS (mg/l) 31900 - 32830 32345
Total fixed residue at 550 oC (mg/l) 2937 - 3200 3150
Total volatile residue (mg/l) 28963 - 29630 29220
Chemical oxygen demand - COD (mgO2/l) 85230 - 90000 87600
Total Kjeldahl nitrogen - TKN (mg/l) 980 - 1005 995
It can be seen from Table 1 that the thick juice spent wash is highly acidic in nature,
and highly loaded with organic matter. It contains high suspended solids, making about
30% of total solids. These suspended solids have poor settleability, which is indicated by
very small amount of settleable matter. Analysis of the data demonstrated that the thick
juice spent wash has a little better characteristics than molasses spent wash (5, 6) - the
large advantage of thick juice spent wash is that it does not contain coloured pigments.
Because these pigments are hardly biodegradable, thick juice spent wash is much easier
for treatment than the molasses one.
Coagulants
Al2(SO4)3x18H2O (alum) was used as 5% sollution. Natural coagulants were obtained
in this way: white common bean seeds were ground and sieved through the sieve with
pore size of 0.4 mm. An amount of 10 g/l of a smaller fraction was suspended in 0.5
mol/l NaCl solution. This suspension was stirred for 10 minutes on a magnetic stirrer in
178
APTEFF, 40, 1-220 (2009) UDC: 628.3.034.2:664.1:628.345
DOI: 10.2298/APT0940177S BIBLID: 1450-7188 (2009) 40, 177-182
order to extract active coagulant components. After that, the suspension was filtered
through arugged filter paper. The obtained filtrate - crude extract was stored in a fridge at
5ºC.
Coagulation test
Jar test was carried out to evaluate coagulation activity of alum and natural coagulant.
A volume of 100 ml of wastewater was filled in four 500 ml bakers. Before coagulation
the pH was adjusted by 1 mol/l NaOH or 1 mol/l HCl to a desired value. The coagulants
were added to the beakers at different doses, and the content was stirred at 200 rpm for
1 minute. The stirring speed was then reduced to 80 rpm and was kept for 30 minutes.
Then, the suspensions were left to allow sedimentation. After 1 hour of sedimentation, an
aliquot of 10 ml of clarified sample was collected from the top of the beaker and pH and
COD were determinated. The residual COD of sample was CODS. The same coagulation
test was performed with no coagulant as the blank. The residual COD in the blank was
CODB. Coagulation activity was calculated as follows:
Coagulation activity (%) = 100⋅(CODB – CODS)/CODB [1]
Analytical methods
pH, settleable matter, suspended solids, total solids, total fixed residue at 550 oC,
COD and TKN were analyzed in conformity to Standard Methods for the Examination of
Water and Wastewater (7).
Coagulation with alum
As the first step in our investigation was studied the influence of pH and alum dose
on efficiency of the coagulation-flocculation process. Table 2 shows theeffect of different
doses of alum on the coagulation activity at the original wastewater pH of 4.64.
The application of alum at the original wastewater pH was not appropriate because of
the very low coagulation activity achieved. This might be expected considering that re-
commended pH range for alum application is from 5.5 to 7. Because of that, the pH value
of wastewater was adjusted to 6.50 before coagulation. The results of these experiments
are presented in Table 3.
Table 3 shows an important improving of removing of colloidal particles from inves-
tigated wastewater by alum. The highest coagulation activity at pH 6.5 is achieved with
analum dose of 1 g/l, but only a little lower coagulation activities were obtained by doses
of 0.05 and 0.10 g/l. Coagulation with low alum dose is very favourable because of lower
treatment costs and lower aluminium concentration in the obtained sludge.
179
APTEFF, 40, 1-220 (2009) UDC: 628.3.034.2:664.1:628.345
DOI: 10.2298/APT0940177S BIBLID: 1450-7188 (2009) 40, 177-182
Table 2. Effect of alum dosage on coagulation activity at pH 4.64
Alum dosage (g/l) Coagulation activity (%) pH of water after treatment

0.050 3.99 4.65
0.100 3.99 4.64
0.150 NE* 4.64
0.250 NE 4.62
0.500 NE 4.58
*
No effect
Table 3. Effect of alum dosage on coagulation activity at pH 6.50
Alum dosage (g/l) Coagulation activity (%) pH in water after treatment

0.013 7.94 6.50
0.025 NE* 6.50
0.050 10.70 6.45
0.100 10.70 6.40
0.150 8.02 6.37
0.250 NE 6.31
0.500 3.78 6.14
1.000 12.30 5.93
2.000 NE 5.52
3.000 4.23 5.25
*
No effect
In the further experiment, natural coagulant was added solely to wastewater at pH

6.50. The optimum pH for coagulation with natural coagulants from common bean was
above 9 (Šćiban et al., 2005), but the adjusment to this pH requires large amount of an
alkaline solution, and this pH is not appropriate for alum coagulation. Table 4 shows the
effect of different doses of natural coagulants on the coagulation activity at pH 6.50.
The highest coagulation activity of natural coagulants at pH 6.5 is achieved with a
dose of 0.5 ml/l, and only a little lower coagulation activity was obtained at a dose of
0.25 ml/l; to obtaine these doses of natural coagulant it was necessary to extract 5 mg and
2.5 mg common bean seed, respectively.
Table 4. Effect of natural coagulant dosage on coagulation activity at pH 6.50
Natural coagulant dosage (ml/l) Coagulation activity (%)

0.250 10.9
0.500 14.3
1.000 NE*
2.000 5.8
5.000 NE
10.00 NE
*
No effect
In the next experiments, the effect of natural coagulant doses was investigated with
simultaneous addition of fixed dose of alum 0.05 g/l (Fig. 1).
180
APTEFF, 40, 1-220 (2009) UDC: 628.3.034.2:664.1:628.345
DOI: 10.2298/APT0940177S BIBLID: 1450-7188 (2009) 40, 177-182
30
Coagulation activity (%) 25
with 0.05 g/l of alum
20
with natural coagulant
15
10 with alum + natural

coagulant
5
0
0.25 0.5 1.0
Natural coagulant dose (ml/l)
Fig. 1. Effect of natural coagulant dosage on coagulation activity at pH 6.50,

with addition of 0.05 g/l of alum
Very good removal efficiency of organic matter from water was achieved when alum
and natural coagulant were applied in combination 0.05 g/l and 0.25 ml/l, respectively.
Moreover, when common bean natural coagulant was used simultaneously with alum, the
higher turbidity removal was achieved in comparison to sole application of alum or
natural coagulant. Although the coagulation activity achieved might seem relatively low
at the first sight, the maximal achievable coagulation activity in this wastewater can be
about 32% (see ratio suspended solids : total volatile residues from Table 1).
CONCLUSION
When natural coagulant from common bean was applied for wastewater treatment,
more efficient removal of suspended solids was achieved. According to the results, com-
mon bean could be used as a substitute for conventional chemical coagulants such as
alum. The highest coagulation activity was obtained when natural coagulant and alum
were added to wastewater simultaneously. Regarding its biodegradability, food grade and
renewable nature, common bean has a bright future as a source of coagulant for both
water and wastewater treatment.
ACKNOWLEDGEMENTS
The financial support of the Ministry of Science and Technological Development of
the Republic of Serbia (Project No. 20009) is greatly acknowledged.
REFERENCES
1. Dodić S., S. Popov, J. Dodić, J. Ranković, Z. Zavargo and R. Jevtić Mučibabić:
Bioethanol production from thick juice as intermediate of sugar beet processing.
Biomass Bioenergy 33 (2009) 822-827.
2. Beltran F.J., P.M. Alvarez, E.M. Rodrigez, J.F. Garcia-Araya and J. Rivas: Treatment
of high strength distillery wastewater (cherry stillage) by integrated aerobic biological
oxidation and ozonation. Biotechnol. Progress 61 (2001) 462-467.
181
APTEFF, 40, 1-220 (2009) UDC: 628.3.034.2:664.1:628.345
DOI: 10.2298/APT0940177S BIBLID: 1450-7188 (2009) 40, 177-182
3. Pant D. and A. Adholeya: Biological approaches for treatment of distillery waste-

water: A review. Biores. Technol. 98 (2007) 2321-2334.
4. Šćiban, M., M. Klašnja and J. Stojimirović: Investigation of coagulation activity of
natural coagulants from seeds of different leguminose species. Acta Periodica
Technol. 36 (2005) 81-87.
5. Preeti, C.S. and A.B. Pandit: Enhancement in biodegradability of distillery wastewa-
ter using enzymatic pretreatment. J. Environ. Menage. 78 (2006) 76–85.
6. Kumar G.S., S.K. Gupta and G. Singh: Biodegradation of distillery spent wash in
anaerobic hybrid reactor. Water Res. 41 (2007) 721-730.
7. APHA, AWWA, WEF, Standard Methods for the Examination of Water and Waste-
water, IWA Publishing, Washington DC, 1992.
ТРЕТМАН ЏИБРЕ ОД ГУСТОГ СОКА ХЕМИЈСКИМ И ПРИРОДНИМ

КОАГУЛАНТИМА
Марина Б. Шћибан, Миле Т. Клашња и Мирјана Г. Антов
У раду је испитивана могућност третмана отпадне воде од производње биоета-

нола алуминијум сулфатом и природним коагулантима екстрахованим из зрна па-
суља. Најбоља коагулациона активност је постигнута на рН 6,5 са дозом алуми-
нијум сулфата од 1 g/l, а само нешто мање коагулационе активности су остварене
са дозама од 0,05 и 0,1 g/l, што је веома погодно како са економског, тако и са
становишта заштите животне средине. Када је примењен само природни коагулант
на рН 6,5, постигнута је коагулациона активност од 14,3% са дозом од 0,5 ml/l.
Када се природни коагулант применио истовремено са алуминијум сулфатом, по-
стигнута је највећа коагулациона активност од 24%, која је боља од коагулационих
активности постигнутих појединачном применом алуминијум сулфата и природног
коагуланта.

182
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
SIMPLE CORRELATIONS FOR BUBBLE COLUMNS AND DRAFT TUBE

AIRLIFT REACTORS WITH DILUTE ALCOHOL SOLUTIONS
Ivana M. Šijački, Radmilo R. Čolović, Milenko S. Tokić and Predrag S. Kojić
Simple empirical correlations were developed to predict gas holdup, liquid circula-
tion time, downcomer liquid velocity and volumetric mass transfer coefficient in dilute
alcohol solutions in bubble columns and draft tube airlift reactors with single orifice
sparger. Also, new experiments were conducted with diluted alcohol solutions to
n-octanol, expanding the experimental data from C1 up to C8. The proposed empirical
correlations include, beside the superficial gas velocity, the alcohol chain length as the
only factor to characterize the liquid phase. The suggested correlations have shown good
agreement between the calculated and the experimental data.
KEY WORDS: Bubble columns, draft tube airlift reactors, dilute alcohol solutions,
hydro-dynamics, mass transfer
INTRODUCTION
Bubble columns and airlift reactors have important applications as bioreactors (in bio-
mass production or production of different metabolites), as chemical reactors and as con-
tactors in the wastewater treatments. In these contactors, the properties of the liquid phase
strongly affect hydrodynamics, bubble behavior and mass transfer rates. Dilute alcohol
solutions are important, as the liquid phase. They can be used to simulate the liquid phase
behavior in bioreactors and in coal liquefaction (1). The only property of these solutions,
which differs considerably from water, is their surface tension (2).
The influence of alcohols on the gas holdup is proportional to their concentration and
to the length of the carbon chain in the alcohol molecule, in the bubble column (BC) (1,
3), continuous BC (1), external loop airlift reactor (EL-ALR) (4), draft tube air lift reactor
(DT-ALR) (5,6) and split rectangular airlift reactor (SR-ALR) (7). Also, the addition of
alcohol influences volumetric mass transfer coefficient (kLa). The increase in both the al-
cohol concentration and the length of the straight chain of alcohol molecule results in the
increase of kLa, in the BC (3), the EL-ALR (4) and in the DT-ALR (5,6).
Ivana M. Šijački, B. Sc., isijacki@uns.ac.rs, University of Novi Sad, Faculty of Technology, Bulevar Cara
Lazara 1, 21000 Novi Sad, Serbia; Radmilo R. Čolović, B. Sc., Food Institute, Bulevar Cara Lazara 1, 21000
Novi Sad, Serbia; Milenko S. Tokić, B. Sc., Predrag S. Kojić, B. Sc., University of Novi Sad, Faculty of
Technology, Bulevar Cara Lazara 1, 21 000 Novi Sad, Serbia
183
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
Only a few studies on the hydrodynamics and mass transfer in dilute alcohol solutions
in bubble columns and draft tube airlift reactors with single orifice sparger have been
published, so far (3, 5, 6, 8).
The correlations for prediction of gas holdup and volumetric mass transfer coefficient
were proposed in various forms based on investigations in different systems. Some of the
correlations can be used for predicting hydrodynamics and mass transfer in aqueous so-
lutions of alcohols with an acceptable error (3, 5, 6, 9). In deriving the correlation, the
main issue was choosing a representative characteristic of the liquid phase. In the case of
addition of alcohols, it is obvious that the only physical property that is to be different
from water is the surface tension (3, 5, 9), as the change in viscosity and density is negli-
gible, or the surface tension gradient can be used in correlations (6, 10), in order to in-
clude the effect of solution concentration on physical properties.
The aim of this paper was to propose simple correlations that could be used to predict
the main characteristics: gas holdup, liquid circulation time, downcomer liquid velocity
and volumetric mass transfer coefficient, in BCs and DT-ALR, with aqueous solutions of
alcohols and with a single orifice sparger as a gas distributor. In order to improve the
correlations and validate their form, the existing experimental data (3, 5, 6, 8) were broa-
dened by additional experiments, conducted in DT-ALR with dilute alcohol solutions
from methanol to n-octanol. The concentration of each alcohol was chosen based on the
research of Keitel (11), as the value of the upper limiting concentration. It has been
shown that increasing alcohol concentration above the limiting value, only enhances the
bubble coalescence and liquid phase frothing (2, 8), with no major influence on hydro-
dynamics. In case of isopropanol, the concentration is chosen equal to n-propanol.
EXPERIMENTAL
The experiments were conducted at 20±1°C and atmospheric pressure in a glass

DT-ALR, with geometrical details presented in Fig. 1. The air, sparged through a single
orifice into the draft tube, was used as the gas phase. Tap water and diluted alcohol solu-
tions from methanol to n-octanol were used as the liquid phase.
The concentrations of the alcohols and the physical properties of the liquid phase at
20°C are summarized in Table 1. Densities of liquids were measured by a densitometer
AP PAAR DMA46 with ±0.1 kg/m3 accuracy. Surface tensions of liquid phases were
obtained by tensiometer (Torsion Balance Model OS) with ±0.0001 N/m accuracy. The
surface tension gradient (-dσ/dCA) was estimated from the slope of the experimental σ
versus CA curve (Fig. 2).
184
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
Fig. 1. Experimental setup
Fig. 2. Evolution of surface tension with concentration of aqueous alcohol solutions
185
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
o
Table 1. Surface tension and surface tension gradient of used liquid phases at 20 C
Liquid Concentration Density Surface Surface Tension

CA, wt % ρ, kg/m3 Tension Gradient
σ, 10-3 N/m - dσ/dCA,10-3 N m2/mol
tap water 0 1000 72.4 -
Methanol 3.2 921.61 66.1 0.006
Ethanol 0.46 988.06 70.4 0.025
n-propanol 0.036 999.10 71.9 0.067
Isopropanol 0.036 998.99 71.7 0.080
n-butanol 0.011 999.74 71.8 0.199
n-pentanol 0.0057 999.87 71.7 1.082
n-hexanol 0.0051 999.89 69.9 1.985
n-heptanol 0.002 999.96 67.8 4.141
n-octanol 0.002 999.96 65.0 15.205
The overall gas holdup was determined by the volume expansion technique with an
error less than 10%. The aerated dispersion height without foam was used for calculating
the overall gas holdup. The gas holdup values along the column were obtained by measu-
ring the differential pressure at five points (in the draft tube and in the downcomer) using
the piezometric tubes. The relative average error of the measurement was up to 2%. A hot
probe method, developed for this purpose, was used to determine the liquid velocity in
the downcomer (12). The mean relative error of this method was ±5%.

The main aim of this paper was to suggest simple empirical correlations to predict
crucial hydrodynamic and mass transfer quantities. The processed data in this investi-
gation, beside the ones obtained in these experiments, were taken from the experiments
of Pošarac and Tekić (3), Albijanić (5), Albijanić et al. (6) and Camarasa et al. (8). In the-
se studies experiments were conducted in BCs and DT-ALR with single orifice as air
sparger and with dilute alcohol solutions as the liquid phase.
As already mentioned, a representative characteristic of diluted alcohol solutions
might be surface tension. Fig. 2 shows evolution of surface tension with the concentration
of alcohol solution. The surface tension gradient was estimated from the slopes of these
experimental curves (Table 1). It is obvious that the surface tension gradient is a function
of CN-value. By applying the regression analysis (13) on these experimental data the
correlation was obtained in the following form:
1.31⋅ C
dσ e N
− = 0.0034 ⋅ [1]
dC A CN
with the coefficient of determination R2=0.99.
Assuming the linear relation: surface tension vs. alcohol solution concentration, with
the surface tension gradient as a slope, and having in mind the connection between sur-
186
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
face tension gradient and the CN-value, it can be concluded that surface tension is also a
function of CN-value. Recently, Zeppieri came to the same conclusion, i.e., he noticed the
relation between the surface tension and CN-value (14). A strong relation between the
CN-value, on one hand, and the surface tension and the surface tension gradient on the
other, suggests that the CN-value could be the only variable which expresses the influence
of dilute alcohol solutions on hydrodynamics and mass transfer.
So, simple correlations were introduced by processing the data obtained in our expe-
riments and the experiments of other authors (3, 5, 6, 8):
y = p1U Gp2 (1 + C N ) p3 [2]

where y represents: ε G , t C , WLD or k L a .
An additional corroboration of the suggested equations was performed through the

comparison of the experimental and calculated values in the case of n-propanol and iso-
propanol, with the same number of C-atoms. A good agreement is achieved, regardless of
the structural differences between these two alcohols. Table 2 contains the values of the
estimated parameters in the proposed equations, along with the standard quantifiers - the
coefficient of determination (R2) and the errors of parameters (expressed as a % of a
value).
Table 2. Values of correlation parameters for the gas holdup, liquid circulation time,
liquid velocity and volumetric mass transfer coefficient
y (Eq.2) Regime p1± Error(p1) p2 ± Error(p2) p3 ± Error(p3) R2 n δ (%)

εG 1.58 ± 5.7% 0.86 ± 2.3% 0.18 ± 5.6% 0.93 349 26.4
I 2.46 ± 4.9% -0.29 ± 3.4% 0.06 ± 16.7% 0.98 24 1.8
tC (s) II 5.07 ± 5.7% -0.11 ± 9.1% 0.16 ± 6.2% 0.87 36 2.9
III 1.15 ± 11.3% -0.56 ± 7.1% 0.16 ± 6.2% 0.94 34 2.6
WLD(m/s) I 3.09 ± 19.1% 0.61 ± 6.6% -0.12 ± 8.3% 0.93 29 3.9
II 0.63 ± 9.5% 0.27 ± 7.4% -0.18 ± 5.5% 0.89 45 3.4
III 3.11 ± 12.8% 0.74 ± 5.4% -0.16 ± 6.2% 0.91 65 4.0
kLa (s-1) 1.45 ± 14.5% 1.24 ± 3.2% 0.41 ± 9.8% 0.90 159 22.6
For the gas holdup, Eq. 2 predicted about 68% of experimental data with an error of
20% or less (Fig. 3). The parameters for liquid circulation time were calculated based on
data of Albijanić (5) and Albijanić et al. (6) as the only available data. The parameters for
liquid circulation time and liquid velocity (Table 2) were calculated for three different
bubble regimes: regime I (small bubbles in the downcomer), regime II (stagnant swarm
of bubbles in the downcomer) and regime III (circulation of bubbles through the column).
187
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
Fig. 3. Comparison between calculated (Eq. 2) and experimental values of gas holdup
Comparison between calculated and experimental values both of circulation time and
liquid velocity is given in Fig. 4 and Fig. 5. The correlation for volumetric mass transfer
coefficient predicts about 52% of experimental data, in a range of 20% error (Fig. 6).
However, a significantly smaller error was achieved for the higher superficial gas velo-
cities.
188
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
Fig. 4. Comparison between calculated Fig.5. Comparison between (Eq. 4) cal-

and experimental values of gas culated (Eq. 4) and experimental
holdup values of downcomer liquid ve-
locity
189
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
Fig. 6. Comparison between calculated (Eq. 4) and experimental values of volumetric

mass transfer coefficient
CONCLUSION
Simple correlations were proposed for prediction of basic hydrodynamic and mass
transfer characteristics of BCs and DT-ALRs with dilute alcohol solutions (from me-
thanol to n-octanol) and with single orifice as a gas distributor. The correlations include
the number of C-atoms in a molecule chain, as the only variable to specify the physical
properties of dilute alcohol solutions. A good agreement between the experimental and
the calculated data was achieved. Based on the simplicity of the proposed correlations it
is to be expected that their useful application in reactors design.
ACKNOWLEDGEMENT
This research was supported by the Ministry for Science and Technological Develop-
ment of the Republic of Serbia. (Project No. 142045).
Notation:
CA = concentration of alcohol, wt%

kLa = volumetric mass transfer coefficient, 1/s
CN = number of C-atoms in alcohol molecule chain
tc = liquid circulation time, s
d = diameter of the orifice, mm
UG = superficial gas velocity, m/s
190
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
D = diameter of the column, m

WLD = downcomer interstitial liquid velocity, m/s
DR = diameter of the riser, m
Greek letters: Subscripts:

δ = average relative error C = circulation
εG = gas holdup D = downcomer
ρ = density, kg/m3 G = gas phase
σ = surface tension, N/m L = liquid phase
R = riser
REFERENCES
1. Kelkar, B.G., S.P. Godbole, M.F. Honath, Y.T. Shah, N.L. Carr and W.D. Deckwer:
Effect of Addition of Alcohol on Gas Holdup and Backmixing in Bubble Column.
AIChE J. 29 (1983) 361-369.
2. Freitas, C. and J. Teixeira: Effect of Liquid-phase Surface Tension on Hydrodyna-
mics of a Three-Phase Airlift Reactor with an Enlarged Degassing Zone. Bioprocess
Eng. 19 (1988) 451–457.
3. Pošarac, D. and M. Tekić: Gas Holdup and Volumetric Mass Transfer Coefficient in
Bubble Columns with Dilute Alcohol Solutions. AIChE J. 33 (1987) 497-499.
4. Pošarac, D.: Investigation of Hydrodynamics and Mass-Transfer in a Three Phase
External-loop Airlift Reactor, Ph.D. Thesis, University of Novi Sad, 1988.
5. Albijanić, B.: Investigation of Alcohol Addition on Hydrodynamics and Mass-
Transfer in a Draft-tube Airlift Reactor, M.Sc. Thesis, University of Novi Sad, 2006.
6. Albijanić, B., V. Havran, D.Lj. Petrović, M. Đurić and M.N. Tekić: Hydrodynamics
and Mass Transfer in a Draft Tube Airlift Reactor with Dilute Alcohol Solutions.
AIChE J. 53 (2007) 2897-2904.
7. El Azher, N., B. Gourich, C. Vial, M. Soulami Bellhaj, A. Bouzidi, M. Barkaoui and
M. Ziyad: Influence of Alcohol Addition on Gas Holdup, Liquid Circulation Velocity
and Mass Transfer Coefficient in a Split–Rectangular Airlift Bioreactor. Biochem
Eng J. 23 (2005) 161-167.
8. Camarasa, E., C. Vial, S. Poncin, G. Wild, N. Midoux and J. Bouillard: Influence of
Coalescence Behaviour of the Liquid and of Gas Sparging on Hydrodynamics and
Bubble Characteristics in a Bubble Column. Chem Eng Process. 38 (1999) 329-344.
9. Albijanić, B., M. Đurić, D. Petrović and M. Tekić: Prediction of Gas Holdup for
Alcohol Solutions in a Draft-Tube Bubble Column. Acta Periodica Technologica. 37
(2006) 71-82.
10. Syeda, S.R., A. Afacan and K.T. Chuang: Prediction of Gas Holdup in a Bubble Co-
lumn Filled with Pure and Binary Liquids. Can J Chem Eng. 80 (2002) 44-50.
11. Keitel, G.: Untersuchungen zum Stoffaustauch in Gas-Flüssig-Dispersionen in Rühr-
schlaufenreaktor und Blasensäule, Ph.D. Thesis, Universität Dortmund, 1978.
12. Tokić, M.: Influence of Distributor Type on Hydrodynamic in a Draft Tube Airlift
Reactor- Perforated Plate, B.Sc. Thesis, University of Novi Sad, 2008.
191
APTEFF, 40, 1-220 (2009) UDC: 66.011:66.069.82:661.722
DOI: 10.2298/APT0940183S BIBLID: 1450-7188 (2009) 40, 183-192
13. Silva Wilton, P., and M.D.P.S. Cleide: LAB Fit Curve Fitting Software (Nonlinear
Regression and Treatment of Data Program) V 7.2.38 (1999-2007).
14. D'Aubetere, A., R. Da Silva and M.E. Aguilera: Experimental Study on Marangoni
Effect Induced by Heat and Mass Transfer. Int Comm Heat Mass Transf. 32 (2005)
677-684.
ПРЕДВИЂАЊЕ ОСНОВНИХ ХИДРОДИНАМИЧКИХ И

МАСЕНОПРЕНОСНИХ КАРАКТЕРИСТИКА У БАРБОТАЖНИМ
КОЛОНАМА СА И БЕЗ УНУТРАШЊЕ ЦЕВИ СА РАЗБЛАЖЕНИМ
РАСТВОРИМА АЛКОХОЛА
Ивана М. Шијачки, Радмило Р. Чоловић, Миленко С. Токић и Предраг С. Којић
У овом раду предложене су корелације за предвиђање садржаја гаса, времена

рециркулације течности, брзине течности у силазној цеви и запреминског коефи-
цијента прелаза масе у барботажним колонама са и без унутрашње цеви, са једно-
струким уводником као дистрибуторoм гаса и разблаженим растворима алкохола
као течном фазом. Расположиви експериментални подаци употребљени за извође-
ње корелација проширени су новим експериментима са разблаженим раствoрима
С1-С8 алкохола. Предложене корелације укључују, поред привидне брзине гаса,
дужину ланца молекула алкохола (број С-атома), као једину величину која карак-
терише течну фазу. Предложене корелације омогућавају врло добро предвиђање
експерименталних података. Захваљујући једноставном облику, може се очекивати
успешна примена изведених корелација при пројектовању оваквих типова реак-
тора.
Received 4 May 2009

192
BIOCHEMICAL AND PHARMACEUTICAL ENGINEERING
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
ANTIMICROBIAL EFFECTS OF SPICES AND HERBS ESSENTIAL OILS
Marija M. Škrinjar and Nevena T. Nemet
Spices and herbs have been used as food additives since ancient times, as flavouring
agents but also as natural food preservatives. A number of spices shows antimicrobial
activity against different types of microorganisms. This article gives a literature review
of recent investigations considering antimicrobial activity of essential oils widely used
spices and herbs, such as garlic, mustard, cinnamon, cumin, clove, bay, thyme, basil, ore-
gano, pepper, ginger, sage, rosemary etc., against most common bacteria and fungi that
contaminate food (Listeria spp., Staphylococcus spp., Salmonella spp., Escherichia spp.,
Pseudomonas spp., Aspergillus spp., Cladosporium spp. and many others). Antimicrobial
activity depends on the type of spice or herb, type of food and microorganism, as well as
on the chemical composition and content of extracts and essential oils. Summarizing re-
sults of different investigations, relative antimicrobial effectiveness can be made, and it
shows that cinnamon, cloves and mustrad have very strong antimicrobial potential, cu-
min, oregano, sage, thyme and rosemary show medium inhibitory effect, and spices such
as pepper and ginger have weak inhibitory effect.
KEY WORDS: Spices, herbs, antimicrobial, antifungal activity
INTRODUCTION
In the recent years, consumers have become more concerned about the processed
food they eat. Syntehtic preservatives, which have been used in foods for decades, may
lead to negative health consequences (1). Besides, the use of synthetic compounds have
significant drawbacks, such as increasing cost, handling hazards, concerns about residues
on food and threat to human environment (2). Therefore, there has been increasing in-
terest to replace synthetic preservatives with natural, effective and nontoxic compounds.
Those are, in the first place, extracts and essential oils (EOs) of spices and herbs (3). As
natural foodstuffs, spices and herbs appeal to all who question safety of synthetic food
additives and demand high-quality products that at the same time are safe and stable (4).
Spices and herbs have been added to food since ancient times, not only as flavouring
agents, but also as folk medicine and food preservatives (5-7). Spices occupy a prominent
place in the traditional culinary practices and are indispensable part of daily diets of mil-
Dr. Marija M. Škrinjar, Prof., skrinjarm@uns.ac.rs, Nevena T. Nemet, B.Sc., nevenan@uns.ac.rs, Faculty of
Technology, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia
195
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
lions of people all over the world. They are essentially flavouring agents used in small
amounts and are reported to have both beneficial effect and antimicrobial properties (8).
Nowadays, plenty of spices and herbs are valued for their antimicrobial activities and
medicinal effects in addition to their flavour and fragrance qualities (9).
Spices and herbs
There is no particular definition of spices, mostly because they are derived from
different parts of the plants, such as cardamom from seed, bay leaf from leafs, clove from
flower bud, pepper from fruit, cinnamon from bark or ginger from rhyzome. Furthermore,
there is no a common method to classify spices. They can be clasiffied by their flavour
and colour, i.e., hot (pepper), pungent (garlic), aromatic (cinnamon, clove), colouring
(turmeric) and herbaceous (rosemary, sage), or according to their taste, such as sweet,
spicy, sour, bitter and astringent (10).
Numerous studies have been published on the antimicrobial activities of plant extracts
against different types of microbes, including foodborne pathogens (5, 11, 12). It has
been reported that spices owe their antimicrobial properties mostly to the presence of
alkaloids, phenols, glycosides, steroids, essential oils, coumarins and tannins (13). As re-
viewed by López-Malo et al. (14), some of antimicrobial components that have been
identified in spices and herbs are: eugenol from cloves, thymol from thyme and oregano,
carvacrol from oregano, vanillin from vanilla, allicin from garlic, cinnamic aldehyde
from cinnamon, allyl isothiocyanate from mustard, etc.
Among these products, particular interest has been focused on essential oils (EOs)
and their components (15), because they are known to be active against a wide variety of
microorganisms, including food-borne pathogens and spoilage bacteria (16, 17). EOs are
aromatic and volatile oily liquids of an aromatic plant’s secondary metabolism. They are
normally formed in special cells or groups of cells, found in leaves and stems, and com-
monly concentrated in one particular region such as leaves, bark or fruit (18). EOs have
long served as flavouring agents in food and beverages, but they are much more impor-
tant because of their antimicrobial activity, which is assigned to a number of small mo-
lecules of terpenoids and phenolic compounds (thymol, carvacrol, eugenol) (19, 20).
Antibacterial activity of EOs
Nedorostova et al. (21) identified antibacterial properties of EOs in vapour phase

against five foodborne bacteria - Escherichia coli, Listeria monocytogenes, Pseudomonas
aeruginosa, Salmonella enteritidis and Staphylococcus aureus. In vitro antibacterial acti-
vity of 27 EOs in vapour phase was evaluated by modified disc volatilization method (22)
at eight different concentrations (0.0042–0.53 µl/cm3), and the minimum inhibitory con-
centrations (MICs) were recorded. The MIC was expressed as microlitres of EOs per vo-
lume unit of atmosphere above the organism growing on the agar surface, and it was
defined as the lowest concentration which made clearly visible inhibition zone. Results
are summarised at Table 1.
Thirteen of the 27 EOs were active at least against one bacterial strain in the range of
tested concentrations. The most effecient was Armoracia rusticana (horseradish), which
196
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
inhibited both Gram-positive and Gram-negative strains with MIC 0.0083 µl/cm3, and
Allium sativum (garlic), which was significantly more active against Gram-positive
(MICs 0.0083 µl/cm3) than against Gram-negative (MICs 0.26–0.53 µl/cm3) bacteria. It is
important to notice that P.aeruginosa was inhibited only by these two, horseradish and
garlic, EOs. S. aureus was the most susceptible bacterium (inhibited by all active EOs),
followed by E. coli > L.monocytogenes > S. enteritidis > P. aeruginosa, which was the
least inhibited. In general, Gram-negative strains were somewhat less inhibited than
Gram-positive strains. Findings of this research suggest that horseradish, garlic, oregano,
marjoram, savory, thyme, large thyme and wild thyme EOs are highly effective in vapour
phase and could be potentially used to fight against foodborne bacterial pathogens.
Table 1. MICs (µl/cm3) of essential oils in vapour phase effective against foodborne
bacteria (21)
Yield Gram-positive Gram-negative

Plant species
% LM SA EC PA SE
Allium sativum 0.33 0.0083 0.0083 0.53 0.53 0.26
Armoracia rusticana 0.03 0.0083 0.0083 0.0083 0.0083 0.0083
Caryopteris x
0.15 - 0.53 - - -
clandonensis
Hyssopus officinalis 0.16 - 0.53 - - -
Mentha x villosa 0.83 - 0.53 - - -
Nepeta x faassenii 0.33 - 0.53 - - -
Ocimum basilicum var.
0.08 - 0.53 - - -
Grant verte
Origanum majorana 0.53 - 0.53 0.26 - -
Origanum vulgare 0.8 0.066 0.017 0.066 - 0.13
Satureja montana 0.28 0.26 0.033 0.033 - 0.26
Thymus pulegoides 0.15 0.26 0.033 0.033 - 0.26
Thymusseryllum 0.27 0.53 0.033 0.033 - -
Thymus vulgaris 0.23 0.26 0.017 0.033 - 0.033
Yield – in % (v/w) of fresh weight; LM – Listeria monocytogenes; SA – Staphylococcus aureus;
EC – Escherichia coli; PA – Pseudomonas aeruginosa; SE – Salmonella enteritidis.
Zhang et al. (23) studied the antibacterial properties of 14 EOs (clove, oregano, rose-
mary, pepper, nutmeg, liquorice, turmeric, aniseed, cassia bark, fennel, prickly ash, round
cardamom, dahurian angelica root and angelica) against four common meat spoilage and
pathogenic bacteria (Listeria monocytogenes, Escherichia coli, Pseudomonas fluorescens
and Lactobacillus sake) and their results showed that individual extracts of clove, ro-
semary, cassia bark and liquorice contained strong antibacterial activity, but the mixture
of rosemary and liquorice extracts was the best inhibitor against all four types of micro-
bes. These herb extracts are widely used in the food industry and are generally regarded
as safe (GRAS). Hence, they may be considered as natural preservatives acceptable by
the food industry.
197
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
There has been several more general studies about antimicrobial activity of EOs. One
of them is the work of Gutierrez et al. (24), who evaluated EOs of lemon balm, marjoram,
oregano and thyme, and applied them on food model media based on lettuce, milk and
meat. Minimum inhibitory concentrations were determined against Enterobacter spp.,
Listeria spp., Lactobacillus spp. and Pseudomonas spp. According to the results, shown
in Table 2, the average efficacy of EOs against Listeria spp. was in the following order:
oregano>thyme> lemon balm, while the efficacy order of EOs against the spoilage bac-
teria was: oregano>thyme> marjoram.
Table 2. MIC of EOs used in this study against the selected bacteria in TSB (A), lettuce
leaf model media (B) or beef extract (C) (24)
Lemon
Microorganism Oregano Thyme Marjoram
balm
(A)
Enterobacter cloacae 400 600 6000 NDa
Listeria innocua NCTC11288 200 200 ND 2500
Listeria monocytogenes IL323 200 200 ND 2500
Listeria monocytogenes NCTC1194 200 200 ND 2500
Pseudomonas fluorescens 2000 2000 50 000 ND
Pseudomonas putida 2000 2000 50 000 ND
(B)
Enterobacter cloacae 250 250 2000 ND
Listeria monocytogenes IL323 20 30 2000 ND
Pseudomonas fluorescens 250 250 2000 ND
(C)
Listeria monocytogenes NCTC1194 60 125 ND 500
Pseudomonas fluorescens 1500 2500 12500 ND
Pseudomonas putida 1500 2500 12500 ND
a
ND, not determined.
As has been shown by some other researchers, the use of antimicrobials can reduce or
eliminate specific microorganisms but it may also produce favourable conditions for
other microorganisms (25). It is recognized that this situation is less likely to develop
towards substances that have more than one mode of action (26), so it is suggested that
the antimicrobial activity of EOs is attributed to more than one mechanism (27). Thus,
combining EOs could lead to useful efficacy against both spoilage and pathogenic target
organisms. Gutierrez et al. (24) tested the synergy of EO combinations and expresed it as
fractional inhibitory concentration (FIC) index, in the lettuce leaf model media. The FIC
indices were calculated as FICA+FICB, where FICA=MICA,combination/MICA,alone and
FICB=MICB,combination/MICB,alone. The results were interpreted as synergy (FIC< 0.5),
addition (0.5<FIC< 1), indifference (1 <FIC<4) or antagonism (FIC> 4). See Table 3.
198
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
Table 3. FIC values of EOs combinations in lettuce leaf model media (24)
Listeria
Enterobacter Pseudomonas Listeria innocua
EO combination monocytogenes
cloacae fluorescens NCTC11288
IL323
Oregano+marjoram 1.75 (I) 2.00 (I) NDa ND
Oregano+lemon balm ND ND 1.50 (I) 1.25 (I)
Oregano+thyme 0.75 (A) 0.88 (A) 1.00 (A) 1.18 (I)
Thyme+marjoram 1.00(A) 1.38 (I) ND ND
Thyme+lemon balm ND ND 0.75 (A) 1.25 (I)
Results are interpreted as synergy (S, FIC< 0.5), addition (A, 0.5 <FIC<1), indifference (I, 1<FIC< 4) or
antagonism (AN, FIC> 4).
a
ND, not detected.
Results from Table 3 show that - with reference to the FIC scale - no synergistic
effect (<0.5) was found, but addition occurred with a number of combinations. More
incidence of additive effects was found with EO combinations against Listeria strains.
Combinations of oregano with thyme or lemon balm were more effective against Listeria
monocytogenes. The combination of thyme with lemon balm had greater efficacy against
Listeria innocua. Only one combination (oregano with thyme) had additive effects
against both spoilage microorganisms. No antagonism was observed for any of the com-
binations evaluated. These results can be observed as an addition to some earlier studies
of the same author, Gutierrez et al. (28), considering the synergistic effect of marjoram
and thyme with some other EOs (Table 4).
Table 4. FIC indices of marjoram and thyme EOs combinations against

L. monocytogenes IL323 (28)
Combinations Marjoram Thyme

Marjoram or thyme +
Basil 0.75 (A) 0.94 (A)
Lemon balm 1.25 (I) 1.25 (I)
Marjoram - 1.55 (I)
Oregano 1.18 (I) 1.18 (I)
Rosemary 1.03 (A) 1.06 (A)
Sage 1.00 (A) 1.00 (A)
Thyme 1.55 (I) -
.
For successful applications of EOs in different food systems, potential interactions
between EOs and food components have to be determined. There is a number of exam-
ples where some studies have shown that plant extracts are useful for reduction of patho-
gens in some food product, while others reported very low antimicrobial activity or no
effect when the same EOs were applied to other product. Thus, the application of plant
EOs requires the evaluation of efficacy within food products or in model systems that
closely mimic food composition, because the efficacy of many antimicrobials may be
reduced by certain food components (29). This was also investigated by Gutierrez et al.
199
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
(28). In this complex task, they studied the effect of food ingredients and pH on the
antimicrobial efficacy of EOs using a number of model media and L.monocytogenes
IL323 as indicator strain. The EOs used in the study were oregano (30 ppm) and thyme
(60 ppm), and they were assessed independently. Results, expressed as lag phase duration
and maximum specific growth rate of L. monocytogenes, are shown in Table 5.
Table 5. Lag phase duration (λ) and maximum specific growth rate (μmax) of
L. monocytogenes IL323 grown in model media containing oregano (30 ppm)
or thyme (60 ppm) (28)
Model media Oregano Thyme Controla

Λ [h] µmax[h-1] Λ [h] µmax[h-1] Λ [h] µmax[h-1]
Beef extract
1.5 7.39 0.034 9.15 0.079 8.01 0.054
3.0 7.81 0.076 11.79 0.146 6.14 0.074
6.0 10.75 0.099 10.94 0.195 6.48 0.117
12.0 10.79 0.195 9.75 0.200 6.33 0.215
Starch media
0.0 15.01 0.185 12.06 0.250 7.80 0.238
1.0 13.96 0.208 11.69 0.271 7.04 0.268
5.0 9.71 0.142 8.87 0.164 7.58 0.147
10.0 8.84 0.098 8.43 0.104 8.04 0.097
Sunflower oil
media
0.0 15.23 0.240 15.05 0.231 7.75 0.279
1.0 14.21 0.226 12.54 0.209 7.31 0.223
5.0 10.50 0.208 9.22 0.235 7.24 0.200
10.0 9.17 0.174 9.47 0.220 7.33 0.170
pH
TSB pH4 0.00 0.000 0.00 0.000 0.00 0.000
TSB pH5 12.43 0.004 10.13 0.016 9.55 0.017
TSB pH6 7.70 0.141 9.48 0.175 6.75 0.173
TSB pH7 9.50 0.284 10.80 0.328 6.88 0.261
a
Listeria monocytogenes grown in model media without any EO was used as the control.
Model media comprised the following: (a) water soluble starch from potato in TSB; (b)
beef extract in deionized water; and (c) sunflower oil in TSB. The pH of each model
medium was adjusted to 7.2. To determine the effect of pH on EO efficacy TSB was
adjusted to pH 4, 5, 6 or 7 with 1 M HCl solution
The lag phase of Listeria monocytogenes grown in beef extract containing oregano
was longer than the control at protein concentrations of 6% and 12%. The efficacy of
oregano and thyme was greater at higher concentrations of protein, probably because
peptones from beef extract may have hydrophobic properties which facilitate dissolution
of EOs in this medium.
In addition, the growth rate of L. monocytogenes decreased at higher starch concen-
trations. The EO efficacy was reduced at high concentrations of starch, in contrast to the
general observation that carbohydrates in foods do not protect bacteria from the action of
EOs as much as fat and protein do (30).
200
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
The growth of L. monocytogenes was monitored in model media containing four

different sunflower oil concentrations (0, 1, 5 and 10%), in order to determine effect of
oil on the antimicrobial activity. High concentrations of sunflower oil had a negative in-
fluence on the antimicrobial activity of oregano and thyme EOs.
Effect of pH was evaluated using TSB at pH 4, 5, 6 and 7. L. monocytogenes did not
grow at pH 4. The lag phase of Listeria grown in pH model media with EO was longer
than in EO free controls, especially at pH 5, however, the lag phase was also greatest at
pH 5 in control media. The growth rate of L. monocytogenes increased at higher pH
values, regardless of the presence or absence of EOs. Some previous studies showed that
the inhibitory effect of plant extracts was greater at acidic pH values (31). The suscep-
tibility of bacteria to EOs appears to increase at lower pH since the hydrophobicity of
EOs increases at low pH, enabling consequently easier dissolution in the lipids of the cell
membrane of target bacteria (32).
Antifungal activity of EOs
All mentioned investigations have demonstrated anibacterial properties of EOs

against pathogenic and spoilage bacteria in food. However, the presence and growth of
fungi in food may also cause spoilage and result in a reduction in quality and quantity
(33-39). As reported by a number of authors (38-45), some Aspergillus species are
responsible for many cases of food and feed contamination, and Aspergillus flavus and
Aspergillus parasiticus are able to produce aflatoxins in food and feedstuffs, which are
known to be potent hepatocarcinogens in animals and humans (46). Aspergillus flavus
Link and Aspergillus parasiticus Speare are major storage fungi found regularly in im-
portant cereal grains cultivated and stored throughout the world (47), which produce afla-
toxins B1, B2, G1 and G2. The biosynthesis of aflatoxins can be inhibited by extracts and
EOs from certain plants toxic to fungi and can control the fungal growth and mycotoxin
production.(48).
Omidbeygi et al. (49) evaluated the antifungal activity of three EOs (thyme, summer
savory and clove) in culture medium and as a real system in tomato paste (in vitro and in
vivo). Results clearly showed that in vitro each EO had notable antifungal activity.
Thyme EO has the highest antifungal activity, followed by summer savory and clove
EOs. Complete inhibition of growth of Aspergillus flavus was observed at 350 and 500
ppm of thyme and summer savory EOs, respectively, while 500ppm of clove oil had
inhibition of 87.5%. In vivo studies was performed in tomato paste, and while in vitro
experiments showed that 500 and 350ppm of thyme oil could inhibit the growth of A.
flavus completely (100%), inhibition in tomato paste were 87% and 42%, respectively.
Also, in other treatments inhibition in tomato paste was lower than in culture medium. It
has probably been related to the more complex matrix of tomato paste than culture
medium. However, the need to use plant EOs at higher concentrations in food than in
laboratory media is believed to be due to the more complex growth environment in food,
which provides microbial cells with greater protection from antimicrobial agents.
Atanda et al. (50), evaluated essential oils of sweet basil (Ocimum basilicum), cassia
(Cinnamomum cassia), coriander (Coriandrum sativum) and bay leaf (Laurus nobilis) for
their potential in the growth control of aflatoxigenic fungus Aspergillus parasiticus CFR
201
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
223 and aflatoxin production. Results, expressed as mycelia dry weight and aflatoxin
production in basal medium, are summarised in Table 6.
Table 6. Effect of different concentrations of spice essential oil on the mycelia

dry weight and aflatoxin production in basal medium (52)
Essential Concentration Mycelia dry Aflatoxin [µg/ml]

oil [%v/v] weight [g/ml] B1 G1
0 0.50 0.050 0.046
1 0.38 0.029 0.027
2 0.40 0.012 0.011
Cassia
3 0.48 0.004 0.003
4 0.52 0.002 0.003
5 0.55 0.001 0.001
1 0.20 0.045 0.042
2 0.20 0.040 0.038
Sweet basil 3 0.20 0.038 0.035
4 0.20 0.024 0.020
5 0.00 0.000 0.000
1 0.50 0.050 0.046
2 0.50 0.050 0.046
Coriander 3 0.50 0.050 0.046
4 0.50 0.050 0.046
5 0.50 0.050 0.046
1 0.55 0.050 0.046
2 0.55 0.045 0.042
Bay leaf 3 0.65 0.040 0.038
4 0.70 0.030 0.035
5 0.75 0.023 0.020
The EOs of sweet basil inhibited completely mycelia growth and prevented aflatoxin
formation at the concentration of 5% (v/v). At the same concentration, oils of cassia and
bay leaf reduced the aflatoxin concentration (B1+G1) of the fungus to 0.002 µg/ml
(97.92%) and 0.043 µg/ml (55.21%), respectively. Cassia and bay leaf EOs stimulated
the mycelia growth of the fungus, while coriander oil did not have any effect on either the
mycelia growth or aflatoxin content of the fungus.
Tzortzakis et al. (51), tested antifungal activity of lemongrass (Cympopogon citratus
L.) oil (range concentration between 25 and 500 ppm) against Colletotrichum coccodes,
Botrytis cinerea, Cladosporium herbarum, Rhizopus stolonifer and Aspergillus niger in
vitro. Oil-enrichment resulted in significant reduction of the subsequent colony develop-
ment for the examined microorganisms. Fungal spore production was inhibited up to 70%
at 25 ppm of lemongrass oil concentration when compared with control samples (equi-
valent plates stored in the ambient air). At the highest oil concentration (500 ppm), fungal
sporulation was completely retarded (Table 7). Lemongrass oil reduced spore germina-
202
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
tion and germ tube length of C. coccodes, B. cinerea, C. herbarum and R. stolonifer with
the effects dependent on oil concentration. This study indicated that EOs may possess
antifungal activity and can be exploited as an ideal treatment for future plant disease
management programs eliminating fungal spread.
Viuda-Martos et al. (52), investigated the effect of the essential oils of lemon (Citrus
lemon L.), mandarin (Citrus reticulata L.), grapefruit (Citrus paradisi L.) and orange
(Citrus sinensis L.) on the growth of moulds commonly associated with food spoilage:
Aspergillus niger, Aspergillus flavus, Penicillium chrysogenum and Penicillium verruco-
sum, using the agar dilution method. The essential oils of lemon, orange, mandarin and
grapefruit at the concentrations assayed (0.27-0.94%) all showed the capacity to reduce
or inhibit the growth of the named moulds. The growth of A. niger was completely inhi-
bited when a concentration of 0.94% of any of the EOs was used. Orange EO produced
the greatest reduction in mycelium growth with this fungus, and it is followed by lemon
EO, then the mandarin, while grapefruit EO caused the lowest percentage of mycelial
reduction in A. niger. In the case of A. flavus, efficacy of EOs was in the following order:
mandarin>lemon>grapefruit>orange, while total inhibition of growth was obtained with
all the EOs at the highest concentrations of 0,94%. The same concentration for growth
inhibition was found for P.verrucosum and P.chrysogenum, and the order of oil efficacy
against these moulds was grapefruit>lemon>orange>mandarin.
Table 7. Impacts of lemongrass (Cympopogon citratus L.) essential oil-enrichment on

spore production (number of spores×106) of tested fungi grown on PDA (51)
EO
Colleotrichum Botrytis Cladosporium Rhizopus Aspergillus
concentration
coccodes cinerea herbarum stolonifer niger
[ppm]
0 12.10 14.28 79.35 14.13 169.50
25 5.13 4.30 47.70 9.23 99.35
50 2.55 5.23 45.38 8.63 62.93
100 1.53 5.75 39.88 5.00 63.13
500 0.00 0.00 0.00 0.00 0.00
In the study of Vilela et al. (53), the Eucalyptus globulus Labill. EO and its major
compound 1,8-cineole were evaluated for antifungal activity against A. flavus and A.
parasiticus, as well as on aflatoxin production. Results of this study showed that
Eucalyptus globulus oil had clear dose-dependent antifungal activity on both fungal
species. Complete fungal growth inhibition was verified at the concentration of 50 ml oil
per millilitre of medium in the contact assay. The antifungal activity offered by 1.8
cineole only showed effects at the highest concentration tested, which indicated the major
oil constituent is not the only component responsible for limiting fungal growth. Inhi-
bition of aflatoxin B1 production required a higher oil dose than was required for inhibi-
tion of fungal growth.
Škrinjar et al. (54) examined an inhibitory effect of various concentrations (0.0, 0.5,
1.0, 1.5 and 2.0%) of mint (Mentha piperita L.) and caraway (Carvum carvi L.) on the
growth of some toxigenic Aspergillus species and aflatoxin B1 production. Mint showed
203
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
stronger inhibitory effect than caraway. Concentrations of 1.0. 1.5 and 2.0% reduced the
growth of all tested Aspergillus species, which was poor and hardly visible, while
concentration of 2% of caraway was needed to achieve the same effect. The applied
concentrations of mint and caraway inhibited completely the production of AB1 by
Aspergillus flavus.
Yeasts are widely distributed in nature and are able to spoil many foods such as
wines, cheese, vinegar, beverages, juices, fruits, salads, sugar and meat, causing changes
in odour, colour, taste and texture (55). Many of data indicate the EOs inhibitory effects
of various spices and herbs on these microorganisms.
The study of Souza et al. (56), aimed at verifying the effectiveness of oreganum
(Origanum vulgare L.) EO to inhibit the growth/survival of various food spoiling yeasts.
Results, expressed in millimeters of yeast growth inhibition halos, are shown in Table 8.
Table 8. Origanum vulgare L. essential oil MIC on food spoiling yeasts

determined by solid medium diffusiona (56)
Origanum vulgare L. essential oil [µl/ml]

Yeasts
160 80 40 20 10 5 2.5 1.25
Candida albicans 38 28 20 12 10 0 0 0
Candida krusei 32 25 15 12 0 0 0 0
Candida tropicalis 35 27 21 14 11 0 0 0
Pichia minuscula 39 36 31 21 16 11 0 0
Pichia ohmeri 33 28 16 13 10 0 0 0
Rodotorula rubra 38 34 30 28 14 0 0 0
Saccharomyces cerevisae 26 22 14 11 0 0 0 0
a
Results expressed in millimeters of yeast growth inhibition halos.
The results showed that the EO had a substantial inhibitory effeect on all assayed
yeast strains, noted by large growth inhibition halos. Most assayed strains showed an
MIC of 10 µL/mL. The highest inhibitory activity was observed against P. minuscula (the
lowest MIC of 5 µL/mL) and the largest growth inhibition halos. On the other hand, S.
cerevisae and C. krusei were the least sensitive yeasts with an MIC of 20 µL/mL;
however S. cerevisae showed the smallest growth inhibition halo diameters when
compared to all other strains. This high antimicrobial activity of O. vulgare EO supports
the results found by other researchers (57,58).
CONCLUSIONS
Food contamination is enormous public health problem, but it could be controlled by

the use of natural preservatives such as essential oils obtained from spices. The fact that
many EOs possess antimicrobial activity has been proved by plenty of investigations in
the recent years. The type and optimal concentration of EO depend on the product used
and against which species of bacteria or fungi it is to be used. But if EOs are expected to
be widely applied as antibacterials and antifungals, the organoleptical impact should be
considered as the use of naturally derived preservatives can alter the taste of food or
exceed acceptable flavour thresholds. The problem may occur if high concentrations
204
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
required to achieve useful EO antimicrobial activity, result in unacceptable levels of

flavours and odours. Therefore, research in this area should be focused on the opti-
mization of EO combinations and applications to obtain effective antimicrobial activity at
sufficiently low concentrations so as not to adversely influence the organoleptic
acceptability of the foods.
REFERENCES
1. Namiki, M.: Antioxidant/antimutagens in food. Food Sci. Nutr. 29 (1990) 273–300.
2. Paster, N. and L. B. Bullerman: Mould spoilage and mycotoxin formation in grains
as controlled by physical means. Int. J. Food Microbiol. 7 (1988) 257−265.
3. Smid, E.J. and L.G.M. Gorris: Natural antimicrobials for food preservation, in
Handbook of Food Preservation. Ed. M.S. Rahman, Marcel Dekker, New York
(1999) pp. 285–308.
4. Brul, S. and P. Coote: Preservative agents in foods: Mode of action and microbial
resistance mechanisms. Int. J. Food Microbiol. 50 (1999), 1-17.
5. Beuchat, L.R.: Antimicrobial properties of spices and their essential oils, in Natural
Antimicrobial Systems and Food Preservation. Eds. Y. M. Dillon and R. G. Board,
CAB International, Oxon, (1994) pp. 167–179.
6. Nakatani, N.: Antioxidative and antimicrobial constituents of herbs and spices, in
Spices, Herbs and Edible Fungi. Ed. G. Charalambous, Elsevier Science, New York,
(1994) pp. 251–271.
7. Cutler, H.G.: Natural product flavor compounds as potential antimicrobials,
insecticides, and medicinals. Agro-Food Ind. Hi-Tech 6, (1995) 19–23.
8. Oluwafemi, F.: Correlation between dietary aflatoxins and human male infertility,
Ph.D. Thesis, University of Benin, 2000.
9. Shan, B., Y. Cai, J. D. Brooks and H. Corke: The in vitro antibacterial activity of
dietary spice and medicinal herb extract. Int. J. Food Microbiol. 117 (2007) 112-119.
10. Ceylan, E. and D.Y.C. Fung: Antimicrobial activity of spices. J. Rapid Meth. Aut.
Mic. 12 (2004) 1–55.
11. Smith-Palmer, A., J. Stewart and L. Fyfe: Antimicrobial properties of plant essential
oils and essences against five important food-borne pathogens. Lett. Appl. Microbiol.
26 (1998) 118–122.
12. Hara-Kudo, Y., A. Kobayashi, Y. Sugita-Konishi and K. Kondo: Antibacterial
activity of plants used in cooking for aroma and taste. J. Food Protect. 67 (2004)
2820–2824.
13. Ebana, R. U. B., B.E. Madunagu, E.D. Ekpe and I. N. Otung: Microbiological
exploitation of cardiac glycosides and alkaloids from Garcinia kola, Borreria
ocymoides, Kola nitida and Citrus aurantiofolia. J. App. Bacteriol. 71 (1991) 398–
401.
14. López-Malo, A., J. Barreto-Valdivieso, E. Palou and F. S. Martin: Aspergillus flavus
growth response to cinnamon extract and sodium benzoate mixtures. Food Control 18
(2006) 1358−1362.
15. Elgayyar, M., F.A. Draughon, D.A. Golden and J.R. Mount: Antimicrobial activity of
essential oils from plants against selected pathogenic and saprophytic microorga-
nisms. J. Food Protect. 64, 7 (2001) 1019–1024.
205
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
16. Hammer, K. A., C.F. Carson and T. V. Riley: Antimicrobial activity of essential oils
and other plant extracts. J. App. Microbiol. 86 (1999) 985–990.
17. Dorman, H. J. D. and S. G. Deans: Antimicrobial agents from plants: antibacterial
activity of plant volatile oils. J. App. Microbiol. 88 (2000) 308–316.
18. Oussalah, M., S. Caillet, L. Saucier and M. Lacroix: Inhibitory effects of selected
plant essential oils on the growth of four pathogenic bacteria: E. coli O157:H7,
Salmonella Typhimurium, Staphylococcus aureus and Listeria monocytogenes. Food
Control 18, 5 (2006) 414–420.
19. Conner, D. E.: Naturally occurring compounds, in Antimicrobials in foods. Eds. P.
Davidson and A. L. Branen, Marcel Dekker, Inc. New York (1993) pp. 441–468.
20. Didry, N., L. Dubreuil and M. Pinkas: Activité antimicrobienne du thymol, du
carvacrol et de l’aldéhyde cinnamique seuls ou associés (Antibacterial activity of
thymol, carvacrol, and cinnalmaldehyde singly or in combinations). Pharmazie 48
(1993) 301–308.
21. Nedorostova, L., P. Kloucek, L. Kokoska, M. Stolcova and J. Pulkrabek:
Antimicrobial properties of selected essential oils in vapour phase against foodborne
bacteria. Food Control 20 (2009) 157-160.
22. Lopez, P., C. Sanchez, R. Batlle and C. Nerin: Solid- and vapor-phase antimicrobial
activities of six essential oils: Susceptibility of selected foodborne bacterial and
fungal strains. J. Agr. Food Chem. 53 (2005) 6939–6946.
23. Zhang, H., B. Kong, Y.L. Xiong and X. Sun: Antimicrobial activities of spice extracts
against pathogenic and spoilage bacteria in modified atmosphere packaged fresh pork
and vacuum packaged ham slices stored at 4°C. Meat Sci. 81 (2009) 686-692.
24. Gutierrez, J., C. Barry-Ryan and P. Bourke: Antimicrobial activity of plant essential
oils using food model media: Efficacy, synergistic potential and interactions with
food components. Food Microbiol. 26 (2009) 142-150.
25. Davidson, P.M. and A.L. Branen: Food antimicrobials – an introduction, in
Antimicrobial in Food. Eds. P.M. Davidson, J.N. Sofos, and A.L. Branen, A.L., CRC
Press, Taylor and Francis Group ed. (2005) pp. 1–9.
26. Ippolito, A. and F. Nigro: Natural antimicrobials in postharvest storage of fresh fruits
and vegetables, in Natural Antimicrobials for the Minimal Processing of Foods. Ed.
S. Roller, Woodhead Publishing LTD (2003) pp. 201–223.
27. Moreira, M.R., A.G. Ponce, C.E. Del Valle and S.I. Roura: Inhibitory parameters of
essential oils to reduce a foodborne pathogen. LWT–Food Sci. Tech. 38, 5 (2005)
565–570.
28. Gutierrez, J., C. Barry-Ryan and P. Bourke: The antimicrobial efficacy of plant
essential oil combinations and interactions with food ingredients. Int. J. Food
Microbiol. 124 (2008) 91-97.
29. Glass, K.A. and E.A. Johnson: Antagonistic effect of fat on the antibotulinal activity
of food preservatives and fatty acids. Food Microbiol. 21 (2004) 675–682.
30. Shelef, L.A., E.K. Jyothi and M.A. Bulgarelli: Growth of enteropathogenic and
spoilage bacteria in sage-containing broth and foods. J. Food Sci. 49 (1984), 737-740.
31. Del Campo, J., M.J. Amito and C. Nguyen-The: Antimicrobial effect of rosemary
extracts. J. Food Protect. 63, 10 (2000) 1359–1368.
206
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
32. Juven, B.J., J. Kanner, F. Schved and H. Weisslowicz: Factors that interact with the
antibacterial action of thyme essential oil and its active constituents. J. App.
Bacteriol. 76 (1994) 626–631.
33. Škrinjar M., I. Kolarović, G. Dimić and O. Jovanović: Pojava plesni i nekih
mikotoksina u sirovinama korišćenim u konditorskoj industriji. Žito hleb 27, 1 (2000)
9-14.
34. Škrinjar, M., I. Vujković, O. Jovanović, I. Kolarević and G. Dimić: Mogućnost
gljivičnog kvara industrijski punjenog kolača sa većim sadržajem vlage. Žito hleb 27,
3 (2000) 93-97.
35. Škrinjar, M., V.Došen-Bogićević and I. Vujković: Mycopopulations, aflatoxins and
ochratoxin in some semi-dry sausage, 47th International Congress of Meat Science
and Technology, Krakow, 21-31 August 2001, Proceedings 7-P27, pp. 272-273.
36. Škrinjar, M: Are toxigenic moulds and mycotoxins frequent contaminants of wheat
flour?, 3rd International Congress «Flour-Bread», Opatija, 26-29 October 2005,
Book of Abstracts, p. 10.
37. Škrinjar, M: Uljarice – mogući nosioci mikotoksina. Uljarstvo 36, 1-2 (2005) 47-50.
38. Škrinjar, M. and D. Tešanović: Hrana u ugostiteljstvu i njeno čuvanje, Prirodno-
matematički fakultet, Univerzitet u Novom Sadu, Novi Sad (2007).
39. Šarić, Lj. and M. Škrinjar: Share of aflatoxigenic moulds from genera Asspergillus
and Penicillium in mycopopulations isolated from spices for meat processing
industry. Proceedings for Natural Sciences 114 (2008) 115-122.
40. Abarc, M. L., M. R. Bragulat, G. Castella, and F. J. Cabanes: Myco-flora and
aflatoxin producing strains in animal mixed feeds. J. Food Protect. 57 (1994) 256–
258.
41. Škrinjar, M., M. Bandu, E. Dimić, K. Bjelobaba and R. Romanić: Infekcija semena
suncokreta žetve 2006. aflatoksigenim gljivama. Uljarstvo 38, 1-2 (2007) 41-48.
42. Šarić, Lj. and M. Škrinjar: Proizvodnja aflatoksina B1 od strane plesni izolovanih iz
začina koji se koriste u industriji mesa, International 54th Meat Industry Conference,
Vrnjačka Banja, 18-20 June 2007, Book of Abstracts, p. 129.
43. Škrinjar, M: Fungal and mycotoxin contamination of some food industry byproducts,
25th Food Technology Days, Ljubljana, 17-18 April 2008, pp. 73-84.
44. Škrinjar, M., M. Ač, M. Krajinović and A. Popović-Vranješ: Zdravstveni i ekonomski
značaj prisustva toksigenih plesni i mikotoksina u stočnoj hrani. Savremena
poljoprivreda 57, 3-4 (2008) 26-34.
45. Guo, B. Z., J.S. Russin, R.L. Brown, T.E. Celveland and N. W. Widstrom: Resistance
to aflatoxin contamination in corn as influenced by relative humidity and kernel
germination. J. Food Protect. 59 (1996) 276–281.
46. Dvorackova, I. : Aflatoxins and human health. Boca Raton, FL: CRC Press (1990).
47. Frisvad, J.C.: Mycotoxins and Mycotoxigenic Fungi, in Storage Grain Ecosystems.
Eds. D.S. Jayas, N.D.G. White, W.E. Muir, Marcel Dekker, New York, (1995) pp.
251–258.
48. Pinto, M.M., E. Gonçales, M.H. Rossi, J.D. Felıćio, C.S. Medina, M.J.B. Fernandes
and L.C. Simoni: Activity of the aqueous extract from Polymnia sonchifolia leaves on
growth and production of aflatoxin B1 by Aspergillus flavus. Braz. J. Microbiol. 32
(2001) 127–129.
207
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
49. Omidbeygi, M., M. Barzegar, Z. Hamidi and H. Naghdibadi: Antifungal activity of

thyme, summer savory and clove essential oils against Aspergillus flavus in liquid
medium and tomato paste. Food Control 18 (2007) 1518–1523.
50. Atanda, O.O., I. Akpan and F. Oluwafemi: The potential of some spice essential oils
in the control of A. parasiticus CFR 223 and aflatoxin production. Food Control 18
(2007) 601–607.
51. Tzortzakis, N. G. and C. D. Economakis: Antifungal activity of lemongrass
(Cympopogon citratus L.) essential oil against key postharvest pathogens. Inn. Food
Sci. Emerg. Technol. 8 (2007) 253–258.
52. Viuda-Martos, M., Y. Ruiz-Navajas, J. Fernández-López and J. Pérez-Álvarez:
Antifungal activity of lemon (Citrus lemon L.), mandarin (Citrus reticulata L.),
grapefruit (Citrus paradisi L.) and orange (Citrus sinensis L.) essential oils. Food
Control 19 (2008) 1130–1138.
53. Vilela, G. R., G.S. Almeida, M. Aparecida Bismara Regitano D’Arce, M.H. Duarte
Moraes, J.O. Brito, M.F. das G.F. da Silva, S.C. Silva, S.M. de Stefano Piedade, M.A.
Calori-Domingues and E. Micotti da Gloria: Activity of essential oil and its major
compound, 1,8 cineole, from Eucalyptus globulus Labill., against the storage fungi
Aspergillus flavus Link and Aspergillus parasiticus Speare. J. Stored Prod. Res. 45
(2009) 108-111.
54. Škrinjar, M., A. Mandić, A. Mišan, M. Sakač, Lj. Šarić and M. Zec: Effect of mint
(Mentha piperita L.) and caraway (Carvum carvi L.) on the growth of some toxigenic
Aspergillus species and aflatoxin B1 production. Proc. Nat. Sci, Matica Srpska Novi
Sad 116 (2009) 131-139.
55. Ray, B.: Fundamental food microbiology. Boca Raton: CRC Press (1996).
56. Souza, E.L., T.L.M. Stamford, E.O. Lima and V.N. Trajano: Effectiveness of
Origanum vulgare L. essential oil to inhibit the growth of food spoiling yeasts. Food
Control 18 (2007) 409–413.
57. Skandamis, P., E. Tsigarida and G. J. Nychas: The effect of oregano essential oil on
survival/death of Salmonella typhimurium in meat stored at 5 °C under aerobic,
VP/MAP conditions. Food Microbiol. 19 (2002) 97–103.
58. Chun, S. S., A.V. Vattern, Y.T. Lin and K.. Shetty: Phenolic antioxidants from clonal
oregano (Origanum vulgare) with antimicrobial activity against Helicobacter pylori.
Process Biochem. 40 (2004) 809–816.
АНТИМИКРОБНО ДЕЛОВАЊЕ ЕСЕНЦИЈАЛНИХ УЉА ЗАЧИНА И

ЛЕКОВИТОГ БИЉА
Марија М. Шкрињар и Невена Т. Немет
Зачини и лековито биље користе се као додаци храни још од давнина, у својству
ароме, као побољшивачи укуса а такође и као природни конзерванси. Велики број
зачина и лековитог биља показује антимикробно и антифунгално деловање према
одређеним микроорганизмима. Овај рад даје литературни преглед нових истражи-
вања која се тичу наведене активности есенцијалних уља широко распрострањених
зачина и лековитог биља, као што су бели лук, слачица, цимет, кумин, каранфилић,
208
APTEFF, 40, 1-220 (2009) UDC: 664.5:582.28+577.18
DOI: 10.2298/APT0940195S BIBLID: 1450-7188 (2009) 40, 195-209
Review
ловор, мајчина душица, босиљак, оригано, бибер, ђумбир, жалфија, рузмарин и др.,
против најчешћих бактерија и гљива које контаминирају храну (Listeria spp., Sta-
phylococcus spp., Salmonella spp., Escherichia spp., Pseudomonas spp., Aspergillus
spp., Cladosporium spp. и многи други). Антимикробна активност зависи од врсте
зачина, врсте намирница и микроорганизама на које се примењује, као и од хемиј-
ског састава и концентрације екстраката или есенцијалних уља зачина. Сумирањем
резултата добијених од стране различитих аутора, може се извести закључак о ре-
лативној антимикробној и антифунгалној ефикасности одређених зачина и лекови-
тог биља, према коме цимет, каранфилић и слачица имају веома јак антимикробни
потенцијал, кумин, оригано, жалфија, мајчина душица и рузмарин имају средњи,
док бибер и ђумбир имају слаб инхибиторни ефекат.

209
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
THE INFLUENCE OF CARBOXYMETHYLCELLULOSE, XANTHAN AND

GUAR-GUM ADDITION IN BREAD DOUGH BEFORE FREEZING ON
METABOLISM AND VIABILITY OF Saccharomyces cerevisiae
Dušanka J. Pejin, Olgica S. Grujić, Jelena D. Pejin, Irena S. Došenović and

Sunčica D. Kocić-Tanackov
Doughs were prepared with different concentrations of carboxymethylcellulose, xan-

than, and guar-gum (0.1, 0.2 and 0.3% in doughs), freezed at -20°C and analyzed after 0,
7, 15 and 30 days. Pure Saccharomyces cerevisiae culture was isolated from dough and
was cultivated under optimal conditions during 24 hrs to determine the following para-
meters: specific growth rate, fermentative activities and cytochromes contents in intact
cells with the aim of determining the respiration intensity. During freezing of dough for
30 days, the percentage of living cells from dough surface was 53.11% and from the
middle 54.95%. Carboxymethylcellulose in concentration of 0.3 and 0.5% increased
number of survived cells on the surface to 70.64, and 70.28% and in the middle to 74.79,
and 76.54%, respectively. Guar–gum increased number of survived cells only in concen-
tration of 0.1% on the surface to 70.17% and in the middle of the dough to 75.26%. The
mean specific growth rate decreased by approximately 10% during 30 days of storage at
-20°C. Content of cytochromes in intact cells decreased in all samples during freezing.
KEY WORDS: Saccharomyces cerevisiae, frozen dough, metabolism, viability
INTRODUCTION
The process of manufacturing bread from frozen dough is widely utilized in the
baking industry (1). In bread making baker’s yeast Saccharomyces cerevisiae encounters
many stresses, such as freezing, heat shock, osmotic stress and air-drying stress. Such
freezing can thus cause cell wall and membrane damage, protein and DNA denaturation
and decreased cell survival. After the dough thaws, yeast cells show dramatically decrea-
sed fermentation activity (2,3).
Survival of frozen yeast cells depends on several genetic, physiological, and en-
vironmental factors (4). A number of factors can affect yeast cells damage, depending on
whether ice is formed intracellularly (high freezing rates) or extracelularly (lower free-
Dr. Dušanka J. Pejin, Prof., Dr. Olgica S. Grujić, Prof., Dr. Jelena D. Pejin, Assist., Sunčica Kocić-Tanackov,
M.Sc., Assist., University of Novi Sad, Faculty of Technology, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia;
Dr. Irena S. Došenović, Institute for Food Technology, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia
211
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
zing rates) (5). Yeast cells are damaged during the freezing process, while the thawing
regime affects insignificantly the survival of yeast cells.
Several ways of decreasing the effect of freezing and frozen storage on yeast cells
survival and fermentative activity can be found, even on improving the characteristics
and quality of frozen dough and obtained bakery products: addition of hydrocolloids,
lower water content in prepared dough, higher amount of yeast compared to traditional
production and shorter dough fermentation before freezing, use of instant yeast, use of
cryotolerant and/or cryoresistent baker’s yeast strains, and use of modified yeast strains
(6). Hydrocolloids have been widely used in food products to modify texture, improve
moisture retention, control water mobility, and maintain overall product quality during
storage (7,8). The effects of hydrocolloids on the functional properties of dough and
bread quality depend on the nature origin and particle size of the hydrocolloid, and the
dosage of the hydrocolloid incorporated into dough formulations. Protein and poly-
saccharide functions are greatly affected by their interaction with each other and with
other components of food system (9). Different hydrocolloids like carboxymethylcellu-
lose, xanthan, and guar-gum have been successfully used in wheat bread production (10).
Hydrocolloids when used in small quantities (<1% (w/w) in flour) are axpected to
increase water retention and loaf volume and to desrease firmness and starch retro-
gradation (11). Ribotta et al. (7) showed that the addition of guar gum (0.5% (w/w) in
flour) improves volume and texture of bread obtained from non-frozen and frozen dough.
Sharadanant and Khan (12) investigated the influence of carboxymethylcellulose (CMC)
addition in three different concentrations (1, 2 and 3% (w/w) in flour) on bread quality.
The doughs were stored frozen for up to 16 weeks. Although the external and internal
characteristics of bread deterorated with storage time addition of CMC improved the
characteristics compared with control after each storage period.
The aim of this research was to investigate the possibility of carboxymethylcellulose,
xanthan, and guar-gum use to protect yeast cells during dough freezing.
EXPERIMENTAL
Average quality commercial T-500 flour was used for the production of dough, which
was frozen later. Quality characteristics were analyzed according to the Regulations on
methods of physical and chemical analyses for quality control of wheat, milling and
bakery products, pasta and fast frozen dough (13). Doughs were prepared according to
the following procedure: flour+water+fresh commercial baker’s yeast (2.5% calculated
on flour) were placed in the spin kneading machine with helical agitators, and mixed for
10 min at 85 rpm (control). Temperature of mixed dough was 20±1°C (14). Carboxy-
methylcellulose, xanthan, and guar-gum (Fluka AG, Buch, Switzerland) were added as a
component into the dough prepared according to the described procedure. Doughs were
prepared with different concentrations of hydrocolloids (0.1, 0.2 and 0.3% in doughs),
divided into portions, frozen at -20°C (freezing rate was 1°C/min), stored at -20°C and
analyzed after 0, 7, 15 and 30 days. The number of living Saccharomyces cerevisiae cells
was determined according to the method given in the Rulebook on methods performing
microbiological analyses and super analyses of food products (15). The samples were
thawed at 4°C for 12 h, and for additional 1.5 h at 20°C. The number of living Saccha-
212
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
romyces cerevisiae cells (viability) was determined by spreading samples on YPD agar
and counting colonies after 3-4 days of incubation at 30°C (16). Colonies of pure Saccha-
romyces cerevisiae culture were transferred into the liquid nutritive medium for yeast
(17) and cultivated under optimal conditions (aeration and temperature) during 24 h. Spe-
cific growth rate was determined according to Pejin (18). Fermentative activity was de-
termined using Einhorn method described by Reiff et al. (19). Cytochrome content in in-
tact cells (with the aim of determining the respiration intensity) was determined accor-
ding to Oura and Suomalainen (20).
During freezing of dough for 30 days, the percentage of living yeast cells from dough
surface was 53.11%, and in the middle 57.87%. These results are in agreement with those
of Ribotta et al. (21). Comparing these results, it can be presumed that the cells in the
middle of the dough were protected from low temperatures and because of that the
number of survived cells was higher. Carboxymethylcellulose in concentration of 0.3 and
0.5% increased the number of survived cells on the surface to 70.64, and 70.28% and in
the middle to 74.79, and 76.54%, respectively (Fig. 1).
On the surface In the middle

80
Percentage of living yeast cells
70
60
50
40
30
20
10
0
Control sample 0.1 0.3 0.5
% of carboxymethylcellulose
Fig. 1. Percentage of living yeast cells in dough compared to initial number before
freezing (on the surface and in the middle) with the addition of carboxymethylcellulose
(0.1, 0.3 and 0.5%) after 30 days of freezing. Values represent means calculated from
three determinations
Addition of xanthan to doughs did not have a great impact on percentage of living
yeast cells after 30 days of freezing (Fig. 2). Guar–gum increased survived cells number
only in concentration of 0.1%: on the surface to 70.17%, and in the middle of the dough
to 75.26% (Fig. 3). Hydrocolloids can modify the dough structure, bind the free water
and control water migration in the dough. Binding immobilization of water decreases the
ice crystal formation and also the damage of gluten and yeast cells (22).
In the further investigations, yeast cells were isolated from the middle of the dough
and propagated for the determination of specific growth rate and cytochrome content
213
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
(dough with 0.5% carboxymethylcellulose, dough with 0.5% xanthan, and dough with
0.1% for guar-gum).
On the surface In the middle

70
60
50
40
30
20
10
0
% of xanthan
freezing (on the surface and in the middle) with the addition of xanthan (0.1, 0.3 and
0.5%) after 30 days of freezing. Values represent means calculated from three
determinations
80 On the surface In the middle
70
60
50
40
30
20
10
0
% of guar-gum
freezing (on the surface and in the middle) with the addition of guar-gum (0.1, 0.3 and
0.5%) after 30 days of freezing. Values represent means calculated from three
determinations
There is today enough evidence to conclude that the exposure to low temperature
protects yeast cells against freeze injury through the cold-induced accumulation of tre-
halose, glycerol and heat-shock proteins (23).
214
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
Specific growth rates of Saccharomyces cerevisiae pure culture decreased constantly

with a longer freezing of dough. The mean specific growth rate decreased by approxi-
mately 10% during 30 days of storage at -20°C (Table 1).
Table 1. Specific growth rate μ (h-1) of pure Saccharomyces cerevisiae culture isolated
from frozen dough samples during 30 days of freezinga
Specific growth rate μ (h-1)

Days Dough sample CMCb Xanthan Guar-gum
Control sample
(0.5%) (0.5%) (0.1%)
on the surface 0.1425 0.1430 0.1430 0.1430
0
in the middle 0.1452 0.1370 0.1390 0.1425
on the surface 0.1510 0.1430 0.1390 0.1395
1
in the middle 0.1540 0.1360 0.1320 0.1360
on the surface 0.1420 0.1360 0.1380 0.1370
7
in the middle 0.1300 0.1353 0.1320 0.1360
on the surface 0.1350 0.1381 0.1370 0.1300
15
in the middle 0.1300 0.1362 0.1290 0.1260
on the surface 0.1340 0.1380 0.1300 0.1300
30
in the middle 0.1300 0.1380 0.1290 0.1290
a
Values represent means calculated from three determinations
b
CMC - Carboxymethylcellulose
Fermentative activities of pure Saccharomyces cerevisiae cultures isolated from

frozen doughs containing carboxymethylcellulose (0.5%), xanthan (0.5%), and guar-gum
(0.1%) are given in Table 2. Fermentative activities decreased during 30 days of freezing.
By addition of hydrocolloids to dough, fermentative activities decreased less compared to
control samples. The best protection of yeast cells was provided by carboxymethylcellu-
lose.
Table 2. Fermentative activity of pure Saccharomyces cerevisiae cultures isolated from

frozen dough samples during 30 days of freezinga
Fermentative activity (cm3 CO2/L g dry matter in 15 minutes)

Days Dough sample Control sample CMCb Xanthan Guar-gum
(0.5%) (0.5%) (0.1%)
on the surface 46.41 47.78 49.50 50.72
0
in the middle 41.62 46.76 47.80 49.27
on the surface 43.80 48.74 49.32 49.80
1
in the middle 42.26 41.98 46.20 48.30
on the surface 43.90 49.50 49.20 49.70
7
in the middle 41.90 49.10 48.20 47.30
on the surface 43.00 47.33 48.30 49.20
15
in the middle 39.80 37.23 40.18 42.76
on the surface 42.04 44.56 45.80 44.30
30
in the middle 34.48 39.80 37.80 35.20
a
Values represent means calculated from three determinations
b
CMC - Carboxymethylcellulose
215
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
Higher fermentative activities were determined in yeast samples with higher specific
growth rates (Tables 1 and 2), which is in accordance with results of Van Hoek et al.
(24).
Contents of aa3, b, and c cytochromes in pure Saccharomyces cerevisiae culture cells
isolated from the surface of doughs and cultivated under optimal conditions for 24 h are
presented in Figs. 4-7. Yeast cells were isolated from the dough middle. Individual and
total cytochromes contents decreased during dough freezing, especially cytochrome aa3.
Cytochromes contents in pure Saccharomyces cerevisiae culture cells isolated from
doughs with the addition of carboxymethylcellulose showed the smallest decrease during
freezing (from 25.85 to 20.12).
30
25
Cytochromes contents
20
a a3
b
15
c
Total
10
0
0 1 7 15 30
Days
Fig. 4. Cytochromes contents (moles 105/kg of yeast with 25% dry matter) in pure
Saccharomyces cerevisiae culture cells isolated from control dough. Values represent
means calculated from three determinations
30
25
20
a a3
b
15
c
Tota l
10
0
0 1 7 15 30
Days
Saccharomyces cerevisiae culture cells isolated from dough with the addition of
carboxymethylcellulose (0.5%). Values represent means calculated from three
determinations
216
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
25
20
a a3
15
b
c
10 Tota l
0
0 1 7 15 30
Days
Saccharomyces cerevisiae culture cells isolated from dough with the addition of xanthan
(0.5%). Values represent means calculated from three determinations
25
20
15 aa3
b
c
10 Total
0
0 1 7 15 30
Days
Saccharomyces cerevisiae culture cells isolated from dough with the addition of guar-
gum (0.1%). Values represent means calculated from three determinations
Codón et al. (16) showed that after prolonged freezing viability decreased in the
frequency of respiratory-deficient (petite) mutant formation. This indicated that mito-
chondria were not stable and were incompatible with the nucleus. Recently, Stoycheva et
al. (25) showed that freezing has mutagenic effect on mitochondrial DNA of the yeast
Saccharomyces cerevisiae, which induces respiration mutants in Saccharomyces cere-
visiae cells. However, in this study petite mutants were not observed. Content of cyto-
chrome, which shows intensity of aerobic metebolism, decreased during freezing. The
decrease of aerobic metobolism leads to lack of energy, which can induce decrease of
specific growth rate.
217
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
CONCLUSIONS
Addition of hydrocolloids to dough protects yeast cells during freezing. Carboxy-

methylcellulose in concentration of 0.3 and 0.5% increased the number of survived cells
on the surface to 70.64, and 70.28% and in the middle to 74.79, and 76.54%, respecti-
vely. Guar–gum increased survived cells number only in concentration of 0.1% on the
surface to 70.17%, and in the middle of the dough to 75.26%. Fermentative activities
decreased during 30 days of freezing. By addition of hydrocolloids to dough fermentative
activities decreased less compared to control samples. The best protection of yeast cells
was provided by carboxymethylcellulose. Individual and total cytochromes contents de-
creased during dough freezing, especially cytochrome aa3. Cytochromes contents in pure
Saccharomyces cerevisiae culture cells isolated from doughs with the addition of car-
boxymethylcellulose showed the smallest decrease during freezing.
REFERENCES
1. Yokoigawa, K., M. Sato and K. Soda: Simple improvement in freee-tolerance of

baker’s yeast with poly-γ-glutamate. J. Biosci. Bioeng. 102 (2006) 215-219.
2. Pejin, D., I. Došenović, S. Popov, Z. Suturović, J. Ranković, S. Dodić, J. Dodić and
V. Vučurović: Influence of dough freezing on Saccharomyces cerevisiae. Matica
Srpska Proc. Nat. Sci. 113 (2007) 293-301.
3. Nakamura, T., H. Takagi and J. Shima: Effects of ice-seeding temperature and intra-
cullular trehalose contents on survival of frozen Saccharomyces cerevisiae cells.
Cryobiology 58 (2009) 170-174.
4. Tanghe, A., A. Teunissen, P. Van Dijck and J. Thevelein: Identification of genes res-
ponsible for improved cryoresistance in fermenting yeast cells. Int. J. Food Microbiol.
55 (2000) 259-262.
5. Giannou, V., V. Kessoglou and C. Tzia: Quality and safety characterisicts of bread
made from frozen dough. Trends Food Sci. Tech. 14 (2003) 99-108.
6. Ribotta, P., G. Pérez, A. León and M. Añón: Effect of emulsifier and guar gum on
micro structural, rheological and baking performance of frozen bread dough. Food
Hydrocolloids 18 (2004) 305-313.
7. Guarda, A., C. Rosell, C. Benedito and M. Galotto: Different hydrocolloids in bread
improvers and antistaling agents. Food Hydrocolloids 18 (2004) 241-247.
8. Ribotta, P., S. Ausar, D. Beltramo and A. León: Interactions of hydrocolloids and
sonicated-gluten proteins. Food Hydrocolloids 19 (2005) 93-99.
9. Bárcenas, M.-E., M. Haros, C. Benedito and C. Rosell: Effect of freezing and frozen
storage on the staling of part-baked bread. Food Research International 36 (2003)
863-869.
10. Collar, C., P. Andreu, J. Martinez and E. Armero: Optimization of hydrocolloid addi-
tion to improve wheat bread dough functionality: a response surface methodology
study. Food Hydrocolloids 13 (1999) 469-475.
11. Sharadanant, R. and K. Khan: Effect of hydrophilic gums on the quality of frozen
dough: II. Bread characteristics. Cereal Chemistry 80 (2003) 773-780.
218
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
12. Regulations on methods on physical and chemical analysis for quality control of
wheat, milling and bakery products, pasta and fast frozen dough, Yugoslav Official
Register No 74/1988.
13. Kenny, S., H. Grau and K. Arendt: Use of response surface methodology to the ef-
fects of processing conditions of frozen dough quality and stability. Eur. Food Res.
Technol. 213 (2001) 323-328.
14. Rulebook on methods performing microbiological analyses and super analyses of
food products, Yugoslav Official Register Nо 25/1980.
15. Codón, A., A. Rinkón, M. Moreno-Mateus, X. Delgado-Jarana, M. Rey, C. Limón, I.
Rosado, B. Cubero, X. Reñate, F. Castrejón and T. Benitetez: New Sacchromyces ce-
revisiae baker´s yeast displayng enhanced resistance to freezing. J. Agric. Food
Chem. 57 (2003) 483-491.
16. Oura, E.: The effect of aeration on the growth energetics and biochemical composi-
tion of baker´s yeast, Ph.D. Thesis, University of Helsinki, 1972.
17. Pejin, D.: Industrijska mikrobiologija, Novi Sad, Tehnološki fakultet Novi Sad
(2003).
18. Reif, F., R. Kautzmann, H. Lüers and M. Lindemann, Die Hefen, Band I, Die Hefen
in der Wissenschaft, Nürnberg, Verlag Hans Carl (1960) 123-125.
19. Oura, E. and H. Suomalainen: Contents of cytochromes in yeast, J. Inst. Brew. 76
(1970) 532-545.
20. Ribotta, P., A. León and M. Añón: Effects of yeast freezing in frozen dough. Cereal
Chem. 80 (2003) 454-458.
21. Sharadanant, R. and K. Khan: Effect of hydrophilic gums on the quality of frozen
dough II . Bread characteristics. Cereal Chem. 80 (2003) 773-780.
22. Aguilera, J., F. Randez-Gil and J. A. Prieto: Cold response in Saccharomyces cere-
visiae: new functions old mechanisms. FEMS Miocrobiol. Rev. 31 (2007) 327-341.
23. Van Hoek, P., J. Van Dijeken. and J. Pronk: Effect of specific growth rate on fer-
mentative capacity of baker’s yeast. Appl. Environ. Microbiol. 64 (1998) 4226-4233.
24. Stoycheva, T., P. Venkov and Ts. Tsvetkov: Mutagenic effect of freezing on mito-
chondrial DNA of Saccharomyces cerevisiae. Crybiology 54 (2007) 243-250.
УТИЦАЈ ДОДАТКА КАРБОКСИМЕТИЛЦЕЛУЛОЗЕ, КСАНТАНА И ГУАР-

ГУМЕ У ХЛЕБНО ТЕСТО ПРЕ ЗАМРЗАВАЊА НА МЕТАБОЛИЗАМ И
ВИЈАБИЛНОСТ Saccharomyces cerevisiae
Душанка Ј. Пејин, Олгица С. Грујић, Јелена Д. Пејин, Ирена С. Дошеновић и

Сунчица Д. Коцић-Танацков
Хлебна теста су припремана са различитим концентрацијама карбоксиметил-

целулозе, ксантана и гуар-гуме (0,1; 0,2 и 0,3% у тесту), замрзавана на -20°C и ана-
лизирана након 0, 7, 15 и 30 дана. Чиста култура Saccharomyces cerevisiae је изо-
лована из теста и култивисана под оптималним условима 24 часа. Следећи пара-
метри су одређивани: специфична брзина раста, ферментативна активност и садр-
жај цитохрома у интактним ћелијама у циљу одређивања интензитета дисања. То-
ком замрзавања у трајању од 30 дана, проценат живих ћелија са површине теста је
219
APTEFF, 40, 1-220 (2009) UDC: 664.64.016.3/.8:664.644.7:664.037.5
DOI: 10.2298/APT0940211P BIBLID: 1450-7188 (2009) 40, 211-220
био 53,11%, а из средине 54,95%. Карбоксиметилцелулоза је при концентрацијама

од 0,3 и 0,5% повећаала број живих ћелија на површини на 70,64 и 70,28% и у
средини на 74,79 и 76,54%. Додатак гуар-гуме је повећао број живих ћелија само у
концентрацији 0,1% на површини до 70,17% и у средини теста до 75,26%. Средња
специфична брзина раста је смањена приближно за 10% током 30 дана чувања на
-20°C. Садржај цитохрома у интактним ћелијама је смањен у свим узорцима током
замрзавања.

220
In memoriam
NIKOLA MARJANOVIĆ
(1945 - 2009)
Prof. Nikola Marjanović was born on August 16, 1945 in

Novi Sad. He spent his childhood and adolescent years in Sa-
lajka, popular old part of Novi Sad. As a talented sportsman,
but also with extraordinary achievements at school, he fini-
shed the prestigious high school “Jovan Jovanović Zmaj” in
1964. After that he enrolled in the Faculty of Technology in
Novi Sad, from which he graduated in 1968 by completing
his diploma work entitled "Separation of Chlorophyll and
Carotenoids on the Starch Sucrose Column". As an extraor-
dinary, prize-winning student he became in 1969 a teaching
assistant of the newly introduced subject Technical Analyses,
which, together with his mentor, the late Prof. Miroslav Tur-
čić he has constantly endeavoured to improve, both in the
academic and research sense. At the same faculty he defended his MS thesis "An Attempt
in Coulometric Determination of L-ascorbic Acid" in 1972 and his PhD thesis entitled
"Kinetic-catalytic Determination by Fast Anodic Polarization" in 1979. From the very
beginning, his scientific interest has been in the field of instrumental analytical methods.
He became an assistant professor in 1980, associate professor in 1985, and full professor
in 1990. In that period he introduced the teaching subjects Instrumental Analysis of Food,
Instrumental Analysis, and Measurement Techniques, the last one being a subject of cru-
cial importance for process engineering.
There are few individuals in our academic environment that would possess such all-
embracing knowledge, both in the theoretical and practical sense, as Prof. Marjanović,
and even a smaller number of those who are ready to share so unselfishly his knowledge
with the others. He was an excellent leader of the Instrumental Analysis Group, founded
by himself, to which he devoted all his professional and organizational capacities, meti-
culousness and readiness for collaboration. Thanking to the well selected and trained
associates, laboratory equipment procured, and the results achieved in both education and
research and in the domain of diversified collaboration, his laboratory has become
recognizable in the present state and the whole space of the former Yugoslavia. To his
associates he offered unlimited help in their academic development and he thus created a
pleiad of followers of respectable capacity and achievements; an imperative for his suc-
cessors being to keep on the inherited high professional and scientific standards.
In the eighties and nineties of the last century, in his full academic maturity and at the
climax of creative activities, Prof. Marjanović implemented the best of his knowledge
and skills into the prospect of the Faculty of Technology and University of Novi Sad. He
was supervisor of numerous diploma works, MS and PhD theses. He authored and co-
authored several books and textbooks and a lot of educational materials, to cover almost
completely all teaching subjects he has taught with his associates. In the last decade of his
223
life he organized the Laboratory for Quality Control, and he was the initiator and genuine
creator of the new majoring profile Quality Control and several new subjects in accor-
dance with the Bologna Declaration.
In addition to the constant care of improving the existing and developing new sub-
jects and major profiles, Prof. Marjanović has enthusiastically worked on the develop-
ment of instrumental methods and techniques and construction of laboratory instruments.
For the development of a versatile system for electrochemical stripping analysis and other
analytical instruments, in 1989 he was awarded the Novi Sad Prize. Besides, he published
numerous papers and presented contributions at scientific meetings from the domain of
research and development projects.
In the course of an almost four-decade long work, in addition to his academic-peda-
gogical, scientific-research and professional activities, Prof. Marjanović has been invol-
ved in the work of the bodies of the Faculty and University, as well as in the other rele-
vant segments of public life. He served three terms (1991-1996) as the Vice-dean for
Finances of the Faculty of Technology, Vice-rector of the University of Novi Sad (1996-
1998) and Dean of the Faculty (2003-2004), and since 2004 he has been Head of the
Chair for Applied and Engineering Chemistries. By nature of things he has also been
active in, at the time inevitable, self-government bodies at the Faculty, University, and
legal-competence self-management associations. He has been bearer of high public
recognition such as the Science Committee Award of the SAP Vojvodina (1990). He has
also been part-time professor at the Faculty of Agriculture in Novi Sad and Faculty of
Technology in Banja Luka.
By his high ethical standards, human qualities, and academic-professional achieve-
ments, Prof. Nikola Marjanovic will permanently remain in our memories, with the unli-
mited gratitude for everything he has done for the affirmation of the Faculty and edu-
cation of numerous generations. His untimely departure from our midst is an irrede-
emable loss, first of all for his family, our Faculty, University of Novi Sad, and wider
academic community. In everything by which he dealed with he set extremely high stan-
dards and system of values, so that those who remain to follow his path must have the
strength and endurance to continue on from where he made an untimely stop.
Prof. Dr. Jovan B. Jakovljević

Prof. Dr. Zvonimir J. Suturović
224
INSTRUCTION FOR MANUSCRIPT PREPARATION
Acta Periodica Technologica, Novi Sad publishes reviews and scientific papers co-
vering all branches of technology: food, chemical, biochemical, pharmaceutical, as well
as process engineering and related scientific fields.
Acta Periodica Technologica is published in English. The journal may include sup-
plements from congresses, meetings or symposiums.
Submission of Papers. All correspondence, including submission of the manuscript,

notification of the Editor’s decision and requests for revision, takes place by e-mail:
sdjilas@tf.uns.ac.rs or bastajab@uns.ac.rs.
Submission of paper implies that
- it has not been published previously (except in the form of an abstract or as a whole
in the proceedings of papers of a scientific meeting, or as part of a published
lecture or academic thesis),
- it is not under consideration for publication elsewhere, and
- it will not be published elsewhere in the same form, in English or in any other lan-
guage, without the written consent of the publisher.
If the paper was given, wholly or in part, at a scientific meeting, this should be stated
in a footnote on the title page.
Preparation of manuscripts
Language: Manuscript should be written in English.

Typing: Manuscript must be written in Word with a font size 10 pt, double spaced
with wide margins (3 cm) on A4 pages (max. 10 pages for scientific papers). Number all
pages of the manuscript. Import tables and figures into the text.
General format. The manuscript should contain the following in this order: Title
page, ABSTRACT and KEY WORDS, INTRODUCTION, EXPERIMENTAL, RE-
SULTS and DISCUSSION, CONCLUSION, ACKNOWLEDGEMENT and REFEREN-
CES as well as ABSTRACT IN SERBIAN LANGUAGE.
Title page: On the first page should be the title without symbols, formulae or abbre-
viations (capital bold letters). Full name (name, initial and surname) of each author and
co-author, without degrees, professional or official titles should be given under the title,
written in italic. Title should be concise and explanatory of the content of the paper.
Abstract of the paper (100-250 words, written in italic) should be given under the tit-
le and authors. Abstracts should contain the aim of investigated work, methods, results
and conclusion.
Key words (normal letters, max. 5 key words) should be listed afterwards.
Introduction should state previous relevant work with appropriate references, the
problem investigated and the aim of work.
Experimental. The methods and materials used should be stated clearly in sufficient
detail to permit the work to be repeated by others. Only new techniques should be descri-
bed in detail; known methods must have adequately references.
Results and Discussion. Results should be presented concisely, with tables or illu-
strations for clarity. The significance of the findings should be discussed without repe-
tition of the material in the Introduction. Adequate number of illustrations, graphs and
chemical formulae used must be kept on minimum.
Conclusion should indicate the significant contribution of the manuscript with its
applications.
Acknowledgements: These should be kept to a minimum.
References cited should be indicated in the text using Arabic numerals in brackets ( ),
in the order of appearing. All publications cited in the text should be presented in a list of
references given on a separate page. Abbreviations of journal titles should be given ac-
cording to the International Codex for Abbreviations of Journal Titles (Chemical Abs-
tracts). The list of references should be arranged according to their appearance in the text.
Give names of all authors (do not use „et.al.“), with their initials before respective surna-
mes. Include article titles in journals. Journal titles should be abbreviated according to the
Chemical Abstracts Service Source Index, 2005 edition, and supplements. The abbrevia-
ted titles should be followed by the volume (in bold), number (if exists), year (in paren-
theses) and first and last page numbers.
Examples:
Journals: E.C. Pascual, B.A. Goodman and C. Yeretzian: Characterisation of Free
Radicals in Solubile Coffee by Electron Paramagnetic Resonance Spectroscopy. J. Agric.
Food Chem. 50, 21 (2002) 6114-6122.
If pagination repeated in each number of one volume it is necessary to add this par-
ticular numeral.
Books: W. Banks and C.T. Greenwood: Starch and its Components, Edinburgh Uni-
versity Press, Edinburgh (1975) p.98.
Book with more chapters: C. Mercier: Extrusion Cooking of Starch, in Polysacchari-
des in Food. Eds. J.M.V. Blanshard and J.R. Mitchell, Butterworth, London (1978) pp.
152-170.
Book of Abstracts: W. Noe, M. Howaldt, R. Ulber and T. Scheper: Immunobase elu-
tion assay for process control, 8th European Congress on Biotechnology, Budapest, 17-21
August 1997, Book of Abstracts WE 163, p. 246.
Thesis: J.B. Linstead: Effects of adding natural antioxidants on colour stability of
paprika, Ph.D. (or M.Sc.) Thesis, University of Glasgow, 2006.
Patent: B.O. Miller: U.S. Pat. 2542356 (1962), Dow Chemical Comp.: Abstr. 51
(1961) 2870.
Unpublished data: Should be cited with one of the following comments „in press“,
„unpublished work“ or „personal communication“.
Online citations: Should include the author, title, website and date of access (exam-
ple: Wright, N.A.: The Standing of UK Histopathology Research 1997-2002.
http://pathsoc.org.uk (accessed 7 October 2004)).
Abstract in Serbian language should be given at the end of manuscript (after re-
ferences), in extended form (max. length 1 page), printed in Cyrillic (normal letters) with
the title (capital letters), full name(s) of each author(s) and affiliation(s) (italic letters).
Chemical nomenclature and units. Authors are requested to use SI units and che-
mical nomenclature following the rules of Chemical Abstracts whenever possible.
Tables. Each Table is numbered with Arabic numeral, followed by the title (Table 1.
Result...).
Figures. Each drawing or figure should also be numbered with Arabic numerals
followed by the title (Fig. 1. Chromatogram of...). Graphs and charts must be prepared by
Microsoft Excel or Origin. Schemes must be prepared by Microsoft Visio or Corel Draw.
It is necessary to submit them in original extension (xls, vdr, cdr). Scanned black&white
schemes should be submitted in tif, wmf, or bpm form. Colour and black&white pho-
tographs should be submitted in jpg form.
Formulae and Equations. Type formulas and mathematical equations clearly, accu-
rately placing superscripts and subscripts. Equations should be indicated in the text using
Arabic numerals in square brackets [ ]. Abbreviations and symbols-notation should be
explained at first appearing, or on a separate list at the end of manuscript.
Additional information
Manuscript should be sent in one hard copy to the address: Faculty of Technology,
Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia, to the editor: Professor Dr. Sonja Djilas
(for Acta Periodica Technologica) and electronic form to the e-mail: sdjilas@tf.uns.ac.rs
or bastajab@uns.ac.rs. Authors are expected to propose the category of manuscript (re-
view, original scientific paper).
Review process. All papers submitted to the journal will be reviewed by at least two
independent referees who will be asked to complete the refereeing job within 4-6 weeks.
Final decision on publication will be made by the Editorial Board. Manuscripts may be
sent back to authors for revision if necessary. Revised manuscript submissions should be
made as soon as possible (within 2 weeks) after the receipt of the referees comments.
Proofs. One set of page proofs will be sent by e-mail to the corresponding Author.
The author may list the corrections and return to the journal in an e-mail. Proofs must be
corrected and returned to the publishers within 48 hours of receipt.
Offprints. The corresponding author, at no cost, will be provided with a 10 free paper
offprints.
Author service. For inquiries relating to the submission of manuscript, please send
an a e-mail to the Editor.
THIS ISSUE OF ACTA PERIODICA TECHNOLOGICA
IS FINANCIALLY SUPPORTED BY:
Ministry of Science and Technological Development of Republic of Serbia,

Belgrade
Cover design Živojin Katić ∗ Typsetting Branislav Bastaja ∗ Printed by VERZAL, Novi Sad
FORMER EDITORS-IN-CHIEF
Prof. Dr. Adalbert Šenborn (1967-1970)

Prof. Dr. Radivoj Žakula (1972-1975)
Prof. Dr. Miroslava Todorović (1976-1994)
Prof. Dr. Biljana Škrbić (1995-1998)

Acta Periodica Technologica 2009-40

Enviado por

Dados do documento

Descrição original:

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Acta Periodica Technologica 2009-40

Enviado por

Direitos autorais:

Formatos disponíveis

FACULTY OF TECHNOLOGY – NOVI SAD

ТЕХНОЛОШКИ ФАКУЛТЕТ – НОВИ САД

APTEFF, 40, 1-220 (2009)

YU ISSN 1450 – 7188 CODEN: APTEFF

Biljana R. Cvetković and Marija R. Jokanović

Gordana R. Dimić, Sunčica D. Kocić-Tanackov, Dušanka J. Pejin,

Ljubica P. Dokić, Marija I. Bodroža-Solarov, Miroslav S. Hadnađev

Aleksandar Z. Fišteš and Đuro M. Vukmirović

Gordana B. Koprivica, Nevena M. Mišljenović, Ljubinko B. Lević

Radomir V. Malbaša, Eva S. Lončar, Spasenija D. Milanović

Anamarija I. Mandić, Sonja M. Djilas, Jasna M. Čanadanović-Brunet,

Dragan V. Palić and Sophia E. Coetzee

Dragan V. Palić and Klaas-Jan Leeuw

Dragana Pešić-Mikulec and Gordana B. Niketić

Slađana M. Savatović, Aleksandra N. Tepić, Zdravko M. Šumić

Mirjana A. Vasić, Biserka L. Vujičić, Aleksandra N. Tepić,

Tanja D. Žugić-Petrović, Nataša M. Joković and Dragiša S. Savić

CHEMICAL TECHNOLOGY AND PROCESS ENGINEERING

Eva S. Lončar, Miroslava M. Radeka, Snežana B. Petrović,

Nataša Lj. Lukić, Svetlana S. Popović and Jelena Dj. Marković

Aleksandar R. Stanić, Saša I. Jovanić, Nikola J. Marjanović

Marina B. Šćiban, Mile T. Klašnja and Mirjana G. Antov

Ivana M. Šijački, Radmilo R. Čolović, Milenko S. Tokić and Predrag S. Kojić

BIOCHEMICAL AND PHARMACEUTICAL ENGINEERING

Marija M. Škrinjar and Nevena T. Nemet

Dušanka J. Pejin, Olgica S. Grujić, Jelena D. Pejin, Irena S. Došenović

Prof. dr Nikola J. Marjanović 223

INSTRUCTION FOR MANUSCRIPT PREPARATION

Биљана Р. Цветковић и Марија Р. Јокановић

Гордана Р. Димић, Сунчица Д. Коцић-Танацков, Душанка Ј. Пејин,

Љубица П. Докић, Марија И. Бодрожа-Соларов, Мирослав С. Хаднађев

Александар З. Фиштеш и Ђуро М. Вукмировић

Гордана Б. Копривица, Невена М. Мишљеновић, Љубинко Б. Левић

Радомир В. Малбаша, Ева С. Лончар, Спасенија Д. Милановић

Анамарија И. Мандић, Соња М. Ђилас, Јасна М. Чанадановић-Брунет,

Драган В. Палић и Sоphia E. Coetzee

Драган В. Палић и Klaas-Jan Leeuw

Драгана Пешић-Микулец и Гордана Б. Никетић

Слађана М. Саватовић, Александра Н. Тепић, Здравко М. Шумић

Мирјана А. Васић, Бисерка Л. Вујичић, Александра Н. Тепић,

Тања Д. Жугић-Петровић, Наташа М. Јоковић и Драгиша С. Савић

ХЕМИЈСКА ТЕХНОЛОГИЈА И ПРОЦЕСНО ИНЖЕЊЕРСТВО

Ева С. Лончар, Мирослава М. Радека, Снежана Б. Петровић, Андреа С. Скапин,

Наташа Љ. Лукић, Светлана С. Поповић и Јелена Ђ. Марковић

Славиша С. Путић, Марина Р. Стаменовић, Бранислав Б. Бајчета

Александар Р. Станић, Саша И. Јованић, Никола Ј. Марјановић

Марина Б. Шћибан, Миле Т. Клашња и Мирјана Г. Антов

Ивана М. Шијачки, Радмило Р. Чоловић, Миленко С. Токић и Предраг С. Којић

БИОХЕМИЈСКО И ФАРМАЦЕУТСКО ИНЖЕЊЕРСТВО

Марија М. Шкрињар и Невена Т. Немет

Душанка Ј. Пејин, Олгица С. Грујић, Јелена Д. Пејин, Ирена С. Дошеновић

Проф. др Никола Ј. Марјановић 223

EFFECT OF PRESERVATION METHOD AND STORAGE CONDITION ON

Biljana R. Cvetković and Marija R. Jokanović

KEY WORDS: L-ascorbic acid, colorimetric method, beverage, storage

Shelf-life study. Samples of beverages thermally pasteurised or preserved with sodi-

RESULTS AND DISCUSSION

According to our results pasteurisation method had high influence on vitamin C

Ascorbic acid content (mg/l)

80 74.16 74.89 100 mg/l added vitamin C

Pasteurisation Preserved with sodium-benzoate

100 mg/l added vitamin C 150 mg/l added vitamin C