Catalysis Today 209 (2013) 122126

Contents lists available at SciVerse ScienceDirect

Catalysis Today
j our nal homepage: www. el sevi er . com/ l ocat e/ cat t od
Key role of common anions on the photocatalytic degradation proles of the
molecular size fractions of humic acids
C.S. Uyguner-Demirel

, C. Birben, M. Bekbolet
Bogazici University, Institute of Environmental Sciences, Bebek, Istanbul 34342, Turkey
a r t i c l e i n f o
Article history:
Received 27 June 2012
Received in revised form
24 November 2012
Accepted 28 November 2012
Available online 15 February 2013
Keywords:
Anions
Molecular size fractions of humic acid
Photocatalytic degradation
a b s t r a c t
Humic acids are the most abundant components of the colloidal and dissolved fraction of natural organic
matter (NOM) and widely known to be representatives of NOM behavior in the aquatic environment.
The objective of this study was to investigate the inuence of common anions on the photocatalytic
degradation of humic acid and its different molecular size fractions. A representative range of anions
were added to humic acid to simulate diverse characteristics of surface water. Induced effects on the
removal of the target organic compound were studied. With the aim of assessing the photochemical
processes under irradiation conditions, the photocatalytic removal efciency of humic acid was deter-
mined taking into account the major background factors. The results revealed that some specic UVvis
parameters agreed well with the information provided by uorescence spectroscopy. Despite the com-
plex nature of humic acids in the presence of a solution matrix, the spectroscopic parameters could be
regarded as indicators for the photocatalytic removal of humic acid and its molecular size fractions.
Depending on the presence of low and high anion matrix, different kinetic model parameters were
attained for the photocatalytic removal of the molecular size fractions of humic acid. The photocatalytic
degradation rates could be expressed by the half-life values ranging from 20 min to 133 min.
2013 Elsevier B.V. All rights reserved.
1. Introduction
Natural organic matter (NOM) is a collective termassigned to a
mixture of chemicals that are present ubiquitously in the aquatic
environment with different molecular sizes ranging from a few
hundred to 100,000daltons (Da). NOM affects water quality by
providing precursor material for disinfection by-products (DBPs),
substrate for biogrowth, and complexation sites for heavy metals.
To understand these processes better and to control their effects
on drinking water quality, it is necessary to study the chemistry of
NOMin detail. Although the components of NOMcan be classied
as humic acids, fulvic acids, polysaccharides, hydrophilic acids, pro-
teins, lipids, amino acids and hydrocarbons, no single molecule
of NOM exists. To a certain extent, different NOM samples can be
considered as having a set of average properties, which depend on
the source material, the degree of subsequent degradation reac-
tions, and transport related fractionation processes. Differences
in these average properties can have a marked effect on processes
involved in water treatment. Therefore, characterization of the

Corresponding author. Tel.: +90 212 3597146; fax: +90 212 2575033.
E-mail addresses: uygunerc@boun.edu.tr, uygunerc@hotmail.com
(C.S. Uyguner-Demirel), cemrebirben@live.com(C. Birben),
bekbolet@boun.edu.tr (M. Bekbolet).
molecular size of humic acid is a valuable tool when assessing
its geochemical origin as well as its effect on the performance of
water treatment systems.
Ultraltration has been frequently used for the physical
fractionation of humic substances [1,2]. An overview of the
fractionation techniques can be found in the recent review of
Matilainen et al. [2] and Sulzberger and Durisch-Kaiser [3]. It has
generally beenreported that structural and spectral properties var-
ied among molecular size fractions, regardless of the fractionation
techniqueemployed. Humic acidfractions that wereobtainedusing
ultraltrationwere characterizedfor their chemical, structural, and
molecular properties [1,4,5].
The chemical analyses showed that the alkyl hydrophobic com-
ponents were mainly distributed in the largest molecular-size
fractions, while phenolic and oxygen-containing groups were pre-
dominant in the low nominal molecular weight fractions. Humic
substances mainly humic acids constitute the major fraction of
natural organic matter in water supplies, therefore, their photo-
catalytic degradation has been studied in detail [68]. Considering
the impact of the solutionmatrix onthe photocatalytic degradation
process, recently the photocatalytic removal efciency of humic
acids under different conditions simulating surface and groundwa-
ter was investigated [9,10]. Since humic acid is a noncharacterized
complex mixture of organic macromolecules, the molecular weight
of which varies over a wide range, the work presented herein
0920-5861/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2012.11.020
C.S. Uyguner-Demirel et al. / Catalysis Today 209 (2013) 122126 123
is a continuation of the efforts to understand the photocatalytic
degradation of its molecular size fractions. Considering the real
case of natural water samples, this work was carried out in an
attempt to mimic natural water conditions and account for the
effect of anions which would have an inuential role on the
humic-TiO
2
binary system by exerting ionic interactions, diverse
conformational and compositional changes in the structure. Spe-
cic UVvis and uorescence parameters were also employed for
the assessment of the photocatalytic degradation efciencies in the
presence of selected anions commonly found in natural waters,
namely uoride, chloride, bromide, nitrite, nitrate, sulfate and
phosphate.
2. Materials and methods
Raw humic acid (humic acid sodium salt, Aldrich) solution
(50mg L
1
) was prepared to account for natural organic matter
containing surface waters. Initially, humic acid solutions were l-
tered through 0.45m (M1) cellulose acetate membrane lter
(Millipore). Subsequently, humic acid was fractionated by ultral-
tration (Amicon 8050) into three nominal molecular size fractions
of 100kDa (M2), 30kDa (M3) and 3kDa (M4).
Standard mix solution containing seven of the anions as nitrate,
nitrite, phosphate, sulfate, uoride, chloride and bromide anions
were obtained fromDionex (Product number: 0575909). To avoid
excessive phosphate concentration and keep an effective load of
anions, difference by one order of magnitude was selected for
the studied range of anion concentrations. Anions of low con-
centration (LC) were: [F

], [Br

] =210
6
mol L
1
, [Cl

], [NO
2

],
[NO
3

] =110
5
mol L
1
; [SO
4
2
], [PO
4
3
] =310
6
mol L
1
whereas anions of high concentration (HC) were as follows: [F

],
[Br

] =210
5
mol L
1
; [Cl

], [NO
2

],[NO
3

] =110
4
mol L
1
;
[SO
4
2
], [PO
4
3
] =310
5
mol L
1
. The quantication of anions
was carried out using Dionex ICS-3000 Ion Chromatography.
To prepare working solutions, each molecular size fraction of
humic acid (M1M4) was added anions of low and high con-
centrations to simulate surface water conditions. Experiments
were conducted at natural pH of the samples (pH6) with-
out any adjustment. All of the experiments were performed in
duplicate and the variability between replicate experiments was
<5%.
Humic acid working solutions were subjected to photocatalytic
degradation in a completely mixed reactor utilizing UV-A light
(125W BLF lamp, I
o
=2.8510
16
quanta s
1
) and TiO
2
Degussa
P-25 in powder form (0.25mg mL
1
). More details about the
experimental procedure have been reported [11]. Prior to analy-
sis, TiO
2
was removed from the solution by ltration (0.45m,
cellulose acetate membrane lter). The extend of mineralization
of organic content was determined by dissolved organic car-
bon (DOC, mg L
1
) measurements using the Total Organic Carbon
Analyzer (Shimadzu TOC-VWP). Humic acid was identied by
spectroscopic parameters (m
1
) such as Color
436
(absorbance at
436nm), UV
365
(absorbance at 365nm), and UV
254
(absorbance at
254nm) by UVvis spectrophotometry (Perkin Elmer lambda 35).
Synchronous scan uorescence spectra were acquired in the exci-
tation wavelength range of 300650nm using the bandwidth of
=18nm between the excitation and emission monochroma-
tors (FI
syn
,) using Perkin Elmer LS 55 Luminescence Spectrometer.
In addition to UVvis absorbance and uorescence intensity
measurements, their DOC normalized values dened as specic
UV absorbance (SUVA

, DOC normalized UV

at =254nm and
365nm (mg CL
1
)
1
m
1
), specic color absorbance (SCOA, DOC
normalized Color
436
(mg CL
1
)
1
m
1
) and specic uorescence
intensity (SFI
syn
, specic synchronous uorescence intensity at

max
: 470nm) were also presented [12].
3. Results and discussion
3.1. UVvis and uorescence spectral features of humic acid and
its molecular size fractions under non treatment and post
treatment conditions
It is known that the molecular size distribution proles of the
humic substances exhibit a general decreasing trend from high
molecular size fractions to lower molecular size fractions irrespec-
tive of the originandsource of the organic matter [5]. Spectroscopic
characterization of the molecular size fractions were presented
with respect to specic absorbance ratios of each fraction derived
fromhumic acid and the natural water sample [5,13]. For simplic-
ity purposes, E termwas used to designate either UV or Color when
presenting the absorbance ratios. The ratio of E
254
/E
436
was pro-
posed to assess the composition of altered NOM provided that the
total removal of color-forming moieties was not required. Under
conditions of total decolorization, the use of E
250
/E
365
was also
recommended relevant to the spectroscopic changes acquired dur-
ing oxidative treatment [5,14]. For comparison purposes selected
spectroscopic parameters of humic acid size fractions in relation to
photocatalytic treatment efciencies were presented in Figs. 13.
Although the UVvis spectral features of humic acid and its size
fractions appearedtobe broadandfeatureless, showing nomaxima
or minima, specic absorbance intensities at 254, 365, and 436nm
as SUVAandSCOAdisplayedvariations among the fractions [5]. The
polydisperse nature of humic acid unifying components of dissim-
ilar molecular weights was clearly observed from the presented
absorbance ratios of each size fraction (Fig. 1). When ultraltra-
tion was applied, it did not involve explicit size or functional group
separation however, it produced fractions of a certain molecular
size distribution that showed spectroscopic evidence of structural
discrepancies. The spectroscopic ratios calculated for the different
molecular size fractions of humic acid showed slight differences
depending on the presence of anions. The ratios for M4 were sig-
nicantly higher than the other fractions studied. It was reported
by several researchers that the quotient E
250
/E
365
(absorbance at
250nm and 365nm), which is a property of aquatic humic sub-
stances, increased as the aromaticity and molecular size decreased
[15]. High molecular weight fractions and unfractionated humic
acid contained a relatively higher amount of aromatic or polyphe-
nolic structures compared to low molecular weight fractions. As a
general trend, the absorbance ratios of the fractionatedhumic acids
increased with decreasing molecular size.
E
254
/E
436
and E
250
/E
365
ratios for raw M1 and M2 fractions of
humic acid showed constant tendency irrespective of the pres-
ence of anions. After treatment for 60min increase of ratios became
Fig. 1. Spectroscopic parameters of the molecular size fractions of humic acids with
respect to treatment conditions (PC: photocatalytic oxidation for 60min).
124 C.S. Uyguner-Demirel et al. / Catalysis Today 209 (2013) 122126
Table 1
Percent photocatalytic removal of the molecular size fractions of humic acid in the
presence of anions (irradiation period of 60min).
Substrate Removal (%)
Color
436
UV
365
UV
254
DOC FIsyn
M1 60 54 43 40 27
M1 +AL 53 49 42 38 29
M1 +AH 38 35 28 26 8
M2 80 65 64 57 68
M2 +AL 77 62 58 45 63
M2 +AH 63 59 45 36 46
M3 96 89 91 71 98
M3 +AL 77 83 82 68 97
M3 +AH 65 67 62 31 76
M4 90 87 85 81 90
M4 +AL 79 81 79 63 99
signicant for the high molecular size fractions but especially for
the fraction of 100kDa. The spectroscopic ratios for the lower
molecular size fractions like M3 and M4 decreased after treatment.
Taking into account the different structural and chemical charac-
teristics of the molecular size fractions and referring to reaction
rate constants given in Table 1, diverse reaction mechanisms could
be assumed for each fraction. The lower molecular size fractions
indicating higher aromaticity andexpressing lower reactivity could
be considered as fulvic like whereas higher molecular size fraction
i.e. 100kDa could be accepted as a more representative form of
humic acid (humic like fraction). The observed increase in absorp-
tion ratios attained in the higher molecular size fractions as M1 and
M2couldbeexplainedbyfast removal regimeof Color
436
andUV
365
with respect to UV
254
rather than possible recombination of UV
absorbing moieties through intramolecular aggregation operating
via hydroxylated functional group bridging.
The uorescence properties of the fractionated humic acid were
also investigated. Fig. 2 shows a comparison of the synchronous
scan spectra of the molecular size fractions of humic acid in the
presence/absence of anions after photocatatalytic degradation for
60min (2671% DOC removal, Table 1).
Synchronous scan spectra of humic acid and its related frac-
tions showed typical uorescence curves with maximumemission
around470nm. Fluorescenceintensities decreasedwithdecreasing
molecular size fraction. As a general trend, lower uorescence
intensities were attained after photocatalytic degradation for
60min. Taking into account the uorescence decrease and ini-
tial shift of emission maxima at 470nm, a subsequent decay
withphotocatalytic irradiationtime was observedindicating major
structural changes which were more prominent in the fractions M2
and M3 (Fig. 2). The decrease in uorescence intensity might be
as a result of the hydroxyl radical attacks to high electron density
regions and break of the chromophoric groups within the struc-
ture. Moreover, preferential break up of high molecular weight
compounds could also occur [16]. A blue shift was associated with
Fig. 3. Specic UVvis and synchronous scanuorescence parameters of the molec-
ular size fractions of humic acids in the presence of anions under non treatment and
post treatment conditions.
decomposition of condensed aromatic moieties and the break up of
the large molecules into smaller fragments such as a reduction in
the degree of the -electron systems and a decrease in the number
of aromatic rings. It was also linked with a reduction in the number
of conjugated bonds in a chain structure, a conversion of a linear
ring systemto a non linear systemand the elimination of particular
functional groups including carbonyl, hydroxyl and amine groups
[17,18]. The formation of a minor intensity peak around 550nm
could be due to the depletion of the aromatic and/or polyphenolic
contents which were mostly of high molecular weight and also
due to the enrichment of its carbohydrate materials which did not
give uorescence (Fig. 2B and C). Additionally, the presence of car-
bonyl containing substituents, hydroxyl, alkoxyl and amino groups
caused a shift in uorescence to longer wavelengths [17]. Under
all experimental conditions investigated, the presence of anions
relatively enhanced the uorescence intensities of each respective
molecular size fraction of humic acid after photocatalytic degrada-
tion.
Normalizing the intensity values with the corresponding DOC
contents revealed higher SFI values for M1 irrespective of anion
concentration (Fig. 3). The specic uorescence parameters of the
low molecular size fraction of humic acid (M4) in the presence
of high anion concentration could not be attained due to very
low intensity values. In case of photocatalytic treatment, the gen-
eral trend was the decreasing uorescence intensity proles with
respect to increasing photocatalytic irradiation time (Fig. 2). The
SFI
syn
values of M1 fraction and its complex system comprising
anions increased after degradation whereas M2M4 molecular size
fractions exhibited a decreasing trend irrespective of the presence
of anions.
SUVA and other related parameters illustrate the degradation
efciencies in relation to the successive removal of the parame-
ters. SUVA>4 reects the presence of highly hydrophobic fractions.
Fig. 2. Synchronous scan spectra of the photocatalytic removal of the molecular size fractions of humic acids (PC: photocatalytic oxidation for 60min).
C.S. Uyguner-Demirel et al. / Catalysis Today 209 (2013) 122126 125
Table 2
Pseudo rst order rate constants of the photocatalytic degradation of molecular size fractions of humic acid (k: rate constant, min
1
; t
1/2
: half-life, min; R
2
>0.64).
Fraction Color
436
UV
365
UV
254
DOC
k 10
2
min
1
t
1/2
min k 10
2
min
1
t
1/2
min k 10
2
min
1
t
1/2
min k 10
2
min
1
t
1/2
min
M1 1.65 42 1.46 47 1.07 65 0.773 90
M1 +AL 1.20 58 1.09 64 0.850 82 0.745 94
M1 +AH 0.789 88 0.697 99 0.521 133 0.907 77
M2 1.70 41 1.67 41 0.932 75 1.31 53
M2 +AL 2.29 30 1.77 39 1.29 54 1.21 57
M2 +AH 2.59 27 2.54 27 2.18 32 0.715 97
M3 2.76 25 3.45 20 3.39 20 2.01 34
M3 +AL 2.55 27 3.12 22 2.99 23 1.92 36
M3 +AH 1.66 42 1.72 40 1.52 46 0.606 116
For SUVA ratios in the range of 24, humic acids are normally
dominated with a mixture of hydrophobic and hydrophilic frac-
tions. SUVA<2 signicantly represents hydrophilic properties [10].
Considering initial SUVA
254
, SUVA
365
and SCOA
436
values of M1
and M2 fractions of humic acid, the addition of anions to these
molecular size fractions did not reveal any signicant change in
the specic UVvis parameters. On the other hand, for the lower
molecular size fractions (M3 and M4), initial SUVA
254
, SUVA
365
and SCOA
436
values increased proportional to anion concentra-
tion. The most signicant change was eminent in the presence of
high anion concentration. However, the presence of anions did not
signicantly change the initial SFI values. For M1 molecular size
fraction irrespective of anion concentration after photocatalytic
degradation, higher SFI values were observed. In all other cases,
specic uorescence intensities were lower thaninitial values after
photocatalysis. The results implied that specic parameters such as
SUVA
365
and SUVA
254
agreed well with the information provided
byuorescencespectroscopy(Fig. 2). Despitethecomplexnatureof
humic acids in the presence of a solution matrix, the spectroscopic
parameters could be visualized as indicators for the photocatalytic
removal of humic acid and its molecular size fractions.
WithreferencetotheabovegivenUVvis anduorescencespec-
troscopic changes attained for raw and photocatalytically treated
humic samples (M1M3), the overall percent removal of the speci-
ed parameters as Color
436
, UV
365
, UV
254
, and DOC were presented
in a comparative manner in the presence/absence of anions after
photocatalytic degradation for 60min (Table 1).
The photocatalytic removal efciency of humic acid increased
withdecreasing molecular size fractionirrespective of the specied
parameters. No signicant difference was attained in the presence
of lowanion concentration; however, increasing anion concentra-
tion had an inhibitory effect on the removal of humic acid in terms
of DOC, and UV
254
. The changes evaluated by uorescence indices
were more prominent. Percent removals of the spectroscopic and
uorescence parameters for M1M3 decreased with increasing
anionconcentration. DOCremoval also followed a decreasing trend
for M1M3. Considering the very lowinitial values of the respective
parameters of the M4 fraction, percent removal in the presence of
high anion matrix were not determined.
The photocatalytic degradation of humic acid resulted in the
removal of DOC accompanied by a decrease in SUVA
254
and
SUVA
365
as well as inSFI
syn
(Fig. 3). Consideringthesurfaceoriented
nature of photocatalysis, initial adsorptive interactions were also
evaluated. The results of the initial dark adsorption studies of the
molecular size fractions of humic acid; M1, M2 and M3 indicated
37%, 41% and 44% removal of Color
436
; 27%, 33% and 28% removal
of UV
254
; 33%, 39% and 38% removal of UV
365
and 30%, 33% and 29%
removal of DOC prior to photocatalysis, respectively. Initial adsorp-
tive removal of the specied parameters revealed that molecular
size distribution did not signicantly affect the functional group
interactions with the TiO
2
surface. Addition of anions of low con-
centration to M1M3 indicated initial adsorption of humic acid
onto TiO
2
as determined by the removal of 34%, 38% and 15%
removal of Color
436
, 31%, 46% and 14% removal of UV
254
and 33%,
29% and 20% removal of DOC. However, the results could not be
considered as signicantly different fromeach other. On the other
hand, inthe presence of highanionsolutionmatrix M1, M2, M3 dis-
played 16%, 72%, 44% removal of Color
436
; 14%, 62%, 40% removal
of UV
254
; 15%, 68%, 44% removal of UV
365
and 23%, 63% and 5%
removal of DOC prior to photocatalysis, respectively. Under oper-
ating neutral pH conditions, while humics were not expected to
bind anions due to charge repulsion, their interactions with oxide
phase could signicantly affect anion binding to the TiO
2
surface
via electrostatic attractions and/or site competition[19]. Therefore,
different photocatalytic removal kinetic model parameters could
be expected for humic molecular size fractions in the presence of
lowand high anion matrix.
3.2. Kinetic modeling of the photocatalytic removal of humic acid
and its molecular size fractions
The photocatalytic degradation kinetics of humic acid in terms
of UVvis parameters as well as DOC removal followed pseudo rst
order kinetic model as previously reported in literature [10,13].
The kinetic evaluation of the experimental data for each molecular
size fraction of humic acid and respective complex system com-
prising anions revealed the diverse behavior of the humic fractions
for Color
436
, UV
365
, UV
254
and DOC as presented in Table 2.
For M1 and M2 molecular size fractions the removal rate
constants followed the usual trend as Color
436
>UV
365
>UV
254
under all conditions. In the presence of anions comparatively lower
removal rate constants were attainedfor M1 andM3molecular size
fractions irrespectiveof thespeciedparameter. Moresignicantly,
the presence of anions both in high and low anion concentra-
tions displayed an enhancement effect on M2 fraction contrary to
the inhibitory effect attained for both M1 and M3. However, DOC
removal rates could be visualized as a slight enhancement for M1
and comparatively inhibitory effect was observed in the case of M2
and M3 molecular size fraction. The photocatalytic removal of M4
fraction displayed a rapid decrease in the early stages of irradiation
followed by a constant non oxidative fraction. Therefore, kinetic
evaluation of M4 could not be attained.
For comparison purposes through the application of a nor-
malization approach of kinetic constants with respect to each
corresponding sole humic acid molecular size fraction, the most
pronounced effect was observed in the presence of high anion con-
centration rather than low anion. In the presence of high anion
matrix, a signicant enhancement was attained in the kinetic
constants that could be ordered as UV
254
>UV
365
=Color
436
for M2
fraction followed by DOC removal for M1 fraction. On the contrary,
a slight inhibitory effect was observed for the UVvis parameters
in the order of UV
254
>UV
365
=Color
436
for M3 fraction. The photo-
catalytic degradation rates could also be expressed by the half-life
values rangingfrom20min to133min. For M1andM3fractions and
126 C.S. Uyguner-Demirel et al. / Catalysis Today 209 (2013) 122126
their complex system comprising anions, increasing half life val-
ues were attained with increasing anion concentration as detected
by the spectroscopic parameters. In contrast, unlikely for the M2
molecular size fraction, the related half life values decreased in
parallel with the anion concentration.
Wiszniowski et al. [20] studied the inuence of the sulfate
(7.75gL
1
) and chloride (2.5gL
1
) mixture on the photomineral-
izationrateof humic acid. Thepresenceof bothanions was reported
to induce relevant reduction in the reaction rate. Similarly, photo-
catalytic degradationof humic acidinarticial seawater containing
a high concentration of anions such as sulfate, chloride, carbonate
and bicarbonate was investigated by Al-Rasheed and Cardin [21].
The degradation of humic acid was reported to be less rapid in high
salinity water than in low salinity water comprising of 20 times
less amount of total dissolved solids (2700mg L
1
) as evidenced
by absorption at 400nm. As a general statement, the presence of
anions induces adecreaseof theratefor therst stepof thereaction,
but nallythe DOCvalues are more or less the same at the endof the
experiments. It can be assumed that there is competitive adsorp-
tionbyanions andhumic acidonthe surface of catalyst especiallyat
lower molecular size fractions of humic acid, together with possi-
ble side reactions in the solution. As reported in literature, chloride
at high concentration may partially block the active site of the cat-
alyst while sulfate may form reactive species such as SO
4

[22].
Bekbolet et al. reported that the presence of chloride, nitrate, sul-
fate anions and signicantly phosphate reduced the color removal
rate of humic acid of terrestrial origin in water as detected by
the parameters Color
436
and Color
400
[9]. Furthermore, by a sim-
ulation approach to surface and groundwaters, diverse effects of
selected anions as well as counter cations were also discussed
by Uyguner and Bekbolet [10]. It was also very well documented
that all investigated anions could specically act as hydroxyl rad-
ical scavergers [23]. Therefore, a signicant competitive reaction
mechanism due to the presence of anions prevailed during pho-
tocatalytic degradation of discrete humic fractions. In accordance
withthe previous ndings fromliterature, the presence of common
anions were expected to inuence the photocatalytic degradation
of humic acids either due to direct effects via mechanistic path-
way or indirect effects via alterations in the structure [10]. The
induced effects could be observed as inhibiting or enhancing fac-
tors. However, considering the diverse nature of the molecular size
fractions, no correlation could be attributed to the photocatalytic
removal rate of humic acids in terms of the type of anions or their
concentration.
4. Conclusions
In the presence of anions, the percent removal of humic
acid decreased. Irrespective of the molecular size fractions of
humic acid, the presence of anions signicantly enhanced initial
adsorption on TiO
2
. No signicant difference was attained in the
photocatalytic removal of humic acid in the presence of lowanion
concentration, however, increasing anion concentration had an
inhibitory effect on the removal of humic acid in terms of DOC, and
UV
254
. The changes evaluated by uorescence indices were more
prominent. The results imply that some easily measured UVvis
parameters correspond well with the information provided by u-
orescence spectroscopy. Despite the complex nature of humic acids
in the presence of a solution matrix, the spectroscopic parameters
could be visualized as indicators for the photocatalytic removal of
humic acid and its molecular size fractions.
Acknowledgement
The nancial support provided by Research Fund of Bogazici
University through project no: 5196 and 6729 is gratefully
acknowledged.
References
[1] L. Li, Z. Zhao, W. Huang, P. Peng, G. Sheng, J. Fu, Organic Geochemistry 35 (2004)
10251037.
[2] A. Matilainen, E.T. Gjessing, T. Lahtinen, L. Hed, A. Bhatnagar, M. Sillanp,
Chemosphere 83 (2011) 14311442.
[3] B. Sulzberger, E. Durisch-Kaiser, Aquatic Sciences 71 (2009) 104126.
[4] O. Francioso, S. Sanchez-Cortes, D. Casarini, J.V. Garcia-Ramos, C. Ciavatta, C.C.
Gessa, Journal of Molecular Structure 609 (2002) 137147.
[5] C.S. Uyguner, M. Bekbolet, Desalination 176 (2005) 4755.
[6] M. Bekbolet, Journal of Environmental Science and Health 31 (1996) 845
858.
[7] B.R. Eggins, F.L. Palmer, J.A. Byrne, Water Research 31 (1997) 12231226.
[8] C.S. Uyguner-Demirel, M. Bekbolet, Chemosphere 84 (2011) 10091031.
[9] M. Bekbolet, Z. Boyacioglu, B. Ozkaraova, Water Science and Technology 38
(1998) 155162.
[10] C.S. Uyguner, M. Bekbolet, Journal of Advanced Oxidation Technologies 12
(2009) 8792.
[11] M. Bekbolet, A.S . Suphandag, C.S. Uyguner, Journal of Photochemistry and Pho-
tobiology Part A 148 (2002) 121128.
[12] C.S. Uyguner, M. Bekbolet, Catalysis Today 101 (2005) 267274.
[13] C.S. Uyguner, M. Bekbolet, Desalination 176 (2005) 167176.
[14] R. Dalrymple, A.K. Carfagno, C. Sharpless, Environmental Science and Technol-
ogy 44 (2010) 58245829.
[15] J. Peuravuori, K. Pihlaja, Analytica Chimica Acta 337 (1997) 133149.
[16] S. Valencia, J. Marn, J. Velsquez, G. Restrepo, F.H. Frimmel, Water Research 46
(2012) 11981206.
[17] N. Senesi, Analytica Chimica Acta 232 (1990) 77106.
[18] P. Coble, Marine Chemistry 41 (1996) 325346.
[19] W.H. van Riemsdijk, L.K. Koopal, D.G. Kinniburg, M.F. Benedetti, L. Weng, Envi-
ronmental Science and Technology 40 (2006) 74737480.
[20] J. Wiszniowski, D. Robert, J. Surmacz-Gorska, K. Miksch, J.V. Weber, Int. J. Inter-
national Journal of Photoenergy 5 (2003) 6974.
[21] R. Al-Rasheed, D.J. Cardin, Chemosphere 51 (2003) 925933.
[22] M. Abdullah, G.K.C. Low, R.W. Matthews, Journal of Physical Chemistry 94
(1990) 68206825.
[23] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Journal of Physical and
Chemical Reference Data 17 (1988) 513886.