Você está na página 1de 11

Construction and Building Materials 164 (2018) 297–307

Contents lists available at ScienceDirect

Construction and Building Materials


journal homepage: www.elsevier.com/locate/conbuildmat

Comparing the pozzolanic activity properties of obsidian to those


of fly ash and blast furnace slag
_
Ilker Ustabasß a,⇑, Ayberk Kaya b
a
Recep Tayyip Erdogan University, Faculty of Engineering, Department of Civil Engineering, 53100 Rize, Turkey
b
Recep Tayyip Erdogan University, Faculty of Engineering, Department of Geological Engineering, 53100 Rize, Turkey

h i g h l i g h t s

 The Ikizdere obsidians were investigated as a pozzolana in cement.


 The obsidians were compared to fly ash and blast furnace slag.
 The chemical and physico-mechanical properties of the obsidian, fly ash and blast furnace slag added cement were determined.
 It was concluded that the Ikizdere obsidians can be standardized as a pozzolana.

a r t i c l e i n f o a b s t r a c t

Article history: The use of pozzolan in cement provides economic advantages and improves the physico-mechanical
Received 7 July 2017 properties of cement. Fly ash and blast furnace slag are widely used in cement as pozzolanic materials.
Received in revised form 20 November 2017 In this study, obsidian known as volcanic glass which crops out in the Ikizdere (Rize) region of NE
Accepted 27 December 2017
Turkey was investigated as a pozzolana in cement. Mainly the pozzolanic activity, chemical and
physico-mechanical properties of the obsidian cement were compared to the properties of the fly ash
and blast furnace slag cement. According to laboratory test results, obsidian was seen to provide more
Keywords:
positive effects compared to the properties of fly ash cement. It was concluded that obsidian was equiv-
Obsidian
Fly ash
alent to blast furnace slag as a pozzolan. As a result, the obsidian located in the Ikizdere region could be
Blast furnace slag used as a pozzolana in cement.
Blended cement Ó 2017 Elsevier Ltd. All rights reserved.
Pozzolanic activity

1. Introduction slags and silica fume. Blending portland cement with pozzolanic
admixtures is an effective way to improve workability, strength
The amount of CO2 released during the production of clinker, and durability (lower permeability, resistance to chlorides and sul-
constitute 5% and 8% of the amount of CO2 emitted into the atmo- phates, mitigation of alkali silica reaction) [10–12]. Pozzolans are
sphere by human [1–3]. Using supplementary cementitious mate- also among the most common substances used to reduce the
rials (SCM) in blended cement production reduces the release of hydration temperature of cement [13–15]. An additional benefit
CO2 into the atmosphere [3]. In addition, substances called poz- of the above approach is that many pozzolanic materials used for
zolan affect many aspects of cement such as the compressive blending today would be otherwise stockpiled or disposed in land-
strength, permeability and chemical durability [4]. Cement sector fills, presenting environmental hazards such as dust contamination
gives close attention. Several researches are done on pozzolanic or leaching of heavy metals in groundwater. As a result, this reuse
matters in nature [5–9]. and recycling approach in blended cement production contribute
Pozzolans do not have a binding property on their own but they to the solution of major environmental problems [16].
gain this property when they are used with lime. The most com- Blast furnace slag is a residual product obtaining pig iron. If
mon SCM is natural pozzolans, generally derived from volcanic ori- cooled rapidly, it will have an amorphous structure and when
gins and artificial pozzolans, which are mainly industrial by grinded it will show excellent binding properties with the portland
products/wastes, such as fly ash, different type of metallurgical cement [17].
Fly ash (FA), which has fine grain property, is another substance
⇑ Corresponding author. with good pozzolanic features [18]. Fly ash and blast furnace slag
_ Ustabasß).
E-mail address: ilker.ustabas@erdogan.edu.tr (I. are among commonly used pozzolans used in the cement. FA is a

https://doi.org/10.1016/j.conbuildmat.2017.12.185
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
298 _ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I.

waste material generated by the coal combustion in thermal power analysis of cement showed more Alite (C3S) than other phases. The XRD pattern is in a
crystallized form which is suitable for a typical portland cement. In the impurity
plants. Fly ash can be used in cement directly or after passing
phase, traces of the crystals of the alkalis, sulphates and other compounds mentioned
through different processes such as activation as well. Binders in Table 1 are observed. It is understood that the XRD pattern of Fig. 2b is a material of
obtained by alkali activation of FA (geopolymers) in general are typical amorphous structure. Besides, silica, alumina and sodium oxide are the most
known of having good strength, good durability in aggressive envi- prominent minerals buried in the obsidian amorphous matrix. In Fig. 2c, the quartz
ronments, low shrinkage and good thermal resistivity [6,19–21]. peak in the fly ash is the most prominent main phase and the impurity phases are
mullite, magnetite, hematite and magnesium ferrite phases. In Fig. 2d, the XRD anal-
However, the limiting factor for wider use of FA for geopolymer
ysis of blast furnace slag shows that the most prominent peaks are silica, magnesium
synthesis is low reactivity and consequent low reaction rate and oxide and aluminium oxide. The XRD analyses of oxide and compounds were found
low strength gain when cured at room temperature [6,22]. suitable for the chemical composition given in Table 1. Through XRD, obsidian was
Obsidian is a volcanic rock which is mostly black, to a lesser found to be a substance with an amorphous structure that could be used as puzzolan
in cement same as fly ash and blast furnace slag.
extent in gray, brown, red and green in hue, glassy bright, fracture
The illustrative detailed information about the materials is given in the follow-
surfaces in the form of shells. Obsidian, which is also called vol- ing chapters.
canic glass because of its amorphous structure, contains very small
amounts of crystalline phases and there is very little amount of 2.2. Fineness of the materials
water (less than 1%) in these crystalline phases. Obsidian formation
from molten magma is controlled by chemical composition of the In Table 2, Sieve fineness and Blaine fineness values of binders used in the study
lava and cooling rate [23]. are shown. According to TS 25 [24] standard, Blaine fineness of natural pozzolan has
to be at least 4000 cm2/g ±%25, at 200 mm sieve remaining by mass % 0.6 at most
There is a lot of information in the literature on the effects of fly and at 90 mm sieve remaining % 8 at most. Obsidians used in the study were pro-
ash and blast furnace slag on properties such as compressive duced in three different fineness considering the TS 25 [24] standard fineness val-
strength, pozzolanic properties, initial-final setting time, and ues by grinding in ball mill for 45, 60 and 75 min. In Table 2, Obsidian45,
hydration heat of cement. However, there is not enough literature Obsidian60 and Obsidian75 indicate obsidians with different fineness which were
named according to their milling time.
information on using obsidian as pozzolana in cement. In this
Size distribution of Obsidian45, Obsidian60, Obsidian75, cement, fly ash and
study, obsidian was compared to fly ash and blast furnace slag blast furnace slag used in this study, which are obtained by laser diffraction particle
on mechanical, chemical and mainly pozzolanic activity properties. size distribution, are shown in Fig. 3.
The feasibility of the obsidians as pozzolana in cement was inves-
tigated with regard to the positive aspects. 2.3. Thermogravimetric analysis

Thermogravimetric analyses were applied to cement, obsidian, fly ash and blast
2. Material and method furnace slag. All the well-defined stages were selected for a study of the decompo-
sition kinetics of the complexes. The kinetics parameters were calculated using the
2.1. Materials used in the study equation of Coats-Redfern [25] given in Eq. (1) to elucidate the kinetic parameters
for each stage [26,27].
The obsidians used in this study were collected from the Ikizdere (Rize) region,    
blast furnace slag was obtained from Karabuk Iron and Steel Works, and fly ash was U AR 2RT Ea
log ¼ log 1  ð1Þ
provided from Tuncbilek Thermal Plant located in Turkey. The sampling locations of T2 /E Ea 2:303RT
the materials are shown in Fig. 1.
where U = [(Wf)/(Wf  Wt)], Wf is the mass loss upon the completion of the reac-
Pozzolan cement was produced by adding obsidian, fly ash, and blast furnace
tion, and Wt is the mass loss up to temperature T. A plot of U versus 1/T gives a
slag to CEM I 42,5 R class portland cement. Furthermore, the obsidians used in
straight line with a slope equivalent to Ea/R.
the study were divided into two groups as black obsidian and red obsidian consid-
ering their color. Table 1 shows the chemical composition of the materials based on
the X-ray fluorescence (XRF) analyses used in this study. The chemical analyses 2.4. Pozzolanic activity and determination of pozzolanic activity index
were performed in ACME laboratories, Bureau Veritas Commodities, Canada.
XRD measurements were taken to determine the structural properties and crystal Hydraulic feature determined through mortar prepared by mixing grinded nat-
structures of the materials studied in this study. The XRD analyses of cement, obsid- ural pozzolan, water, slaked lime [Ca(OH)2] and sand in terms of compressive
ian, fly ash and blast furnace slag used in this study are shown in Fig. 2. In Fig. 2a, XRD strength is called pozzolanic activity [24]. Pozzolanic activity of pozzolans used in

Fig. 1. Location map of the study areas.


_ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I. 299

Table 1
Chemical composition values of the materials.

Cement Obsidian (red) Obsidian (black) Fly ash Blast furnace slag
CaO 64.60 0.88 0.87 4.5 35.30
SiO2 19.30 73.80 74.10 52.70 38.60
Al2O3 5.17 13.80 13.80 18.00 11.40
Fe2O3 3.55 1.57 1.43 12.60 0.58
MgO 1.05 0.16 0.14 4.79 7.92
K2O 1.16 4.86 4.88 1.87 1.45
Na2O 0.46 4.11 4.08 0.74 0.29
P2O5 0.077 0.017 0.015 0.126 0.003
SO3 2.860 0.006 0.005 1.64 3.43
TiO2 0.14 0.15 0.14 0.76 0.44
Sr 0.009 0.015 0.014 0.026 0.058
Cr2O3 – 0.09 0.07 0.10 –
MnO 0.23 0.06 0.06 0.15 1.40
LOI 2.81 0.21 0.03 1.47 1.08
Reactive SO2 content 40.18 40.13 49.48 40.74
Total 101.45 99.40 99.55 99.56 100.77

LOI: Loss on ignition.

Fig. 2. XRD analyses for (a) cement, (b) obsidian, (c) fly ash, and (d) blast furnace slag.

the study was determined according to TS 25 [24] standard. Mortar samples pre- samples prepared with only reference cement of the same age is called pozzolanic
pared with the amount of materials given in Table 3 were measured in compressive activity index. In this study, TS EN 450-1 [29] and TS EN 15167-1 [30] standards
strength values according to TS EN 196-1 [28] standard after kept at 55 ± 2 °C for were used to compare pozzolanic activity indexes of obsidian to fly ash and blast
6 days. furnace slag. While % 75 cement and % 25 pozzolan mixture by mass are used for
According to TS EN 196-1 [28] standard, the ratio (in percentage) of compres- calculating pozzolanic activity index according to TS EN 450-1 [29] standard, the
sive strength obtained from samples prepared by mixing any proportion of refer- % 50 cement and % 50 pozzolan mixture are used as binder for pozzolanic activity
ence cement with natural pozzolan to compressive strength obtained from index according to TS EN 15167-1 [30] standard.
300 _ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I.

Table 2
Fineness values of the materials.

Material Remaining on the sieve (%) Blaine fineness (cm2/g)


200 lm 90 lm 45 lm
Obsidian45 0 6 42 4746
Obsidian60 0 3 33 5497
Obsidian75 0 2 27 6122
Cement 0 0 20 4424
Fly ash 0 0 9 4254
Blast furnace slag 0 0 10 4666

temperature was determined over two samples, one for the calorimeter and the
other for the measurement. The temperature increase (ht) of the sample was deter-
mined from the difference between the temperature (Ts) of the sample and the
temperature (Tr) of the reference sample in calorimeter at each temperature read-
ing [31]. The hydration temperature (Q) was calculated from the following
equation.

c 1 Xi¼1
Q¼ ht þ aixhixDti ð2Þ
mc mc i¼1

where mc is mass of cement in sample (g), t is hydration time (h), c is thermal capac-
ity of calorimetry (JK1) and a is thermal loss coefficient of calorimetry (Jh1K1).

2.6. Geological properties of the Ikizdere region that obsidian was provided

The region shown in Fig. 4 is in the north of the Eastern Black Sea Mountains,
covers 10 km2 north-northeast of Ikizdere region (Rize, Turkey), on the location
there are plenty of obsidian deposits to create economic value. This obsidian rocks
are not in use for the time being. Investigated obsidians of Ikizdere region were
found interbedded with rhyolitic volcanic rocks in layered structure in base levels,
in the upper levels partly lens-shaped in vulcanite and partly cover up the rhyolite.
Covering volcanic and granitic rocks with rhyolites in the surrounding, obsidian is
the youngest volcanic product in the region. Obsidian exists in layered structure
in base levels, brecciated in the middle, lens-shaped and in massive structure in
the upper levels. Obsidians in the region are black, red, black banded red and red
banded black in color. In the last phase of volcanism, rhyolites and obsidians were
formed during Upper Pliocene Pleistocene period [32]. The total thickness of the
obsidian rhyolite intercalation was measured as 6 m by Heide et al. [33]. Under
these circumstances, the amount of obsidian in this region is economically viable.
The visible reserve approximately is (10 km2 * 6 m * 2.35 t/m3) 14.1 * 106 tonnes.
The region has obsidian deposits in large and thick layers as shown in Fig. 5. In
Fig. 5a and b, inhabitable areas for plants and animals have occurred where the
obsidian accumulates. Due to the fact that obsidian is a volcanic glass, its sharp
structure keeps animals out of these parts of the area. Removal of the material from
this area will enable plants and animals to live in these parts of the land.

3. Discussion
Fig. 3. Size distribution curves of Obsidian45, Obsidian60, Obsidian75, cement, fly
ash, and blast furnace slag. The chemical compositions of the black and red obsidians are
presented in Table 1. However, their chemical composition is con-
siderably similar. In this context, although obsidian varies in color,
2.5. Measurement of hydration heat in case of using as pozzolana, it will not cause a significant differ-
ence on properties of pozzolana of cement. That point has also
Hydration heat of Cement I 42.5 R and mineral added cements obtained by mix-
been encountered in chemical and mechanical experiments.
ing this cement with obsidian, fly ash and blast furnace slag used in the study were
measured considering the TS EN 196-9 [31] standard. According to TS EN 196-9 [31] According to TS EN 197-1 [34] standard, reactive silica content of
standard, samples were prepared by mixing 360 g of cement, 180 g of water and pozzolans used in the cement has to be above 25 %. Reactive silica
CEN standard sand. The total amount of material was 1575 g and the hydration content according to TS EN 196-2 [34] standard was determined %

Table 3
The amount of materials used for the pozzolanic activity experiment.

Material Amount to be used (g)


Samples with added obsidian Blast furnace slag samples Samples with added fly ash TS 25 standard
Slaked lime 150 150 150 150
[Ca(OH)2]
Natural pozzolan 308.38 373.6 222.7 T = 2  150 (natural pozzolan density/
slaked lime density)
Standard sand 1350 1350 1350 1350
(TS EN 196-1)
Water 229.19 261.6 268 + addition 82 = 350 0.5 (150 + T)
_ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I. 301

Fig. 4. Geological map of the Ikizdere (Rize) region [32].

40.43 in obsidian, 40.74 % in blast furnace slag and 49.48 % in fly


ash (Table 1). As shown in Table 1, reactive silica content of obsid-
ian (40.13 %) is high enough to be used as pozzolan. The reactive
silica content of obsidians is higher than the minimum value of
cement standard. The Ikizdere region obsidians, same as fly ash
and blast furnace slag used in this study, are chemically suitable
to be used as a pozzolana in cement.
In Fig. 6, the pozzolanic activity values of obsidian, fly ash, and
blast furnace slag are indicated. In order to show its pozzolanic
property, the compressive strength of the mortar produced from
lime and pozzolan mixture should be above 4 N/mm2 [24]. In
Fig. 6, the pozzolanic activity values of all three pozzolans are
above 4 N/mm2. This indicates that obsidian has pozzolanic activ-
ity similar to those of fly ash and blast furnace slag. In Fig. 6, the
greater the fineness of the obsidian is, the higher the pozzolanic
activity value is. In this study, among the pozzolans used, blast fur-
nace slag was determined to have the highest pozzolanic activity
value, followed by Obsidian75, Obsidian60, Obsidian45, and fly
ash, respectively. If the fineness of Obsidian75 shown in Fig. 2 is

Fig. 5. (a) Slope by the road containing obsidian and (b) accumulated obsidian in Fig. 6. Compressive strength of mortars including obsidian, fly ash, and blast
the field. furnace slag.
302 _ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I.

equalized with the fineness of blast furnace slag used in this study, index of the fly ash must be 28 days 75% and 90 days 85%. All poz-
the pozzolanic activity value of obsidian will increase further, zolans in Fig. 8 are suitable for the standard of TS EN 450-1 [29].
which can be seen from the increase in the pozzolanic activity val- The 90-day pozzolanic activity index of Obsidian75 is higher than
ues of obsidians with three different levels of fineness, as shown in that of fly ash but lower than that of blast furnace slag. The mortar
Fig. 6. In this study six day compressive strength of fly ash pro- strength obtained from Obsidian75 produced with a mixture of
duced with lime cement mortar was determined 7.2 N/mm2 75% cement and 25% obsidian is very close to the strength of mor-
(Fig. 6). Uzal et al. [4] in a similar way found 6.2 N/mm2 of com- tars produced with 100% cement. As a consequence, obsidian from
pressive strength in seven days fly ash. The content of reactive sil- the Ikizdere region could be used as pozzolan in cement. Since the
ica of fly ash used in the study by Uzal et al. [4] was 41.15% and the compressive strength of mortars prepared with Obsidian 90-day is
Blaine fineness was 3880 cm2/g. Compressive strength value of fly very close to the compressive strength of mortars made with port-
ash used in this study is higher compared to compressive strength land cement only, it can be concluded that obsidian is a good
value of fly ash pozzolana blend mortars used by Uzal et al. [4]. In pozzolan.
this study, reactive silica content of fly ash was 49.48% and 4254 Fig. 9 shows 7, 28 and 90 day compressive strength values of
cm2/g of Blaine fineness is the reason for having high value com- Cement I 42.5 R cement and 50% cement with obsidian and blast
pressive strength of 6 day lime pozzolana mortar samples. This furnace slag. 50% Obsidian addition to cement resulted in a
finding shows that higher reactive silica content and fineness for decrease in 90 day compressive strength values approximately
the same kind of pozzolana cause high compressive strength of 25% compared to unblended cement samples. In Fig. 10, the poz-
the mortar sample prepared with lime and pozzolana. Uzal et al. zolanic activity indices also indicate the speed of compressive
[4] did not find a direct relationship between the compressive strength of pozzolana added cementitious mortars compared to
strength and the amount of lime in the reaction. The fact that the normal cementitious mortars.
pozzolana used has a strong relationship between the amorphous The obsidian red, black or red-black banded does not cause any
silica type and the compressive strength and that the zeolite, change in compressive strength and pozzolanic activity index of
which is another pozzolana used in the study, has a crystal struc- cement used as pozzolana. In Fig. 10, the pozzolanic activity
ture causes the zeolite samples to have low compressive strength indexes of mortars prepared considering the TS EN 15167-1 [30]
[4]. Pozzolanic activity depends on the chemical structure, amor- standard with a mixture of 50% cement and 50% pozzolan are
phisity and fineness of the material [17]. In this study, the chemical shown. According to TS EN 15167-1 [30] standard, the pozzolanic
composition, reactive silica content and fineness of pozzolan were activity indexes for 7- and 28-day blast furnace slag must be 45%
found to be the most important factors affecting the pozzolanic and 70%, respectively. When cement is substituted by 50% obsid-
activity of the mortar prepared with lime and pozzolana. ian, an activity index of 71% is found in 90-day strength as seen
The compressive strengths in Fig. 7 show 7, 28 and 90 day com- compared to the pozzolanas in this study. It is seen that blast
pressive strengths obtained from the mortar samples prepared furnace slag has the highest rate of compressive strength gain.
according to Cement I 42,5 R cement and mineral added cements Schöler et al. [3] note that blast furnace slag mortars gain more
prepared by adding obsidian, fly ash and blast furnace slag to compressive strength compared to fly ash mortars is due to high
25% of the same cement according to TS EN 196-1 [28] standard. speed C-S-H formation in blast furnace slag because of its Al2O3
In Fig. 7, it was observed among these cements that the highest content. Compared to the pozzolanas in this study, the fact that
compressive strength was in unblended cement and high blast- high furnace slag has the lowest Al2O3 content is seen. CaO content
furnace cement, followed by obsidian – milled for 75 min – added of blast furnace slag is higher than other pozzolans. According to TS
cement; 25% fly ash added cement was less than Obsidian 75 and EN 197-1 [33] standard, the proportion of CaO/SiO2 in portland
same level as Obsidian 60. The pozzolanic activity indices in Fig. 8 cement must not be less than 2, proportion of (CaO + MgO)/SiO2
were obtained by proportioning the mineral-added cements in in blast furnace slag must not be less than 1. In this study, the
Fig. 7 to the compressive strength values obtained from unblended fact that blast furnace slag has a high rate of compressive
cement samples. strength is considered as blast furnace slag blended cement which
In Fig. 8, the pozzolanic activity indexes of mortar samples pre- has higher ratio of CaO/(SiO2 + Al2O3 + Fe2O3) than all other
pared considering the TS EN 450-1 [29] standard are indicated. pozzolanic cements. Marjanović et al. [6] indicate that as the
According to TS EN 450-1 [29] standard, the pozzolanic activity Ca/Si ratio increases, the compressive strength of the blast
furnace slag increases. Celik et al. [35] reported that compres-
sive strength values of concretes produced with pozzolana
blended cements according to the chemical analyses proportion

Fig. 7. The compressive strength values of CEM I 42.5 R cement with 25% obsidian,
fly ash, and blast furnace slag added cement. Fig. 8. Pozzolanic activity index according to TS EN 450-1.
_ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I. 303

Fig. 9. The compressive strength values of CEM I 42.5 R cement and 50% obsidian
and blast furnace slag added cement.
Fig. 11. Standard consistency of obsidian added to cement.

Fig. 10. Pozzolanic activity index according to TS EN 15167-1.

Ca/(Al + Si) = 0.4–0.6 and Na + Mg + K + Ti + Fe/(Al/Si) = 0.6–0.7 are Fig. 12. Standard consistency of fly ash added to cement.
higher compared to the compressive strength values of concretes
produced with portland cement. Robayo-Salazar et al. [36] indicate
that the use of natural pozzolana by 20% according to SiO2/Al2O3
ratio improves the compressive strength of the cement. In Table 4,
regression analysis results in Backward method between compres-
sive strength values of 2, 7 and 28 day mortar samples produced
with obsidian, fly ash, blast furnace slag and unblended portland
cement and chemical analysis values of used cements in Table 1,
are shown. In Table 4, regression analysis indicates that the SiO2/
Al2O3 ratio is only effective on the compressive strength of 2 day
cements. In Table 4, the fact that ratio of CaO/(SiO2 + Al2O3 +
Fe2O3) in cement is the most effective parameter on 2, 7 and 28
day compressive strength, is figured. The ratio of CaO/(SiO2 +
Al2O3 + Fe2O3) of the blended cement can be used in order to deter-
mine the ratio of pozzolana added to the cement. The amount of
pozzolana determined as 1 or higher/CaO/(SiO2 + Al2O3 + Fe2O3)
Fig. 13. Initial setting time of cements with added obsidian, fly ash, and blast
ratio will not decrease the compressive strength of the cement. furnace slag.

Table 4
Regression analysis of 2, 7 and 28 days compressive strength with chemical composition.

2 days* 7 days** 28 days***


Beta Std. Err. Sig. Beta Std. Err. Sig. Beta Std. Err. Sig.
(Constant) – 8.120 0.002 – 5.833 0.093 – 5.526 0.008
CaO/(SiO2 + Al2O3 + Fe2O3) 4.204 15.857 0.003 0.789 1.680 0.000 0.827 1.591 0.000
SiO2/Al2O3 0.708 2.007 0.001 0.080 2.358 0.656 0.221 2.234 0.247
CaO/SO3 5.284 11.132 0.001 0.436 0.558 0.032 0.331 0.529 0.093
*
Adj R2 = 0.815; F3.12 = 18.573. p < .001.
**
Adj R2 = 0.893; F3.12 = 34.261. p < .001.
***
Adj R2 = 0.894; F3.12 = 34.776. p < .001.
304 _ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I.

This will also ensure that the cement’s pozzolanic activity index standard consistency. Pozzolanas used in cement such as silica
value meets the standards. fume, fly ash and metakaolin increase the standard water consis-
The standard consistency values of obsidian added to cement tency of cement [20]. Nikolic et al. [37] indicate in their studies
are indicated in Fig. 11. In Fig. 11, as the rate of obsidian increases, that fly ash increases the water demand of cement but geopolymer
the amount of water required to achieve the standard consistency binding decreases the water demand of fly ash. Low standard con-
decreases. The fact that the addition of obsidian reduces the sistency of cement enables to use less water to produce concrete
amount of water required for the standard consistency is a benefi- in desired consistency which is a favorable property. As seen in
cial property for blended cement. The reduction in the amount of this study, the obsidian cement reduces the standard water
water needed to obtain the standard consistency of cement due consistency.
to the effect of obsidian will also cause a decrease in the amount In Fig. 13, the initial setting time of cements with added obsid-
of water required for concrete. ian, fly ash, and blast furnace slag is indicated. Fly ash used in the
In Fig. 12, the standard consistency values of cement with the study prolongs the initial setting time of the cement more than
addition of fly ash are shown. An increase in the fly ash addition obsidian and blast furnace slag do, as shown in Fig. 13. Increasing
increases the amount of water required to achieve the standard the amount of obsidian and blast furnace slag does not lead to a
consistency. As it can be seen in the scanning electron microscopy significant increase in the initial setting of cement. Since obsidian
(SEM) images (Fig. 16), the porous structure of fly ash is the reason does not prolong the initial setting, its use as pozzolan confers ben-
for the increase in the amount of water required to achieve the eficial properties on the cement.

Fig. 14. TGA curves for non-hydrated cement, hydrated cement, and 20% mineral added hydrated cements.
_ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I. 305

Fig. 15. Hydration temperatures of (a) unblended cement (b) 20% cement with obsidian75, (c) blast furnace slag, and (d) fly ash.

Fig. 16. SEM images of (a) cement, (b) obsidian, (c) fly ash, and (d) blast furnace slag (15,000).

In Fig. 14, thermal gravimetric analyses of non-hydrated that the mass losses occur and cause degradation in ettringite,
cement, hydrated cement and 20% mineral added hydrated portlandite and calcium carbonate from low temperature to high
cements are seen. The hydrated cement in Fig. 14 is prepared by temperature. In Fig. 14, the thermal gravimetric analysis curve of
adding 20% water into cement mass. In Fig. 14, hydrated cement non-hydrated cement is shown. The non-hydrated cement is
mass losses at 114.6 °C, 462.3 °C and 701.8 °C are seen. Schöler CEM I 42.5 R type portland cement. So it is composed of clinker
et al. [3] and Lemonis et al. [16] also found mass losses in hydrated and gypsum mixture. In the TGA curve of the cement, mass losses
cement at similar temperatures in their own studies. It is stated occur at temperatures of 133.8 °C, 448.6 °C and 685.1 °C is seen
306 _ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I.

(Fig. 14). When the clinker in cement is considered to be cooked at ash, lower than blast furnace slag. If grain size of obsidian is
about 1500 °C, the mass losses occurring in Fig. 14 must be due to reduced, it may cause a further increase in the pozzolanic charac-
the deterioration of the gypsum. Data obtained from the thermal teristic and pozzolanic activity indices shown in Fig. 6. In Fig. 16,
gravimetric analysis of gypsum of Engbrecht et al. [38] and the fly ash has the highest number and boundary of grain size; it can
thermal analysis data obtained from the non-hydrated cement in be said that the reason why the standard consistency value of
this study are in perfect correspondence. The chemical formula of cement increases as the fly ash ratio increases in Fig. 12 may be
gypsum is CaSO42H2O and in its thermal analysis turns into due to the water settling at these grain boundaries. Fig. 16b shows
CaSO42H2O by losing some of the water at 133 °C and finally turns that as the obsidian grain size is smaller than the cement in
into anhydrite (CaSO4) by losing other water up to 250 °C [34]. At Fig. 16a, and as the obsidian ratio in Fig. 11 increases, it may lead
448.6 °C, of other stranger minerals Ca(OH)2 and at 685 °C, CaCO3 to a decrease in water content required for the standard consis-
admixed as a stranger mineral to gypsum cause mass losses. When tency of the obsidian added cement. It is determined that the high
thermal gravimetric analysis curves of Fig. 14 are compared to amount of water required to obtain the standard consistency of fly
non-hydrated cement and hydrated cement the hydration of ash is caused by the surface structure of fly ash.
cement causes only small changes in mass loss temperatures. In
Fig. 14, mass loss occurs at 114.6 °C whereas in Fig. 14 mass loss 4. Results
occurs at 133.8 °C. The hydration of the cement and the rise of
the hydration heat reduce the initial mass loss value of the In this study, the outcomes obtained from laboratory experi-
hydrated cement. In the hydration of cement ettringite (C-A-S-H) ments can be summarized as follows;
(Afm) and Monosulfates (C-A-S-H) (Aft) occurs with reaction of tri-
calcium aluminate (C3A), water and gypsum. Schöler et al. [3] and 1) The mineralogical structure of obsidian from the Ikizdere
Lemonis et al. [16] reported that loss of mass at 110 °C occurred region (Rize, NE Turkey) makes it suitable to use as poz-
because of degradation of ettringite. Very similar data were found zolana in cement. The reactive silica content of obsidian
in this study. Mass loss in Fig. 14 is believed to be caused by the from the Ikizdere region is close to those of blast furnace slag
abandonment of chemical water in gypsum coming from Afm or and fly ash.
Aft. During the hydration, the increase in temperature in the gyp- 2) When the pozzolanic activity index value of obsidian from
sum causes the loss of mass in the TGA at lower temperature com- the Ikizdere region is examined, it is found to be higher than
pared to non-hydrated cement. In Fig. 14, the mass losses at 462 that of the fly ash used in the study, and over a long period it
and 701 °C were due to deterioration in Ca(OH)2 and CaCO3. In is found to be equal to that of blast furnace slag. Therefore,
the thermal analysis, it was understood that the materials coming obsidian could be used as a pozzolan in cement such as fly
with gypsum were effective in mass losses in hydrated cement as ash and blast furnace slag.
well. In Fig. 14, it is observed that the thermal gravimetric analysis 3) The ratio of CaO/(SiO2 + Al2O3 + Fe2O3) on the compressive
curves of obsidian, fly ash and blast furnace slag cement are similar strength of the cement is an effective parameter which can
to hydrated cement. Since obsidian, fly ash and blast furnace slag be used to determine the amount of pozzolana to be added
are materials which are formed at high temperature effect, their into the cement.
addition to the cement did not cause any difference in the thermal 4) It was found that fly ash increases the standard consistency
gravimetric analyses of the blended cements. of cement and blast furnace slag maintains the standard
Fig. 15 shows 160 h hydration temperatures of unblended consistency of cement. The fact that obsidian reduces the
cement and 20% obsidian, fly ash and blast furnace slag added amount of water required for cement of the standard consis-
cements. The addition of 20% mineral addition did not make a sig- tency more than fly ash and blast furnace slag do but does
nificant difference in hydration temperatures of 160 h compared to not significantly prolong the initial and final setting times
unblended cement. Han et al. [13] calculated hydration tempera- was identified as a superior property of obsidian.
tures of approximately 300 J/g for 80 h of hydration heat in portland 5) No adverse effects of obsidian, fly ash, or blast furnace slag
cement, 285 J/g in 30% blast furnace slag cement, and 180 J/g in 35% on cement were encountered in the microstructural or ther-
fly ash cement. Baran et al. [39] calculated values between 374 and mogravimetric analyses. Addition of 20% obsidian, fly ash
404 J/g for the CEM I 42.5 R cement. The hydration temperatures and blast furnace slag does not cause a significant change
calculated in this study were similar to Baran et al. [39]. The reduc- in hydration temperature of the cement up to 160 h.
tion in hydration heat of cement with 30% blast furnace slag com- 6) It was found that the obsidian from Ikizdere is a material
pared to the hydration heat with 5% more fly ash is more in that can be standardized as a pozzolana.
proportion to the contribution rate. Merzouki et al. [14] noted that
the slag reduces the hydration temperature after 140 h measures. In
this study, 20% fly ash, blast furnace slag and obsidian addition did Acknowledgements
not cause a significant change in hydration heat. This suggests that
the pozzolan additive ratio and the duration of hydration are Authors would like to express their sincerest gratitude to the
important for hydration heat. As the pozzolan ratio in cement editor and reviewers. Thanks are due to the Academic Writing
increases, the delay of hydration may cause the hydration temper- and Translation Office of RTE University for improving the lan-
ature to fall. In this study, the hydration heat of 20% obsidian added guage of the manuscript. The authors gratefully acknowledge the
cement exhibited a similar behavior compared to fly ash and blast Scientific and Technological Research Council of Turkey (TUBITAK)
furnace slag. – Turkey for funding this work through research project no:
Electron microscope measurements were performed to exam- 214M023.
ine the surface morphology and microstructures of the samples.
In Fig. 16, 15,000 magnified images obtained by SEM of cement References
(16a), obsidian (16b), fly ash (16c) and blast furnace slag (16d)
are shown, respectively. In Fig. 16, it is clear that the fly ash has [1] J.S. Damtoft, J. Lukasik, D. Herfort, D. Sorrentino, E.M. Gartner, Sustainable
the smallest grain size. It is followed by blast furnace slag, cement development and climate change initiatives, Cem. Concr. Res. 38 (2) (2008)
115–127.
and obsidian samples, respectively. As shown in Fig. 7, obsidian has [2] M. Schneider, M. Romer, M. Tschudin, H. Bolio, Sustainable cement production
the largest grain size and has higher compressive strength than fly at present and future, Cem. Concr. Res. 41 (7) (2011) 642–650.
_ Ustabasß, A. Kaya / Construction and Building Materials 164 (2018) 297–307
I. 307

[3] A. Schöler, B. Lothenbach, F. Winnefeld, M. Zajac, Hydration of quaternary mechanical properties and structure of the fly ash based geopolymers,
portland cement blends containing blast-furnace slag, siliceous fly ash and Constr. Build. Mater. 41 (2013) 570–579.
limestone powder, Cem. Concr. Compos. 55 (2015) 374–382. [22] F. Winnefeld, A. Leemann, M. Lucuk, P. Svoboda, M. Neuroth, Assessment of
[4] B. Uzal, L. Turanlı, H. Yücel, M.C. Göncüoğlu, A. Çulfaz, Pozzolanic activity of phase formation in alkali-activated low and high calcium fly ashes in building
clinoptilolite: a comparative study with silica fume, fly ash and a non-zeolitic materials, Constr. Build. Mater. 24 (2010) 1086–1093.
natural pozzolan, Cem. Concr. Res. 40 (2010) 398–404. [23] T. Ercan, Z. Yeğingil, G. Bigazzi, Obsidian definition and characteristics,
[5] D. Caputo, B. Liguori, C. Colella, Some advances in understanding the distribution and geochemical characteristics of those of the central
pozzolanic activity of zeolites: the effect of zeolite structure, Cem. Concr. Anatolian obsidian in Anatolia, J. Geomorphol. 17 (1989) 71–83.
Compos. 30 (2008) 455–462. [24] TS 25. Natural Pozzolan (Trass) for Use in Cement and Concrete – Definitions,
[6] N. Marjanović, M. Komljenović, Z. Baščarević, V. Nikolić, R. Petrović, Physical– Requirements and Conformity Criteria, Turkish Standard, December 2008.
mechanical and microstructural properties of alkali-activated fly ash–blast [25] A.W. Coats, J.P. Redfern, Kinetic parameters from thermogravimetric data,
furnace slag blends, Ceram. Int. 41 (2015) 1421–1435. Nature 201 (1964) 68.
[7] M.C.G. Juenger, F.E. Winnefeld, J.L. Provis, J.H. Ideker, Advances in alternative [26] A.H. Kianfar, V. Sobhani, M. Dostani, M. Shamsipur, M. Roushani, Synthesis,
cementitious binders, Cem. Concr. Res. 41 (2011) 1232–1243. spectroscopy, electrochemistry and thermal study of vanadyl unsymmetrical
[8] H. Tchakoute Kouamo, J.A. Mbey, A. Elimbi, B.B. Kenne Diffo, D. Njopnouo, Schiff base complexes, Inorg. Chim. Acta 365 (2011) 108–112.
Synthesis of volcanic ash-based geopolymer mortar by fusion method: effects [27] P.A. Lorenzo Luis, P. Martín-Zarza, A. Sanchez, C. Ruiz-Pérez, M. Hernández-
of adding metakaolin to fused volcanic ash, Ceram. Int. 39 (2013) 1613–1621. Molina, X. Solans, P. Gili, Anderson-Evans type molybdotellurates with
[9] E. Ul Haq, S.K. Padmanabhan, A. Licciulli, Synthesis and characteristics of fly diazolium cations: preparation, structure and properties of (H2imz)6
ash and bottom ash based geopolymers – a comparative study, Ceram. Int. 40 [TeMo6O24]4H2O and (H2pyz)6 [TeMo6O24]Te(OH)6, Inorg. Chim. Acta 277
(2014) 2965–2971. (1998) 139.
[10] D. Kastis, G. Kakali, S. Tsivilis, M.G. Stamatakis, Properties and hydration of [28] TS EN 196-1, Methods of Testing Cement – Part 1: Determination of Strength,
blended cements with calcareous diatomite, Cem. Concr. Res. 36 (10) (2006) Turkish Standard, March 2009.
1821–1826. [29] TS EN 450-1, Fly Ash for Concrete – Part 1: Definition, Specifications and
[11] S.J. Barnett, M.N. Soutsos, S.G. Millard, J.H. Bungey, Strength development of Conformity Criteria, Turkish Standard, January 2008.
mortars containing ground granulated blast-furnace slag: effect of curing [30] TS EN 15167-1, Ground Granulated Blast Furnace Slag for Concrete, Mortar and
temperature and determination of apparent activation energies, Cem. Concr. Grout – Part 1: Definition, Specifications and Conformity Criteria, Turkish
Res. 36 (3) (2006) 434–440. Standard, December 2006.
[12] M. Bohac, M. Palou, R. Novotny, J. Masilko, D. Vsiansky, T. Stanek, Investigation [31] TS EN 196-9, Methods of Testing Cement – Part 9: Heat of Hydration – Semi-
on early hydration of ternary portland cement-blast-furnace slag-metakaolin Adiabatic Method, Turkish Standard, July 2011.
blends, Constr. Build. Mater. 64 (2014) 333–341. [32] A. Hanedan, Büyükyayla Obsidian Genes (Ikizdere-Rize) Petrography,
[13] F. Han, X. He, Z. Zhang, J. Liu, Hydration heat of slag or fly ash in the composite Mineralogy, Mineral Chemistry and Geochemical Characteristics (PhD
binder at different temperatures, Thermochim. Acta 655 (2017) 202–210. Thesis), KTÜ, Institute of Science, 2008, p. 241.
[14] T. Merzouki, M. Bouasker, N.E.H. Khalifa, P. Mounanga, Contribution to the [33] K. Heide, M.B. Sadıklar, K. Gerth, G. Völksch, E. Hartman, Obsidian from
modeling of hydration and chemical shrinkage of slag-blended cement at early Büyüksulata and Sırıklı tepe, Eastpontides, Turkey: a glass-chemical study,
age, Constr. Build. Mater. 44 (2013) 368–380. Chemie der Erde 56 (4) (1996) 285–312.
[15] S.T. Erdogan, A.Ü. Saglık, Early-age activation of cement pastes and mortars [34] TS EN 197-1, Cement – Part 1: Composition, Specification and Conformity
containing ground perlite as a pozzolan, Cem. Concr. Compos. 38 (2013) 29–39. Criteria for Common Cements, Turkish Standard, April 2012.
[16] N. Lemonis, P.E. Tsakiridis, N.S. Katsiotis, S. Antiohos, D. Papageorgiou, M.S. [35] K. Celik, M.D. Jackson, M. Mancio, C. Mera, A.H. Emwas, P.K. Mehta, P.J.M.
Katsiotis, M. Beazi-Katsioti, Hydration study of ternary blended cements Monteiro, High-volume natural volcanic pozzolan and limestone powder as
containing ferronickel slag and natural pozzolan, Constr. Build. Mater. 81 partial replacements for portland cement in self-compacting and sustainable
(2015) 130–139. concrete, Cem. Concr. Compos. 45 (2014) 136–147.
[17] E. Özbay, M. Erdemir, D.H. Ibrahim, Utilization and efficiency of ground [36] R.A. Robayo-Salazar, R.M. de Gutiérrez, F. Puertas, Effect of metakaolin on
granulated blast furnace slag on concrete properties – a review, Constr. Build. natural volcanic pozzolan-based geopolymer cement, Appl. Clay Sci. 132–133
Mater. 105 (2016) 423–434. (2016) 491–497.
[18] T. Yamamoto, T. Kanazu, M. Nambu, T. Tanosaki, Pozzolanic reactivity of fly [37] I. Niklioć, S. Marković, I. Janković-Častvan, V.V. Radmilović, L. Karanović, B.
ash – API method and K-value, Fuel 85 (2006) 2345–2351. Babić, V.R. Radmilović, Modification of mechanical and thermal properties of
[19] D. Panias, I.P. Giannopoulou, T. Perraki, Effect of synthesis parameters on the fly ash-based geopolymer by the incorporation of steel slag, Mater. Let. 176
mechanical properties of fly ash-based geopolymers, Colloids Surf. A 301 (2016) 301–305.
(2007) 246–254. [38] D.C. Engbrecht, D.A. Hirschfeld, Thermal analysis of calcium sulfate dihydrate
[20] M.R. Wang, D.C. Jia, P.G. He, Y. Zhou, Microstructural and mechanical sources used to manufacture gypsum wallboard, Thermochim. Acta 639
characterization of fly ash cenosphere/metakaolin-based geopolymeric (2016) 173–185.
composites, Ceram. Int. 37 (2011) 1661–1666. [39] T. Baran, P. Pichniarczyk, Correlation factor between heat of hydration and
[21] Z. Baščarević, M. Komljenović, Z. Miladinović, V. Nikolić, H. Marja-nović, Z. compressive strength of common cement, Constr. Build. Mater. 150 (2017)
Žujović, R. Petrović, Effects of the concentrated NH4NO3 solution on 321–332.

Você também pode gostar