www.elsevier.com/locate/wasman

Waste Management 27 (2007) 161–167

Strength properties of concrete incorporating coal bottomash and granulated blast furnace slag

Omer Ozkan a,*, Isa Yuksel b, Ozgur Muratoglu b

a Alaplı Vocational School, Zonguldak Karaelmas University, 67850 Alaplı, Zonguldak, Turkeyb Department of Civil Engineering, Engineering Faculty, Zonguldak Karaelmas University, Zonguldak, Turkey

Accepted 10 January 2006Available online 3 April 2006

Abstract

Coal bottom ash (CBA) and fly ash (FA) are by-products of thermal power plants. Granulated blast-furnace slag (GBFS) is devel-oped during iron production in iron and steel plants. This research was conducted to evaluate the compressive strength property andsome durability characteristics of concrete incorporating FA, CBA, and GBFS. FA is used as an effective partial cement replacement;CBA and GBFS are used as partial replacement for fine aggregate without grinding. Water absorption capacity, unit weight and com-pressive strengths in 7, 28, and 90-day ages were assessed experimentally. For these experiments, concrete specimens were produced in thelaboratory in appropriate shapes. The samples are divided into two main categories: M1, which incorporated CBA and GBFS; and M2,which incorporated FA, CBA, and GBFS. Remarkable decreases are observed in compressive strength and water absorption capacity ofthe concrete; bulk density of the concrete is also decreased. It can be concluded that if the content of CBA and GBFS is limited to areasonable amount, the small decreases in strength can be accepted for low strength concrete works.� 2006 Published by Elsevier Ltd.

1. Introduction

CBA is a mineral by-product obtained from the com-bustion of coal used for power-generation purposes. Thephysical and chemical properties of CBA may vary depend-ing on the type, source and fineness of the parent fuel, aswell as the operating conditions of the power plant. It iscollected from the bottom of the combustion chamber ina solid granular form. It is necessary to recycle industrialwastes due to a shortage of natural resources, economicproblems and environmental regulations (Churcill andAmirkhanian, 1999). In Turkey, 100,000 tons/yr of CBAis produced at the Catalagzı Thermal Power Plant(CATES) alone.

Many researchers (Malhotra and Ramezanianpour,1994; Shannag and Yeginobali, 1995; Duchesne and Ber-

0956-053X/$ - see front matter � 2006 Published by Elsevier Ltd.

doi:10.1016/j.wasman.2006.01.006

* Corresponding author. Tel.: +90 372 3782005/105; fax: +90 37237822205.

E-mail addresses: omerozkan@hotmail.com, ozkan@karaelmas.edu.tr(O. Ozkan).

ube, 1995; Paya et al., 1996; Kuroda et al., 2000; Shannag,2000; Colak, 2002) have reported on the use of GBFS, FAand CBA to produce concrete because of their positiveeffects on durability. Appropriate use of FA in concretecan prevent expansion due to the alkali–silica reaction(Shehata and Thomas, 2000) and can reduce heat genera-tion, as well as produce better durability properties.

Relatively few detailed studies have been conductedusing CBA, as a low-cost replacement material in the pro-duction of concrete. A study by Cheriaf et al. (1999)pointed out that the pozzolanic activity of CBA can beimproved with adequate grinding, so that it can be usedin Portland cement and concrete. However, grindingCBA results in a prolonged setting time and causes a reduc-tion in the workability of pastes, produced in the case of35% replacement with cement (Shannag and Yeginobali,1995). A study by Berg and Neal (1998) pointed out thatmunicipal solid waste bottom ash (MSWBA) was foundto be a marginal concrete aggregate. This aggregate isfound to be conformable to the ASTM C 331. They alsoreported that, due to its high angularity and brittle nature,

Table 1Chemical composition of materials used in the investigation

Chemical analysis(wt%)

SC 32.5 GBFS CBA FA

SiO2 28.12 35.11 59.53 58.67Al2O3 8.28 17.54 20.12 25.22Fe2O3 2.45 – 13.08 5.81CaO 51.90 37.80 2.02 1.46MgO 3.84 5.54 3.20 2.22SO3 1.71 0.70 Trace 0.10Na2O – 0.38 – 0.61K2O – 0.98 0.06 4.05Loss on ignition 1.14 9.4 9.81 1.13Free CaO 0.49 0.66 – 0.12Cl 0.0076 – – 0.015

Table 2Physical analysis of SC 32.5 and FA

Property SC 32.5 FA

Fineness (wt%)+45 lm 29.5 38.74+90 lm 2.2 21.50

Specific surface (cm2/g) 3200 3820Specific gravity (g/cm3) 3.01 2.02Compressive strength (3:7:28 days) (MPa) 11:23:40 –Activity index (7:28:90 days) (%) – 75:80:93

Table 3Physical properties of GBFS, CBA, sand and crushed stone

Property Sand Crushed stone GBFS CBA

Loose unit weight (kg/m3) 1930 1850 1052 620Dense unit weight (kg/m3) 1950 1910 1236 660Specific gravity (g/cm3) 2.60 2.78 2.08 1.39Water absorption (%) 2.30 1.80 8.30 6.10Amount of clay (%) 4 0.24 1 2Organic impurities Yellow Yellow – –Loss on ignition (%) 5.00 5.20 1.8 2.4Proportion of lightweight

particles (%)4.00 2.00 3.00 7.00

162 O. Ozkan et al. / Waste Management 27 (2007) 161–167

MSWBA would significantly affect concrete workability.Besides, it may lead to potential sulfate attack and corro-sion problems.

The coarse, fused, glassy texture of CBA normallywould make an ideal substitute for natural aggregates(Ramme et al., 1998). An important use of bottom ashis as a base layer material in road construction (Churcilland Amirkhanian, 1999). Also, there has been researchconcerning whether CBA could partially or entirelyreplace natural gravel in the production of a concrete pre-senting a 28-day compressive strength of 25 MPa (Peraet al., 1997).

Ground granulated blast-furnace slag (GGBFS) is usedas a partial replacement in Portland cement and concrete.Much research has been conducted on the hydraulic bind-ing properties of GGBFS (Smolczyk, 1978; Mantel, 1994;Chang and Hou, 2003; Topcu and Ugurlu, 2003). Theseresearch efforts conclude that the hydraulic binding prop-erty of blast-furnace slag depends on the chemical compo-sition and fineness of slag, hydration temperature andalkali concentration of the system, and the amount of vit-reous structure in slag. The fineness of slag, activity indexand slag/cement ratio are the parameters affecting thestrength of slag concrete (Pal et al., 2003). Although grind-ing of GBFS adds many good properties to concrete, itshould be kept in mind that grinding is an expensive andtime consuming process.

In many other research efforts, GGBFS is used forreplacing cement in concrete. If GGBFS replaces 100% ofcement, the total amount of GGBFS used for replacementwill be about 15% of the total concrete weight. In ourresearch, we have utilized GBFS for replacing the fineaggregate, not cement. Thus, the weight of replaced mate-rial over the total concrete weight can increase to about40%, which is an upper limit value theoretically. Moreover,apart from using abground form, using a non-ground formwill give opportunity to another use for this by-product orwaste.

This paper presents the physical and mechanical proper-ties of concrete incorporating FA, CBA and GBFS. Effectsof these materials on the properties of concrete were stud-ied by assessing compressive strength, unit weight (bulkdensity), water absorption capacity, surface hardness andultrasonic pulse velocity measures. In other words, thisstudy mainly focuses on the practical percentages of FA,GBFS and CBA in concrete. Also, some non-destructivetest results are compared with the destructive test resultsregarding GBFS, CBA and FA replacement.

2. Materials and experimental procedures

2.1. Materials

A kind of special cement, called slag cement (SC 32.5)was used in this study. It is equivalent to CEM III/A typein EN 197-1 (TS EN 197-1, 2002). The chemical composi-tions of the cement, FA, CBA and GBFS are shown in

Table 1. Physical properties of SC 32.5 and FA are givenin Table 2 separately.

CBA and FA are supplied by CATES in Turkey. FA isF-type according to ASTM C 618. 0–7 mm fraction of nat-ural river sand is used as fine aggregate. Only fine aggregateis replaced with GBFS and CBA in M1-group mixtures (forexplanations of groups used in the tests, see Section 2.2)where the GBFS/CBA ratio is the same in all cases. Inthe M2-group mixtures, cement is replaced by FA and sandis replaced by GBFS and CBA with the same ratio. A mid-range plasticizer concrete admixture was added to all mix-tures at 0.7% of the cement weight. This chemical admix-ture was conformable to ASTM C 494-81.

Physical properties of sand, crushed stone, GBFS andCBA are determined according to TS-706 (TS-706 EN12620, 2003) and shown in Table 3. Particle size distribu-tions are shown in Fig. 1.

0

10

20

30

40

50

60

70

80

90

100

0.01 0.1 1 10 100

Sieve size (mm)

% p

assi

ng

GBFSCBAcrushed stone0-7 mm sand

Fig. 1. Gradation curves of the GBFS, CBA, sand and crushed stone used in experiments.

O. Ozkan et al. / Waste Management 27 (2007) 161–167 163

2.2. Method

All of the tests are conducted on two main groups ofconcrete (Table 4). The first main group is called M1,which does not contain FA, and the second main groupcalled M2 with FA. Meanwhile the control concrete iscoded with M0, depicting a zero amount of replacement.Each main group is divided into 6 sub-groups in terms ofGBFS and/or CBA content. These sub-groups are denotedwith small case letters (a, b, c, d, e and f) added to the maingroup code (M1 and M2); for example, M1c or M2f. Theletter ‘a’ is used for 10% replacement of fine aggregate withGBFS and CBA equally (i.e., 5% GBFS and 5% CBA), andthe letter ‘b’ for 20% replacement and so on, with 10%increments between the letter codes. (To give an exampleto clearly grasp the coding system, M1c denotes that thereplacement of 0–7 mm aggregate with 15% GBFS and

Table 4Mixture proportions

Mixture Mix proportions Cement F(kFor cement For aggregate

M0 100% SC 100% aggregate 350 –M1a 100% SC 5% GBFS + 5% CBA 350 –M1b 100% SC 10% GBFS + 10% CBA 350 –M1c 100% SC 15% GBFS + 15% CBA 350 –M1d 100% SC 20% GBFS + 20% CBA 350 –M1e 100% SC 25% GBFS + 25% CBA 350 –M1f 100% SC 30% GBFS + 30% CBA 350 –M2a 95% SC + 5% FA 5% GBFS + 5% CBA 333M2b 90% SC + 10% FA 10% GBFS + 10% CBA 315M2c 85% SC + 15% FA 15% GBFS + 15% CBA 298M2d 80% SC + 20% FA 20% GBFS + 20% CBA 280M2e 75% SC + 25% FA 25% GBFS + 25% CBA 263M2f 70% SC + 30% FA 30% GBFS + 30% CBA 245 1

15% CBA is carried out in the first main group, totaling30% replacement.)

For only compression tests, each sample (sub-group) isdivided into three additional categories in terms of concreteage, namely A7, A28 and A90. The first category is A7which denotes 7-day aging, the second category (A28)denotes 28-day aging, and the third category (A90), 90-day aging. However, apart from the compression tests,the only aging category used is A28. Schmidt and ultra-sonic tests are conducted before the compression test onthe same specimens.

All types of concrete mixture were produced accordingto the mix design prepared earlier (Table 4). At first, aslump test is conducted to measure the workability of freshconcrete. Then the molds were filled with concrete in twolayers. Each layer is compacted using a vibrating table.After the specimens were removed from molds they were

Ag)

Water(kg)

Sand(0–7 mm)(kg)

GBFS(kg)

CBA(kg)

Crushedstone(kg)

Ch. adm.(kg)

175 716 – – 1150 2.7175 644 36 36 1150 2.7175 572 72 72 1150 2.7175 502 107 107 1150 2.7175 431 143 143 1150 2.7175 299 180 180 1150 2.7175 286 215 215 1150 2.7

18 175 644 36 36 1150 2.735 175 573 72 72 1150 2.753 175 502 107 107 1150 2.770 175 431 143 143 1150 2.788 175 299 180 180 1150 2.705 175 286 215 215 1150 2.7

Table 5The results of the compressive strength, and water absorption tests

Mixture Compressive strength(MPa)

Water absorptionby mass (%)

Measuredslump (cm)

A7 A28 A90

M0 24.25 37.77 43.70 4.14 14M1a 20.02 31.38 40.60 4.35 12M1b 17.45 30.08 38.42 5.24 10M1c 17.34 27.54 36.50 6.11 10M1d 15.78 26.06 35.93 6.34 9M1e 15.63 25.67 35.04 6.66 6M1f 14.14 21.91 31.34 6.87 6M2a 21.30 33.22 44.73 4.4 13M2b 19.80 30.80 46.33 5.35 11M2c 17.98 27.78 42.60 6.05 10M2d 16.42 26.63 42.11 6.45 8M2e 15.91 25.87 38.73 6.6 6M2f 14.44 22.60 34.32 7 6

164 O. Ozkan et al. / Waste Management 27 (2007) 161–167

placed in a curing tank where the water temperature was22 ± 1 �C until the 28th day. The specimens were removedfrom the tank and stored in the laboratory where the meantemperature was maintained at 20 �C ± 2 �C with a relativehumidity of 60 ± 5%. Three specimens were prepared foreach experiment to form a sub-group. Then arithmeticalaverages of results from these specimens were used as thefinal result of the corresponding sub-group.

Compressive strength, water absorption capacity ratio,saturated unit weight, surface hardness and ultrasonicpulse velocity (UPV) experiments were conducted. Thecompression strength test was conducted with a force-con-trolled compression machine that has 2000 kN loadingcapacity; A 4.5 kN/s pace rate was applied on all speci-mens. The Schmidt test and UPV measurements were alsoconducted before the compressive strength test. The aim ofthese non-destructive tests is to compare the surface hard-ness and elastic properties of new concrete to referenceconcrete. An N-type rebound hammer, used for non-destructive measurement of the concrete compressivestrength, is used for conducting rebound index. The mea-surement range of the hammer is 10–70 MPa. The UPVmethod is based on the fact that the pulse velocity of com-pressional waves in a concrete body may be related to theelastic properties. The pulse velocity depends only on theelastic properties of the material, and this is a very conve-nient technique for evaluating concrete quality (Mindesset al., 2003). Transmission time is measured and pulsevelocity is calculated in the ultrasonic test. The ultrasonictest equipment generates a pulse, transmits this to the con-crete by a transducer, and then receives the pulse with atransducer and measures the transmission time. Pulsevelocity is calculated by dividing the path length to thetransmission time. The test apparatus used in this study

15

20

25

30

35

40

45

0 10 20

Replac

I d

nu

obe

Rxe

dn

Fig. 2. Results of Schmidt T

includes a pulse generator with the frequency of 54 kHz.The procedures outlined in ASTM C 597 and ASTM C805 are followed for the UPV and Schmidt tests,respectively.

3. Results and discussion

The workability of fresh concrete is decreased withincrease in replacement ratio (Table 5), but there is noremarkable difference between M1 and M2 pertaining toworkability. While FA improves the workability of con-crete (GBFS + CBA) has an opposite effect, due to the par-ticle shape of these materials. The particles of CBA werespherical in shape and improve workability (Bai and Bash-aer, 2003). However, GBFS particles have an irregularshape and increase water demand. The porosity of the

30 40 50 60

ement ratio (%)

M1-7 day

M1-28 day

M1-90 day

M2-7 day

M2-28 day

M2-90 day

ests of concrete samples.

O. Ozkan et al. / Waste Management 27 (2007) 161–167 165

aggregates may also affect workability. Blast-furnace slag(GBFS) has a honeycombed structure and, consequently,workability of the concrete is decreased.

Compressive strengths for M2 concrete are higher thanfor M1 concrete in all cases. This can be seen clearly fromFig. 4 for the M2-A90 series. The reason for this is that FAreplaces cement in M2 concrete (FA replacement is themain difference between M1 and M2 concretes). It is previ-ously been reported that FA contributes to the strengthand improves durability of concrete (Haque and Kayali,1998; Li and Zhao, 2003).

Compressive strength decreases are observed in the M2-group for all three age categories (A7, A28 and A90). How-ever, as the age of concrete increases, a decrease in strengthis naturally inevitable, even though a minor increase hasbeen detected at a strength of about 10% FA replacement(M2b) for A90 (Fig. 4). No significant difference isobserved between the 7-day and 28-day cumulative com-pressive strength decreases for M1 and M2 concretes.However, there is a striking slower decrease for 10% and20% (GBFS + CBA) replacement in M2. An evident dis-tinction stands out when considering 90-day compressivestrengths. A 3.6% decrease can be a tolerable ratio forthe M2d group. To some extent, FA is preventing thedecreases of compressive strength caused by GBFS andCBA.

As the replacement ratio increases, the unit weight ofconcrete decreases. This is an expected result because theunit weights of GBFS and CBA are very low comparedto that of sand. There is not a clear difference betweenM1 and M2 concretes for the decrease of unit weight.The maximum decrease percentages with respect to the ref-erence concrete are 8.95% and 10.15% for M1 and M2,respectively. The small difference is caused by FA replace-

3200

3400

3600

3800

4000

4200

4400

0 10 20

Repla

PU

/m(

Vs)

Fig. 3. Results of ultrasonic test of M1 a

ment; the FA specific gravity is about two-thirds that ofcement.

Variations in the strength values of M1 and M2 con-cretes are supported by the rebound index and ultrasonicpulse velocity measurements conducted. Alteration ofrebound index values for A90 is very similar to alterationof compressive strength for A90 between M1 and M2 con-cretes. The other age categories (A7 and A28) show smalldifferences. However, the general shape of the curves inFigs. 2 and 3 is similar to the curves in Fig. 4. Small undu-lations are observed in the results of UPV test results inFig. 3, especially for the A7 and A28 age categories. It isconcluded that non-destructive testing methods such asUPV can be applied to concrete incorporating solid wastes(e.g., GBFS and CBA) on the condition that necessary cal-ibration procedures be carried out.

The differences in unit weights between the replacedmaterials and sand are shown in Table 3. As can be seenin Fig. 5, the water absorption ratio by mass is increasingwith replacement ratio in both of the concretes M1 andM2. The maximum increase is 70% with respect to refer-ence concrete. This is an unacceptably high level for waterabsorption. Permeability and porosity affects the waterabsorption capacity of concrete. The permeability ofconcrete is important because the replaced-materials havea higher water absorption capacity than sand. Porosity isimportant for the surface of the concrete. The surfacetexture of GBFS and CBA particles is different than thatof sand. A rougher texture helps the formation of astronger bond between aggregate and the cement paste.Consequently, the possible replacement ratio for(GBFS + CBA) in concrete should be low or some newprecautions should be considered to minimize waterabsorption capacity.

30 40 50 60

cement ratio (%)

M1-A7

M1-A28

M1-A90

M2-A7

M2-A28

M2-A90

nd M2 concretes for different ages .

10

15

20

25

30

35

40

45

50

0 10 20 30 40 50

Replacement ratio (%)

Com

p. s

tren

gth

(MP

a)

60

M1-A7M1-A28M1-A90M2-A7M2-A28M2-A90

A90

A28

A7

a fedcb

Fig. 4. Comparison of compression strength results.

20.50

21.00

21.50

22.00

22.50

23.00

23.50

0 10 20 30 40 50 60

GBFS+CBA replacement ratio (%)

Un

itw

ieg

ht

( k/

Nm

3 )

M1

M2

Fig. 5. Relative changes of unit weight in M1 and M2 concretes.

166 O. Ozkan et al. / Waste Management 27 (2007) 161–167

4. Conclusions

The effects of partial replacement of non-ground GBFSand/or CBA on concrete are experimentally investigated.Firstly, GBFS and CBA are used as a partial replacementfor aggregate, and then FA is partially substituted withcement in addition to these materials. From the experimen-tal data obtained during this research, the following con-clusions can be drawn:

� Workability of fresh concrete decreases as (GBFS +CBA) content increases.� Replacement of GBFS and CBA as fine aggregate in

concrete generally decreases the compressive strength.

� Compressive strength of M2 concrete is higher than thecompressive strength of M1 concrete. FA is balancing,to some extent, the decreases coming from GBFS andCBA substitution.� Non-destructive testing methods such as Schmidt ham-

mer and UPV can be used to estimate the strength prop-erties of (GBFS + CBA)-replaced concrete on thecondition that necessary calibration procedures be car-ried out.� The concrete incorporating GBFS, CBA, and FA can be

used in low-strength works. The M2b concrete, includ-ing 10% FA instead of Portland cement and 20%(GBFS + CBA) instead of fine aggregate, may be sug-gested for low-strength works.

O. Ozkan et al. / Waste Management 27 (2007) 161–167 167

� Concrete unit weight is proportional to the negativeslope of the replacement ratio. This is true for bothM1 and M2 groups. The maximum decrease rate at unitweight is at about a 10% level.� The water absorption capacity of (GBFS + CBA)-

replaced concrete increases as the replacement ratioincreases. The replacement ratio should be limited orsome precautions should be considered from the pointof view of concrete durability.� Producing concretes utilizing by-products instead of fine

aggregate will generate savings of about 20% of thesand.

The positive results from research of this nature will helppreserve the environment by recycling large amounts of by-products that are discharged every year and by findingadditional uses for the by-products.

References

ASTM C 331. Standard Specification for Lightweight Aggregates forConcrete Masonry Units. ASTM International.

ASTM C 494. Standard Specification for Chemical Admixtures forConcrete. ASTM International.

ASTM C 597. Standard Test Method for Pulse Velocity ThroughConcrete. ASTM International.

ASTM C 618. Standard Specification for Coal Fly Ash and Raw orCalcined Natural Pozzolan for Use in Concrete. ASTM International.

ASTM C 805. Standard Test Method for Rebound Number of HardenedConcrete. ASTM International.

Bai, Y., Bashaer, P.A., 2003. Influence of furnace bottom ash onproperties of concrete. Proceedings of the Institution of Civil Engineers– Structures and Buildings 156 (1), 85–92.

Berg, E.R., Neal, F.A., 1998. Municipal solid waste bottom ash asportland cement concrete ingredient. Journal of Materials in CivilEngineering 10 (3), 168, 176.

Chang, P.K., Hou, W.M., 2003. A study on the hydration properties ofhigh performance slag concrete analyzed by SRA. Cement andConcrete Research 33 (2), 183–189.

Cheriaf, M., Rocha, J.C., Pera, J., 1999. Pozzolanic properties ofpulverized coal combustion bottom ash. Cement and ConcreteResearch 29, 1387–1391.

Churcill, V.E., Amirkhanian, S.N., 1999. Coal ash utilization in asphaltconcrete mixtures. Journal of Materials in Civil Engineering 11 (4),295–297.

Colak, A., 2002. The long-term durability performance of gypsum–portland cement–natural pozzolan blends. Cement and ConcreteResearch 32, 109–115.

Duchesne, J., Berube, M.A., 1995. Effects of supplementary cementingmaterials on the composition of cement hydration products. AdvancedCement Based Materials 2, 43–52.

Haque, M.N., Kayali, O., 1998. Properties of high-strength concrete usinga fine fly ash. Cement and Concrete Research 28, 1445–1452.

Kuroda, M., Watanabe, T., Terashi, N., 2000. Increase of bond strengthat interfacial transition zone by the use of fly ash. Cement andConcrete Research 30, 253–258.

Li, G., Zhao, X., 2003. Properties of concrete incorporating fly ash andground granulated blast-furnace slag. Cement and Concrete Compos-ites 25, 293–299.

Malhotra, V.M., Ramezanianpour, A.A., 1994. Fly Ash in Concrete,second ed. CANMET MSL 94-95 (IR). CANMET, Ottawa, Canada.pp. 307.

Mantel, D.G., 1994. Investigation into hydraulic activity of five granu-lated blast furnace slags with eight different portland cements. ACIMaterial Journal 91 (5), 471–477.

Mindess, S., Young, J.F., Darwin, D., 2003. Concrete. Prentice Hall, NewJersey.

Pal, S.C., Mukherjee, A., Pathak, S.R., 2003. Investigation of hydraulicactivity of ground granulated blast furnace slag in concrete. Cementand Concrete Research 33 (9), 1481–1486.

Paya, J., Monzo, J., Borrachero, M.V., Peris-Mora, E., Gonzalez-Lopez,E., 1996. Mechanical treatment of fly ashes: Part II. Particlemorphologies in ground fly ash (GFA) and Workability of GFA –cement mortars. Cement and Concrete Research 26, 225–235.

Pera, J., Coutaz, L., Ambroise, J., Chababbet, M., 1997. Use ofincinerator bottom ash in concrete. Cement and Concrete Research27 (1), 1–5.

Ramme, B.W., Lingle, J.W., Naik, T.R., 1998. Coal Combustion ProductsUtilization in Wiscosin – an Environmental Approach. In: Proceedingsof the Three-Day CANMET/ACI International Symposium onSustainable Development of the Cement and Concrete Industry.October 21–23, Ottawa, Canada, pp. 349–383.

Shannag, M.J., 2000. High strength concrete containing natural pozzolanand silica fume. Cement and Concrete Composite 22, 399–406.

Shannag, M.J., Yeginobali, A., 1995. Properties of pastes, mortars andconcretes containing natural pozzolan. Cement and Concrete Research25, 647–657.

Shehata, H., Thomas, M.D.A., 2000. The effect of fly ash composition onthe expansion of concrete due to alkali– silica reaction. Cement andConcrete Research 30, 1063–1072.

Smolczyk, H.G., 1978. The effect of chemistry of slag on the strength ofblast furnace slags cements. Zement–Kalk-Gibs 31 (6), 294–296.

TS EN 197-1, 2002. Cement-Part 1: Composition, Specification andConformity Criteria for Common Cements. Turkish Standards Insti-tution, Ankara.

TS-706 EN 12620, 2003. Aggregates for Concrete. Turkish StandardsInstitution, Ankara.

Topcu _I, B.I., Ugurlu, A., 2003. Effect of the use of the mineral filler on theproperties of concrete. Cement and Concrete Research 33 (7), 1071–1075.