Construction and Building Materials 116 (2016) 15–24

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Investigation of coal bottom ash and fly ash in concrete as replacementfor sand and cement

http://dx.doi.org/10.1016/j.conbuildmat.2016.04.0800950-0618/� 2016 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: mohdrazman@utm.my (M.R. Salim).

Mahdi Rafieizonooz a, Jahangir Mirza b, Mohd Razman Salim c,⇑, Mohd Warid Hussin b, Elnaz Khankhaje a

a Faculty of Civil Engineering, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor Bahru, MalaysiabUTM Construction Research Centre (UTM CRC), Faculty of Civil Engineering, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor Bahru, MalaysiacCentre for Environmental Sustainability and Water Security, Faculty of Civil Engineering, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor Bahru, Malaysia

h i g h l i g h t s

� Concrete with bottom ash and fly ash added as replacement for sand and cement, respectively.� Compressive, flexural and tensile strengths of the concrete are determined.� Pulse velocity, drying shrinkage and micro-structural tests are performed.� Relationship between mechanical properties and pulse velocity is discussed.

a r t i c l e i n f o

Article history:Received 26 January 2016Received in revised form 5 April 2016Accepted 20 April 2016Available online 30 April 2016

Keywords:Coal ashConcreteMechanical propertiesUPVSEMDrying shrinkage

a b s t r a c t

Malaysia produces about 8.5 million tons of coal ash as waste which comprises of bottom ash and fly ash.Reusing such waste which is otherwise sent to landfills is an environment-friendly option. Hence, themajor aim of this research study was to investigate their use in concrete to replace sand with bottomash waste and cement with fly ash. Concrete specimens were prepared incorporating 0, 20, 50, 75 and100% of bottom ash replacing sand and 20% of coal fly ash by mass, as a substitute for OrdinaryPortland cement. Fresh and hardened state properties of the experimental specimens were determined.Results revealed that concrete workability reduced when bottom ash content increased replacing sand.On the other hand, at the early age of 28 d, no significant effect was observed in compressive, flexuraland tensile strengths of all concrete samples. After curing at 91 and 180 d ages, compressive strengthof both the experimental and control concrete samples increased significantly but remained almost sim-ilar. However, flexural and splitting tensile strengths of the experimental mix containing 75% bottom ashand 20% fly ash exceeded much more than the control sample. Moreover, drying-shrinkage of experimen-tal concrete mixtures containing 50%, 75% and 100% bottom ash and 20% fly ash was lower than the con-trol mix. It is concluded that those experimental concrete mixes can be used in several structures(foundations, sub-bases, pavements, etc.) which will minimize the cost, energy and environmental prob-lems to a great extent.

� 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Concrete is one of the most important materials in building con-struction and other infrastructure works. About 2.7 billion m3 ofconcrete was generated in 2002 worldwide, which is more than0.4 m3 of concrete generated per person once a year [1]. It is antic-ipated that the need for concrete will increase further to almost7.5 billion m3 (about 18 billion tons) a year by 2050 [2] . Such an

enormous utilization of concrete calls for higher use of naturalaggregates and cement, thus taking toll on the environment. Atleast three-quarters of the total volume of concrete consists ofcoarse and fine aggregates. Obviously, natural resources such asriver sand are getting depleted [3]. The prohibition on mining insome areas and the growing need for natural environment conser-vation further exacerbate the problem of river sand availability.

Finding new alternative materials for sustainable developmentso as to substantially decrease the consumption of naturalresources became imperative to safeguard the interests of futuregenerations. High consumption of natural resources led to greateramount of industrial wastes and environmental degradation [4].

16 M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24

Such factors have driven researchers to come up with solutionsleading to much needed sustainable development.

Coal fired thermal power plants have created tremendous vol-umes of coal bottom ash (CBA) and coal fly ash (CFA) (20–80%respectively) for years. CBA and CFA are by-products of pulverizedcoal combustion. Using them together, increase the use of disposalwastes which can reduce the environmental impact.

The projected forecast for electricity usage in peninsular Malay-sia will be produced from coal and gas (58% and 25%) by the year2024. Even though some difference exists between these figuresand those of 2014, it appears that dependence on fossil fuels maybe reduced but won’t be eliminated altogether. In fact coal con-sumption will likely mount from 43% to 58% [5]. As mentioned ear-lier, electricity production in Malaysia leads to a whopping amountof coal fly ash (CFA) some 6.8 million tons and coal bottom ash(CBA) roughly 1.7million tons.While CFA is being used tomanufac-ture pozzolanic Portland cement, CBA is not commonly used at all.

The coal ash content depends upon the non-combustible matterpresent in coal. Rock detritus filled in the fissures of coal becomesseparated from the coal during pulverization. In the furnace, car-bon and other combustible matter burn, whereas the non-combustible matter results in coal ash. Swirling air carries ash par-ticles out of hot zone where it cools down. CBA displaced fromunder the furnace accounts for nearly 20% which is directed to sus-pension ponds that take over several acres of countryside land.

The particle size distribution and appearance of CBA is compa-rable to that of river sand. CBA is comprised of mostly silica, ironand alumina, trace amount of sulphate, magnesium, and calcium,etc. These chemical constituents in and grading of CBA make itmore feasible for the production of concrete. It has been substanti-ated by previous researchers who came up with quite reassuringresults when CBA was used partially or totally replacing sand inconcrete because of its fine aggregate quality.

The flue gases carry away the finer and lighter ash particles. Inthe electrostatic precipitators installed prior to the stack, the ashparticles are extracted from the flue gases. The coal ash obtainedfrom the electrostatic precipitators is termed as CFA. It is used inconstruction industry worldwide as cement substitute in concreteand in the generation of cement as additive mineral in huge quan-tity. Using different sources of supplementary cementing materials(SCMs), especially CFA could lead to sharp reduction in overall CO2

footprint related to the final concrete production [6]. The use ofCFA in concrete has proven to improve long term strength andworkability.

There have been a substantial number of studies on concreteproduction incorporating coal ash either as cement replacements,fine and coarse aggregate. Cheriaf et al. [7] studied the pozzolanicproperty of CBA and found that strength activity indexes of CBAwith Ordinary Portland cement at 28 d and 90 d of hydration werehigher than that specified in European code EN 450 for pozzolanicmaterial to be used in concrete. Their findings confirm that CBA haspozzolanic property and is suitable for use in concrete manufactur-ing. It was also reported that [8] CBA can be used as aggregates(fine and coarse) in high-strength concrete. They studied the work-ability and mechanical properties of high-strength concrete andfound that CBA had more effect on the flexural strength than com-pressive strength. Singh and Siddique [9] in their review reportedthat CBA is a potential substitute material for sand in concrete.Singh and Siddique [10] also investigated the properties of con-crete incorporating high volume of CBA as sand replacement. Theyfound that at 28 d of curing, pulse velocity and compressivestrength were not affected by CBA used in concrete. Aggarwaland Siddique [11] investigated the microstructure and propertiesof concrete containing CBA and waste foundry sand as replacementof natural sand in concrete. Singh and Siddique [12] studied thedrying shrinkage and compressive strength of concrete containing

CBA as total or partial replacement of fine aggregate. They reportedthat after 90 d of curing period, the compressive strength of CBAconcrete outstripped that of normal concrete. Moreover, theyfound that drying shrinkage of CBA concrete mixtures reducedwith increase in CBA content in concrete.

Concrete made with low calcium CBA as a replacement of riversand displayed strength properties comparable to that of conven-tional concrete [13]. The major gain from CBA as fine aggregatein concrete is reduced dead weight of structure as well as allevia-tion of environmental hazards. Due to low specific gravity of CBA,concrete made with it has low density as compared to control con-crete. Sua-iam and Makul [14] have reported that the use of wastematerials either as cement supplementary material or as sandreplacement in concrete can result in cost savings and help inreducing the environmental problems. Topçu et al. [15] observedthat CBA can be used in production of durable geopolymer con-crete without cement.

Limited research studies have been reported on mechanicalproperties, microstructure, drying shrinkage and pulse velocity ofconcrete containing both CBA as fine aggregate and CFA as cementreplacement.

The objective of current research work was to investigate theeffects of using CBA and CFA as replacement of sand and cement,respectively, on the compressive, tensile and flexural strengthproperties of concrete. They were then compared with those ofnormal concrete. The materials chosen were carefully studied withrespect to their properties such as fineness modulus, specific grav-ity, particle size distribution and chemical composition. Moreover,the effect of using CBA and CFA on microstructure, drying shrink-age and pulse velocity properties of fly ash-bottom ash concretemixtures were also investigated in this study. The long-term dura-bility properties of fly ash-bottom ash concrete may be analysed infuture study.

2. Experimental program

2.1. Materials

Ordinary Portland cement (OPC) used in this research achieved the require-ments of ASTM C150-07 [16]. The chemical compositions of OPC, CFA and CBAare given in Table 1. OPC used had a blain surface area, specific gravity, soundness,initial and final setting times of 3990 cm2/g, 3.15, 1.0 mm, 125 min and 210 min,respectively. A single source of CFA conforming to ASTM C618-15 [17] was obtainedfrom Tanjung Bin coal power plant located in Johor, Malaysia. CFA was used as 20%replacement of OPC in all fly ash-bottom ash concrete mixtures with specific grav-ity, blain fineness and soundness of 2.45, 3450 cm2/g and 1.0 mm, respectively.

The FESEM image shows that CFA has spherical and regular shape and smallerparticles compared to CBA (Fig. 1). Chemical analysis showed that CFA is mostlycomposed of Silica, Iron, and Alumina. The percentage sum of SiO2, Al2O3 andFe2O3 in CFA is about 78.82% showing that it is a Class F according to ASTMC618-15 [17].

Sand was collected from Sungai Sayong River quarry near Johor, Malaysia. Thesand used in this research was as per the specification of ASTM C778-13 [18] andwas graded in accordance with the specification of ASTM C33/C33M-13 [19]. Theresults of fineness modulus, water absorption and specific gravity of this River sandare presented in Table 2.

Coal bottom ash (CBA) was collected from Tanjung Bin coal power thermal plantJohor, Malaysia. Particle size distribution of CBA and river sand are shown in Fig. 2.CBA was graded in accordance with the specification of ASTM C33/C33M-13 [19]. Inthis research, CBA was sieved through 4.75 mm sieve before use as replacement ofsand. The chemical properties of CBA are presented in Table 1. The chemical anal-ysis of CBA was carried out using XRF. The chemical analysis shows that CBA ismostly comprised of Silica, Iron and Alumina with small quantities of Sulphate,Magnesium and Calcium etc. Summation of the percentage of SiO2, Al2O3 andFe2O3 present in CBA was 83.24%. Loss on Ignition (LOI) of CBA was less than0.1%. The physical properties of CBA are shown in Table 2.

Fig. 3 shows the FESEM image of CBA. The FESEM image shows that CBA hasirregular and spherical shaped, porous particles and complicated texture. Fig. 4 pre-sents three different sizes of CBA; coarse, fine and ultrafine CBA.

The crushed stone aggregate was obtained from Bukit Namu quarry. The phys-ical properties of coarse aggregate are mentioned in Table 2 below. The maximumsize of coarse aggregate was 20 mm.

Table 1Chemical composition of cement, CFA and CBA.

Material SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O TiO2 P2O5 MnO SO3 BaO LOI

Cement 20.4 5.20 4.19 62.39 1.55 – 0.005 – – – 2.11 – 2.36CFA 47.6 23.8 7.42 10.7 1.50 2.16 1.68 2.92 1.16 0.120 0.759 0.154 –CBA 45.3 18.1 19.84 8.70 0.969 – 2.48 3.27 0.351 0.248 0.352 0.311 –

Fig. 1. FESEM image of CFA.

Table 2Physical properties of CBA, river sand and coarse aggregate.

Material Specific gravity Water absorption Fineness modulus

CBA 1.88 11.61 3.44River sand 2.62 7.4 2.67Coarse aggregate 2.69 0.61 6.48

0102030405060708090

100

0.01 0.1 1 10 100

Perc

ent p

assi

ng (%

)

Particle size (mm)

CBA Sand

Fig. 2. Grading curve of CBA and sand.

M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24 17

2.2. Mix proportions

Initially, sixteen batches of concrete with different percentages of CBA as fineaggregate and CFA as cement were prepared. As a result of initial samples’ compres-sive strength and workability, concrete with 20% of CFA as cement replacement anddifferent percentage of CBA as fine aggregate were selected for further analysis. Sat-urated Surface Dry (SSD) CBA and river sand was used and fixed quantity of water tocement ratio (w/c) was added in all of the concrete mixtures. The CBA was used bymass in concrete as river sand replacement. The effective w/c was 0.55 for all cases.The amount of cement (375 kg/m3) and the range of slump (6–18 cm) were thesame for all manufactured concretes. The British (DOE) [20] method was used tocalculate the mixture proportions shown in Table 3.

2.3. Casting and curing of specimens

Concrete cubes of 100 mm � 100 mm � 100 mm sizes were cast to determinethe compressive strength, water absorption, density and pulse velocity. Cylindricalspecimens of size 100 mm � 200 mm were cast to determine the split tensile

strength and 75 mm � 285 mm prism specimen for drying shrinkage of concretemixtures. 100 mm � 100 mm � 500 mm size beams were cast to determine theflexural strength of concrete. The specimens were demoulded after 24 ± 1 h of add-ing water to concrete mixture and were water cured at room temperature up to aspecified age of the test. The casting and curing of samples were performed inaccordance with the specification of BS EN 12390-02 [21].

2.4. Testing procedure

Compressive strength of concrete samples was determined at 7 d, 28 d, 91 d and180 d curing age as per BS EN 12390-03 [22]. The dry cubes with well saturated sur-face were subjected to 3000 kN compression testing machine. A consistent load wasthen applied to all the experimental specimens without any shock, thereby addingmore at the rate of 5.0 kN/s till the samples couldn’t take it. Flexural strength underfour point loading was assessed at 7 d, 28 d, 91 d and 180 d curing as per BS EN12390-05 [23] using beams of 100 mm � 100 mm � 500 mm.

Splitting tensile strength of cylinders (100 mm � 200 mm), at the age of 7 d,28 d, 91 d, and 180 d was determined as per ASTM C496-11 [24]. Pulse velocitythrough concrete was determined at 7 d, 28 d, 91 d and 180 d of curing age asper procedure ASTM C597-09 [25]. Battery operated Portable Ultrasonic Non-destructive Digital Indicating Tester was used to measure the pulse velocitythrough concrete. The pulse velocity (V) is calculated by dividing the length ofthe specimen (L) by transit time (T). Shrinkage due to drying of concrete mixtureswas evaluated as per ASTM: C157/C157M-08 [26]. It was measured on75 mm � 285 mm prism specimen in one direction; using stainless steel pins fixedover 100 mm gauge length on two opposite long sides. Test specimens were kept inupright position in the drying storage room. The comparator readings wereacquired after curing of 7 d, 14 d, 21 d, 28 d, 60 d, 91 d, 120 d, 150 d and 180 d ofair storage.

3. Results and discussions

3.1. Workability

The workability of fresh concrete is a multiple issue whichincludes the diverse requirements of compatibility, stability, andmobility. Using industrial by-products in concrete as total or par-tial substitute of sand by CBA mixed with a partial substitute ofcement paste by CFA could affect the fresh concrete properties ofthe mix. Slump is a measure indicating the consistency or worka-bility of concrete. The effect of CFA as cement replacement andCBA as replacement of sand in concrete mixtures on slump valueswith similar w/c, are tabulated in Table 3.

Fig. 3. FESEM image of CBA.

Fig. 4. Different size of CBA.

Table 3Mix proportion.

Codes Mix proportions (Kg/m3) Slump (mm)

Cement Fine aggregate W (w/c = 0.55) Coarse aggregate

% of CFA CFA Cement % of CBA CBA Sand

Control 0 0 375 0 0 780 205 1035 73FBC1 20 75 300 25 141 585 205 1035 92FBC2 20 75 300 50 281 390 205 1035 76FBC3 20 75 300 75 422 195 205 1035 53FBC4 20 75 300 100 562 0 205 1035 37

18 M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24

Slump values of control mix C0, FBC1, FBC2, FBC3 and FBC4were 73 mm, 92 mm, 76 mm, 53 mm and 37 mm, respectively.Since the CBA is known to possess much higher water absorptionratio in comparison to river sand particles, some water is absorbedinternally by the porous CBA particles. Up to 50% replacement levelof CBA, fly ash-bottom ash concrete mixtures FBC1 and FBC2 dis-played more increase in slump values as compared to that of con-trol mix C0. It may be because of existence of CFA with lowpercentage of CBA (less than 50%) in these concrete mixtures. Onthe other hand, by increasing CBA content in fly ash-bottom ashconcrete mixtures FBC3 and FBC4, considerable decrease occurredin slump values as compared to C0 mix control concrete. Moreover,Fig. 2 shows that particles of CBA carry rough texture and irregularshape. Use of CBA as fine aggregate actually enhances the con-crete’s texture with many more irregular and fine-shaped, porousparticles that are usually very rough. Hence, it enhances the interparticle friction which is responsible for obstructing the flow offresh concrete. Therefore, for fixed w/c, the workability of concretereduces with increasing use of CBA as replacement of river sand.

These results are comparable to those reported by Singh and Sid-dique [13]. Bong et al. [27] also obtained similar trend showingreduced workability of concrete containing CBA.

3.2. Compressive strength

Compressive strength test results are illustrated in Fig. 5. It isevident that compressive strength development pattern for flyash-bottom ash concrete after curing, is almost the same as thatof control concrete. At curing period of 7 d, compressive strengthof FBC mixtures reduced with increase in CBA content as substituteof sand. At curing period of 7 d, fly ash-bottom ash mixture FBC1(25% CBA – 20% CFA) achieved 90.1%, FBC2 (50% CBA – 20% CFA)achieved 81.4%, FBC3 (75% CBA – 20% CFA) achieved 77.2% andFBC4 (100% CBA – 20% CFA) achieved 74.2% compressive strengthof control concrete mix C0. With increase in curing age, compres-sive strength of control concrete increased at a slower rate thanthe experimental mixtures. At 28 d curing age, compressivestrength of the experimental mixtures namely, FBC1, FBC2, FBC3

0

5

10

15

20

25

30

35

40

7 28 91 180

Com

pres

sive

stre

ngth

(MPa

)

Age (days)

Control FBC1 FBC2 FBC3 FBC4

Fig. 5. Variation of compressive strength with age for C0 and FBC mixes.

M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24 19

and FBC4 was lower than that of control concrete. Compressivestrength of fly ash-bottom ash concrete mixtures FBC1, FBC2,FBC3 and FBC4 was 26.49 MPa, 26.33 MPa, 25.01 MPa and24.59 MPa respectively, as compared to 31.03 MPa of control con-crete mixture. At curing period of 91 d, fly ash-bottom ash concretemixture FBC1 gained 96.98%, FBC2 gained 100.2%, FBC3 gained99.18%, and FBC4 gained 98.17% compressive strength of the con-trol specimen. It can be easily deduced that significant compressionstrength of the experimentalmixtures at 28 dmaturity was attribu-table to pozzolanic activity of CBA and CFA. Cheriaf et al. [7] andSingh and Siddique (2015) observed that the consumption of port-landite by pozzolanic action of CBA between 28 d and 91 d wasnotable. At curing age of 91 d, the Scanning Electron Micrographs(SEM) of concrete mixtures FBC1, FBC2, FBC 3 and FBC4 (Fig. 6) con-firm increase in number and size of voids when CBA is used in con-crete. The encircled portions in micrographs represent the voidsand the rest is calcium silicate hydrate (C-S-H) gel, calcium hydrox-ide crystals (CH), ettringite (E) and aggregate. Low portlandite crys-tals are seen in SEM images of fly ash-bottom ash concretemixtures. Both the SEM images of control and fly ash-bottom ashconcrete mixtures show C-S-H gel formation.

As shown in Fig. 6, fly ash-bottom ash concrete mixtures FBC1,FBC2, FBC 3 and FBC4, C-S-H gel is not as monolithic and compactas in control concrete mixture. Formation of an imprecise C-S-H geland higher percentage of voids in fly ash-bottom ash concrete mix-tures might have affected their compressive strength at early cur-ing age. At curing period of 91 d and 180 d, the fine spread of C-S-Hgel and formation of extra C-S-H gel due to consumption of port-landite by pozzolanic action of CBA and CFA resulted in highercompressive strength of fly ash-bottom ash concrete mixtures. At180 d curing age, fly ash-bottom ash concrete mixtures FBC1,FBC2, FBC 3 and FBC4 gained compressive strength 99.73%,100.27%, 101.92% and 100.49% respectively of control concrete.At curing period of 180 d, all the experimental concrete mixtureswith the exception of FBC3, exhibited the compressive strengthmuch like the control concrete mix C0.

The outcomes of present research work are comparable to thatreported by Singh and Siddique [12], Kim and Lee [8] and Ghafooriand Bucholc [28] who also found no significant change in compres-sive strength with the utilization of CBA as replacement of sand inconcrete.

3.3. Flexural strength

Flexural strength test data showed that all fly ash-bottom ashconcrete mixtures showed almost similar strength as compared

to that of reference control mix C0. The values obtained for flexuralstrength are shown in Fig. 7. Similar to splitting tensile strength,flexural strength of concrete mixes varied slightly with theincrease in CBA content. At 7 d curing age, fly ash-bottom ash con-crete mixtures FBC1, FBC2, FBC 3 and FBC4 gained flexural strength99.43%, 99.81%, 98.11% and 94.75% respectively of control mix C0.The 28-day flexural strength of control mix C0 was observed as3.98 MPa, whereas mixes FBC1, FBC2, FBC3 and FBC4 achieved92.11%, 94.58%, 103.06% and 96.33% of control mix C0 respectively.At curing period of 91 d, flexural strength of fly ash-bottom ashconcrete mixtures FBC1 and FBC2 was slightly lower than that ofcontrol concrete mix C0. On the other hand, flexural strength ofmixtures FBC3 and FBC4 was slightly higher than that of controlconcrete at same age of curing time. The same trend was observedat the age of 180 d with FBC1, FBC2, FBC3 and FBC4 and achieved94.83%, 99.07%, 105.51% and 103.05% of control concreterespectively.

From the results given in Fig. 7, it is also evident that flexuralstrength of fly ash-bottom ash mixes increased with age. The delayin hydration and slow pozzolanic activity of CFA and CBA at earlycuring period may be the possible explanation for decrease in flex-ural strength of fly ash-bottom ash concrete at earlier ages. At cur-ing period of 91 d and 180 d, the fine spread of C-S-H gel andformation of extra C-S-H gel due to consumption of portlanditeby pozzolanic action of CBA and CFA resulted in higher flexuralstrength of fly ash-bottom ash concrete mixtures with highamount of CBA as fine aggregate. The outcomes of present researchwork are comparable to that reported by Singh et al. [29].

3.4. Tensile strength

The results of splitting tensile strength of fly ash-bottom ashmixes are summarized in Fig. 8. The decrease of 3.55%, 8.87%,and 5.32% in tensile strength was observed for the mixes FBC1,FBC2 and FBC4 and increase of 1.99% was observed for the mixFBC3 at 7 d age curing time with regards to control mix C0. Simi-larly, at age of 28 d curing period, the decrease of 9.76%, 14.88%and 2.48% for mixes FBC1, FBC2 and FBC4 was observed withincrease of 4.96% for the mix FBC3 in comparison to the controlmix C0. The decrease of 2.88% and 4.09% for FBC1 and FBC2 mixeswas observed at age of 91 d curing period, with increase of 23.67%and 11.21% for the FBC3 and FBC4 mixes in comparison to the con-trol mix C0. Similarly, at age of 180 d curing time, the decrease of1.29% and 2.51% for mixes FBC1 and FBC2 was observed withincrease of 24.07% and 12.68% for the mixes FBC3 and FBC4 in com-parison to the control mix C0.

FBC1

FBC2

FBC3

FBC4

Fig. 6. Scanning Electron Micrographs (SEM) of concrete mixtures FBC1, FBC2, FBC3 and FBC4 at 90 d of curing age.

20 M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24

0

1

2

3

4

5

6

7 28 91 180

Flex

ural

stre

ngth

(MPa

)

Age (days)

Control FBC1 FBC2 FBC3 FBC4

Fig. 7. Variation of flexural strength with age for C0 and FBC mixes.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

7 28 91 180

Tens

ile st

reng

th (M

Pa)

Age (days)

Control FBC1 FBC2 FBC3 FBC4

Fig. 8. Variation of tensile strength with age for C0 and FBC mixes.

Table 4Pulse velocity through concrete mixtures.

Mix Pulse velocity (m/s) Concrete quality grading as perNeville 2012

7 d 28 d 91 d 180 d Pulse velocity Concrete quality

C0 4021 4209 4313 4387 Less than 3000 DoubtfulFBC1 3973 4057 4218 4326 3000–3500 MediumFBC2 3967 4076 4263 4354 3500–4500 GoodFBC3 4015 4204 4339 4412 Above 4500 ExcellentFBC4 3996 4186 4301 4389

M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24 21

At 7 d and 28 d, all fly ash-bottom ash mixes showed the tensilestrength lower than C0 mix except FBC3. However, as the ageincreased to 91 d and 180 d, all mixes showed almost comparabletensile strengths to that of control mix C0. It could have been aresult of fine spread of C-S-H gel and extra C-S-H gel due to con-sumption of portlandite by pozzolanic action of CFA and CBA inconcrete with high amount of CBA as sand replacement. The out-come of present study corroborated with Aggarwal and Siddique[11] and Singh and Siddique [13].

3.5. Pulse velocity

Table 4 illustrates the effect of CFA and CBA on pulse velocitythrough fly ash-bottom ash concrete mixtures evaluated on thebasis of percentage decrease/increase over that of control mix. At

curing period of 7 d, the pulse velocity values of all experimentalconcrete mixtures were surely found to be lower than that of con-trol concrete mix. The pulse velocity values through the concretemixtures FBC1, FBC2 FBC3 and FBC4 were lower by 1.19%, 1.34%,0.15% and 0.62% respectively in comparison to the control mixC0. At 28 d of curing age, the pulse velocity of mixtures FBC1 andFBC2 when compared to the control mix C0, decreased by 3.61%and 3.16% respectively. Whereas, the pulse velocity through theconcrete mixtures FBC3 and FBC4 over that of control mix C0 were0.12% and 0.55% respectively. Pulse velocity of bottom ash concretemixtures increased with age as well as with the increase in CBAcontent.

Comparing pulse velocity values obtained in this study withpulse velocity values given in Neville [30], the quality of concretemade with CBA and CFA can be graded as good. The difference inpulse velocity values for all ages of maturing were less than3.61%. Higher values of pulse velocities obtained in this study indi-cated that the quality of fly ash-bottom ash concrete mixtures wasgood in terms of density, homogeneity and uniformity. At curingperiod of 91 d, the pulse velocity values for concrete mixturesFBC1, FBC2 and FBC4 were lower by 2.20%, 1.16% and 0.28% respec-tively when compared to the control mix C0. On the other hand,the pulse velocity values for FBC3 were higher by 0.60% comparedto that of control mix C0. Significant decrease in permeable porespace in fly ash-bottom ash concrete mixtures resulted in highervalues of pulse velocities. At curing period of 180 d, the pulsevelocity values for fly ash-bottom ash concrete mixture FBC1,

y = 0.0525e1.0297x

R² = 0.953

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

3.9 4 4.1 4.2 4.3 4.4 4.5

Flex

ural

stre

ngth

(MPa

)

Pulse velocity (km/s)

Fig. 10. Relationship between flexural strength and pulse velocity.

y = 0.0048e1.4882x

R² = 0.9331

1.0

1.5

2.0

2.5

3.0

3.5

4.0

3.9 4 4.1 4.2 4.3 4.4 4.5

Tens

ile st

reng

th (M

Pa)

Pulse velocity (km/s)

Fig. 11. Relationship between tensile strength and pulse velocity.

0

100

200

300

400

500

600

0 50 100 150 200

Dry

ing

shrin

kage

(×10

-6)

Age (days)

Control F G H I

22 M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24

FBC2 and FBC3 increased by 6.63%, 6.82% and 5.09% respectively. Incase of FBC4 the values increased by 4.85% over pulse velocity val-ues at curing period of 28 d as compared to 4.22% increase seen forcontrol mix C0.

Fig. 9 shows the relationship between compressive strength andultrasonic pulse velocity of concrete obtained in the study. Theequation showing the relationships between compressive strength(fcu) and the ultrasonic pulse velocity (V), together with the coeffi-cients of determination (R2) derived is given below. The empiricalparameters of the equation obtained from the present researchwork are almost similar to that reported by Singh and Siddique[10] for CBA concrete and P. Turgut [31] for normal concrete. How-ever, the coefficient of determination R2 is higher than the onereported by them. Higher value of coefficient of determinationindicates good relevance between the regression curve and datapoints.

fcu ¼ 0:0111e1:8593V R2 ¼ 0:8556 ðAuthorÞ

fcu ¼ 1:0741e0:8102V R2 ¼ 0:9493 ½10�

fcu ¼ 1:146e0:77V R2 ¼ 0:80 ½31�where

V = Pulse velocity in km/s.fcu = Compressive strength of cube in MPa.

Fig. 10 shows the relationship between flexural strength andultrasonic pulse velocity of concrete. The equation showing therelationships between flexural strength (ff) and the ultrasonicpulse velocity (V), together with the coefficients of determination(R2) derived is given below.

ff ¼ 0:0525e1:0297V R2 ¼ 0:953

whereV = Pulse velocity in km/s.ff = Flexural strength of concrete in MPa.

Fig. 11 shows the relationship between tensile strength andultrasonic pulse velocity of concrete. The equation showing therelationships between tensile strength (ft) and the ultrasonic pulsevelocity (V), together with the coefficients of determination (R2)derived is given below.

ft ¼ 0:0048e1:4882V R2 ¼ 0:9331

whereV = Pulse velocity in km/s.ft = Tensile strength of concrete in MPa.

y = 0.0111e1.8593x

R² = 0.8556

10

15

20

25

30

35

40

45

3.9 4 4.1 4.2 4.3 4.4 4.5

Com

pres

sive

stre

ngth

(MPa

)

Pulse velocity (km/s)

Fig. 9. Relationship between compressive strength and pulse velocity.

Fig. 12. Variation of drying shrinkage with age for C0 and FBC mixes.

3.6. Drying shrinkage

The result of drying shrinkage performance of control concreteand fly ash-bottom ash concrete mixtures are shown in Fig. 12. Atconstant w/c, the amount of porous particles in concrete increaseswith the increase of CBA quantity, that gradually allows the watercontent to flow during the drying process of concrete and there-fore, cause decreased drying shrinkage. It has been surmised thatthe porosity of CBA results in decreased drying shrinkage of con-crete. Hence it exerted reduced shrinkage strain on the drying pro-cess of experimental mixtures. The shrinkage strains of concretemixtures FBC1, FBC2, FBC3 and FBC4 were 361.38 � 10�6,324.77 � 10�6, 334.32 � 10�6 and 222.88 � 10�6 respectively at28 d of maturity. Nevertheless, shrinkage strain of control mix C0

y = 1.1231e0.0283x

R² = 0.8579

1.0

1.5

2.0

2.5

3.0

3.5

4.0

10 15 20 25 30 35 40

Tens

ile st

reng

th (M

Pa)

Compressive strength (MPa)

Fig. 14. Relationship between compressive and tensile strength.

y = 0.5662e0.3708x

R² = 0.9573

1.0

1.5

2.0

2.5

3.0

3.5

4.0

2.5 3.0 3.5 4.0 4.5 5.0 5.5

Tens

ile st

reng

th (M

Pa)

Flexural strength (MPa)

Fig. 15. Relationship between tensile and flexural strength.

M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24 23

was 369.97 � 10�6 at the same curing age. At 91 d drying age, theshrinkage strains of fly ash-bottom ash concrete mixtures FBC1,FBC2, FBC3 and FBC4 were 469.64 � 10�6, 450.54 � 10�6,394.02 � 10�6 and 296.11 � 10�6 respectively. On the other hand,the control mix C0’s shrinkage was 462.75 � 10�6. At 180 d of dry-ing period, fly ash-bottom ash concrete mixtures FBC2, FBC3 andFBC4 experienced 4.26%, 15.77%, and 37.89% respectively lowershrinkage strain as compared to control mix C0. Moreover, theexperimental mixture FBC1 containing 25% CBA as replacementof sand and 20% CFA as cement replacement experienced 1.54%more shrinkage than that of control mix C0. In fact, the dryingshrinkage of all experimental mixtures, with the exception ofFBC1 mixture, were recorded lower than the control concretemix C0 during all the tests’ drying age. These results are compara-ble to Singh and Siddique [12], Ghafoori and Bucholc [28,32] andKou and Poon’s [33] work. All of them have reported that in thegiven slump range, drying shrinkage standards of all the experi-mental mixes were lower than the control concrete.

4. Relationship between compressive, flexural and tensilestrength

Fig. 13 demonstrates the relation between flexural and com-pressive strength of fly ash-bottom ash concrete mixtures. Theequation showing the relationship between compressive and flex-ural strength together with the coefficients of determination R2

obtained from the present research is given below.

ff ¼ 2:2335e0:0202f cu R2 ¼ 0:9337

whereff = Flexural strength in MPa.fcu = Compressive strength of cube in MPa.

The elevated coefficient of determination R2 points to favour-able relationship between regression curve and data points.

Fig. 14 demonstrates the association of compressive and split-ting tensile strength of fly ash-bottom ash concrete mixtures. Theequation showing the relationship between compressive and splittensile strength together with the coefficients of determination R2

derived from test results of the present research is given below.

ft ¼ 1:1231e0:0283fcu R2 ¼ 0:8579

whereft = Splitting tensile strength in MPa.fcu = Compressive strength of cube in MPa.

Fig. 15 demonstrates the relation between flexural and splittingtensile strength of fly ash-bottom ash concrete mixtures. The equa-tion showing the relationship between spilt tensile and flexuralstrength together with the coefficients of determination R2 is givenbelow.

y = 2.2335e0.0202x

R² = 0.9337

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

10 15 20 25 30 35 40

Flex

ural

stre

ngth

(MPa

)

Compressive strength (MPa)

Fig. 13. Relationship between compressive and flexural strength.

ft ¼ 0:5662e0:3708ff R2 ¼ 0:9573

whereft = Splitting tensile strength in MPaff = Flexural strength of cube in MPa

5. Conclusions

The outcome of present experimental research work led to thefollowing conclusions:

1. CBA shows low density, high water absorption, irregular andspherical shaped and complicated texture. On the other hand,due to suitable particle size distribution of CBA it can be con-cluded that it is possible to utilize CBA as a fine aggregate sub-stitute for natural sand.

2. The workability of fly ash-bottom ash concrete was reduced dueto the utilization of CBA as total or partial substitute of fineaggregate in concrete. The descending values of experimentalconcrete mix showed a downswing at fixed w/c, with increasein CBA content as substitute of sand in concrete. The CBA parti-cles are found to have rough texture and irregular shape. Use ofCBA as fine aggregate actually enhances the concrete’s texturewith many more irregular and fine-shaped, porous particlesthat are usually very rough. It, therefore, enhances the inter par-ticle friction which is responsible for obstructing the flow offresh concrete.

3. The phenomenon of compressive strength development of flyash-bottom ash concrete with curing period is almost similarto that of control concrete. At curing age of 7 d there was con-siderable reduction in compressive strength in all fly ash-bottom ash concrete mixtures compared with that of controlconcrete. With progress in curing period, considerable increasein compressive strength of fly ash-bottom ash concrete mix-tures was noticed. It can be concluded with some certainty thatnotable increase in compressive strength of fly ash-bottom ash

24 M. Rafieizonooz et al. / Construction and Building Materials 116 (2016) 15–24

concrete mixtures after 28 d curing period, was because of poz-zolanic activity of CBA and CFA. At curing period of 91 d, com-pressive strength of fly ash-bottom ash concrete mixtures wasalmost the same as the control concrete and after 180 d, itexceeded compare to control concrete.

4. Increase in curing age resulted in flexural strength of fly ash-bottom ash mixes to rise at a higher rate than of control con-crete. The delay in hydration and slow pozzolanic activity ofCFA and CBA at early curing period may be the possible expla-nation for decrease noted in flexural strength of fly ash-bottomash concrete at earlier ages. Flexural strength of these mixturesgot better at 91 d maturing and 180 d on use of CBA as partial ortotal replacement of fine aggregate. At the age of 180 d mixturesof FBC3 and FBC4 achieved 5.51% and 3.05% higher flexuralstrength as compared to the control concrete respectively.

5. The same trend of flexural strength was observed for the split-ting tensile durability of the experimental mixtures in all curingages. At 7 d and 28 d, all fly ash-bottom ash mixes showed thetensile strength lower than C0 mix except FBC3, but as the ageincreased to 91 d and 180 d, all mixes showed almost compara-ble tensile strengths to that of control mix C0. At the age of180 d, mixtures of FBC3 and FBC4 achieved 24.07% and 12.68%higher split tensile strength than that of control concreterespectively.

6. With increase in curing, pulse velocity through fly ash-bottomash mixes increased at a higher rate than control concrete.The pulse velocity results of this study were in line with flexuraland split tensile strength results. The relationship derived fromthis research work presents good agreement between flexuraland split tensile strength and pulse velocity. This can be usedto calculate approximately the flexural and split tensilestrength of concrete.

7. Concrete mixtures incorporating CBA as fine aggregate for par-tial or total replacement of natural sand, displayed betterdimensional stability. It is, therefore, obvious that the porousparticle structure of CBA is beneficial for decreasing the dryingshrinkage of concrete. At 180 d of drying period, fly ash-bottomash concrete mixtures FBC2, FBC3 and FBC4 experienced21.79%, 34.62%, and 37.17% respectively less drying shrinkageas compared to control mix C0.

Acknowledgement

Thanks for the financial support provided by Universiti Tekno-logi Malaysia through the Ministry of Education Malaysia underOTR grant, Vot. R.J1300000.7301.4B145. Appreciation is also tothe support given by the Japan International Cooperation Agency(JICA) under the scheme of SATREPS Program (Science and Tech-nology Research Partnership for Sustainable Development) forthe project Development of Low Carbon Society Scenarios for AsianRegion.

References

[1] T.R. Naik, Sustainability of concrete construction, Pract. Period. Struct. Des.Constr. 13 (2008) 98–103, http://dx.doi.org/10.1061/(ASCE)1084-0680(2008)13:2(98).

[2] M.P. Monteiro, Concrete: Microstructure, Properties, and Materials, 2015.<http://www.sidalc.net/cgi-bin/wxis.exe/?IsisScript=SUV.xis&method=post&formato=2&cantidad=1&expresion=mfn=005676> (accessed November 25,2015).

[3] A. Mardani-Aghabaglou, M. Tuyan, K. Ramyar, Mechanical and durabilityperformance of concrete incorporating fine recycled concrete and glassaggregates, Mater. Struct. 48 (2014) 2629–2640, http://dx.doi.org/10.1617/s11527-014-0342-3.

[4] E. Khankhaje, M.W. Hussin, J. Mirza, M. Rafieizonooz, M.R. Salim, H.C. Siong,et al., On blended cement and geopolymer concretes containing palm oil fuelash, Mater. Des. (2015), http://dx.doi.org/10.1016/j.matdes.2015.09.140.

[5] S. Tenaga, Malaysia Electricity Supply & Voltage, 2014. <http://www.malaysia-trulyasia.com/tourism/electricity_supply.htm> (accessed November 25, 2015).

[6] M.T. Junaid, A. Khennane, O. Kayali, Performance of fly ash based geopolymerconcrete made using non-pelletized fly ash aggregates after exposure to hightemperatures, Mater. Struct. 48 (2015) 3357–3365, http://dx.doi.org/10.1617/s11527-014-0404-6.

[7] M. Cheriaf, J.C. Rocha, J. Péra, Pozzolanic properties of pulverized coalcombustion bottom ash, Cem. Concr. Res. 29 (1999) 1387–1391, http://dx.doi.org/10.1016/S0008-8846(99)00098-8.

[8] H.K. Kim, H.K. Lee, Use of power plant bottom ash as fine and coarse aggregatesin high-strength concrete, Constr. Build. Mater. 25 (2011) 1115–1122, http://dx.doi.org/10.1016/j.conbuildmat.2010.06.065.

[9] M. Singh, R. Siddique, Effect of coal bottom ash as partial replacement of sandon properties of concrete, Resour. Conserv. Recycl. 72 (2013) 20–32, http://dx.doi.org/10.1016/j.resconrec.2012.12.006.

[10] M. Singh, R. Siddique, Properties of concrete containing high volumes of coalbottom ash as fine aggregate, J. Clean. Prod. 91 (2015) 269–278, http://dx.doi.org/10.1016/j.jclepro.2014.12.026.

[11] Y. Aggarwal, R. Siddique, Microstructure and properties of concrete usingbottom ash and waste foundry sand as partial replacement of fine aggregates,Constr. Build. Mater. 54 (2014) 210–223, http://dx.doi.org/10.1016/j.conbuildmat.2013.12.051.

[12] M. Singh, R. Siddique, Compressive strength, drying shrinkage and chemicalresistance of concrete incorporating coal bottom ash as partial or totalreplacement of sand, Constr. Build. Mater. 68 (2014) 39–48, http://dx.doi.org/10.1016/j.conbuildmat.2014.06.034.

[13] M. Singh, R. Siddique, Strength properties and micro-structural properties ofconcrete containing coal bottom ash as partial replacement of fine aggregate,Constr. Build. Mater. 50 (2014) 246–256, http://dx.doi.org/10.1016/j.conbuildmat.2013.09.026.

[14] G. Sua-iam, N. Makul, Utilization of high volumes of unprocessed lignite-coalfly ash and rice husk ash in self-consolidating concrete, J. Clean. Prod. 78(2014) 184–194, http://dx.doi.org/10.1016/j.jclepro.2014.04.060.

[15] _I.B. Topçu, M.U. Toprak, T. Uygunoglu, Durability and microstructurecharacteristics of alkali activated coal bottom ash geopolymer cement, J.Clean. Prod. 81 (2014) 211–217, http://dx.doi.org/10.1016/j.jclepro.2014.06.037.

[16] ASTM C150-07, Standard Specification for Cement, vol. i, 2011, pp. 1–4, http://dx.doi.org/10.1520/C0010.

[17] ASTM C618-15, Standard Specification for Coal Fly Ash and Raw or CalcinedNatural Pozzolan for Use, Annu. B. ASTM Stand, 2010, pp. 3–6, http://dx.doi.org/10.1520/C0618.

[18] ASTM C778-13, Standard Specification for Sand 65.198.187.10., 2014, pp. 1–3,http://dx.doi.org/10.1520/C0778-13.2.

[19] ASTM C33/C33M-13, Standard Specification for Concrete Aggregates, ASTMInt, 2003, p. 11, http://dx.doi.org/10.1520/C0033.

[20] BS EN 206:2013, Concrete – Specification, Performance, Production andConformity, 2013. <http://shop.bsigroup.com/ProductDetail/?pid=000000000030297896> (accessed March 17, 2016).

[21] BS EN 12390-02, Testing Hardened Concrete Curing, 2009, pp. 420–457.[22] BS EN 12390-03, Testing Hardened Concrete Compressive Strength, 2009, pp.

420–457.[23] BS En 12390-05, Testing Hardened Concrete Flexural Strength, 2009, pp. 420–

457.[24] ASTM C496-11, Standard test method for splitting tensile strength, Fuel i

(2014) 1–5, http://dx.doi.org/10.1520/C0496.[25] ASTM C597-09, Standard Specification for Pulse Velocity Through Concrete,

2009, pp. 6–9, http://dx.doi.org/10.1520/C0597-09.2.[26] ASTM:C157/C157M-08, Standard Test Method for Length Change of Hardened

Hydraulic-Cement Mortar and Concrete, vol. 8, ASTM Int, 2008, pp. 1–7, http://dx.doi.org/10.1520/C0157_C0157M-08.

[27] L. Bong, C.-S. Researcher, K.J. Sung, K. Tae, C. Seongtae-Manager, A Study on theFundamental Properties of Concrete Incorporating Pond-Ash in Korea, (n.d.).<http://www.researchgate.net/publication/268419297_A_STUDY_ON_THE_FUNDAMENTAL_PROPERTIES_OF_CONCRETE_INCORPORATING_POND-ASH_IN_KOREA> (accessed November 30, 2015).

[28] N. Ghafoori, J. Bucholc, Properties of high-calcium dry bottom ash concrete,ACI Mater. J. 94 (1997) 90–101. <http://www.scopus.com/inward/record.url?eid=2-s2.0-0031100545&partnerID=tZOtx3y1>.

[29] M. Singh, R. Siddique, K. Ait-Mokhtar, R. Belarbi, Durability properties ofconcrete made with high volumes of low-calcium coal bottom ash as areplacement of two types of sand, J. Mater. Civ. Eng. (2015) 04015175, http://dx.doi.org/10.1061/(ASCE)MT.1943-5533.0001464.

[30] A.M. Neville, Properties of Concrete, Prentice Hall, 2012.<https://books.google.com/books?id=WW1wUvwQ0sUC&pgis=1> (accessedNovember 28, 2015).

[31] P. Turgut, Research into the correlation between concrete strength and UPVvalues, J. Nondestr. Test. (2004). http://www.ndt.net/article/v09n12/turgut/turgut.htm (accessed December 1, 2015).

[32] N. Ghafoori, J. Bucholc, Investigation of lignite-based bottom ash for structuralconcrete, J. Mater. Civ. Eng. 8 (1996) 128–137, http://dx.doi.org/10.1061/(ASCE)0899-1561(1996) 8:3(128).

[33] S.-C. Kou, C.-S. Poon, Properties of concrete prepared with crushed fine stone,furnace bottom ash and fine recycled aggregate as fine aggregates, Constr.Build. Mater. 23 (2009) 2877–2886, http://dx.doi.org/10.1016/j.conbuildmat.2009.02.009.