Embed Size (px)
Citation preview
Construction and Building Materials 38 (2013) 455–464
Contents lists available at SciVerse ScienceDirect
Construction and Building Materials
journal homepage: www.elsevier .com/locate /conbui ldmat
Utilization of limestone powder to improve the properties of self-compactingconcrete incorporating high volumes of untreated rice husk ash as fine aggregate
Gritsada Sua-iam a,1, Natt Makul b,⇑a The Project Office of Consortium on Doctoral Philosophy Program of Rajabhat University, Phranakhon Rajabhat University, 9 Changwattana Road, Bangkhen, Bangkok10220, Thailandb Faculty of Industrial Technology, Phranakhon Rajabhat University, 9 Changwattana Road, Bangkhen, Bangkok 10220, Thailand
h i g h l i g h t s
" Limestone powder was used to produce self-compacting concrete incorporating RHA." The T50cm flow times increased in mixtures containing RHA or RHA and LS." Combinations of LS and RHA improved the workability of SCC mixtures." The hardened properties of the concretes were progressively improved with addition of LS.
a r t i c l e i n f o
Article history:Received 7 May 2012Received in revised form 10 August 2012Accepted 16 August 2012Available online 2 October 2012
Keywords:Self-compacting concreteUntreated rice husk ashLimestone powderFine sand replacement
0950-0618/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.conbuildmat.2012.08.016
⇑ Corresponding author. Tel./fax: +66 2 544 8275.E-mail address: [email protected] (N. Ma
1 Ph.D. candidate, Technology Management Program,Doctoral Program, Phranakhon Rajabhat University, 9 CBangkok, 10220, Thailand.
a b s t r a c t
Self-compacting concrete (SCC) is a relatively recent development in the construction industry. It flowsunder its own weight while remaining homogeneous in composition. We examined the feasibility ofusing limestone powder (LS) as a modifying agent in self-compacting concrete in which a portion ofthe fine aggregate was replaced with untreated rice husk ash (RHA). The mixtures were designed to pro-duce a controlled slump flow. The Portland cement content was 550 kg/m3 for all of the mixtures. The fineaggregate was replaced with up to 100% RHA and LS by volume. The T50 slump flow, J-ring flow, blockingassessment, V-funnel, air density, and compressive strength of the SCC mixtures were tested. The freshproperties of the RHA-containing mixtures were improved in mixtures containing less than 60 vol.%RHA. SCCs containing LS exhibited superior hardened properties, and the fresh and hardened propertiesof SCCs made using RHA were substantially improved when combined with LS. Limestone powder has thepotential to improve self-compacting concrete mixtures in which untreated RHA is used as a partial fineaggregate replacement.
� 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Self-compacting concrete (SCC) has gained wide acceptance inthe construction industry since its introduction in 1988 in Japanto address a lack of skilled workers. Its properties are the resultof modifications to the composition of ordinary concrete. A com-parison of the compositions of SCC and conventional concrete isshown in Fig. 1. SCC flows under its own weight while maintainingresistance to segregation [1,2]. Fresh SCC must be stable to ensurehomogeneity and mechanical strength in the finished structure.Several problems may occur in some formulations, includingbleeding, settlement, and segregation [3]. Superplasticizers are
ll rights reserved.
kul).The Project for Consortium onhangwattana Road, Bangkhen
used to improve the flowability without causing deformation orsegregation problems [4]. Mixtures containing moderate amountsof cementitious materials and fine fillers decrease the coarseaggregate volume and reduce the risk of blockage while simulta-neously increasing the segregation resistance and reducing thecosts associated with high volumes of Portland cement andsuperplasticizer [5]. Mineral admixtures such as fly ash, limestonepowder, blast furnace slag, silica fume, brick powder, kaolinite,bagasse ash, and rice husk ash (RHA) have been used in attemptsto improve the properties of SCC [6–12].
Ground RHA is highly pozzolanic and may be used as a supple-mentary cementitious material in concrete [13,14]. Previous studieshave indicated that up to 20% ground RHA may be advantageouslyblended into the mixture without adversely affecting the strengthand durability of the resulting concrete [15]. In addition, it is possi-ble to use residual RHA without grinding by adapting the mixingprocess to optimize the ash particle size [16]. A preliminary study
Conventional Concrete
Self-Compacting Concrete
Cement WaterAir Fine aggregate Coarse aggregateAdditive
Fig. 1. Comparison of conventional and self-compacting concrete mix proportions [1].
456 G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464
examining the utilization of rice husk ash (RHA) as a cementreplacement has demonstrated that incorporation of RHA in SCCdecreases the unit weight, flowability, water absorption, totalporosity, compressive strength, ultrasonic pulse velocity, and cost[17–20]. Limestone (LS) is a potentially valuable resource producedduring stone crushing operations, and is the most common additivefor improving the flowability of SCC [21,22]. The addition of lime-stone reduces the initial and final concrete setting times whileincreasing the total shrinkage only slightly compared to conven-tional concrete [23]. The limestone filler also acts as a viscosityenhancer, increasing the workability [24,25].
The objective of this study was to investigate the use of as-re-ceived residual RHA as a partial fine aggregate replacement. Satis-factory RHA particle size was obtained by mixing RHA with theremaining fine and coarse aggregates. The elimination of grindingcosts increases the feasibility of using RHA in concrete production,particularly for projects near rice production zones and for smallcontractors.
2. Experimental program
2.1. Materials
The mixtures were prepared using Type 1 Portland cement (OPC) complyingwith ASTM C150 [26]. The rice husk ash (RHA) was obtained from an electric powerplant in Chainat Province, Thailand, and the only treatments prior to use were dry-ing and homogenization. Limestone powder (LS) was obtained from a rock crushingplant located in Saraburi Province, Thailand. The chemical compositions and phys-ical properties of the cement, rice husk ash, and limestone powder are listed inTable 1. The morphologies of the materials were examined using scanning electronmicroscopy (SEM). Fig. 2 contains SEM images of the materials obtained at approx-imately 1000� magnification. RHA, OPC, and LS are crystalline materials (Fig. 3).The major phases are cristobalite in RHA and calcite in LS.
A polycarboxylate-based high range water reducing admixture [HRWR] con-forming to ASTM C494 [27] standard type F was added to the mixtures at a concen-tration of 2.0 wt.% of the binder materials. The solids content and specific gravity ofthe HRWR were 42% and 1.05%.
Table 1Chemical composition and physical properties of SCC components.
Type 1 Portlandcement
Rice huskash
Limestonepowder
Chemical composition (% by mass)Silicon dioxide (SiO2) 16.39 93.00 8.97Aluminum oxide (Al2O3) 3.85 0.35 1.02Ferric oxide (Fe2O3) 3.48 0.23 0.37Magnesium oxide (MgO) 0.64 0.41 2.38Calcium oxide (CaO) 68.48 1.31 46.77Sodium oxide (Na2O) 0.06 0.15 0.02Potassium oxide (K2O) 0.52 1.61 0.13Sodium oxide (SO3) 4.00 0.09 0.33
Physical propertiesLoss on Ignition (% by mass) 1.70 1.90 39.54Particle size distribution (lm) 23.32 84.32 15.63Specific gravity 3.2 2.2 2.76Specific surface area (m2/kg) 610 240 1300
The particle size distributions of the Portland cement and limestone powderwere measured using a Malvern Instruments Mastersizer 2000 particle size ana-lyzer. The limestone powder particles were slightly smaller than the Portland ce-ment particles. The fine aggregates included river sand with a fineness modulusof 2.67 and untreated rice husk ash with a fineness modulus of 0.69. The particlesize distributions of all aggregate materials conformed to the requirements of ASTMC33 [28] (Fig. 4).
2.2. Mixture proportions
The compositions of the SCC mixtures are listed in Table 2. Several mixtureswere prepared containing various fine aggregate replacement amounts. The cementcontent was 550 kg/m3 and the coarse aggregate content was 708 kg/m3 in all mix-tures. RHA and LS were used to replace the river sand at levels of 0%, 10%, 20%, 40%,60%, 80% or 100% by volume. The SCC mixtures were identified using the formsRHAx, LSy, and RHAxLSy, in which x and y are the volume percentages of river sandreplaced by RHA or LS.
2.3. Specimen testing
Effective mixing was critical to concrete performance. The concrete was pre-pared in 35 l batches using a tilting mixer. The addition of superplasticizer was de-layed until 1–2 min after the addition of water, resulting in a higher flowabilitymixture [29]. The procedure is illustrated in Fig. 5.
The controlled slump flow diameter was maintained at 70 ± 2.5 cm. The unitweight of the freshly-prepared SCC was measured as specified in ASTM C29 [30]and the air content was measured as specified in ASTM C231 [31]. Slump flow testswere performed using an inverted mold without compaction in accordance withASTM C1611 [32]. The reported spread diameters are the averages of four measure-ments. The passing ability was tested using a J-ring according to the procedure inASTM C1621 [33]. The filling ability was tested using a V-funnel according to theprocedure outlined in EFNARC [34] and illustrated in Figs. 6–8. The hardened prop-erties were determined using ultrasonic pulse velocity and compressive strengthtests using triplicate cylinders of 150 mm diameter and 300 mm length for eachtest. Samples were tested after aging for 1, 7, 28, and 91 days in accordance withASTM C597 [35] and ASTM C39 [36].
2.4. General acceptance criteria
The acceptance criteria for the self compacting concrete mixtures are describedin Table 3. The slump flow, T50 slump flow, and V-funnel tests were performedaccording to EFNARC procedures [34]. The J-ring and blocking assessments wereadopted from ASTM C1621 [33].
3. Results
3.1. Properties of fresh SCC
The water requirement, unit weight, slump flow, J-ring flow,blocking assessment, V-funnel, and air density test results arelisted in Table 4.
3.1.1. Water requirementThe SCC water/binder ratios (w/c) resulting in controlled slumps
of 70 ± 2.5 cm diameter are provided in Fig. 9. In order to maintainthe desired slump flow, SCC mixtures containing RHA required morewater than those containing only LS or a combination of RHA and LS.In general, the water requirement increased with increasing RHA
(b) Rice husk ash
(c) Limestone powder
(a) Type 1 Portland cement
Fig. 2. SEM micrographs (1000�) of: (a) Type 1 Portland cement, (b) rice husk ash, and (c) limestone powder.
0
500
1000
1500
2000
2500
3000
3500
0
100
200
300
400
500
600
700
800
5 10 15 20 25 30 35 40 45 50 55 60 65 70
LS
Two – Theta (deg)
Inte
nsity
(ar
bitr
ary
units
)
C
CC C
Ca
Ca Ca
Ca Ca
Ca Calcite
Inte
nsity
(ar
bitr
ary
units
)
OPC
RHA
C3SC2SC3S
C3SC3S
C3S
Cristobalite
C3S – 3CaOSiO2
C2S – 2CaOSiO2
C
Fig. 3. X-ray diffraction results for LS, OPC and RHA.
0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50
60
70
80
90
100
0.1 1 10 100 1000
RHAFine AggregateUpper gradationLower gradationLimestoneOPC
Diameter size (μm)
Fine
r (%
)
Cum
ulat
ed f
iner
(%)
D [4,3] = 23.32 m
D [4,3] = 15.63 m
Upper ASTM C33 Lower ASTM C33
μ
μ
Fig. 4. Particle size distributions of rice husk ash, fine aggregate, limestone powder,and Type 1 Portland cement.
G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464 457
content, mostly due to the increased specific surface area and highunburned carbon content of RHA [15]. Use of RHA combined withLS substantially reduced the water requirement. The water demandreduction was attributed to reduced flocculation due to interactionbetween the RHA particles and the oppositely-charged cementparticles. The effective dispersion of the LS particles retained large
amounts of water, resulting in a reduction of water requirementsand improved particle packing [25].
3.1.2. Unit weightThe unit weight of the SCC decreased with increasing RHA
replacement and increased with increasing LS content (Fig. 10).The greater unit weight of mixtures containing LS was due to thehigher density of the LS particles (2.76 versus 2.67 for sand and2.2 for RHA) and void-filling by the finer particles.
3.1.3. T50cm slump flowThe times required for the SCC mixtures to reach a slump flow
diameter of 50 cm were all within 3–7 s, the range consideredacceptable in EFNARC guidelines [34] (Table 3). All of the SCC
Table 2Mix proportions.
Mixsymbol
Materials (kg/m3) SuperplasticizerType F (%)
Type 1 Portland cement Fine aggregate Coarse aggregate
Sand RHA LS
Control 550 813 – – 708 2.0
RHA10 550 731 67 – 708 2.0RHA20 550 650 135 – 708 2.0RHA40 550 488 270 – 708 2.0RHA60 550 325 405 – 708 2.0RHA80 550 163 540 – 708 2.0RHA100 550 – 677 – 708 2.0
LS10 550 731 – 85 708 2.0LS20 550 650 – 169 708 2.0LS40 550 488 – 339 708 2.0LS60 550 325 – 508 708 2.0LS80 550 163 – 677 708 2.0LS100 550 – – 846 708 2.0
RHA5LS5 550 731 34 42 708 2.0RHA10LS10 550 650 67 85 708 2.0RHA5LS15 550 650 34 127 708 2.0RHA15LS5 550 650 101 42 708 2.0RHA20LS20 550 488 135 169 708 2.0RHA10LS30 550 488 67 254 708 2.0RHA30LS10 550 488 202 85 708 2.0RHA30LS30 550 325 202 254 708 2.0RHA20LS40 550 325 135 339 708 2.0RHA40LS20 550 325 270 169 708 2.0RHA40LS40 550 163 270 339 708 2.0RHA50LS50 550 – 337 423 708 2.0
Mineral admixtures(RHA and LS) WaterCement SuperplasticizerCoarse aggregateFine aggregate Stop mixing
minutes150 41 106 9
Fig. 5. Mixing procedure time line.
t = 7s, ∅ = 50 cm t = 14s, ∅ = 71 cmt =0s, ∅ = 0 cm
Fig. 6. Slump flow testing.
t =0s, ∅ = 0 cm t = 10s, ∅ = 50 cm t = 24s, ∅ = 69 cm
Fig. 7. J-ring testing.
458 G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464
t = 0s t = 10sDimension in mm
Fig. 8. V-funnel testing.
Table 3General acceptance criteria for SCC from Refs. [33,34].
Workability test Slump flow (mm) T50cm (s) V-funnel (s) Blocking assessment (mm)
No Minimal Extreme
Requirement 650–800 3–7 8–12 0–25 25–50 >50
Table 4Properties of fresh SCC mixtures.
Mix Slump flow J-ring test V-funnel time (s) w/b Air content (%)
Diameter (cm) T50cm (s) Diameter (cm) Blocking
Control 70 6 68 No 7 0.22 1.65
RHA10 69 6 67 No 9 0.31 1.85RHA20 70 6 67 Minimal 11 0.46 1.60RHA40 70 7 66 Minimal 14 0.75 1.15RHA60 70 8 62 Extreme N/Aa 1.17 1.50RHA80 60 15 50 Extreme N/Aa 1.80 1.95RHA100 N/Aa N/Aa N/Aa N/Aa N/Aa 2.18 2.30
LS10 69 8 66 Minimal 8 0.22 2.20LS20 71 10 68 Minimal 10 0.26 1.10LS40 73 15 68 Minimal 34 0.36 2.00LS60 69 20 62 Extreme 42 0.49 2.10LS80 70 14 62 Extreme 24 0.74 2.00LS100 68 12 64 Minimal 14 0.90 1.80
RHA5LS5 70 6 68 No 9 0.20 1.95RHA10LS10 71 7 69 No 10 0.33 2.35RHA5LS15 68 7 66 No 8 0.38 1.65RHA15LS5 70 8 68 No 14 0.43 2.00RHA20LS20 70 8 68 No 10 0.48 1.00RHA10LS30 68 10 66 No. 13 0.45 1.60RHA30LS10 68 5 66 No 15 0.64 2.00RHA30LS30 70 9 68 No 11 0.64 1.85RHA20LS40 70 14 68 No 16 0.53 1.80RHA40LS20 68 7 64 Minimal 5 0.85 2.25RHA40LS40 70 9 65 Minimal 8 0.94 1.95RHA50LS50 68 16 62 Extreme N/Aa 1.17 2.15
a N/A indicates ‘‘not applicable’’.
G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464 459
mixtures exhibited satisfactory average slump flows of 70 ± 2.5 cmdiameter (Table 4), which is an indication of good workability. Theslump flow time increased with increasing RHA and LS content.The slump flow time increased varied in the range of 6–15 s, 8–20 s and 6–16 s for SCC mixtures containing RHA, LS and RHA withLS, respectively (Fig. 11a–c). The longer slump flow times may havebeen due to the increased surface area of RHA, which increased theviscosity of the paste. Increased viscosity in SCC mixtures reducesthe risk of segregation during concrete placement operations [5].
3.1.4. V-funnel testThe V-funnel test measures the time required for a concrete
mixture to flow through a funnel and provides a means of evaluat-
ing the viscosity and segregation resistance of concrete mixtures.Acceptable V-funnel flow times according to EFNARC guidelinesare between 8 and 12 s [34]. The flow times increased in propor-tion to the water requirement and amount of RHA added. Accept-able flow times were obtained for mixtures RHA10 and RHA20(Fig. 11a). The RHA particles absorbed water, resulting in a highlyviscous mix and reducing bleeding. V-funnel times were longerfor samples containing LS due to lower water requirements, com-pactness of the mixture, and greater viscosity (Fig. 11b).
SCC mixtures containing a combination of RHA with LS pro-vided acceptable flow results (Fig. 11c) according to EFNARC spec-ifications. In 1:1 blends of RHA and LS, replacement levels of lessthan 60% met flow time guidelines (Table 3).
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
0 10 20 30 40 50 60 70 80 90 100
RHALSRHA+LS
45%
36%
25%28%
21%
23%
35%
-10%
Fine aggregate replacement (%)
Req
uire
d w
/c
48%
21%
46%
23%
Fig. 9. Required water/binder ratios for SCC mixtures.
70
75
80
85
90
95
100
105
0 20 40 60 80 100
LS
RHA+LS
RHA
14%
4%11%
4%
1%
5%
9%
2%
Increased
Decreased
Fine aggregate replacement (%)
Uni
t wei
ght
(% c
ontr
ol m
ixtu
re)
16%
11%
19%
12%
Fig. 10. Unit weights of SCC mixtures.
(a) Rice husk ash.
(b) Limestone powder.
0
4
8
12
16
20
24
28
32
36
40
44
0
2
4
6
8
10
12
14
16
Control RHA10 RHA20 RHA40 RHA60 RHA80 RHA100
Flow timeV-funnel
048121620242832364044
0
5
10
15
20
25
Control LS10 LS20 LS40 LS60 LS80 LS100
Flow timeV-funnel
* Not applicable
Slum
p fl
ow ti
me
(sec
onds
)
SCC Mixtures
V-f
unne
l flo
w ti
me
(sec
onds
)
V-funnel Target range
Slump flow Target range
** *
Slum
p fl
ow ti
me
(sec
onds
)
SCC Mixtures
V-f
unne
l flo
w ti
me
(sec
onds
)
V-funnel Target rangeSlump flow Target range
(c) Rice husk ash and limestone powder
0
4
8
12
16
20
24
28
32
36
40
44
0
2
4
6
8
10
12
14
16
18
Flow time
V-funnel
Slum
p fl
ow ti
me
(sec
onds
)
SCC MixturesV
-fun
nel f
low
tim
e (s
econ
ds)
V-funnel Target range
Slump flow Target range
0% 10% 40%20% 80%60% 100%
Fig. 11. Slump flow and V-funnel times for SCC mixtures.
460 G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464
3.1.5. J-ring test and blocking assessmentThe J-ring and slump flow tests provide a means of determining
the passing ability, or the ability of the concrete to flow under itsown weight to completely fill all voids. The differences in theslump flow and J-ring flow diameters were used to obtain theblocking assessment. The blocking assessment criteria conformedto ASTM C1621 [33], in which 0–25 mm [0–1 in.] is defined as novisible blocking, 25–50 mm [1–2 in.] is defined as minimal tonoticeable blocking, and greater than 50 mm [2 in.] is defined asnoticeable to extreme blocking. From the results in Table 4 therewas either no blocking or minimal apparent blocking in samplescontaining both RHA with LS except for RHA50LS50, which exhib-ited extreme blocking. A small degree of blocking was evident inthe control and in mixtures containing 10%, 20%, or 40% RHA orLS. Extreme blocking was observed in samples containing morethan 60% RHA or LS, except for the 100LS mixture that experiencedminimal blocking. The J-ring flow results for mixtures containing acombination of RHA and LS indicated adequate passing ability andsufficient resistance to segregation around congested reinforce-ment areas due to the combined influence of a decrease in RHAcontent and an increase in water–powder ratio. Decreasing theriver sand content by replacement with the finer LS could resultin increased viscosity and lower segregation.
3.2. Properties of hardened SCC
The compressive strength and ultrasonic pulse velocity weretested at 1, 7, 28, and 91 days, and the results reported are themeans of measurements on three specimens.
3.2.1. Compressive strengthThe average compressive strengths of the SCC samples are pre-
sented in Table 5. The compressive strength continued to increaseover the 91-day curing period. The 28-day compressive strengthvaried from 2.0 to 67.5 MPa (Table 5), while the 91-day compres-sive strength ranged from 2.60 to 82.8 MPa. The greatest compres-sive strength was achieved using the LS10 mixture at 28 days andthe control mixture at 91 days. The lowest compressive strength atall ages occurred in the mixture in which the fine aggregateconsisted of 100% RHA. All mixtures decreased in compressivestrength with increasing replacement level due to the greaterporosity as indicated by the higher water requirement of RHAand LS particles.
RHA addition increased the compressive strength of SCC at 1, 7,28, and 91 days mostly due to the microfilling ability andpozzolanic activity of RHA [17]. Addition of LS also improved the
Table 5Compressive strength of SCC mixtures.
Mix Compressive strength, MPa (% control mixture)
1-day 7-days 28-days 91-days
Control 33.1 (100%) 55.9 (100%) 65.0 (100%) 82.8 (100%)
RHA10 21.9 (66%) 48.4 (87%) 54.8 (84%) 72.6 (88%)RHA20 9.4 (28%) 21.2 (38%) 28.0 (43%) 39.6 (48%)RHA40 5.1 (15%) 13.8 (25%) 19.1 (29%) 26.4 (32%)RHA60 2.8 (8%) 8.3 (15%) 10.4 (16%) 14.8 (18%)RHA80 1.3 (4%) 2.8 (5%) 4.1 (6%) 5.7 (7%)RHA100 0.5 (2%) 1.5 (3%) 2.0 (3%) 2.6 (3%)
LS10 36.4 (110%) 61.9 (111%) 67.5 (104%) 79.8 (96%)LS20 30.6 (92%) 45.2 (81%) 51.0 (78%) 65.7 (80%)LS40 28.0 (85%) 38.0 (68%) 42.0 (65%) 50.9 (61%)LS60 17.3 (52%) 22.9 (41%) 26.8 (41%) 34.8 (42%)LS80 12.2 (37%) 20.9 (37%) 25.0 (38%) 32.2 (39%)LS100 7.60 (23%) 16.1 (29%) 20.8 (32%) 27.1 (33%)
RHA5LS5 29.3 (88%) 53.5 (96%) 61.2 (94%) 79.1 (96%)RHA10LS10 25.5 (77%) 43.3 (77%) 51.0 (78%) 72.4 (87%)RHA5LS15 25.5 (77%) 40.8 (73%) 49.4 (76%) 68.2 (82%)RHA15LS5 19.6 (59%) 31.6 (57%) 39.8 (61%) 54.7 (66%)RHA20LS20 24.0 (72%) 36.9 (66%) 45.9 (71%) 58.0 (70%)RHA10LS30 26.8 (81%) 40.8 (73%) 52.0 (80%) 64.4 (78%)RHA30LS10 10.4 (31%) 16.3 (29%) 21.2 (33%) 28.5 (34%)RHA30LS30 15.0 (45%) 24.0 (43%) 33.1 (51%) 44.4 (54%)RHA20LS40 21.4 (65%) 32.6 (58%) 39.2 (60%) 48.7 (59%)RHA40LS20 7.9 (24%) 10.7 (19%) 13.5 (21%) 16.6 (20%)RHA40LS40 7.9 (24%) 11.2 (20%) 14.5 (22%) 18.1 (22%)RHA50LS50 5.1 (15%) 7.9 (14%) 10.4 (16%) 12.8 (15%)
G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464 461
compressive strength of RHA mixtures, as the smaller particleswere able to fill micro-voids within the cement particles. The mainconstituent of LS is calcium carbonate, which reacts very little withcement hydrates. The primary effect of LS addition is most likelytherefore a ‘‘filler effect’’ [37], improving the microstructure inthe bulk paste matrix and transition zone and leading to increasedcompressive strength.
Table 6Ultrasonic pulse velocities of SCC mixtures.
Mix Pulse velocity, km/s (% control mixture)
1-day 7-days
Control 2.5 (100%) 3.6 (100%
RHA10 2.1 (84%) 3.4 (94%)RHA20 2.0 (80%) 2.8 (78%)RHA40 0.8 (32%) 1.4 (39%)RHA60 0.7 (28%) 1.1 (31%)RHA80 0.5 (20%) 0.7 (19%)RHA100 0.4 (16%) 0.6 (17%)
LS10 3.2 (128%) 3.5 (97%)LS20 2.8 (112%) 3.3 (92%)LS40 2.7 (108%) 3.1 (86%)LS60 2.6 (104%) 2.9 (81%)LS80 2.5 (100%) 2.7 (75%)LS100 2.1 (84%) 2.6 (72%)
RHA5LS5 3.1 (124%) 3.2 (89%)RHA10LS10 2.8 (112%) 3.0 (83%)RHA5LS15 2.9 (116%) 3.1 (86%)RHA15LS5 2.7 (108%) 3.0 (83%)RHA20LS20 2.6 (104%) 2.9 (81%)RHA10LS30 2.6 (104%) 2.9 (81%)RHA30LS10 2.3 (92%) 2.5 (68%)RHA30LS30 2.1 (84%) 2.5 (68%)RHA20LS40 2.5 (100%) 2.8 (78%)RHA40LS20 1.3 (52%) 2.0 (56%)RHA40LS40 2.0 (80%) 2.2 (61%)RHA50LS50 0.8 (32%) 1.7 (47%)
3.2.2. Ultrasonic pulse velocity testsUPV testing is a non-destructive technique to evaluate the
homogeneity of the concrete. The velocity of ultrasonic pulses trav-eling in a solid depends on the density and elastic properties of thematerial. The trend in UPV measurements was similar to that ofcompressive strength, with UPV increasing with increasing com-pressive strength for all of the mixtures.
28-days 91-days
) 4.4 (100%) 5.2 (100%)
4.2 (95%) 5.0 (96%)3.5 (80%) 4.2 (81%)2.2 (50%) 3.0 (58%)1.8 (41%) 2.6 (50%)0.9 (20%) 1.3 (25%)0.7 (16%) 1.0 (19%)
4.1 (93%) 4.9 (94%)3.5 (80%) 4.1 (79%)3.3 (75%) 3.8 (73%)3.2 (73%) 3.6 (69%)2.9 (66%) 3.2 (62%)2.7 (61%) 3.0 (58%)
3.9 (89%) 4.6 (88%)3.4 (77%) 4.1 (79%)3.4 (77%) 4.2 (81%)3.2 (73%) 3.9 (75%)3.4 (77%) 4.0 (77%)3.6 (82%) 4.4 (85%)2.6 (59%) 3.0 (58%)3.0 (68%) 3.5 (67%)3.2 (73%) 3.7 (71%)2.5 (57%) 2.9 (56%)2.4 (55%) 2.9 (56%)2.0 (45%) 2.5 (48%)
462 G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464
The average ultrasonic pulse velocities at 1, 7, 28, and 91 daysare presented in Table 6. The ultrasonic pulse velocity ranged from0.7 to 4.4 km/s at 28 days, while the 91-day ultrasonic pulse veloc-ity ranged from 1.0 to 5.2 km/s. The variations corresponded to thedegree of densification within the internal structure of the SCCmixtures, and higher velocities generally indicated better qualitySCC. The highest pulse velocity was achieved in the control con-crete, while the lowest velocity at all ages occurred in samples con-taining 100% RHA. The velocity decreased in all mixtures withincreasing amounts of RHA or LS.
Addition of RHA increased the ultrasonic pulse velocity in SCCsamples at 1, 7, 28, and 91 days mostly due to microfilling andpozzolanic effects on the physical and chemical properties of theconcrete pore structure. Furthermore, addition of LS to the RHAmixtures increased the ultrasonic pulse velocity due to porerefinement and porosity reduction in both the bulk paste matrixand transition zone of the concrete. The ultrasonic pulse velocitywas not significantly increased as the increase in compressivestrength resulted in the fine aggregate content. The reduction offine aggregate content was decreased the ultrasonic pulse velocityof SCC [17].
3.2.3. Relationship between compressive strength and ultrasonic pulsevelocity
The relationship between the compressive strength and theultrasonic pulse velocity of the SCC mixtures is described inFig. 12. UPV values increased with increasing compressive strengthfor all of the SCC mixtures. Similar results were obtained inprevious studies [8,19]. There was a good correlation betweencompressive strength and ultrasonic pulse velocity (R2 = 0.9119
y = 4.0284x1.833
R² = 0.9119
0
10
20
30
40
50
60
70
80
90
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5
Ultrasonic pulse velocity (km/s)
Com
pres
sive
str
engt
h (M
Pa)
Fig. 12. Relationship between compressive strength and ultrasonic pulse velocity.
y = 4.2678x1.7152
R² = 0.9403
y = 1.1614x2.8875
R² = 0.8839
y = 3.3162x2.0229
R² = 0.8256
0
20
40
60
80
100
120
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5
RHALSRHA+LS
Ultrasonic pulse velocity (km/s)
Com
pres
sive
str
engt
h (M
Pa)
Fig. 13. Relationship between compressive strength and ultrasonic pulse velocity ofrice husk ash, limestone powder, and rice husk ash/limestone powder mixtures.
for all mixtures). The relationships for each mineral admixtureare considered separately in Fig. 13. The correlation between com-pressive strength and ultrasonic pulse velocity was 0.9403 in theRHA mixtures, 0.8839 in the LS mixtures, and 0.8256 in the com-bined mixtures.
As shown in Fig. 12, there was a strong correlation (R2 = 0.9119)between ultrasonic pulse velocity and compressive strength for allof the SCC mixtures. The effect of limestone powder on Portlandcement hydration suggests two distinct actions: the fine limestoneparticles act as nucleation sites for calcium silicate hydrates, there-by accelerating the strength development at early ages, and car-boaluminate compounds are produced by reaction with thealuminate phases (particularly the C3A), and these have somecementitious properties. The magnitude of this effect depends onthe C3A content of the cement. There is also some reaction withthe C3S and C2S phases producing carbosilicate hydrates [22]. How-ever, addition of limestone reduces the compressive strength dueto dilution of pozzolanic reactions [38].
4. Discussion
In order to meet workability targets, self-compacting concreteusing high volumes of untreated rice husk ash as a fine aggregaterequired large amounts of water, mainly because of the increasedvolume fraction and surface area of the binder in the presence ofRHA. In particular, the RHA-induced increase in binder surface areawas substantial. The increased surface area adsorbs a greateramount of water, thus decreasing the quantity of free water avail-able in the mixture. However, the extent of flow spreading is alsodependent on particle characteristics such as size, shape, surfacearea, and porosity [18]. Limestone powder has a greater surfacearea than RHA, and the smooth texture and spherical shape ofthe LS particles are also important in determining the workabilitycharacteristics. Utilization of limestone powder with untreated ricehusk ash significantly improved the properties of the concrete.
The use of fillers is intended to enhance the particle distributionof the powder skeleton, reducing interparticle friction and ensuringgreater packing density. This can promote release of a portion ofthe mixing water that would otherwise be entrapped in the system[25]. The physical effect of limestone filler in highly fluid mortarmixtures such as those investigated in this paper is related to thewater–binder ratio (Fig. 14a and b). Mixtures containing morelimestone powder than rice husk ash had lower water–binder ra-tios and unit weights. The water–binder ratio controls the amountof free space in the system in terms of void volume and the amountof fine material required to fill the voids. Void filling in packed sys-tems may improve the particle arrangement, ensuring better waterdistribution and adequate fluidity. However, substantial increasesin viscosity and unit weight occur at the concentration at whichclose packing is reached. The increase of viscosity beyond this limitmay be explained by an increase in inter-particle friction due to in-creased solid–solid contact. In summary, the flow properties ofself-compacting concrete depend heavily on powder particle size,shape, surface morphology, and internal porosity in addition to fac-tors such as mixing regimen, sequence of admixture addition, andwater/superplasticizer content [39].
With suitable proportions of amorphous rice husk ash and lime-stone powder the compressive strength was increased can in-creases the strength, mostly due to the micro-filling ability andpozzolanic activity of RHA [7,40]. The calcium carbonate in LS re-acts very little with cement hydrates, and improvement in strengthare essentially due to void filling and acting as nucleation sites forcement hydrate crystals [24], mechanically improving the micro-structure of the bulk paste matrix and transition zone and leadingto increased compressive strength.
(a) Water-binder ratios and ash contents.
(b) Unit weight and ash content compared to control mixture.
05
101520253035404550
0.2
0.33
0.38
0.43
0.48
0.45
0.64
0.64
0.53
0.85
0.94
1.17
RHA
LS
05
101520253035404550
100
99.5
97.9
102.9
91.9
98.6
107.289.6
100.6
109.5
88.7
100.0
97.6RHA
LS
Fig. 14. Schematic depiction of relationship between rice husk ash and limestonepowder.
y = 86.088e-1.764x
R² = 0.9593
0
10
20
30
40
50
60
70
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4
Water-binder ratio (w/b)
Com
pres
sive
str
engt
h (M
Pa)
Fig. 15. Relationship between compressive strength and water–binder ratio.
G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464 463
To achieve a satisfactory combination of high workability andhardened properties, SCC requires high powder volumes at rela-tively low water–binder ratios (Fig. 15). There was a strong corre-lation between strength and water–binder ratio (R2 = 0.9593).
5. Conclusions
Based on our investigation of fine aggregate replacement withuntreated rice husk ash and limestone powder under conditionsof controlled slump, we drew the following conclusions:
� To maintain constant flowability, addition of RHA alonerequired an increase in water–binder ratio. Combinations ofRHA with LS decreased the water–binder ratio by more than28%.� The unit weight of the SCC decreased with increasing RHA con-
tent and increased with increasing LS content. Combinations ofRHA and LS were lighter than the control and LS mixtures.� Combined LS/RHA mixtures exhibited some blocking in the V-
funnel test and an increase in J-ring step height, but overall pro-vided improved workability.� The compressive strength decreased at higher water–binder
ratios and increased RHA or LS content. Suitable levels of fineaggregate replacement by RHA and LS provide development ofhigh compressive at early ages due to filling effects and pozzo-lanic reactions.
References
[1] Okamura H, Ouchi M. Self-compacting concrete. J Adv Concr Technol2003;1(1):5–15.
[2] Barbhuiya S. Effects of fly ash and dolomite powder on the properties of self-compacting concrete. Constr Build Mater 2011;25:3301–5.
[3] Schwartzentruber D, Roy RL, Cordin J. Rheological behaviour of fresh cementpastes formulated from a self compacting concrete (SCC). Cem Concr Res2006;36:1203–13.
[4] Chandra S, Björnström J. Influence of cement and superplasticizers type anddosage on the fluidity of cement mortar – Part I. Cem Concr Res 2002;32:1605–11.
[5] Khayat KH. Workability, testing, and performance of self-consolidatingconcrete. ACI Mater J 1999;96(3):346–53.
[6] Gesoglu M, Güneyisi E, Özbay E. Properties of self-compacting concretes madewith binary, ternary, and quaternary cementitious blends of fly ash, blastfurnace slag, and silica fume. Constr Build Mater 2009;23:1847–54.
[7] Liu M. Self-compacting concrete with different levels of pulverized fuel ash.Constr Build Mater 2010;24:1245–52.
[8] Uysal M, Yilmaz K. Effect of mineral admixtures on properties of self-compacting concrete. Cem Concr Compos 2011;33:771–6.
[9] S�ahmaran M, Christianto HA, Yaman ÏÖ. The effect of chemical admixtures andmineral additives on the properties of self-compacting mortars. Cem ConcrCompos 2006;28:432–40.
[10] Akram T, Memon SA, Obaid H. Production of low cost self compacting concreteusing bagasse ash. Constr Build Mater 2009;23:703–12.
[11] Gowda MR, Narasimhan MC, Karisiddappa CR. Development and study of thestrength of self-compacting mortar mixes using local materials. J Mater CivEng 2011;23:526–32.
[12] Memon SA, Shaikh MA, Akbar H. Utilization of rice husk ash as viscositymodifying agent in self compacting concrete. Constr Build Mater2011;25:1044–8.
[13] Nehdi M, Duquette J, Damatty AE. Performance of rice husk ash producedusing a new technology as a mineral admixture in concrete. Cem Concr Res2003;2003(33):1203–10.
[14] Zain MFM, Islam MN, Mahmud F, Jamil M. Production of rice husk ash for usein concrete as a supplementary cementitious material. Constr Build Mater2011;25:798–805.
[15] Chao-Lung H, Anh-Tuan BL, Chun-Tsun C. Effect of rice husk ash on thestrength and durability characteristics of concrete. Constr Build Mater2011;25:3768–72.
[16] Zerbino R, Giaccio G, Isaia GC. Concrete incorporating rice-husk ash withoutprocessing. Constr Build Mater 2011;25:371–8.
[17] Safiuddin MD, West JS, Soudki KA. Hardened properties of self-consolidatinghigh performance concrete including rice husk ash. Cem Concr Compos2010;32:708–17.
[18] Safiuddin MD, West JS, Soudki KA. Flowing ability of the mortars formulatedfrom self-compacting concretes incorporating rice husk ash. Constr BuildMater 2011;25:973–8.
[19] Sua-iam G, Makul N. The use of residual rice husk ash from thermal powerplant as cement replacement material in producing self-compacting concrete.Adv Mater Res 2012;415–417:1490–5.
[20] Memon SA, Shaikh MA, Akbar H. Production of low cost self compactingconcrete using rice husk ash, first int conf on const in devctries. Pakistan: Karachi; 2008. p. 260–9.
464 G. Sua-iam, N. Makul / Construction and Building Materials 38 (2013) 455–464
[21] Domone PL. Self-compacting concrete: an analysis of 11 years of case studies.Cem Concr Compos 2006;28:197–208.
[22] Domone PL. A review of the hardened mechanical properties of self-compacting concrete. Cem Concr Compos 2007;29:1–12.
[23] Valcuende M, Marco E, Parra C, Serna P. Influence of limestone filler andviscosity-modifying admixture on the shrinkage of self-compacting concrete.Cem Concr Res 2012;42:583–92.
[24] Felekoglu B. Utilisation of high volume of limestone quarry wastes in concreteindustry (self-compacting concrete cases). Resour Conserv Recycl2007;51:770–91.
[25] Yahia A, Tanimura M, Shimoyama Y. Rheological properties of highly flowablemortar containing limestone filler-effect of powder content and W/C ratio.Cem Concr Res 2005;35:532–9.
[26] ASTM C150/C150M-11. Standard specification for Portland cement. Annualbook of ASTM standards, vol. 04.01. Philadelphia (USA): American Society forTesting and Materials; 2011.
[27] ASTM C494/C494M-11. Standard specification for chemical admixtures forconcrete. Annual book of ASTM standards, vol. 04.01. Philadelphia (USA):American Society for Testing and Materials; 2011.
[28] ASTM C33/C33M-11. Standard specification for concrete aggregates. Annualbook of ASTM standards, vol. 04.02. Philadelphia (USA): American Society forTesting and Materials; 2011.
[29] Liu M. Incorporating ground glass in self-compacting concrete. Constr BuildMater 2011;25:919–25.
[30] ASTM C29/C29M-09. Standard test method for bulk density (‘‘unit weight’’)and voids in aggregate. Annual book of ASTM standards, vol. 04.02.Philadelphia (USA): American Society for Testing and Materials; 2009.
[31] ASTM C231/C231M-10. Standard test method for air content of freshly mixedconcrete by the pressure method. Annual book of ASTM standards, vol. 04.02.Philadelphia (USA): American Society for Testing and Materials; 2011.
[32] ASTM C1611/C1611M-09b. Standard test method for slump flow of self-consolidating concrete. Annual book of ASTM standards, vol. 04.02.Philadelphia (USA): American Society for Testing and Materials; 2011.
[33] ASTM C1621/C1621M-09b. Standard test method for passing ability of self-consolidating concrete by J-ring. Annual book of ASTM standards, vol. 04.02.Philadelphia (USA): American Society for Testing and Materials; 2011.
[34] EFNARC. Specifications and guidelines for self-compacting concrete. February;2002.
[35] ASTM C597/C597M-09. Standard test method for pulse velocity throughconcrete. Annual book of ASTM standards, vol. 04.02. Philadelphia (USA):American Society for Testing and Materials; 2009.
[36] ASTM C39/C39M-1. Standard test method for compressive strength ofcylindrical concrete specimens. Annual book of ASTM standards, vol. 04.02.Philadelphia (USA): American Society for Testing and Materials; 2009.
[37] Makhloufi Z, Kadri EH, Bouhicha M, Benaissa A. Resistance of limestone mortarwith quaternary binders to sulfuric acid solution. Constr Build Mater2012;26:497–504.
[38] Heikal M, El-Didamony H, Morsy MS. Limestone-filled pozzolanic cement. CemConcr Res 2000;30:1827–34.
[39] Rizwan SA, Bier TA. Blends of limestone powder and fly-ash enhance theresponse of self-compacting mortars. Constr Build Mater 2012;27:398–403.
[40] Qijun Y, Sawayama K, Sugita S, Shoya M, Isojima Y. The reaction between ricehusk ash and Ca(OH)2 solution and the nature of its product. Cem Concr Res1999;29:37–43.
