You are on page 1of 10

Construction and Building Materials 23 (2009) 1201–1210

Contents lists available at ScienceDirect

Construction and Building Materials


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

A preliminary concrete mix design for SCC with marble powders


Kürsßat Esat Alyamaç, Ragip Ince *
Firat University, Engineering Faculty, Civil Engineering Department, 23279 Elazig, Turkey

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

Article history: The marble has been commonly used as a building material since ancient times. Disposal of the waste
Received 27 September 2007 materials of the marble industry, consisting of very fine powders, is one of the environmental problems
Received in revised form 15 August 2008 worldwide today. However, these waste materials can be successfully and economically utilized to
Accepted 16 August 2008
improve some properties of fresh and hardened self-compacting concrete (SCC).
Available online 24 September 2008
The aim of this study is to find some relationship between properties of the fresh SCC and the hardened
SCC containing marble powder. For this purpose, the mix design approach based on monogram developed
Keywords:
by Monteiro and co-workers for normal vibrated concrete was adapted to SCC mixes. In order to obtain
Self-compacting concrete
Marble powder
this monogram, a series of SCC mixes with different water/cement ratios and water/powder ratios were
Waste prepared. Several tests such as slump-flow, T500 time, L-box, V-funnel and sieve segregation resistance
Strength were applied for fresh concrete and tests such as compressive strength and split-tension strength at 7,
28 and 90 days were performed for hardened concrete. In conclusion, the mix design method based on
monogram can be suggested for preliminary design in SCC.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction contributing to economy, decreasing the waste materials and


investing for the future [2]. The self-compacting concrete technol-
All natural stones that industrially can be processed as cut to ogy has a big potential for this type solid waste materials.
size, polished, used for decorative purposes and economically valu- Self-compacting concrete (SCC) is a special very liquid concrete
able are called as marble. USA, Belgium, France, Spain, Sweden, type that can settle in to the heavily reinforced, narrow and deep
Italy, Egypt, Portugal and Greece are among the countries with sections by its own weight, and can consolidate itself without
considerable marble reserve [1]. Turkey has the 40% of total marble necessitating internal or external vibration, and while providing
reserve in the world. 7,000,000 tons of marble have been produced with these features can keep its cohesion (stability) without
in Turkey annually and 75% of these production have been pro- leading segregation and bleeding. This concrete type developed
cessed in nearly 5000 processing plants. It can be apparently seen in Japan in 1980s with the progressions in the concrete technolo-
that the waste materials of these plants reach millions of tons. gies has become widespread in all over the world. Especially the
Stocking of these waste materials is impossible. developments in the superplasticizer technology have contributed
In marble quarries, the stones are being cut as blocks via differ- considerably to formation and progression of the self-compacting
ent methods (Fig. 1a). These blocks are being moved to processing concrete [3,4].
plants. In these plants, the blocks with 15–20 tons weight are being Different from the classical concrete design, the self-compact-
cut to size as decorative tiles and being polished. During the cut- ing concrete needs the superplasticizers, viscosity increasing
ting process, the dust of the marble and water mixes together addition and inert or pozzolanic mineral additions in big quantity
and become waste marble mud. The material that become dry all together or partly. New experiment techniques, design meth-
mud after being refined within the refinement facilities are too ods and ergo standards relating the selection of these materials
big for stocking and becoming harmful for the environment day and usage of them in proper ratios in concrete design is being
by day (Fig. 1b). During the cutting process 20–30% of the marble developed [5–10]. The expected performance criteria for the
block become dust. self-compacting concrete are self-compacting when it is fresh,
These type solid waste materials should be inactivated properly high early age strength that can stand for the early negative ef-
without polluting the environment. The most suitable inactivating fects and durability to the all external effects in hardened
method nowadays is recycling. Recycling provides with some situation.
advantages such as protecting the natural resources, energy saving, Numerous experimental tests on SCC revealed that the SCC
mixes containing inert fine powders such as limestone and chalk
* Corresponding author. Tel.: +90 424 2370000x5402; fax: +90 424 2415526.
have good fresh properties, excellent surface finish and hardening
E-mail address: rince@firat.edu.tr (R. Ince). strength higher than expected such as compressive strength and

0950-0618/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2008.08.012
1202 K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210

splitting tensile strength same as the pozzolanic fine powders such This method has based on three laws: the Abrams’ law [12],
as fly ash, blast-furnace slag and silica fume. This is mainly due to Lyse’s law [13] and Molinari’s law [14]. In the experimental pro-
improved particle packing and water retention of the fresh mixes gram of the study, a series of SCC mixes with different water-to-
and chemical reactions involving cement hydrates and calcium cement and water-to-powder ratios were prepared by using three
carbonate [8]. type marble powder: cherry (Rosso Levanto), gold and white.
This paper presents a study combining the properties of fresh Then, several tests such as slump-flow, T500 time test, L-box, V-
SCC and hardened SCC in one graph. For this purpose, the mix de- funnel and sieve segregation resistance were applied for fresh
sign method based on monogram devised by Monteiro and co- concrete and tests such as compressive strength and split-tension
workers [11] for ordinary concrete was modified to SCC mixes. strength at 7, 28 and 90 days were performed for hardened con-
crete cube specimens. Consequently, a monogram was developed
using the data obtained from 47 experimental programs for pre-
liminary design in SCC.

2. Concrete mix design monogram

There are many methods for concrete mix design with reference
to compressive strength. Since it is combining the properties of
fresh and hardened cementitious material in one graph, the mix
design method proposed by Monteiro and co-workers [11] is
highly useful for preliminary design. Fig. 2 illustrates a typical
mix design monogram for constant water/cement ratio. This
monogram utilizes the three relationships below:

Fig. 1. (a) A marble quarry and (b) marble powder waste. Fig. 3. Gradation curves of granular materials.

Fig. 2. Mix design monogram for a given water/cement ratio [11].


K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210 1203

2.1. Abrams’ law K1


fc ¼ w=c
ð1Þ
K2
One of the most important parameters influencing the behavior
of the concrete is water–cement ratio (w/c) used in a concrete mix. where fc is the compressive strength at fixed age and w/c is by
Basically, concrete made with high w/c ratios will have low proper- weight. K1 and K2 in Eq. (1) are the empirical constants which de-
ties such as low compression strength and low tension strength. pend usually on curing, test conditions, test age and cement proper-
Considering the compressive strength, this relation was first ob- ties [15,16].
tained by Abrams [12] in 1918 and has common use all over the
world. The so-called Abrams’ law is expressed as 2.2. Lyse’s law

This law essentially gives a relationship between water content


and maximum aggregate size for a specific workability value
Table 1
Physical and chemical properties of marble powders used
(slump) [13]. It is well known that the cement requirement in con-
crete decreases with increasing maximum aggregate size. On the
Properties Cherry Gold White other hand, for a particular water/cement ratio, the compressive
Specific gravity 2.71 2.71 2.71 strength of concrete tends to increase with increasing aggregate
Specific surface area (cm2/g) 3924 5106 4372 cement ratio. Consequently, the following linear relationship be-
CaO (%) 40.45 49.53 54.55
tween water–cement ratio (w/c) and aggregate-cement ratio by
SiO2 (%) 28.35 1.25 0.14
Fe2O3 (%) 9.70 0.32 0.32 weight (m) can be written as:
MgO (%) 16.25 0.40 4.17
m ¼ K 3 ðw=cÞ þ K 4 ð2Þ

Fig. 4. SEM of cherry marble powder.

Fig. 5. SEM of gold marble powder.


1204 K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210

in which K3 and K4 are the empirical constants which depend on the segregation is to remain homogeneity of concrete without separat-
workability properties of cementitious materials. ing of grout from the mix. Although several methods have been
used for self-compacting concrete in order to characterize the fresh
2.3. Molinari’s law state of the resulting concrete, there is no single test that can ade-
quately measure for workability properties. On the other hand,
It is necessary to develop a relationship between aggregate/ce- there is no unique standard test method measuring the workability
ment ratio (m) and cement content (C) in order to complete the properties of SCC. In this study, the standard test methods used
monogram in Fig. 2 [14]. Molinari has proposed the following in- according to EFNARC [9] are given:
versely proportional relationship;
3.2.1. Slump-flow and T500 time test
1000
C¼ ð3Þ This test method, based on the traditional slump test, is utilized
K5m þ K6
to determine the flowability and flow rate of SCC. The test equip-
where K5 and K6 are the material constants. ments are one slump cone and one flow table, as shown in Fig. 7.

3. Experimental studies

3.1. Materials

According to EN 197-1, CEM I 42.5 N was used in all mixes. Its


specific gravity, specific surface area by Blain, and 28 days com-
pressive strength were 3.1, 3393 cm2/g and 49.2 MPa respectively.
The maximum aggregate size was 16 mm. The maximum sand
grain size was 4 mm. Mineralogically, the aggregate consisted of
river. The gradation curves of the granular materials are shown
in Fig. 3. The aggregate and sand were air-dried prior to mixing.
The superplasticizer ViscoCrete-3075 was used in order to pro-
duce SCC for all mixes. Three types of marble powder were utilized
to obtain SCC mixes. Their physical and chemical properties are gi-
ven in Table 1. Typical SEM photographs are illustrated at 5000
and 10,000 magnifications for each types of marble powder in
Figs. 4–6. As shown in the Figs. 4–6, these marble powders have
angular shapes with a rough surface texture unlike the fly ash par-
ticles that have spherical shape. On the other hand, cherry and
white marble powders seem to be coarser than the gold marble
powder. This can be also seen from the specific surface area values
of these marble powders in Table 1.

3.2. Test equipment

Self-consolidation characteristics are related to the workability


properties: filling ability, passing ability and, the resistance to seg-
regation. Filling ability is the capability of completely filling all
spaces without vibration. Passing ability is the aptitude to flow
through reinforcement bars without any blocking. Resistance to Fig. 7. Test equipments according to EFNARC [9].

Fig. 6. SEM of white marble powder.


K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210 1205

The slump cone is filled with concrete and then lifted vertically. diameter to reach 500 mm (T500) and the final mean diameter of
Two characteristic values are measured; time for the concrete concrete (dm).

Table 2 3.2.2. V-funnel test


Classes according to fresh SCC properties expressed by different test methods [9]
The test is utilized to determine filling ability and viscosity of
Slump-flow Viscosity (flow Viscosity (V- Passing Segregation SCC. The test equipment is a V-shaped funnel, as detailed in
(flow test) test) funnel test) ability (L-box resistance Fig. 7. The V-funnel is filled with concrete and then its gate is
test) (Sieve test)
opened. Time for the concrete to flow out of the funnel (tv) is
Class dm (cm) Class T500 (s) Class tv (s) Class PA Class SR (%) determined.
SF1 55–65 VS1 62 VF1 68 PA1 P0.8a SR1 620
SF2 66–75 VS2 >2 VF2 9–25 PA2 P0.8b SR2 615 3.2.3. L-box test
SF3 76–85 The test is utilized to determine flowability and passing ability
a
With 2 rebars. of SCC. The test equipment is an L-shaped box, as detailed in Fig. 7.
b
With 3 rebars. The vertical compartment of L-box is filled with concrete and then

Table 3
Experimental results: SCC mixes with cherry marble powders

Mix C (kg/m3) P (kg/m3) w/c sp (%) G (kg/m3) S (kg/m3) dmb (cm) T500 (s) tv (s) PA SR (%) fc (MPa) ft (MPa)
7d 28d 90d 7d 28d 90d
C1C 300 100 0.55 1 722 1095 30 – – – – 25.8 35.6 38.7 2.19 2.94 3.19
C2Ca 300 100 0.55 2 722 1095 57 2.92 4.1 0.89 1 30.7 39.1 41.8 2.63 3.19 3.47
C3Ca 300 100 0.55 3 722 1095 58 2.45 3.5 0.91 1 22.2 29.3 32.9 1.90 2.45 2.70
C4C 300 100 0.60 1 706 1071 43 – – – – 25.8 33.3 36.0 2.19 2.80 3.01
C5Ca 300 100 0.60 2 706 1071 60 2.38 3.4 0.95 7 28.9 37.8 41.3 2.41 3.12 3.44
C6Ca 300 100 0.63 2 696 1055 66 2.34 3.1 0.96 12 27.6 36.4 39.6 2.34 3.05 3.26
C7Ca 300 100 0.67 2 685 1039 68 2.22 2.7 0.98 16 26.2 34.2 36.9 2.27 2.87 3.01
C8C 300 100 0.70 2 675 1023 71 1.8 2.3 0.99 33 23.1 31.6 33.8 1.98 2.63 2.87
C9C 300 200 0.58 1 672 1020 [2] – – – – 29.3 40.0 45.8 2.48 3.33 3.71
C10C 300 200 0.58 2 672 1020 51 3 – – – 28.4 38.2 42.7 2.45 3.16 3.54
C11Ca 300 200 0.60 2 667 1012 62 3.2 6.1 0.91 4 29.3 40.9 44.9 2.45 3.33 3.75
C12Ca 300 200 0.63 2 657 996 68 2.9 5.5 0.94 7 28.0 38.7 41.8 2.37 3.16 3.50
C13Ca 300 200 0.63 3 657 996 69 2.5 5.2 0.93 7 28.0 37.8 41.3 2.37 3.16 3.37
C14Ca 300 200 0.67 2 646 980 71 2.7 4.7 0.96 9 26.7 37.3 40.4 2.23 3.09 3.33
C15Ca 300 200 0.70 2 636 964 73 2.4 4.4 0.97 13 25.3 34.7 38.2 2.19 2.91 3.19
C16C 300 250 0.57 2 658 999 50 5.4 – – – 24.0 35.6 37.3 2.01 2.94 3.12
C17Ca 300 250 0.60 2 648 983 58 4 7 0.89 2 25.3 37.8 39.1 2.16 3.16 3.23
C18C 300 250 0.63 1 637 967 [2] – – – – 23.6 34.7 36.4 2.05 2.87 2.98
C19Ca 300 250 0.63 2 637 967 65 3.6 6.4 0.91 4 24.4 37.3 39.6 2.09 3.09 3.33
C20Ca 300 250 0.63 3 637 967 66 3 6.2 0.92 5 25.8 37.3 39.1 2.16 3.05 3.30
C21C 300 250 0.67 1 627 951 [4] – – – – 24.0 35.6 37.8 2.05 2.94 3.19
C22Ca 300 250 0.67 2 627 951 66 3.4 5.7 0.94 7 24.4 35.6 38.2 2.09 2.98 3.12
C23Ca 300 250 0.67 3 627 951 68 2.9 5.4 0.94 7 23.6 35.1 36.4 1.98 2.91 3.01
C24Ca 300 250 0.70 2 616 935 69 3.2 5.3 0.97 10 23.1 33.8 35.6 2.01 2.84 2.94
C25Ca 300 250 0.70 3 616 935 71 2.7 5.2 0.97 16 21.8 32.0 33.8 1.90 2.70 2.80
C26C 300 500 0.70 2 519 787 [20] – – – – 18.2 25.3 26.7 1.57 2.12 2.30
C27Ca 350 50 0.60 2 677 1027 63 2.2 3.6 0.96 8 27.6 38.7 41.3 2.30 3.16 3.44
C28Ca 350 150 0.60 2 638 968 66 2.9 4.2 0.93 5 29.3 40.9 43.6 2.48 3.37 3.54
C29Ca 350 200 0.60 2 619 938 62 3.3 4.9 0.92 4 25.3 36.4 39.1 2.16 3.01 3.19
C30Ca 400 0 0.45 2 711 1079 62 2.75 5.4 0.92 9 38.7 55.6 61.8 3.26 4.47 5.01
C31Ca 400 0 0.48 2 701 1063 67 2.1 5.1 0.95 12 34.7 52.0 60.0 2.91 4.23 4.81
C32Ca 400 0 0.50 2 690 1047 68 1.5 4.5 0.97 14 33.8 49.3 54.7 2.80 4.06 4.40
C33Ca 400 50 0.51 2 667 1013 66 1.8 5 0.94 11 35.1 50.2 56.0 2.91 4.06 4.60
C34Ca 400 100 0.45 2 672 1020 55 2.69 6.5 0.88 8 42.7 53.8 64.4 3.57 4.37 5.25
C35Ca 400 100 0.48 2 662 1004 58 2.3 6.1 0.91 12 40.9 50.7 63.1 3.40 4.13 5.08
C36Ca 400 100 0.53 2 641 972 69 1.8 4.9 0.97 14 36.9 46.2 56.9 3.12 3.82 4.60
C37Ca 400 100 0.55 2 630 956 73 1.1 4.2 0.99 17 30.7 39.6 43.6 2.55 3.23 3.61
C38Ca 400 150 0.45 2 653 990 58 3.7 5.3 0.88 2 42.7 57.3 65.8 3.54 4.64 5.31
C39C 400 150 0.50 1 632 958 44 – – – – 29.3 40.0 45.8 2.45 3.26 3.82
C40Ca 400 150 0.50 2 632 958 66 3.1 4.1 0.93 4 41.3 54.2 61.3 3.44 4.40 5.01
C41Ca 400 150 0.50 3 632 958 71 2.2 3.9 0.94 5 39.1 52.0 58.7 3.30 4.26 4.71
C42Ca 400 150 0.53 2 621 942 68 2.8 3.6 0.94 9 38.2 51.1 57.8 3.16 4.20 4.67
C43Ca 400 150 0.55 2 611 926 72 2.2 3.3 0.96 14 37.3 48.9 52.9 3.12 3.95 4.30
C44C 400 150 0.58 2 600 910 76 1.7 2.7 0.99 23 32.0 42.7 45.8 2.63 3.50 3.75
C45C 400 400 0.53 2 524 795 51 6.6 – – – 20.9 30.7 33.3 1.76 2.52 2.84
C46Ca 450 50 0.44 2 654 992 67 2.4 5.9 0.95 13 39.6 52.4 61.8 3.23 4.20 5.05
C47Ca 450 100 0.44 2 634 962 70 3.1 6.4 0.93 10 42.2 55.1 64.4 3.47 4.43 5.15
C48Ca 500 0 0.42 2 646 979 70 1.13 3.5 0.96 14 43.6 57.8 67.1 3.64 4.64 5.42
C49C 500 50 0.36 2 658 998 52 4.47 – – 11 54.7 67.1 73.8 4.47 5.38 5.88
C50Ca 500 50 0.38 2 647 982 68 2.03 4.6 0.93 13 52.0 65.3 71.1 4.30 5.28 5.65
C51Ca 500 50 0.40 2 637 966 71 1.56 4.3 0.96 16 50.7 62.2 67.6 4.13 5.01 5.48
C52Ca 500 50 0.42 2 626 950 73 1.37 3.9 0.97 18 46.7 60.0 64.4 3.85 4.88 5.15
C53C 500 50 0.44 2 616 934 77 0.8 3.3 0.99 27 39.1 49.8 53.3 3.19 4.09 4.30
C54C 500 300 0.42 2 529 802 56 5.6 – – – 31.6 44.4 48.4 2.63 3.61 3.95
a
These mixes are provided with the properties of SCC.
b
The values in bracket indicate slump values of fresh concrete.
1206 K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210

the sliding gate is lifted. The concrete flows through the steel bars contents varied between 0 and 400 kg/m3 and the water/cement
into the horizontal part. The distances H1 and H2 are measured ratios varied from 0.36 to 0.70. The superplasticizer was used in
when the concrete is stopped. The value of passing ability different proportions as 1%, 2% and 3% of the cement content, ex-
PA = H2/H1 is calculated. cept for references mixes.
All specimens in each series were cast from the same batch of
3.2.4. Sieve segregation resistance test concrete. The experimental results for SCC mixes with cherry pow-
This is utilized to determine resistance of SCC to segregation. ders and for SCC mixes with white and gold powders are given in
The test equipment is the sieve with diameter frame = 500 mm Table 3 and Table 4, respectively. The reference mixes are pre-
and sieve size = 5 mm. After placing 4.8 ± 0.2 kg concrete sample sented in Table 5. The first column in Table 3 and 4 gives the
into the sieve, the fresh SCC is allowed to stand for 2 min on the mix numbers, namely C1 to C54 for SCC trial mixes and REF1 to
sieve. The segregated portion (SR) is calculated as the proportion REF5 for reference mixes. The letters following the mix numbers
of the sample passing through the sieve. in the trial mixes in the first column are depicting the marble pow-
The slump-flow test is quick while the sieve segregation resis- der types used in the batches; C for Cherry, W for White and G for
tance test is slow. On the other hand, these tests can be performed Gold. The mixture proportions are listed on the columns of Tables 3
by single operator. There is good correlation between the sieve seg- and 4 from second to seventh, where C, P, w/c, sp, G and S indicate
regation resistance test and the behavior in full scale structures. cement content, powder content, water/cement ratio, percent of
Moreover the slump-flow test has good correlation with L-box test. superplasticizer by weight of cement, aggregate content and sand
V-funnel and L-box tests are good indicator of viscosity [17]. In Ta- content, respectively. The properties of fresh concrete prepared
ble 2, the classifications recommended by EFNARC [9] are given for according to the procedure proposed by EFNARC committee proce-
the properties of fresh SCC obtained from different test methods. dure are shown in columns of Tables 3 and 4 from eighth to
twelfth. According to the criterias in Table 2, 53 of the 67 total
3.3. Mix details and test procedure SCC trial mixes are provided with the properties of SCC. These
mixes are represented with an asterisk in Table 3. Besides, all of
Experimental studies have revealed that the properties of a the SCC mixes containing gold and white marble powders are pro-
fresh SCC are particularly influenced by two material parameters: vided with the properties of SCC in Table 4. None of the trial mixes
water–cement ratio (w/c) and water–powder ratio (w/p) [8–10]. containing 1% superplasticizer were illustrated the properties of
Several mix design methods have been proposed at academic mili- self-compatibility, as shown in Table 3.
eu and institutions for SCC [4–6,8–10]. Not only the properties of To determine mechanical properties of hardened concrete, com-
fresh concrete but also the properties of hardened concrete are pressive strength and split-tension strength tests were applied at
hereby considered in order to find a general expression of the 7, 28 and 90 days. The test specimens were used as 150 mm cubes.
SCC mix design. Specimens in each series were cast in plastic moulds. Specimens
Totally seventy-three series of mix proportions, of which five were removed from the mold after 1 day and subsequently, the
series were the references mixes, were tested concerning two ones with test age of 7 days were cured at 7 days and the ones with
essential material parameters (w/c and w/p) in different combina- test age of 28 and 90 days were cured at 28 days in a moist room of
tions and for three different marble powder types. The cement con- 95% relative humidity and temperature of about 23 °C. All the spec-
tents varied from 300 kg/m3 to 500 kg/m3 whereas the powder imens were tested in a testing machine with the capacity of

Table 4
Experimental results: SCC mixes with white and gold marble powders

Mix C (kg/m3) P (kg/m3) w/c sp (%) G (kg/m3) S (kg/m3) dm (cm) T500 (s) tv (s) PA SR (%) fc (MPa) ft (MPa)
7d 28d 90d 7d 28d 90d
C7W 300 100 0.67 2 685 1039 66 2.4 3 0.93 12 25.7 34.0 36.5 2.21 2.72 2.97
C7G 300 100 0.67 2 685 1039 68 2.3 2.8 0.96 15 25.2 33.7 36.3 2.09 2.60 2.72
C12W 300 200 0.63 2 657 996 65 3.2 5.9 0.91 5 27.8 38.5 41.4 2.21 2.97 3.34
C12G 300 200 0.63 2 657 996 67 3 5.7 0.94 6 27.2 38.4 41.3 2.21 2.85 3.10
C24W 300 250 0.70 2 616 935 67 3.7 5.7 0.93 7 22.8 33.5 35.4 1.83 2.85 2.97
C24G 300 250 0.70 2 616 935 68 3.4 5.4 0.96 9 22.6 33.0 35.1 1.83 2.72 2.72
C35W 400 100 0.48 2 662 1004 56 2.5 6.4 0.93 9 40.7 50.4 62.7 3.10 3.96 4.80
C35G 400 100 0.48 2 662 1004 57 2.3 6.3 0.91 11 40.5 50.3 62.1 3.10 3.83 4.68
C38W 400 150 0.45 2 653 990 55 3.4 5.6 0.87 1 42.7 57.0 65.4 3.47 4.44 5.16
C38G 400 150 0.45 2 653 990 58 3.7 5.4 0.88 1 42.4 56.9 65.2 3.22 4.32 4.92
C43W 400 150 0.55 2 611 926 72 2.6 3.6 0.93 12 36.8 48.3 52.4 3.10 3.83 4.20
C43G 400 150 0.55 2 611 926 69 2.3 3.3 0.95 13 36.4 47.9 52.2 2.97 3.83 4.08
C50W 500 50 0.38 2 647 982 65 2.6 5 0.9 12 51.5 64.5 70.3 4.08 4.92 5.40
C50G 500 50 0.38 2 647 982 68 2.2 4.7 0.92 13 50.9 63.9 69.8 3.96 4.92 5.16

Table 5
Experimental results: reference mixes

Mix C (kg/m3) w/c G (kg/m3) S (kg/m3) Slump (cm) fc (MPa) ft (MPa)


7d 28d 90d 7d 28d 90d
REF1 300 0.65 729 1106 15 20.4 30.2 32.4 1.76 2.52 2.73
REF2 350 0.60 697 1057 13 23.1 33.3 36.4 1.94 2.73 3.05
REF3 400 0.53 680 1031 9 27.1 37.3 39.1 2.23 3.01 3.30
REF4 450 0.47 663 1005 5 29.8 43.1 46.7 2.48 3.50 3.82
REF5 500 0.42 646 979 2 37.8 48.9 54.7 3.16 3.99 4.37
K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210 1207

2000 kN. The specimens were loaded monotonically until failure was 25% higher than the references concrete. A similar conclusion
and care was taken to apply a constant loading rate. Typically, it was observed by Zhu and Gibbs [8] for SCC mixes containing lime-
took about 5 min (±30 s) to reach the maximum load for each spec- stone and chalk powders.
imen. The compressive strength, fc and the split-tension strength, ft
at 7, 28, and 90 days of the specimens are given in the last six col- 5. Analysis of test results
umns of Table 3–5. The values of strength in these tables were ob-
tained from mean of at least two test results. Fig. 10 illustrates typical volume proportions of the ordinary
concrete and of the self-compacting concrete comparatively [10].
4. Experimental results As shown, both the aggregate content (fine + coarse) and the
water-to-powder ratio of SCC are smaller than that of traditional
The compressive strength and the split-tension strength mea- vibrated concrete. The powder contribution consists of cement
sured, as shown in Table 3, are slightly lower for SCC mixes con- and inert or pozzolanic additions in SCC. In the present work, mar-
taining 3% superplasticizer than for the ones containing 2% ble powder is used as additional material while it is assumed as the
superplasticizer. For this reason, the values of the mixes only con- mineral filler material in fine aggregate in the ordinary concrete.
taining 2% superplasticizer (a total of 47 series) were utilized in the Moreover, the properties of fresh SCC are, as already mentioned
analysis of mix design. Three powder types should have different above, particularly influenced by the two material parameters:
water requirements because they have different specific surface water–cement ratio and water–powder ratio. Nevertheless, the
areas. Nevertheless the flow values of the concretes obtained from water/powder ratio is not used directly since it is possible to pro-
these powder types are almost identical in the same water-to-ce- duce SCC mixes without powder. Deducting from the above discus-
ment ratio, same cement content, and same powder content, as sion, a relationship between powder/cement ratio and cement
illustrated in Fig. 8. This means that the effect of the powder type content should be constituted. Consequently, the Molinari’s law
can be negligible for the practical aims. (Eq. (3)), which is used for estimating the cement content in the
Fig. 9 shows the compressive strength at 28 days against the normal concrete, can be modified to the following form for SCC
water/cement ratio, for the considered 47 SCC mixes and 5 refer- mix:
ences mixes. In the same figure, the best-fit curves expressed by
1000
an exponential function are also given. This revealed that SCC C¼ ð4Þ
K5m þ K6n þ K7
mixes containing marble powder pointed out a higher strength at
a given water/cement ratio than the traditionally concrete mix. where n is the powder cement ratio by weight and, K5, K6 and K7 are
Furthermore, the mean strength of SCC mixes with marble powder the material constants. The constants in Eq. (4) can be determined
from the multiple linear regression made on Y = K5X1 + K6X2 + K7
with Y = 1000/C, X1 = m and X2 = n. Fig. 11 shows the individual test
data and the modified Molinari’s law calculated via Eq. (4) for differ-
ent ratios of powder/cement in the aggregate/cement ratio versus
cement content plane. It is derived from Fig. 11 that the considered
test results are very close to the modified Molinari’s law because
the determination coefficient R2 = 0.988.
The Lyse’s law in the mix design based on monogram proposed
by Monteiro et al. [11] uses the slump test for workability of con-
crete. Similarly, the slump-flow test, which is commonly proposed
for determining the workability properties of fresh SCC such as
flowability and flow rate at academic milieu and institutions for
SCC, was utilized in this study. The flow values in Table 3 and 4 var-
Fig. 8. Comparison of flow values of SCC with different marble powder. ied from 55 to 75 for the considered 47 data. According to EFNARC,
these are classified as: (1) SF1 (55–65 mm) and (2) SF2 (66–
75 mm). Fig. 11 illustrates the individual test data and Lyse’s law
calculated via Eq. (2) for two different slump-flow classes in the
aggregate/cement ratio versus water/cement ratio plane. The cor-
relation coefficients of Eq. (2) are r = 0.833 for flow 55–65 mm
(SF1) and r = 0.911 for 66–75 mm (SF2).
Abrams’ law (Eq. (1)) has been applied to some properties of
concrete such as the compressive strength, Young’s modulus and

Fig. 9. Comparison of SCC mixes with reference mixes. Fig. 10. Comparison of mix proportioning between SCC and ordinary mix.
1208 K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210

Fig. 11. Mix design monogram for SCC mixes with marble powder.

fracture energy [11]. In the present research, this law was also ap- powders available in the literature, the study by Zhu and Gibbs [8]
plied to the split-tension strength of concrete. The constants in Eq. has been simulated. The mixes in their study were designed for
(1) can be calculated as K 1 ¼ a and K 2 ¼ expðbÞ from the expo- water/cement ratios: 0.42 and 0.57 with crushed coarse aggregate
nential regression made on Y ¼ a ebX with Y = fc, X = w/c. The results of maximum size of 20 mm. In addition, CEM I 42.5 Portland cement
of analysis based on Abrams’ law for the compressive strength (fc) with a Blaine specific surface area of 3850 cm2/g, three types of
and the split-tension strength (ft) of the specimens at 7, 28 and 90 limestone powder and two types of chalk powder were utilized as
days are given in Fig. 11. In Fig. 11, the equations’ correlation coef- filler in their SCC mixes. Corresponding to the investigated series
ficients, at 7, 28 and 90 days, are respectively 0.965, 0.974 and of w/c values, values of slump-flow were ranging from 60 to
0.966 for the compressive strength and, 0.959, 0.964 and 0.958 65 cm. The compressive cube strength tests were performed at 7,
for the split-tension strength. 28 and 90 days and splitting cylindrical tests at 28 days. Since the
compressive cube strengths at 7 and 90 days are not reported clearly
6. Comparison with the previous studies in Ref [8], these mixes (SCC1 and SCC2) and only their average 28-
day strength values are summarized in Table 6 according to w/c val-
In the previous investigations, the effects of water/cement ra- ues. Consequently, it is found that the contributions to strength
tios, water/powder ratios, the powder type (inert or pozzolanic), gains were significantly greater for the limestone powder than for
the volume fraction of the solids in the mix, and the content of vis- the chalk powder. Besides, the diversity in the mean strength values
cosity modifier agents on the properties of fresh and hardened SCC are owing to usage of five different powders in Table 6.
were evaluated. However, general relationships between proper- Table 6 and Fig. 12 give results of the experiments and the ap-
ties of the fresh SCC and the hardened SCC containing inert/pozzo- proach based on the presented monogram, comparatively. It is
lanic powder have not been proposed. In this study, the mix design noted that only the SF1 slump-flow class and the strength proper-
approach based on monogram developed by Monteiro and co- ties at 28 days are taken into consideration in Fig. 12 different from
workers [11] for vibrated concrete was adapted to SCC mixes. Fig. 11. Although test data obtained from different laboratory for
In order to investigate how well the monogram proposed in this SCC with different powder types were utilizing in the comparison,
study would simulate the behavior of SCC with different types of Fig. 12 shows that Eq. (2) and Eq. (4) agree with previous test re-
K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210 1209

Table 6
Comparison design based on the monogram and the study of Zhu and Gibbs [8]

Mix Agg. (kg/m3) Cement (kg/m3) Powder (kg/m3) Experimental Predicted


a/c w/c fc (MPa) ft (MPa) a/c w/c fc (MPa) ft (MPa)
SCC1 1670 300 240 5.57 0.57 44–52 3.2–4.7 5.30 0.60 41 3.3
SCC2 1685 405 135 4.16 0.42 67–74 4.6–5.9 3.94 0.43 58 4.6

Fig. 12. Comparison of developed mix design monagram with study by Zhu and Gibbs [8].

sults quite well. Consequently, the modified Molinari’s Law (Eq. gram proposed by Monteiro and co-workers, the following
(4)) developed in this study seems to be useful for determining conclusions can be extracted:
the contribution of powder/cement ratio on the properties of fresh
SCC. 1. Although the marble have been commonly used as a building
Nevertheless, experimental results of the study by Zhu and material since ancient times, disposal of the waste materials
Gibbs have greater strength values than results of the monogram of the marble industry is one of the major problems for the
presented in this study. This is mainly due to the strength is influ- environmental concerns all over the world. However, this study
enced by not only water/cement ratio and aggregate/cement ratio emphasizes that these waste materials can be successfully and
but also properties of aggregate particles such as grading, surface economically utilized as additional inert filler material in SCC
texture, shape, strength, stiffness and maximum size [18]. How- technology.
ever, if the further researches will be performed, more reliable 2. The presented monogram is highly useful since its ability to
findings can be proposed. combine the properties of fresh SCC such as flowability and
properties of hardened SCC such as compressive strength and
7. Conclusions split-tension strength at 7, 28 and 90 days in one graph.
3. The results obtained from the comparison of the modified
From the findings of these experimental and statistical investi- Molinari’s Law (Eq. (4)) developed in this study and the previ-
gations for self-compacting concrete mix design based on mono- ous tests depicts that such a relationship seems to be useful
1210 K.E. Alyamaç, R. Ince / Construction and Building Materials 23 (2009) 1201–1210

for determining the contribution of powder/cement ratio on the solution to urban infrastructure problems through civil engineering
technology, Bangkok; 1992. p. 16.
properties of fresh SCC.
[4] Ozawa K, Maekawa K, Kunishima M, Okamura H. High-performance concrete
4. This monogram designed for two slump-flow classes: (1) SF1 based on the durability of concrete structures. In: Proceedings of the second
(55–65 mm) which is suitable for unreinforced/slightly rein- East Asia Pacific conference on structural engineering and construction; 1989.
forced concrete structures and concretes cast by a pomp injec- [5] Ghezal A, Khayat KH. Optimizing self-consolidating concrete with limestone
filler by using statistical factorial design methods. ACI Mater J
tion system, (2) SF2 (66–75 mm) which is appropriate for many 2002;99(3):264–72.
applications such as columns. In addition, the monogram allows [6] Su N, Hsu K, Chai H. A simple mix design method for self-compacting concrete.
performing mixes of SCC with different powder to cement Cem Concr Res 2001;31:1799–807.
[7] Bui V, Montgomery D, Hinczak I, Turner K. Rapid testing method for
ratios. Consequently, this also gives opportunity to obtain eco- segregation resistance of self-compacting concrete. Cem Concr Res
nomic mix design. 2002;32:1489–96.
[8] Zhu W, Gibbs JC. Use of different limestone and chalk powders in self-
compacting concrete. Cem Concr Res 2005;35:1457–62.
Acknowledgements [9] EFNARC. European guidelines for self-compacting concrete: specification,
production and use; 2005. p. 68.
[10] Okamura H, Ouchi M. Self-compacting concrete. J Adv Concr Tech
The authors would like to acknowledge the Scientific Research 2003;1(1):5–15.
Projects Administration Unit of Firat University (Project No: FU- [11] Monteiro PJM, Helene PRL, Kang SH. Designing concrete mixtures for strength,
BAP-1370) for the financial support. elastic modulus and fracture energy. Mater Struct 1993;26:443–52.
[12] Abrams DA. Design of concrete mixtures. Lewis Institute Bulletin 1918:1.
[13] Lyse I. Test on consistence and strength of concrete having constant water
References content. ASTM Proc 1932;32(1):629–36.
[14] Helene PRL. Contribuiçao ao estudo do controle e dosagem dos concretos de
[1] Onargan T, Köse H, Deliormanlı AH. Marble. 4th ed. Union of Chambers of cimento Portland. PhD Thesis. Sao Paulo University; 1987.
Turkish Engineers and Architects (UCTEA), Chamber of Mining Engineers of [15] Mindess SS, Young JF. Concrete. UK: Prentice-Hall International; 1981.
Turkey; 2006. p. 1–5 [in Turkish]. [16] Ince R, Alyamaç KE. Determination of fracture parameters of concrete based on
[2] Kaseva ME, Gupta SK. Recycling-an environmentally friendly and income water–cement ratio. Indian J Eng Mater Sci 2008;15:14–22.
generating activity-towards sustainable solid wastes management, case study- [17] Bartos PJM. Testing-SCC: towards new European standards for fresh SCC. In:
Dar es Salaam City, Tanzania. Resour Conserv Recycl 1996;17(4):299–309. Proceedings of the first international symposium on design, performance and
[3] Okamura H, Ozawa K, Maekawa K, Tangtermsinikul S. High-performance use of self-consolidating concrete, China; 2005.
concrete mechanism of super-fluidized concrete. EIT-JSCE-AIT joint seminar on [18] Neville AM. Properties of concrete. 4th ed. London: Longman; 1995.