Materials and Design 30 (2009) 3056–3065

Contents lists available at ScienceDirect

Materials and Design

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

Experimental investigation of some fresh and hardened properties

of rubberized self-compacting concrete
_
Ilker Bekir Topçu, Turhan Bilir *
Eskisßehir Osmangazi University, Department of Civil Engineering, 26480 Eskisßehir, Turkey

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

Article history: In this study, the usage of ground elastic wastes such as rubber in SCC is investigated. Rubber has
Received 2 July 2008 replaced aggregates at the contents of 60, 120 and 180 kg/m3 in SCC by weight. Four different mixture
Accepted 10 December 2008 proportions have been prepared. Moreover, 24 series have been produced by using six different viscosity
Available online 16 December 2008
agents in both SCC and rubberized self-compacted concrete (RSCC). By using these agents, it is attempted
to see the effects of them on the properties of RSCC. Fly ash (FA) is used as filler material. The slump–flow,
Keywords: V-funnel, compressive strength, high temperature and freezing–thawing resistances of RSCC have been
Composites
compared to the properties of ordinary SCC. At the end, it is observed that increase in RA content leads
Concrete
Elastomers and rubber
to increase in fresh properties of RSCC such as workability because of the existence of viscosity agents in
mixtures. It decreases the hardened properties such as compressive strength and durability. However, the
different viscosity agents can provide appropriate results for RSCC containing the same rubber aggregate
(RA) content and the hardened properties of RSCC are better than the properties of ordinary concrete
even if they are lower than the ones of SCC.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction compacted, however, this time, the amount of capillarity pores

and permeability may increase due to higher water–cement ratio.
Along with the development of ready-mixed concrete industry Naturally, the compressive strength and the durability of concrete
throughout the world, other industries have also shown progress decrease likewise the low water content case. Even if desired con-
related to this industry. One of the most important of them is the sistency or water demand is satisfied, the inelaborate transporting,
admixture sector. Concrete has gained improved properties with pouring, casting, placing, compacting and curing can also lead to
chemical and mineral admixtures. By the way, workability is de- decreased hardened concrete properties. In this manner, all these
fined in terms of the amount of mechanical work, energy required processes have to be carefully achieved during concrete production
to produce full compaction of the concrete without segregation [1]. by skilled workers [3,4]. After all, the potential for segregation and
If concrete is workable, it can be transported and poured easily loss of entrained air void can be said as the main problems in con-
without leading to segregation. It can also be cast, placed, com- crete production [3].
pacted and finished easily without causing any void in fresh con- Most of the problems such as mentioned above encountered by
crete. Herein, water content to obtain desired consistency in ready-mixed concrete producers and consumers during the pour-
fresh concrete is an important parameter to produce workable con- ing, casting, placement, compacting and finishing processes have
crete. Consistency is determined in terms of flow and increasing been solved by using such admixtures. Before the usage of chemi-
water content cause increase in flow. The concrete which flows cal admixtures in concrete, concrete mixture had high water–ce-
can also be easily cast and placed into formworks [1,2]. If the water ment ratios in order to develop the workability of concrete. This
content is low, the concrete cannot be transported, poured, cast, application has reduced the quality of concrete. At the early
placed and compacted without segregation due to the lack of water 1970s, it is known that concrete requiring little vibration or com-
for lubricating the particles [2]. Therefore, the amount, the distri- paction has begun to be used in Europe [4]. Besides, the plasticizers
bution, the shapes and the sizes of voids in concrete may change or super plasticizers such as lignosulfonate and potassium carbon-
and the compressive strength of concrete may decrease [1–3]. If ate to increase the workability of cement pastes with low water–
the water content is high, concrete can be easily cast and cement ratios have begun to be investigated [5,6]. Increasing the
flow rate of concrete has not completely solved the compaction
problems especially in intensively equipped reinforced concrete.
* Corresponding author. Tel.: +90 222 239 3750; fax: +90 222 239 3613. Higher workability has been obtained but the self-compacting
E-mail address: [email protected] (T. Bilir). and spreading properties could not be. Hence, in 1986, researches

0261-3069/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2008.12.011
 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B. 3057

have begun to be conducted on the self-compacting and diffused 2. Materials and methods
concrete at University of Tokyo. Therefore, the first step for design
of SCC was taken. The effects of chemical admixtures on the 2.1. Materials
development of SCC have been mentioned above. Besides these,
the filler materials have been used in SCC to increase the self-com- 2.1.1. Cement
pacting property by increasing the amount of finer material [7–10]. In this experimental study, CEM II/AL-LL 42.5 N Portland Calcar-
These materials are generally used as mineral admixtures in ordin- eous Cement corresponding TS EN 197-1 [22] standards has been
ary concrete and the usage of mineral admixtures can also improve used. The chemical and physical properties of cement used in
workability, compressive strength and durability [11,12]. experiments are shown in Table 1. The chemical and physical prop-
In recent years, the durability of concrete, using environmental erties are in accordance with the limit values specified in TS EN
friendly waste or industrial by-product materials and resource effi- 197-1 [22].
ciency are critical issues to achieve economical concrete and con-
crete structures with longer service lives corresponding 2.1.2. Fly ash
sustainable development [11–14]. According to Okamura and Ou- It is important to increase the amount of pastry in SCC because
chi, when SCC becomes so widely used, then the concrete industry it is an agent to carry the aggregates. As a consequence, FA has
will have succeeded in creating durable and reliable concrete been used in order to increase the amount of pastry. Some chemi-
structures that require very little maintenance work [15]. Using cal and physical properties of FA used in experimental study are gi-
FA, limestone powder, marble dust, etc. as filler material or par- ven in Table 1.
tially in powder material improves the durability of SCC [12,16]
and thus, it improves the effect of SCC on constructing durable 2.1.3. Aggregate
and reliable structures. As mentioned above, using wastes or by- In this study, the crushed stone aggregates have been used and
products in ordinary or special concretes as filler material, aggre- the maximum aggregate size is 16 mm. These aggregates are in
gate, mineral admixture, etc. also makes it possible to produce three different types and sizes as sand (0–4 mm), fine aggregate
durable, economical and environmental friendly concrete [13,14]. (4–8 mm) and coarse aggregate (8–16 mm), respectively. The abra-
In this way, RSCC with FA filler material satisfying the regular fresh sion-proof test (Los Angeles) has been carried out on the aggre-
and hardened concrete properties could be more economical, dura- gates and test results are shown in Table 2.
ble and environmental than any other concrete types.
When the mechanical properties of SCC are discussed, it can be 2.1.4. Rubber
seen that RSCC has regular, homogeneous compressive–tensile The granulated rubbers that are prepared for rubberized mix-
strength distribution, better elastic properties as greater ductility, tures have been sieved with square-mesh sieve of 4 mm size and
lower toughness and higher deflection compared to ordinary con- the rubbers sieved have been used in RSCC mixtures.
crete and furthermore, the performance and applicability of SCC
used in reinforced or prestressed structural concrete is similar to 2.1.5. High plasticizer admixture
ordinary vibrated concrete [17]. At 1990s, SCC was currently being In concrete mixtures, six different types of high performance
used in structures in order to shorten the construction period of viscosity agents have been used in the production of SCC as chem-
large-scale structures like prestressed concrete bridges, LNG tanks, ical admixtures. These admixtures have been coded as S-1, S-2, C-1,
etc. [15]. Nowadays, it can be said that SSC is commonly being em- C-2, C-3 and C-4. These admixtures were produced by two different
ployed in many structures including reinforced concrete buildings. international companies. The first company was coded as S and the
The reasons can be generally given as to shorten the construction second one was as C. Thus, two different types of admixtures were
period, to ensure compaction in the structure especially in con- provided from S company and four different ones were provided
fined zones where compaction by vibration is difficult, to eliminate from C company. Besides, these admixtures can also be called as
noise due to vibration especially at concrete products plants [15]. hyperplasticizers due to the property of increasing viscosity and
Moreover, there are many studies about SSC investigating fresh flow of concrete higher than superplasticizers. Therefore, these
and hardened properties like flow–slump, mix design, mechanical,
physical, thermal, hydric and elastic properties. For instance, Aly-
_ Table 1
amaç and Ince have investigated mix design for SCC with marble
Properties of cement and fly ash.
powders [18], Topçu et al. have studied the effects of marble dust
Chemical properties (%) CEM II/A-LL 42.5 Fly ash
powder on compressive strength, porosity, capillary coefficient,
flow and flow time properties of SCC [16], Türkmen and Kantarcı SiO2 20.74 55–58
Al2O3 5.13 23–25
have studied such properties of SCC incorporating expanded perlite
Fe2O3 2.78 6–8
aggregates [19] and Khatib has discussed the strength, absorption CaO 60.28 10–12
and shrinkage properties of SCC containing FA [20]. Felekoğlu et al. Free CaO 0.78 0.60
has determined the optimum water/cement ratio range for SCC in MgO 1.68 3.86
order to lead sufficient performance [21]. Consequently, it can be Na2O 0.27 0.5–0.8
K2O 0.59 1.4–1.8
said that most of the properties of SCC are known in recent times SO3 2.56 0.3–0.6
and there is enough knowledge for contributing the common use Insoluble residue 3.28 –
of SCC. Ignition loss 5.94 –
In this study, rubber obtained from used automobile tires has Physical and mechanical properties CEM II/A-LL 42.5
been ground, granulated and used in SCC mixture. First of all, con- Specific gravity 3.06
trol SCC mixture providing self-compactibility criteria has been ob- Specific surface (cm2/g) 4061
Mass stability (mm) 1
tained in accordance with the Europe SCC Specifications [4]. The
Setting period start (min) 165
fresh and hardened (mechanical and durability) properties such Setting period stop (min) 210
as slump–flow time and diameter, V-funnel flow time, compressive Strength of 1 day (MPa) 11.9
strength; high temperature and freezing–thawing resistances of Strength of 2 days (MPa) 23.4
RSCC mixtures have been compared to control SCC mixtures with- Strength of 7 days (MPa) 39.9
Strength of 28 days (MPa) 51.2
out RA content.
3058 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B.

Table 2
Abrasion loss values of aggregates.

First weight of Weight of aggregate after Weight of aggregate after Abrasion loss of aggregate after Abrasion loss of aggregate after
aggregate (g) 100 cycle (g) 500 cycle (g) 100 cycle (%) 100 cycle (%)
5000 4723 4255 5.54 14.9

Table 3 until the mixture has had the consistency of self-compactibility.

Some properties of chemical admixtures. This duration was not less than 3 min.
Properties C-1 C-2 C-3 C-4 S-1 S-2
Specific surface (±0.02 g/cm3) 1.09 1.04 1.05 1.045 1.10 1.075 3. Results and discussion
pH (±1) 6.5 7 7 7 3–7 3–7
Chloride content (%) <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 3.1. Properties of fresh concrete
Solid material (%) 20 21.5 20.2 21 – –
NaO (%) <1 <1 <1 <1 – <4
Dosage (%) 0.3–3 0.3–3 0.3–3 0.3–3 1–2 1–2
The experiments required for the self-compacting concrete are
generally carried out worldwide under laboratory circumstances.
Some of these experiments test the liquidity, segregation, place-
ment and compacting of fresh concrete. Conventional workability
types of chemical admixtures are generally used in SCC. The usage
experiments are not enough for the workability evaluation of
dosage given by the producers of these admixtures are 1–2% of SCC. Some experiment methods have been developed previously
total binding material amount for S-1 and S-2 by weight and are
by several research groups in order to measure the liquidity, segre-
0.3–3% of total binding material amount for C-1, C-2, C-3 and C-4 gation, placement and compaction of SCC. These experiment meth-
products by weight. In this study, admixture in the rate of 2% has
ods are defined in The European guidelines for self-compacting
been used. Some properties of these admixtures are given in Table concrete: Specification, production and use as slump–flow, V-fun-
3. The tap water provided from municipal city waterworks has
nel, U-box, L-box and fill box tests [4]. According to the results of
been used as the mixing water of mixture. many studies conducted, SCC containing different fillers, light-
weight aggregates can be defined as a type of high performance
2.2. Mixture proportions concrete due to its improved workability, high durability, low per-
meability and higher compressive strength [15,16,20,24–26].
One control SCC mixture and three RSCC mixtures have been Using limestone, stone or marble dust as filler material in SCC also
used in this study. RSCC mixtures containing rubber in the leads to better workability and higher compressive strengths
amounts of 60, 120 and 180 kg/m3 have been prepared. The usage [16,26]. Pozzolans like FA as filler also increase workability and,
of rubber aggregate is detrimental for compressive strength of SCC on the other hand, they decrease the early age compressive
[23]. Thus, these contents were chosen to attempt keeping com- strength however, increase later age strength, water absorption
pressive strength similar to ordinary SCC. On the other hand, these and decrease shrinkage in SCC [12,20,26]. Even if the greater usage
limited replacement contents would make it able to consider the amount of FA increases permeability, FA increases durability prop-
effects of RA incorporation on SCC properties. Thus, this paper erties compared to ordinary SCC incorporating limestone as filler
would contribute the limited studies about the RSCC. The cement [12,26]. When RA in SCC is considered, it decreases the fresh prop-
and fly ash content are kept constant as 380 and 150 kg/m3, erties of SCC and in this way, using superplasticizer, viscosity and
respectively. After then, four different types of mixtures have been entraining-air agents makes it available to produce workable RSCC
obtained including control and rubberized mixtures. These are due the results of slump–flow test [23]. In addition to this, the
called as: Control SCC mixture (C), Rubberized Mixture-1 (RAM- compressive strength, flexural strength and the modulus of elastic-
1), Rubberized Mixture-2 (RAM-2) and Rubberized Mixture-3 ity generally decrease; however, strain and restraint shrinkage
(RAM-3). The mixture proportions of these mixtures are presented cracking can be reduced [23,25]. In this study, slump–flow and
in Table 4. During the production of fresh SCC, all aggregates (0–4, V-funnel tests have been conducted to determine fresh concrete
4–8, 8–16 mm) have firstly been mixed in dry state. After then, FA properties of RSCC. Additionally, it has been observed that whether
and cement have been added. After a homogeneous dry mixture the concrete segregates or not during the tests. The experimental
has been obtained, the 80% of mixture water by volume has been
results obtained from the tests are presented in Table 5.
added to the mixture. Afterwards, chemical admixture has
been dissolved in the rest of the mixing water and this solution
3.1.1. Slump–flow test
has been added to the mixture. Mixing process has been continued This test includes observation of fresh SCC flow time observa-
tion and measurement of flow diameter after spreading of speci-
men by its self-weight. A slump funnel and a spandrel having
Table 4
dimensions of 80  80 cm are used as test devices. This method
The mixture proportions and design parameters of materials used.
is the most common used in order to measure self-flow properties
Component K RAM-1 RAM-2 RAM-3 of concrete. In other words, this test also determines the filling
Cement (kg/m3) 380 380 380 380 quality of SCC. The concrete is filled to the slump funnel on the
Fly ashes (kg/m3) 150 150 150 150 flow spandrel without any concrete spitting and then the slump
Water (l/m3) 170 170 170 170 funnel is pulled up and the concrete is supposed to flow by its
Fine aggregate (0–4 mm) (kg/m3) 780 750 710 670
Fine aggregate (4–8 mm) (kg/m3) 340 330 320 310
own weight without any effect. The time in which the concrete
Coarse aggregate (8–16 mm) (kg/m3) 520 500 490 480 comes to the diameter of 50 cm (T50) is measured by using a chro-
Rubber (kg/m3) – 60 120 180 nometer. After the end of flow, final flow diameter is also measured
Admixture (%) 2 2 2 2 in two directions perpendicular to each other. And then, the test is
Water/(FA + C) 0.32 0.32 0.32 0.32
completed [16,24]. The flow diameter values obtained from the
Unit weight (kg/m3) 2340 2340 2340 2340
test are presented in Fig. 2. Slump–flow test and T50 time have been
 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B. 3059

Table 5
Fresh concrete experimental results.

Mixture Admixture Fresh concrete experiments Observation

V-Funnel (s) T50 (s) Flow diameter (mm)
C C-1 9.04 3.94 63 No segregation
C-2 10.44 4.01 59 No segregation
C-3 14.30 4.12 61 No segregation
C-4 8.45 2.84 68 No segregation
S-1 15.25 3.72 64 No segregation
S-2 9.83 3.48 66 No segregation
RAM-1 C-1 8.12 3.38 69 Slight segregation
C-2 8.83 3.59 67 No segregation
C-3 11.65 3.65 65 No segregation
C-4 9.21 3.44 66 No segregation
S-1 14.96 3.98 61 No segregation
S-2 8.17 2.82 70 Slight segregation
RAM-2 C-1 9.22 3.72 67 No segregation
C-2 7.37 2.95 71 Slight segregation
C-3 10.71 3.84 62 Slight segregation
C-4 8.28 3.01 68 Slight segregation
S-1 9.47 3.22 69 No segregation
S-2 6.43 2.60 76 Slight segregation
RAM-3 C-1 6.12 2.89 76 Segregation
C-2 5.68 2.42 74 Segregation
C-3 4.46 2.77 70 Segregation
C-4 4.89 2.64 75 Segregation
S-1 5.18 2.81 72 Segregation
S-2 4.37 2.36 80 Segregation

measured by using required amount of fresh SCC during the mixing

process before the preparation of hardened specimens (Fig. 1). The
values of T50 time obtained from the test are shown in Fig. 3. The
flow times for different types of SCC in slump–flow test to reach
flow diameter of 50 cm were generally less than 5 s. The mixtures
having flow diameter between 60 and 80 cm have been accepted as
appropriate in slump–flow test regarding the European SCC Speci-
fications [4].
As a result of slump–flow test, the flow diameter was between
59 and 80 cm. These values were in accordance with the flow
diameter values mentioned in Europe Self-Compacting Concrete
Specifications. The lowest flow diameter (59 cm) has been mea-
sured for Control (C) SCC mixture containing C-2 chemical admix-
ture. As for the highest flow diameter (80 cm), it has been
measured for RAM-3 mixture containing S-2 chemical admixture.
The time to reach 50 cm flow diameter varies between 2.36 and
4.12 s. The lowest T50 time has been measured for RAM-3 mixture
Fig. 1. Slump–flow test and T50 time measurement. containing S-2 chemical admixture. The highest T50 time has been
measured for Control SCC mixture containing C-3 chemical admix-

C RAM-1 RAM-2 RAM-3

100
80
76 76
Flow diameter (cm)

71 74 75
80 69 67 67 70 68 66 68 69 72 66
70
63 65 62 64 61
59 61
60

40

20

0
C -1 C -2 C -3 C -4 S -1 S -2
Admixtures
Fig. 2. The effects of RA and admixture type on the flow diameter.
3060 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B.

C RAM-1 RAM-2 RAM-3

5
3,94 4,01 4,12 3,98
3,72 3,84 3,72
4 3,59 3,65 3,48
3,38 3,44
3,22
2,95 3,01

T50 (sec)
2,89 2,77 2,84 2,81 2,82
3 2,64 2,6
2,42 2,36

0
C-1 C-2 C-3 C-4 S-1 S-2
Admixtures
Fig. 3. The effects of RA and admixture type on the T50 time.

ture. By the way, it seems like the increase in the rubber amount This funnel is called as V-funnel. This test is a self-compactibility
leads to the decrease in T50 time and increase in workability. How- test to measure the capacity of concrete to pass into the spaces.
ever, as it is known from the literature, rubber decreases workabil- The test suggests an opinion about the viscosity and passing capac-
ity of ordinary concrete due to its physical properties and besides ity of SCC. Fresh SCC is filled to the funnel without any compaction
this, a combination of super plasticizer and viscosity agent makes or vibration and then the sliding cover under the funnel is opened
available it to produce SCC containing rubber having sufficient and the filling time of all concrete in the funnel to the pail is re-
fresh concrete properties [23]. Therefore, the type of chemical corded [16,24].
admixture used was the main parameter increasing flow diameter The V-funnel flow time values obtained from the test are shown
and T50 of SCC in this study. On the other hand, the coarser aggre- in Fig. 4. According to the results of V-funnel test, the V-funnel flow
gate amount in SCC decreased with the increasing replacement ra- time varies between 4.37 and 15.25. The highest flow time has
tio. Thus, the amount of finer material has increased relatively, been measured for Control SCC mixture containing S-1 chemical
higher fine aggregate amount also increased the flow diameter admixture and the lowest flow time has been measured for
and flow time. RAM-3 mixture produced with S-2 chemical admixture. The same
In a previously carried out study about RSCC, it was observed reasons and mechanism mentioned for slump–flow test results are
that the rubber waste has not negatively affected the fresh con- also commanding the explanations for the results of V-funnel test.
crete properties. As a result of slump–flow test, the flow diameter It has generally been observed that when the RA in SCC has in-
value was measured more than 60 cm. The flow time for SCC in creased, the fluidity of concrete also increases. However, increased
slump–flow test flow diameter of 50 cm was measured less than fluidity or flow speed also causes segregation.
5 s. There was not any adhesion or segregation in mixtures. It
was observed that when the fine material amount and admixture 3.2. Properties of hardened concrete
types are kept constant, the mixtures containing rubber wastes
were flowable enough [25]. In this study, it can be said that suffi- Fresh concretes have been cast into the cubic molds having each
cient results could be obtained due to the mechanism mentioned dimension as 15 cm. The specimens have been taken out from the
above. molds after 24 h. After then, the standard water cure has been ap-
plied. The 3rd, 7th and 28th day compressive strengths of SCC
3.1.2. V-funnel flow time measurement specimens have been determined. Compressive strength tests have
In this test, the flow of SCC per unit time is determined. It in- been conducted on two specimens for each mixture and an average
cludes the discharge time measurement of fresh SCC with its value of them has been taken into consideration. Considering the
self-weight from the narrow mouth of funnel designed specially. 3rd, 7th and 28th day compressive strengths of four different types

C RAM-1 RAM-2 RAM-3

18
15,25
14,96
14,3
15
V funnel (sec)

11,65
12 10,44 10,71
9,47 9,83
9,04 9,22 8,83 9,21
8,45 8,28
9 8,12 8,17
7,37
6,12 6,43
5,68 5,18
6 4,46 4,89 4,37

0
C-1 C-2 C-3 C-4 S-1 S-2
Admixtures
Fig. 4. The effects of RA and admixture type on V-funnel flow time.
 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B. 3061

Table 6
Hardened concrete properties

Mixture Admixture Compressive strengths (MPa)

3 days 7 days 28 days 400 °C 800 °C Freezing–thawing
C C-1 29.0 39.6 49.5 41.6 9.4 44.3
C-2 27.2 37.9 48.6 41.3 10.2 43.2
C-3 30.5 41.8 51.1 42.4 9.7 46.9
C-4 24.7 35.7 46.4 37.6 8.3 41.8
S-1 33.6 43.2 52.4 46.1 11.0 48.6
S-2 26.9 36.2 50.3 43.3 10.1 45.6
RAM-1 C-1 23.4 30.9 38.1 30.1 6.1 32.1
C-2 22.0 28.3 34.9 26.5 6.6 29.6
C-3 18.7 25.0 39.8 32.2 6.8 33.3
C-4 19.6 26.2 36.5 28.1 5.1 30.2
S-1 20.4 29.6 37.0 22.2 6.7 31.5
S-2 21.1 27.2 38.7 22.5 5.8 32.8
RAM-2 C-1 13.0 19.2 27.2 18.5 3.3 21.5
C-2 12.3 20.4 26.6 17.3 3.7 20.3
C-3 16.3 22.1 29.8 21.5 3.2 24.4
C-4 14.3 17.8 26.0 18.0 2.4 19.6
S-1 15.7 21.5 28.0 15.3 3.6 22.6
S-2 13.6 20.1 30.2 18.7 2.8 21.7
RAM-3 C-1 6.7 12.8 17.2 10.3 1.1 12.4
C-2 8.1 11.2 16.7 9.7 0.9 16.4
C-3 9.7 14.1 18.3 11.3 1.5 13.4
C-4 7.8 10.5 15.1 9.5 0.6 10.5
S-1 9.1 14.6 19.9 12.7 1.6 14.8
S-2 8.0 11.9 14.4 8.8 1.0 10.3

of SCC mixtures; increase of RA in SCC have led to decrease in com- sive strengths with respect to the increase in RA are shown in
pressive strength compared to the control SCC mixture. The poros- Fig. 5. Considering Fig. 5, it is seen that compressive strengths of
ity has also increased with the increase in rubber amount. Under SCC decrease with the increase in the RA content similar to the lit-
the compressive load, the rubber granules separate from cement erature [23,25]. In Fig. 5, it can be said that the compressive
pastry and make spaces unavoidable [25]. The 3rd, 7th and 28th strength of RSCC seems to be independent on concrete age. First
day compressive strength results of four SCC mixtures obtained of all, Fig. 5 presents the relative or residual compressive strengths
from the test are given in Table 6. The 3rd day compressive in order to compare the strengths of RSCC to control SCC. Thus, the
strength values were between 6.7 and 33.6 MPa, 7th ones were ob- compressive strength of controls SCC has been considered as 100%
tained between 10.5 and 43.2 MPa. Finally, 28th day compressive in Fig. 5. Another reason can be said as the inert RA for producing
strengths were determined in the range between 14.4 and C–S–H gels. In other words, rubber does not have any hydraulic or
52.4 MPa. When the 28 days of compressive strength results are pozzolanic activity to contribute to hydration, occur more C–S–H
discussed, it is seen that the highest compressive strength of Con- gels and gain compressive strength in concrete with elapsed time.
trol SCC mixture and RAM-3 mixture has been obtained with the Besides, cement capable for hydraulic reaction and pozzolans such
usage of S-1 chemical admixture, the highest compressive strength as FA having pozzolanic activity improves hydration and strength
of RAM-1 mixture has been observed with C-3 chemical admixture gain related to concrete age. Therefore, it is exactly obvious that
usage and the highest compressive strength of RAM-2 has been ob- compressive strength is exactly dependant on concrete age due
served with S-2 chemical admixture. For the mixture prepared to the existence of cement and pozzolans. Even if the pozzolans
with S-1 chemical admixture, the 3rd, 7th and 28th day compres- are used in ordinary concrete or self-compacting concrete as fine

120
3 Day 7 Day 28 Day
Residual Compr. Str. (%)

100 100

80
71
68,5
60 60.7
53,4
49.8
46,7
40 37.9
33,8
27.1
20

0
C RAM-1 RAM-2 RAM-3
Mixtures
Fig. 5. The compressive strengths with respect to increased RA.
3062 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B.

aggregate or filler, they still contribute to the compressive strength temperature in the kiln with 1200 °C heating capacity. The changes
gain. Herein, in this study, RA content in SCC were changed, cement of compressive strength and specimen weights have been observed.
and pozzolan contents were kept constant. In addition, it can be The specimens have shown pink color after exposure to 400 °C
mentioned that cement and the pozzolanic active FA as filler mate- and the specimens have turned into light grey after 800 °C. How-
rial still contributes to strength gain. As it is seen in Fig. 5, relative ever, all SCC specimens have not shown exhibited expected failure
compressive strength of RSCC mixtures at 28 days age is always modes for conventional concrete cubes subjected to compression.
greater than 3rd and 7th day compressive strengths. If the constant As mentioned above, the structure of RSCC may not be homoge-
cement and FA content are considered, compressive strength neous. Besides, the RA in SCC leave spaces in the structure of con-
behaviour is similar for different ages. However, compressive crete after exposing to high temperature. The failure modes of the
strength gain is still continuous compared to the previous age. specimens after being exposed to high temperature is presented in
On the other hand, the decrease in compressive strengths of differ- Fig. 6.
ent mixtures were occurred due to the increase in the RA replace- When concrete is exposed to high temperatures, the free water
ment. Besides, the decrease in the amount of natural coarser in capillary pores of the concrete, the water in C–S–H gel and chem-
aggregates may improve the intense of decrease in compressive ical bond, and the water in C–S–H and sulphoaluminate evaporate.
strength. Therefore, the relatively big differences in residual com- Consequently, this causes shrinkage in concrete around 300 °C
pressive strengths with respect to RA content presented in Fig. 5 [11,27]. When the temperature above 400 °C is employed, C–S–H
may have occurred due to this reason. Since rubber is a lightweight gels decompose. Around 530 °C, Ca(OH)2 transforms to the anhy-
aggregate, the structure of the concrete produced may not be drate lime. Thus, the high temperature leads to cracking and de-
homogeneous if an air entraining agent is not used [23]. Thus, creases the compressive strength of concrete due to these reasons
the structure of RSCC may not be homogeneous due to the exis- [11]. After mentioning the mechanism of high temperature effect
tence of RA and the increase in RA content may increase this. This on compressive strength of concrete, the compressive strength of
can be said as another reason for decrease in compressive strength SCC exposed to high temperatures can now be discussed. The com-
mentioned previously. Additionally, the type of viscosity agent pressive strength values of SCC specimens exposed to 400 and
used can also affect and change the compressive strength of SCC 800 °C are given in Figs. 7 and 8. The specimens have been kept in
incorporating same or different amounts of RA. Consequently, it water for 28 days and exposed to 400 and 800 °C temperature at
can be said that the RA decreases the compressive strength. 28th day. After then, their compressive strengths have been deter-
High temperature test has been conducted on the SCC and RSCC mined at the same day. The residual compressive strength values
specimens. The specimens were exposed to 400 and 800 °C high have been calculated with respect to the strengths of specimens

Fig. 6. The failure modes of specimens exposed to high temperature.

C RAM-1 RAM-2 RAM-3

100 88
84 85 86
Residual Compr. Str. (%)

79 83 81 81 77 80
76 78
80 68 72 69 70
65 63 64 67
60 58 62 61
60

40

20

0
C-1 C-2 C-3 C-4 S-1 S-2
Admixtures
Fig. 7. The compressive strengths of specimens exposed to 400 °C.
 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B. 3063

C RAM-1 RAM-2 RAM-3

25

Residual Compr. Str. (%)

21 21 20
19 19 19 18 18
20 17
16 15
14 14 13
15 12 11 10
8 9 8
10 7
6 5 4
5

0
C-1 C-2 C-3 C-4 S-1 S-2
Admixtures
Fig. 8. The compressive strengths of specimens exposed to 800 °C.

determined before the exposure of 400 and 800 °C. In other words, The same mechanism is valid for RSCC. Thus, the deceleration of
the initial compressive strength of the mixture type before expo- loss in compressive strength is greater in SCC containing higher
sure of the high temperature was considered as 100% (400 and contents of RA. When the RA in RSCC mixture has increased, the
800 °C), after then, the residual strength of the same mixture after amount of voids increased after the exposure of high temperature
the exposure was calculated in percents of its initial strength. [29]. Finally, the compressive strength also decreased and the big
As a result of high temperature effect, the highest compressive differences has occured between the compressive strengths of
strength loss at 400 °C has been determined for RAM-3 mixture SCC and RSCC mixtures. Besides, as mentioned before, the amount
containing C-2. The highest compressive strength loss at 800 °C of coarser aggregates decreased with increase in RA content. This
has been measured for RAM-3 mixture containing C-4. might have also increased the deceleration of loss in residual com-
The remaining compressive strengths of specimens after high pressive strength. Additionally, the type of chemical admixture can
temperature effect decreases to 58–88% for 28 days compressive lead to changes in residual compressive strength after high tem-
strength at 400 °C and to 4–21% at 800 °C. The highest compressive perature exposure. However, the effect of increase in RA content
strength has been seen for Control SCC mixture and it can be said can be said as the significant parameter compared to the effect of
that, the loss in compressive strength increases with the increase chemical admixture.
in RA content due to the exposure of high temperature. The com- The initial weights of control SCC and RSCC specimens have
pressive strengths have decreased due to mechanism of high tem- been measured just after they have been taken out of the water
perature effect as the decomposition of C–S–H gels. The cure at 28th day. After the 400 and 800 °C temperatures exposure,
deceleration of residual compressive strength due to high temper- their weights have been measured again and the losses in their
ature increases with the increase in exposed temperature. After weights due to high temperature effect have been calculated. The
then, very low compressive strengths are generally obtained with loss in weight values can be seen in Fig. 9. The values in Fig. 9 in-
the increase in temperature especially above 400 °C. Besides, the cludes only the results of mixtures prepared with S-1 chemical
relatively big residual compressive strength differences between admixture. Similarly, the weight loss increased with increase RA
series may have occurred because of the existence of RA and in- content. The same mechanism has caused this results likewise
crease in RA content. The RA burns after exposure to high temper- residual compressive strengths obtained because of the high tem-
ature, disappears and leaves voids in the structure of concrete [28]. perature effect.

15
400 0C 800 0C 13.8

12
10.8
Loss in Weight (%)

9 9.1
8.5
7.1
6
5.5

3.7
3 2.6

0
C RAM-1 RAM-2 RAM-3
Mixtures
Fig. 9. Losses in weights of specimens after 400 and 800 °C.
3064 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B.

C RAM-1 RAM-2 RAM-3

35
29,73 30,2

Loss in Compr. Str. (%)

30 27,61 28,12
26,87
24,6 25,42
23,63 22,48
25
20,84
18,27 19,14
20 16,27 17,36
15,81 15,11 14,82 15,33
15
10,53 11,02
9,81 9,24
10 8,17 7,32

0
C-1 C-2 C-3 C-4 S-1 S-2
Admixtures
Fig. 10. Residual strength after freezing and thawing test.

The freezing–thawing test has been conducted on Control SCC ever, the performance of RSCC is still better than ordinary
and RSCC specimens after they have been kept in standard water concretes.
cure for 28 days. The 28th day compressive strength and the resid-  The viscosity agents as chemical admixtures can decrease or
ual compressive strengths after freezing–thawing test have been improve the fresh and hardened concretes related to its proper-
compared to each other. The losses in compressive strengths of ties. Nevertheless, the mechanism could not be explained in this
specimens because of freezing–thawing effect are presented in study. In order to explain the mechanism and effects of the
Fig. 10. The highest loss in compressive strength has been seen chemical admixtures used, the microstructure of RSCC due to
for RAM-3 mixture containing C-4 chemical admixture. mechanic and durability effects should be investigated.
According to the results of freezing–thawing test, the losses in  According to the results of this study, the optimum usage RA
compressive strengths are between 7.32% and 30.2% compared to content in SCC can be said as 180 kg/m3 in order to obtain suffi-
the 28th day compressive strengths. The lowest loss in compres- cient performance for both of fresh and hardened properties of
sive strength has been measured for Control SCC mixture contain- SCC.
ing S-1 chemical admixture and the highest loss in compressive  Consequently, it can be reported that a new usage method for RA
strength has been measured for RAM-3 mixture containing C-4 in SCC formulation is developed. RA content used about 180 kg/
chemical admixture. The losses in compressive strengths are great- m3 has also beneficial effects on both fresh and hardened prop-
er for the series incorporating higher RA contents. In other words, erties of SCC compared to ordinary concretes. Thus, RA can be
increase in RA decreases the residual compressive strengths after recycled at higher amounts, the environmental pollution can
freezing–thawing effect. The type of viscosity agent changes and be prevented and economical advantages can be provided. This
may improve freeze–thawing resistance likewise the results ob- is also an important conclusion for sustainable development
tained from other mechanical properties. and concrete industry similar to use of other by-products.

The conclusions of this study are in accordance with the conclu-

4. Conclusions sions of a previous study by Bignozzi and Sandrolini [25]. Similarly,
it can be pointed out that controlling the microstructure of the rub-
The conclusions of this study can be said as follows: berized SCC to obtain improved mechanical and durability behav-
ior for SCC should be the main objective. Besides, they have
 When the properties of fresh SCC such as slump–flow, T50 and reported that further RSCC formulations are needed to strongly
V-funnel times are considered as a criterion to determine the confirm the effective superiority of RSCC in literature [25]. The
optimum usage ratio of RA in SCC, it can be said that usage aim of this study is to contribute this innovation by using RA in dif-
amount up to 180 kg/m3 content is suitable for improving all ferent sizes (0–4 mm), different types of chemical admixtures (vis-
of these properties especially related to the properties of the cosity agents) and different rubber contents. Besides, durability of
agent used when it is used in SCC with the viscosity agents. RSCC has not sufficiently been discussed in literature. This study
Besides, all fresh SCC properties satisfy the EFNARC SCC Specifi- also contributes to the RSCC progress by discussing durability
cation [4]. Consequently, it can be said that RA improves the properties. However, there is still a need to investigate the effects
workability of SCC produced with different viscosity agents. of new RSCC formulations on different mechanical, durability,
 If the mechanical and durability properties of hardened SCC physical and thermal properties of SCC.
such as compressive strength, high temperature and freezing–
thawing resistances are considered, the usage amount can also
be said as 180 kg/m3 content again. Concrete compressive References
strength and durability properties decrease with increasing RA
[1] Neville AM. Properties of concrete. 4th ed. England: Pearson Education Limited
content in the mixture. Besides, the residual compressive
Press; 2002.
strengths after high temperature and freezing–thawing effects [2] Mindess S, Young JF, Darwin D. Concrete. 2nd ed. United States of
also decrease. On the other hand, the obtained values are still America: Prentice Hall; 2003.
[3] Özyıldırım Ç, Lane DS. Evaluation of self-compacting concrete. United States of
higher than those of ordinary concretes mixed with similar
America: Transportation Research Board; 2003.
amounts of RA. Finally, it can be said that RA in SCC decrease [4] EFNARC. Specification and guidelines for self-compacting concrete. Fernham:
the mechanical and durability properties of ordinary SCC. How- EFNARC; 2002.
 _ Topçu, T. Bilir / Materials and Design 30 (2009) 3056–3065
I.B. 3065

[5] Yudenfreund M, Odler I, Brunauer S. Hardened portland cement pastes of low [17] Domone PL. A review of the hardened mechanical properties of self-
porosity. I: Materials and experimental methods. Cem Concr Res compacting concrete. Cem Concr Compos 2007;29:1–12.
1972;2:313–30. _
[18] Alyamaç KE, Ince R. A preliminary concrete mix design for SCC with marble
[6] Yudenfreund M, Skalny J, Mikhail RS, Brunauer S. Hardened portland cement powders. Construct Build Mater 2009;23:1201–10.
pastes of low porosity. II: Exploratory studies, dimensional changes. Cem _
[19] Türkmen I, Kantarcı A. Effects of expanded perlite aggregate and different
Concr Res 1972;2:331–48. curing conditions on the physical and mechanical properties of self-
[7] Okamura H, Ouchi M. Self-compacting concrete: development, present use and compacting concrete. Build Environ 2007;42:2378–83.
future. In: Skarendahl A, Petersson O, editors. Proceedings of the first [20] Khatib JM. Performance of self-compacting concrete containing fly ash.
international RILEM symposium on SCC. Bagneux: RILEM Publications SARL; Construct Build Mater 2008;22:1963–71.
1999. p. 3–14. [21] Felekoğlu B, Türkel S, Baradan B. Effect of water/cement ratio on the fresh and
[8] Ozawa K, Maekawa K, Kunishima M, Okamura H. Development of high hardened properties of self-compacting concrete. Build Environ
performance concrete based on the durability design of concrete structures. In: 2007;42:1795–802.
Nishino F, editor. Proceedings of the second East-Asia and Pacific conference on [22] TSI. TS EN 197-1/A1 Cement. Part 1: Compositions and conformity criteria for
struct. eng. and constr. Tokyo: JCI Publications; 1989. p. 445–50. common cements. Ankara: Springer; 2005.
[9] Ozawa K, Tagtermsirikul S, Maekawa K. Role of materials on the filling capacity [23] Turatsinze A, Garros M. On the modulus of elasticity and strain capacity of self-
of fresh concrete. In: Malhotra VM, editor. Proceedings of the fourth CANMET compacting concrete incorporating rubber aggregates. Resour Conser Recyl
and ACI international conference on fly ash, silica fume, slag and natural 2008;52:1209–15.
pozzolans in concrete. Farmington Hills: ACI Publications; 1992. p. 121–37. [24] Wu Z, Zhang Y, Zheng J, Ding Y. An experimental study on the workability of
[10] Khayat KH, Bickley J, Lessard M. Performance of self-consolidating concrete for self-compacting lightweight concrete. Construct Build Mater, in press.
casting basement and foundation walls. ACI Mater J 2000;97:374–80. [25] Bignozzi MC, Sandrolini F. Tyre rubber waste recycling in self-compacting
_ Bilir T. Effects of slag fineness on durability of mortars. J Zheijang
[11] Topçu IB, concrete. Cem Concr Res 2006;36:735–9.
Univ Sci A 2007;8:1725–30. [26] Mnahoncakova E, Pavlikova M, Grzeszczyk S, Rovnanikova P, Cerny R. Hydric,
[12] Dinakar P, Babu KG, Santhanam M. Durability of high volume fly ash self- thermal and mechanical properties of self-compacting concrete containing
compacting concretes. Cem Concr Compos 2008;30:880–6. different fillers. Construct Build Mater 2008;22:1594–600.
[13] Yüksel I,_ Bilir T, Özkan Ö. Durability of concrete incorporating non-ground _ Yalaman B. Concrete cover effect on reinforced concrete
[27] Ünlüoğlu E, Topçu IB,
blast furnace slag and bottom ash as fine aggregate. Build Environ bars exposed to high temperatures. Construct Build Mater 2007;21:1155–60.
2007;42:2651–9. _ Demir A. Durability of rubberized mortar and concrete. ASCE J Mater
[28] Topçu IB,
_ Boğa AR, Bilir T. Alkali–silica reactions of mortars produced by using
[14] Topçu IB, Civil Eng 2007;19:173–8.
waste glass as fine aggregate and admixtures such as fly ash and Li2CO3. Waste [29] Baylavlı H. The effect of different additives on the properties of rubberized
Manage 2008;28:878–84. aggregate self-compacting concrete. MSc thesis, Eskisßehir Osmangazi
[15] Okamura H, Ouchi M. Self-compacting high performance concrete. Prog Struct University, Turkey; 2008.
Eng Mater 1998;4:378–83.
_ Bilir T, Uygunoğlu T. Effect of waste marble dust content as filler on
[16] Topçu IB,
properties of self-compacting concrete. Construct Build Mater, in press.