Effect of filler types on physical, mechanical and microstructure of self compacting concrete and Flow-able concrete Academic research paper on "Civil engineering"

A] -

Alexandria Engineering Journal (2014) xxx, xxx-xxx

FACULTY OF ENGINEERING ALEXANDRIA UNIVERSITY

Alexandria University Alexandria Engineering Journal

www.elsevier.com/locate/aej www.sciencedirect.com

ORIGINAL ARTICLE

Effect of filler types on physical, mechanical and microstructure of self compacting concrete and Flow-able concrete

Hafez E. Elyamany a, Abd Elmoaty M. Abd Elmoaty a *, Basma Mohamed b

a Structural Engineering Department, Alexandria University, Egypt b Civil Engineering, Alexandria University, Egypt

Received 13 January 2014; revised 4 March 2014; accepted 13 March 2014

KEYWORDS

Self compacting concrete; Flow-able concrete; Pozzolanic fillers and porosity

Abstract The objective of this study is to evaluate the effect of various filler types on the fresh and hardened properties of self-compacting concrete (SCC) and Flow-able concrete. For this purpose, two groups of fillers were selected. The first group was pozzolanic fillers (silica fume and metakaolin) while the second group was non-pozzolanic fillers (limestone powder, granite dust and marble dust). Cement contents of 400 kg/m3 and 500 kg/m3 were considered while the used filler material was 7.5%, 10% and 15%. Slump and slump flow, T50, sieve stability and bleeding tests were performed on fresh concrete. The studied hardened properties included unit weight, voids ratio, porosity, and water absorption and cube compressive strength. In addition, thermo-gravimetric analysis, X-ray diffraction analysis and scanning electronic microscope were performed. The test results showed that filler type and content have significant effect on fresh concrete properties where non-pozzolanic fillers improve segregation and bleeding resistance. Generally, filler type and content have significant effect on unit weight, water absorption and voids ratio. In addition, non-pozzolanic fillers have insignificant negative effect on concrete compressive strength. Finally, there was a good correlation between fresh concrete properties and hardened concrete properties for SCC and Flow-able concrete.

© 2014 Production and hosting by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria

University.

Corresponding author. Tel.: +20 34355004. E-mail addresses: Abduo76@yahoo.com, Dr_abdelmoaty76@ yahoo.com (Abd Elmoaty M. Abd Elmoaty).

Peer review under responsibility of Faculty of Engineering, Alexandria University.

Production and hosting by Elsevier

1. Introduction

Self-compaction is often described as the ability of the fresh concrete to flow under its own weight over a long distance without segregation and without the need to use vibrators to achieve proper compaction. This saves time, reduces overall cost, improves working environment and opens the way for the automation of the concrete construction [1-4].

1110-0168 © 2014 Production and hosting by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria University. http://dx.doi.Org/10.1016/j.aej.2014.03.010

Self compacting concrete (SCC) mixes always contain a powerful superplasticizer and often use a large quantity of powder materials and/or viscosity-modifying admixtures. The superplasticizer is necessary for producing a highly fluid concrete mix, while the powder materials or viscosity agents are required to maintain sufficient stability/cohesion of the mixture, hence reducing bleeding, segregation and settlement [4]. Benefits of using SCC also include: improving homogeneity of concrete production and the excellent surface quality without blowholes [5].

In Flow-able concrete, introduction of high volumes of mineral admixtures to concrete mixtures is limited due to their negative effects on water demand and strength of the hardened concrete. However, these mineral admixtures can be efficiently utilized as viscosity enhancers particularly in powder-type SCC. Thus, successful utilization of lime powder (LP), basalt powder (BP) and marble powder (MP) in SCC could turn these materials into a precious resource. Moreover, these mineral admixtures can significantly improve the workability of self-compacting [6,7]. When used in SCC, these mineral admixtures can reduce the amount of superplasticizer necessary to achieve a given property [8]. It should be noted that the effect of mineral admixtures on admixture requirements is significantly dependent on their particle size distribution as well as particle shape and surface characteristics. From this viewpoint, a cost effective SCC design can be obtained by incorporating reasonable amounts of LP, BP and MP [9]. The addition of MP is the best mineral admixture among LP, BP and MP, to improve the properties of fresh SCC such as slump-flow, T50 time, L-box ratio, air content and unit weight. All the mineral admixtures have shown significant performance differences and the highest compressive strength has been obtained for the MP mixtures. Incorporation of mineral admixtures reduced the cost per unit compressive strength of these SCC [9].

A lot of researches were performed to study the effects of filler materials on the properties of SCC. These studies showed that the use of filler materials improves workability with reduced cement content. By this way, low heat of hydration and decreased thermal and shrinkage cracking can be achieved [10,11]. Belaidi et al. stated that at a constant water/binder ratio and superplasticizer content, the use of both natural pozzolana and marble powder by substitution to cement has no negative effects on the workability of SCC [12]. Industrial by-products such as fly ash (FA), stone dust, silica fume and blast furnace slag are generally used as filler materials in SCC [6,13]. This helps to provide economical benefits and reduce environmental pollution [14].

Chloride ion permeability decreased considerably when mineral admixtures were used in the production of SCC. Pozzolanic admixtures exhibited better performance than fillers. The SCC mixtures were assessed as "low" chloride permeability concretes as per ASTM C 1202-94 assessment criteria, with less than 2000 coulombs of total charge passing, so the durability of SCC enhances due to the decrease in permeability [7,15].

Ho et al. [3] demonstrated that the granite fines, as supplied, could be used successfully in the production of SCC. Compared to the use of limestone powder, both paste and concrete studies confirmed that the incorporation of granite fines required a higher dosage of superplasticizer for similar yield stresses and other rheological properties. However, it is important to point out that as a waste material, the properties of granite fines are expected to vary over time. Furthermore,

the fineness of granite fines could promote durability problems, such as alkali-silica reactions.

The marble has been commonly used as a building material since ancient times. Disposal of the waste materials of the marble industry, consisting of very fine powders, is one of the environmental problems worldwide today. However, these waste materials can be successfully and economically utilized to improve some properties of fresh and hardened self-compacting concrete (SCC) [16].

Valeria Corinaldesi et al. [17] stated that due to its quite high fineness, marble powder proved to be very effective in assuring very good cohesiveness of mortar and concrete. It is used for other ultra-fine mineral additions (such as silica fume) that are able to confer high cohesiveness to the concrete mixture. Moreover, an even more positive effect of marble powder is evident at early ages, due to its filler ability.

The use of fillers is intended to enhance the particle distribution of the powder skeleton, reducing inter-particle friction and ensuring greater packing density. This can promote release of a portion of the mixing water that would otherwise be entrapped in the system [18]. The water-binder ratio controls the amount of free space in the system in terms of void volume and the amount of fine material required to fill the voids. Void filling in packed systems may improve the particle arrangement, ensuring better water distribution and adequate fluidity. However, substantial increases in viscosity and unit weight occur at the concentration at which close packing is reached. The increase in viscosity beyond this limit may be explained by an increase in inter-particle friction due to increased solid-solid contact. In summary, the flow properties of self-compacting concrete depend heavily on powder particle size, shape, surface morphology, and internal porosity in addition to factors such as mixing regimen, sequence of admixture addition, and water/ superplasticizer content [19,20].

In general, Dehwah reported that the mechanical properties of SCC incorporating quarry dust powder (QDP) are better than those of SCC incorporating silica fume (SF) plus QDP or only fly ash (FA). The use of quarry dust powder alone results in a significant cost saving in regions where SF and FA are not available locally and have to be imported from other regions [21]. The use of mineral admixtures in various combinations can provide excellent mechanical properties of SCC. As pozzolanic materials FA and granulated blast furnace slag (GBFS) increased the late age compressive strengths of SCC mixtures [22].

2. Experimental programs

The present work aims to study the effect of filler types on fresh and hardened properties of SCC and Flow-able concrete. Fresh and hardened concrete properties such as slump, slump flow, sieve stability, bleeding, porosity, compressive strength and scanning of microstructure for Flow-able concrete (FAW) with slump of 220 ± 20 mm and self compacting concrete (SCC) were considered in this study.

2.1. Materials

Two groups of filler were selected. The first group was pozzo-lanic fillers); silica fume (SF), metakaolin (MK) while the second one was non-pozzolanic fillers; limestone powder (LP),

Granite dust (GD) and marble dust (MD). These five types of fillers passing from sieve No. of 200 (125 im) were used. Portland cement classified as CEM I N 42.5 according to E.S.S 4756/2006 was considered in this study. The chemical compositions of these fillers and cement are presented in Table 1. Also the grading of LP, GR and MP is shown in Fig. 1. The Toxic-ity test was also carried out to evaluate the toxicity of used fillers and cement. Table 2 shows the test results of toxicity test for cement and other different fillers. The results show the satiation of testing mice after each day of exposure throughout the testing period (6 days). From these test results, one mouse died on the third day when they were exposed to cement fume where one mouse was subjected to infer breathing (low rate of breathing) in case of metakaolin fume. In addition, there was no change observed in testing mice subjected to other fillers.

Natural siliceous sand with 2.66 fineness modulus and pink lime stone with nominal maximum size of 9.5 mm were used.

100 90 £ 80

tLi -------

u 30 --------

№ -----------

^ 20 --------

0 |llllll |ll. 100 10

Type G of high performance super plasticizer concrete admixture based on poly- carboxylic material was used.

2.2. Mix proportions and test procedure

The used water/binder ratio (w/b) for Flow-able and self compacting concrete was kept constant as 0.415, while the dose of used superplasticizer was changed to obtain the desired slump (220 ± 20 mm) for Flow-able concrete and constant slump flow (720 ± 30 mm) for self compacting concrete. For all the used concrete mixtures, the coarse/fine aggregate ratio was 1.0. Cement contents of 400 kg/m3 and 500 kg/m3 were considered in this study. For cement content of 400 kg/m3, the additional percentages of used filler materials were 7.5%, 10% and 15% while for cement content 500 kg/m3, only 10% was considered. By using different types of fillers, filler percentages and cement contents, twenty-six concrete mixes were cast and Table 3 shows the mixture proportions of these mixes.

Slump flow and T50 tests according to ACI 237R-07 were performed on fresh concrete, also sieve stability test was done. Sieve stability test measures the ability of SCC to remain uniform under both dynamic and static conditions. To perform the test, 10 liters sample of concrete is placed in a sealed bucket and left undistributed for 15 min to allow segregation to occur and then, approximately 2 liters from the top of the concrete sample is poured from a height of 500 mm onto a 5-mm sieve. Mortar from the sample allowed to flow through the sieve into a lower pan for a period of 2 min. The ratio of the mass of material in the pan to the total mass of concrete poured over

¿3 80

-!J r 50

1 0.1 0.01 0.001 0.0001

Diameter (mm)

(a) Lime powder

100 10 1 0.1 0.01 0.001 0.0001 Diameter (mm) (b) Granite Powder

100 90 80 70 60 50 40 30 20 10 0

0.001 0.0001

100 10 1 0.1 0.01 Diameter (mm)

(c) Marble Powder Figure 1 Grading of lime, granite and marble powder.

Table 1 Chemical compositions of cement and filler materials.

Component % Cement SF MK LP GP MP

SiO2 21.92 95.32 79.02 3.30 92.4 1.12

AhOs 3.30 0.88 5.96 0.82 1.25 0.73

Fe2O3 1.20 0.39 0.44 0.58 0.40 0.05

CaO2 63.0 0.90 12.3 92.9 1.00 83.22

SO3 2.1 1.03 0.40 1.18 3.80 0.56

Loss of ignition 1.2 1.40 1.20 1.20 1.10 2.50

Table 2 Toxicity test results.

Type of filler Exposure time/survivals

First day Second day Third day Fourth day

Silica fume 6 6 6 6

Metakaolin 6 6 6 6a

Lime stone powder 6 6 6 6

Marble dust 6 6 6 6

Granite dust 6 6 6 6

Cement 6 6 5b 5b

a Inferred breathing for metakaolin on 4th day.

1 mouse died on the 3rd day in case of Portland cement.

Table 3 Mixture proportions of concrete mixes.

Type of concrete Mix no. Filler type Mixture proportions (kg/m3) Flow time Diameter

Cement Filler Water Coarse Fine Admixture T50 (s) of sump

content aggregate aggregate (Lit) flow (mm)

Self compacting concrete 1 Silica fume (SF) 400 30 178.5 836 836 14.3 2.5 740

2 400 40 182.6 825 825 14.4 2.5 750

3 400 60 190.9 804 804 14.6 2.3 740

4 500 50 228.3 725 725 14.7 2.2 750

5 Meta-kaolin (MK) 400 30 178.5 835 835 15.8 3.5 720

6 400 40 182.6 823 823 16.0 3.2 710

7 400 60 190.9 802 802 16.2 3.2 720

8 500 50 228.3 724 724 16.4 3.0 740

9 Lime powder (LP) 400 30 178.5 838 838 16.0 3.5 710

10 400 40 182.6 825 825 16.1 3.0 720

11 400 60 190.9 805 805 16.3 2.8 720

12 500 50 228.3 723 723 16.3 2.5 750

13 Granite dust (GD) 400 30 178.5 835 835 15.9 3.0 720

14 400 40 182.6 825 825 16.0 3.0 730

15 400 60 190.9 805 805 16.1 2.5 750

16 500 50 228.3 723 723 16.2 2.5 750

17 Marble dust (MD) 400 30 178.5 835 835 15.9 3.5 720

18 400 40 182.6 825 825 16.0 3.0 740

19 400 60 190.9 805 805 16.1 2.7 750

20 500 50 228.3 724 724 16.2 2.4 750

Flow-able concrete Slump (mm)

21 - 400 - 166.0 870 870 4.0 200

22 SF 400 40 182.6 830 830 4.0 215

23 MK 400 40 182.6 830 830 4.1 200

24 LP 400 40 182.6 832 832 4.1 205

25 GP 400 40 182.6 832 832 4.1 210

26 MP 400 40 182.6 832 832 4.1 200

the sieve is taken as the segregation ratio [23]. Bleeding test according to ASTM C 232 was performed.

The hardened properties included unit weight, voids ratio, water absorption and cube compressive strength. Unit weight, voids ratio and water absorption were determined according to ASTM C 642, while concrete compressive strength was determined according to BS 1881: Part 3. 150 mm cubes were used. Each results represented in this section is the average of three tested specimens.

In addition, thermo-gravimetric analysis (TGA), X-ray diffraction (XRD) and scanning electronic microscope (SEM) were performed.

3. Test results and discussion

3.1. Fresh concrete

3.1.1. Slump, slump flow and T 50 cm

Results of fresh concrete properties of all self compacting and Flow-able concrete mixes are illustrated in Table 3. Slump flow and T 50 cm were used to measure the workability performance of SCC while slump test was considered for Flow-able concrete. Comparing the obtained results of slump flow and T 50 cm with the SCC criteria, it can be concluded that all mixes

Silica Metakaolin Limestone Granite Marble fume powder powder dust dust Filler Type (7.5 % filler content) 7.5 % Filler content (400 kg/m3 cement content)

Silica Metakaolin Limestone Granite Marble fume powder powder dust dust

Filler Type (10.0 % filler content)

(b) 10.0 % Filler content (400 kg/m3 cement content)

Silica fumeMetakaolin Limestone Granite Marble powder powder dust dust Filler Type (15.0 % filler content)

(c) 15.0 % Filler content (400 kg/m3 cement content)

Silica fume Metakaolin Limestone Granite Marble dust powder powder dust Filler Type (10.0 °/o filler content)

(d) 10.0 % Filler content (500 kg/m3 cement content)

Figure 2 Sieve stability test results of SCC.

ä? 16

I 14 1

Control Silica fume Metakaolin Limestone Granite dust Marble dust (without tiller) powder powder

Filler Type ( 10.0 % filler content)

Figure 3 Sieve stability test results of Flow-able concrete.

satisfy the requirement of SCC limits. For Flow-able concrete, no specific slump was required but it depends on the casting method. 200 mm slump was chosen as a recommended slump for casting using pump.

3.1.2. Sieve stability test results

Sieve satiability test is conducted to evaluate the resistance of fresh concrete to segregation. The effect of filler type and content on fresh concrete segregation resistance of SCC is presented in Fig. 2. From this figure, for concrete with 400 kg/m3 cement content, it is clear that filler type has a significant effect on the segregation resistance of SCC at different contents of filler. From these results, the non-pozzolanic fillers have a segregation resistance higher than that of pozzolanic fillers. The test results confirm that the use of granite dust and marble dust improves the segregation resistance of SCC compared with other types of fillers. The very fine silica fume SCC yields less segregation resistance compared other types of fillers. For example, at 10.0% filler content, the segregated portion is 14.0%, 9.6%, 9.6%, 4.4% and 3.8% for SCC with silica fume, metakaolin, lime powder, granite dust and marble dust, respectively. This trend is generally the same at different filler content. The previous behavior is the same for concrete with 500 kg/m3 cement content.

« 0.2 ca

♦ Silica fume

— - Limestone powder

— A -Granitedust • • • • Marble dust

• Metakaolin powder

6.0 8.0 10.0 12.0 14.0 Filler content (%)

Silica Metakaolin Limestone Granite Marble fume powder powder dust dust

Filler Type (10.0 % filler content)

(a) 400 kg/m3 cement content (b) 500 kg/m cement content

Figure 4 Effect of filler content and type on bleeding percent of SCC.

Figure 5 Effect of filler type on bleeding percent of Flow-able concrete.

8.0 - 7.0

Ï" I 3.0

< 3.0 2.0

Control Silica fume Metakaolin Limestone Granite dust Marble dust (without filler) powder powder

Filler Type ( 10.0 % Aller content)

Figure 7 Effect of filler type on concrete absorption of Flow-able concrete on.

For Flow-able concrete, the effect of using filler on segregation resistance is presented in Fig. 3. From this figure, it is obi-vious that the use of filler enhances the segregation resistance of fresh Flow-able concrete. The non-pozzolanic fillers still yield higher segregation resistance compared with pozzolanic fillers. The use of granite dust and marble dust has still good resistance to concrete segregation compared with other types of fillers. The use of silica fume, metakaolin, lime stone powder, granite dust and marble dust decreases the segregation resistance by 4.3%, 25.6%, 23.1%, 33.3% and 40.2% compared with control mix, respectively.

3.1.3. Bleeding test results

Fig. 4 shows the effect of filler type and content on the bleeding resistance of SCC with 400 kg/m3 cement content. From this figure, the filler type has a significant effect on the

bleeding resistance of SCC where the use of non-pozzolanic fillers specially granite dust and marble dust showed good bleeding resistance compared with other types of filler. Also, the increase in filler content decreases the bleeding resistance of SCC. The significant effect of filler type is more clear at high level of filler content, 15.0%, whereas there is no oblivious effect of filler type on bleeding resistance at lower filler content, 7.5%. The pervious trend is the same for SCC with 500 kg/m3 as shown in Fig. 5.

The effect of 10.0% filler on the bleeding resistance of Flow-able concrete is presented in Fig. 5. From this figure, the use of 10.0% filler enhances the bleeding resistance of Flow-able concrete compared with concrete without filler. This behavior may be due to the cohesion of concrete with fillers. Also, granite dust and marble dust (non-pozzolanic fillers) still show the good bleeding resistance compared with types of

Table 4 Bulk density of SCC and Flow-able concrete test results.

Type of concrete Cement content (kg/m3) Filler content (%) Type of filler

SF MK LP GD MD

Unit weight (gm/cm3)

SCC 400 7.50 2.29 2.28 2.29 2.29 2.29

10.0 2.29 2.29 2.29 2.30 2.29

15.0 2.34 2.30 2.33 2.34 2.31

500 10.0 2.32 2.33 2.31 2.32 2.31

Flow-able 400 0.0 2.28

10.0 2.29 2.29 2.29 2.29 2.28

Silica fiimc Metakaolin Limestone Granite dustMarble dust powder powder

Filler Type (10.0 % filler content)

(a) 400 kg/m3 cement content (b) 500 kg/m3 cement content

Figure 6 Effect of filler content and type on absorption percent of SCC.

Table 5 Voids ratio of SCC and Flow-able concrete test results.

Type of concrete Cement content (kg/m3) Filler content (%) Type of filler

SF MK LP GD MD

Voids ratio (%)

SCC 400 7.50 4.00 4.47 4.36 3.54 3.68

10.0 3.56 3.70 3.98 3.25 3.35

15.0 3.22 3.67 3.68 3.14 2.88

500 10.0 3.47 3.62 3.75 3.14 3.08

Flow-able 400 0.0 14.52

10.0 8.04 9.30 9.85 8.10 7.79

fillers. The reduction in bleeding resistance is 53.0%, 51.5%, 56.1%, 66.7% and 65.2% for concrete with silica fume, metakaolin, lime stone powder, granite dust and marble dust compared with control mix, respectively.

3.2. Hardened concrete

3.2.1. Unit weight

Table 4 presents the values of measured dry unit weight of hardened SCC and Flow-able concrete at 56 days. For SCC, the test results showed that the type of filler has insignificant effect on hardened unit weight. On the contrary, the filler con-

ane Metakaolin Limestone Granite dust Marble dust powder powder

Filler type

(a) Concrete compressive strength at 7 days

tent has a noticeable effect on dry unit weight of SCC where the increase in filler content increases the dry unit weight of hardened SCC. This may be due to the voids - filling by the fine used fillers. For example, for silica fume SCC, the reduction in hardened unit weight for concrete with 7.5% silica fume is 2.2% lower than SCC with 15.0% silica fume. Also, the increase in cement content increases the unit weight of hardened SCC. For Flow-able concrete, the hardened unit weight of Flow-able concrete is filler independent.

3.2.2. Water absorption and voids ratio test results Fig. 6 shows the effect of filler type and content on the water absorption of hardened SCC after 56 days. The test results showed that the filler type has a significant effect on hardened SCC water absorption. The used non-pozzolanic fillers (GD and MD) except lime powder yield lower water absorption and voids ratio compared with pozzolanic fillers (SF and MK). This trend is the same for concrete with 400 kg/m3 and 500 kg/m3 cement. For SCC with 500 kg/m3, the water absorption of 10% MK and LP is higher than that of 10% SF by 6.7% and 14.7, respectively, while the use of 10% GD and MD decreases the water absorption by 10% compared with SCC with 10% silica fume. Moreover, the filler content has a

fume Metakaolin Limestone Granite dust Marble dust powder powder

Filler type

(b) Concrete compressive strength at 28 days

« -£

^ Silica fume Metakaolin Limestone Granite dust Marble dust

powder powder

Filler type

(c) Concrete compressive strength at 56 days

Figure 8 Concrete compressive strength for SCC with 400 kg/m3 cement content.

Figure 9 Concrete compressive strength for SCC with 500 kg/m3 cement content and 10% filler content.

Figure 10 Concrete compressive strength for Flow-able concrete with 400 kg/m3 cement content.

significant effect on water absorption. For example, for silica fume SCC with 400 kg/m3 cement content, the reduction in water absorption, for SCC with 10.0% and 15.0% silica fume is about 10.8% and 21.6% compared with SCC with 7.5% silica fume, respectively.

For Flow-able concrete as shown in Fig. 7, the use of 10.0% filler reduces significantly the water absorption of Flow-able hardened concrete. The use of GD and MD still has the lowest values of water absorption for Flow-able concrete.

Test results of voids ratio of SCC and Flow-able concrete confirm the test results of water absorption as given in Table 5.

3.2.3. Cube compressive strength

Fig. 8 shows the variation in concrete compressive strength of SCC at the different ages of curing for different types of fillers

for concrete mixes with 400 kg/m3 cement content. From this figure, generally one can obliviously observe that there is no significant variation between compressive strength of pozzolanic fillers and that of non-pozzolanic fillers. This trend is the same at 7 days, 28 days and 56 days. The SCC containing 500 kg/m3 cement content shows the same pervious observation as presented in Fig. 9. The good performance of used non-pozzolanic filler is due to the micro-filling ability, improving the microstructure of the bulk paste matrix and transition zone. The calcium carbonate in LP reacts very little with cement hydrates to substitute the pozzolanic activity [20].

Further, it is clear that the increase in filler content from 7.5% to 15.0% has not any significant effect on concrete compressive strength for all types of fillers except silica fume. There is a noticeable increase in concrete compressive strength

Bleeding (%)

(a) Self compacting concrete

Bleeding (%)

(b) Flow-able concrete

Figure 11 Correlation between concrete bleeding and concrete compressive strength.

2.26 2.28 2.30 2.32 2.34 2.36 2.27 2.28 2.29 2.3 2.31

Unit weight (t/m3) Unit weight (t/m3)

(a) Self compacting concrete (b) Flow-able concrete

Figure 12 Correlation between concrete unit weight and concrete compressive strength.

3 4 5 6 Voids ratio (%)

(a) Self compacting concrete

6 6.5 7 7.5 Voids ratio (%)

(b) Flow-able concrete

Figure 13 Correlation between concrete voids ratio and concrete compressive strength.

Position (2 Thêta)

(e) Marble dust cement paste

Figure 14 XRD test results of cement pastes modified with different fillers.

as silica fume content increase. This may be due to the high pozzolanic effect of silica fume mineral admixture. Also, Fig. 10 shows the concrete compressive of flow-concrete at the different ages of curing for different types of fillers for concrete mixes with 400 kg/m3 cement content.

3.2.4. Correlations between bleeding, unit weight and voids ratio versus compressive strength

Fig. 11 shows the correlation between concrete bleeding and 28 days concrete compressive strength for SCC and traditional concrete. From this figure, it is clear that there is a negative correlation between bleeding and concrete compressive strength whereas the increase in concrete bleeding decreases the concrete compressive strength. The increase in bleeding

allows moving water upward to the concrete top layer. This is due to the increase in w/c ratio at the top layer which decreases the concrete compressive strength. This conclusion confirms the clear correlation of fresh concrete properties, bleeding, and hardened concrete properties, compressive strength. This behavior is the same for SCC and traditional concrete.

Fig. 12 presents the correlation between hardened concrete unit weight and 28 days concrete compressive strength. From this figure, there is a clear positive correlation between hardened concrete unit weight and concrete compressive strength where the increase in unit weight increases the concrete com-pressive strength. The positive effect of high concrete unit weight may be due to lower voids ratio which increases the

0 200 400 600 800 1000 Temperature (C°) (a) Silica fume cement paste

200 400 600 800 1000

Temperature (C°) (b) Metakaolin cement paste

0 200 400 600 800 1000 Temperature (C°) (c) Limestone cement paste

0 200 400 600 800 1000 Tempemperature (C°) (d) Granite cement paste

19 18 17

16 f 15

•o 13

BD 5 12 £ 11 10

1 \ __,

V/ 1 I

w \ I

0.002 £ E

0.000 I»

-0.002 ^ H

-0.004 O -0.006

-0.008

400 600 800 Temperature (C°)

(e) Marble dust cement paste

Figure 15 TG, DTA test results of cement pastes modified with different fillers.

concrete compressive strength. This trend is confirmed in Fig. 13. From this figure, the decrease in voids ratio increases the concrete compressive strength. This trend is the same for SCC and traditional concrete.

3.3. X-ray diffraction analysis

Although X-ray qualitative diffractometry does not provide any reliable quantitative information, it is considered as a sensitive technique which gives acceptable information about the most probable phases. Fig. 14 shows the X-ray diffraction patterns of cement pastes specimens with 10% fillers as cement addition. From these patterns, there is no obvious change in the tested samples where type of filler, generally, has insignificant effect. The only clear effect is the low calcium hydroxide and ettringite in silica fume and metakaolin cement pastes, pozzolanic fillers, compared with non-pozzolanic fillers. This behavior is due to the pozzolanic effect of silica fume and metakaolin mineral admixtures. The pozzolanic effect consumes a part of calcium hydroxide.

3.4. TGA, DTA analysis

Thermo-gravimetric analysis uses heat to force reactions and physical changes in materials. TGA reveals changes due to weight, whereas DTA reveals the changes due to phase transitions. TGA provides quantitative measurement of mass change in materials associated with transition and thermal

degradation. TGA records change in mass from dehydration, decomposition and oxidation of a sample with time and temperature. It was reported that zone one between 100 and 300 0C is attributed to the dehydration of C-S-H and ettringite and zone two includes from 290 to 350 0C characterize the decomposition of calcium aluminates silicate hydrate, calcium aluminates hydrate and calcium chloro-aluminate. The third zone ranging from 450 to 510 0C is attributed to the dehydration of calcium hydroxide. An endo-therm around 700 0C indicates the de-carbonation of calcium carbonate in the hydrated compound [24].

Fig. 15 shows the relation between the weight of specimen and the applied temperature. From these test results, there is no clear difference is observed in zone one and two. In the third zone, ranging from 450 to 510 0C which attributed to the dehydration of calcium hydroxide, weight loss drop easily observed within the third zone in the cement paste containing limestone, granite or marble. On the other hand, for the samples containing silica fume or metakaolin, almost the weight loss is hardly observed within zone between 450 and 510 0C which also confirms the earlier recorded pozzolanic reactivity of silica fume and metakaolin. It means that calcium hydroxide was consumed by pozzolanic reaction. Thus, calcium hydroxide content is lower in case of silica fume and metakaolin, while other samples results may give a hint about its non-reactivity. This observation confirms the test results of X-ray diffraction.

Moreover, at temperature around 700 0C, the specimens containing limestone and marble dust indicate greatly

(c) Limestone cement paste (d) Granite cement Paste

(e) Marble dust cement paste Figure 16 SEM analysis results of cement pastes modified with different fillers.

observed weight loss by TGA and endo-therm transition by DTA than other samples. It meets with logic whereas this specimen containing the higher content of calcium carbonate.

3.5. SEM analysis

Fig. 16 shows the microstructure of the investigated pastes. The sample containing silica fume shows a refined pore structure. The infinitesimal volume of silica fume particles reduces the porosity of the paste as a result of filler effect. The sample containing metakaolin or granite shows a well defined ettring-ite fibers which match with the aforementioned result of X-ray diffraction analysis. Moreover, the comparison between these images shows that the ettringite needled are clearly observed in limestone, granite and marble cement pastes whereas traces of ettringite are observed in silica fume and metakaolin cement pastes. The micrograph showed a good agreement with XRD, TGA and mechanical test results previously shown.

4. Conclusions

Based on the findings of the current study, the following conclusions may be drawn:

• Filler type has a significant effect on segregation resistance and bleeding resistance of SCC and Flow-able concrete. The use of non-pozzolonic fillers (granite dust and marble dust) decreases the segregation and bleeding compared with pozzolanic fillers (silica fume and metakaolin).

• The increase in filler content improves bleeding resistance of SCC. The significant effect of filler type on bleeding resistance is obvious at high level of filler content, 15.0%, whereas there is no obvious effect of filler type on bleeding resistance at lower filler content.

• Filler type and content have significant effect on hardened SCC water absorption. For example, for silica fume SCC with 400 kg/m3 cement content, the reduction in water absorption, for SCC with 10.0% and 15.0% silica fume is about 10.8% and 21.6% compared with SCC with 7.5% silica fume, respectively.

• For Flow-able concrete, the use of 10.0% filler reduces significantly the water absorption of Flow-able hardened concrete. In addition, filler type has insignificant effect on water absorption.

• There is no negative effect of non-pozzolanic fillers on concrete compressive strength compared to that of pozzolanic fillers.

• There is a negative correlation between bleeding and concrete compressive strength whereas the increase in concrete bleeding decreases the concrete compressive strength.

• X- ray diffraction patterns showed that there is no obvious change in the tested samples where type of filler, generally, has insignificant effect. The only clear effect is the low calcium hydroxide and ettringite in silica fume and metakaolin cement pastes, pozzolanic fillers compared with non-pozzolanic fillers.

• Micrograph scanning and TGA showed a good agreement with XRD and mechanical test results.

References

[1] H. Okamura, M. Ouchi, Self-compacting concrete. Development, present use and future, in: Proceedings of first international RILEM symposium on self-compacting concrete, RILEM Publications, S.A.R.L., Stockholm, 1999, pp. 3-14.

[2] P.J.M. Bartos, M. Grauers, Self-compacting concrete, Concrete 33 (4) (1999) 9-13.

[3] D.W.S. Ho, A.M.M. Sheinn, C.C. Ng, C.T. Tam, The use of quarry dust for SCC applications, Cem. Concr. Res. 32 (2002) 505-511.

[4] Wenzhong Zhu, J.C. Gibbs, Use of different limestone and chalk powders in self-compacting concrete, Cem. Concr. Res. 35 (2005) 1457-1462.

[5] M. Ouchi, S. Nakamura, T. Osterson, S. Hellberg, M. Lwin, Applications of self compacting concrete in Japan, ISHPC, Europe and the United States, 2003, p. 1-20.

[6] M. Sahmaran, H.A. Christianto, I.O. Yaman, The effect of chemical admixtures and mineral additives on the properties of self-compacting mortars, Cem. Concr. Compos. 28 (5) (2006) 432-440.

[7] Mucteba Uysal, Kemalettin Yilmaz, Metin Ipek, The effect of mineral admixtures on mechanical properties, chloride ion permeability and permeability of self-compacting concrete, Constr. Build. Mater. 27 (2012) 263-270.

[8] M. Sonebi, L. Svermova, P.J.M. Bartos, Factorial design of cement slurries containing limestone powder for self-consolidating slurry-infiltrated fiber concrete, ACI Mater. J. 101 (2) (2004) 136-145.

[9] Mucteba Uysal, Kemalettin Yilmaz, Effect of mineral admixtures on properties of self-compacting concrete, Cement Concr. Compos. 33 (2011) 771-776.

[10] G. Ye, X. Liu, G. De Schutter, A.-M. Poppe, L. Taerwe, Influence of limestone powder used as filler in SCC on hydration and microstructure of cement pastes, Cem. Concr. Compos. 29 (2) (2007) 94-102.

[11] A.M. Poppe, G.D. Schutter, Cement hydration in the presence of high filler contents, Cem. Concr. Res. 35 (12) (2005) 22902299.

[12] A.S.E. Belaidi, L. Azzouz, E. Kadri, S. Kenai, Effect of natural pozzolana and marble powder on the properties of self-compacting concrete, Constr. Build. Mater. 31 (2012) 251-257.

[13] B. Felekoglu, K. Tosun, B. Baradan, A. Altun, B. Uyulgan, The effect of fly ash and limestone fillers on the viscosity and compressive strength of self-compacting repair mortars, Constr. Build. Mater. 36 (9) (2006) 1719-1726.

[14] Ilker Bekir Topcu, Turhan Bilir, Tayfun Uygunoglu, Effect of waste marble dust content as filler on properties of self-compacting concrete, Constr. Build. Mater. 23 (2009) 19471953.

[15] S. Assie, G. Escadeillas, V. Waller, Estimates of self-compacting concrete 'potential' durability, Constr. Build. Mater. 21 (10) (2007) 1909-1917.

[16] Alyamac Kurs_at Esat, Ince Ragip, A preliminary concrete mix design for SCC with marble powders, Constr. Build. Mater. 23 (2009) 1201-1210.

[17] Valeria Corinaldesi, Giacomo Moriconi, Tarun R. Naik, Characterization of marble powder for its use in mortar and concrete, Constr. Build. Mater. 24 (2010) 113-117.

[18] A. Yahia, M. Tanimura, Y. Shimoyama, Rheological properties of highly flowable mortar containing limestone filler-effect of powder content and W/C ratio, Cem. Concr. Res. 35 (2005) 532539.

[19] S.A. Rizwan, T.A. Bier, Blends of limestone powder and fly-ash enhance the response of self-compacting mortars, Constr. Build. Mater. 27 (2012) 398-403.

[20] Gritsada Sua-iam, Natt Makul, Utilization of limestone powder to improve the properties of self-compacting concrete incorporating high volumes of untreated rice husk ash as fine aggregate, Constr. Build. Mater. 38 (2013) 455-464.

[21] H.A.F. Dehwah, Mechanical properties of self-compacting concrete incorporating quarry dust powder, silica fume or fly ash, Constr. Build. Mater. 26 (2012) 547-551.

[22] Mucteba Uysal, Mansur Sumer, Performance of self-compacting concrete containing different mineral admixtures, Constr. Build. Mater. 25 (2011) 4112-4120.

[23] First International Symposium on Design, Performance and Use of Self-Consolidating Concrete, SCC 2005 - China, Changsha, Hunan, China, 26-28 May 2005, p. 39-42.

[24] R. Vedalakshmi, S.A. Raj, S. Srinivasan, G.K. Babu, Quantification of hydrated cement products of blended cements in low and medium strength concrete using TG and DTA technique, Thermochim. Acta 407 (1-2) (2003) 49-60.