3.0 DEVELOPMENT OF M 25 GRADE OF SELF COMPACTING CONCRETE

Transcription

1 63 3. DEVELOPMENT OF M 25 GRADE OF SELF COMPACTING CONCRETE This investigation is mainly focused on the development of cost-effective normal strength M 25 grade of SCC with moderate fines for the use of normal building constructions. This chapter describes the constituent materials used to make SCC, the test methods that are used for testing fresh mortar, fresh, hardened mechanical, drying shrinkage and microlevel properties of SCC. In this study, a simple tool has been developed for SCC mix design on the basis of key proportions of the constituents of SCC. This chapter investigated the use of mini slump cone test along with the graduated glass plate to obtain the optimization of superplasticiser (SP) and viscosity modifying agent (VMA) in self compacting mortar (SCM). The effect of coarse aggregate blending on fresh properties of SCC has been studied and thereby optimization of coarse aggregate blending in a given coarse aggregate content of SCC was determined. The effect of coarse aggregate blending on mechanical properties of SCC has been studied. The mechanical properties of M 25 grade of SCC and CC were compared at different curing periods. Drying shrinkage of M 25 grade of SCC and CC was measured at different drying periods. The effect of class F fly ash on the microlevel properties of SCC was investigated and the correlation of micro and macrolevel (mechanical) properties of concrete (SCC and CC) was also studied. 3.1 MATERIALS AND METHODS This section describes the materials used to make SCC and test methods for testing the fresh mortar and SCC fresh and hardened properties.

2 Materials Constituent materials used to make SCC can have a significant influence on the fresh and hardened characteristics of the SCC. The following sections discuss constituent materials used for manufacturing SCC. Chemical and physical properties of the constituent materials are presented in this section Cement Ordinary Portland Cement 53 grade (Dalmia) was used corresponding to IS (1987)86. The chemical properties of the cement as obtained by the manufacturer are presented in the Table 3.1. Table 3.1. Chemical composition of cement Particulars Test Requirement as per result IS: Chemical Composition % Silica(SiO2) % Alumina(Al2O3) 5.67 % Iron Oxide(Fe2O3) 4.68 % Lime(CaO) % Magnesia(MgO).84 % Sulphuric Anhydride (SO3) 2.48 % Chloride content.3 Max..1%.92.8 to Min..66 Not more Than 6.% Max. 3.% when C3A>5. Max. 2.5% when C3A<5. Lime Saturation Factor CaO.7SO3/2.8SiO2+1.2Al2O3+.65Fe2O3 Ratio of Alumina/Iron Oxide Summary of physical properties and various tests conducted on cement as per IS 431(1988)8, 81, 82, 83 are presented in the Table 3.2.

3 65 Table 3.2 Physical Properties of Cement Physical properties Test result Requirement as Test method/ Remarks per IS (1987) Specific gravity 3.15 IS 431(1988) part 11 - Fineness (m2/kg) Manufacturer data Min.225 m2/kg Normal consistency 3% IS 431 (1988)- part 4 - Initial setting time (min) 9 IS 431 (1988)- part 5 Min. 3 min Final setting time (min) 22 IS 431 (1988)- part 5 Max. 6 min.8 Manufacturer data Max. 1 mm Soundness Lechatelier Expansion (mm) Autoclave Expansion (%).1 Max..8% MPa Compressive strength (MPa) 3 days 7 days 28 days 39 IS 431 (1988)- part MPa 53 MPa Additive or Mineral Admixture Fly Ash Class F fly ash produced from Rayalaseema Thermal Power Plant (RTPP), Muddanur, A.P is used as an additive according to ASTM C 618( 23)11. As per IS 456 (2)76, cement is replaced by 35% of fly ash by weight of cementitious material. The properties of fly ash as obtained by RTPP are presented in the Table 3.3.

4 66 Table 3.3 Chemical and physical properties of class F fly ash Particulars Class F ASTM C 618 class F fly ash Fly ash Chemical composition % Silica(SiO2) 65.6 % Alumina(Al2O3) 28. % Iron Oxide(Fe2O3) 3. % Lime(CaO) 1. % Magnesia(MgO) 1. % Titanium Oxide (TiO2).5 % Sulphur Trioxide (SO3).2 Max. 5. Loss on Ignition.29 Max. 6. SiO2+ Al2O3+ Fe2O3>7 Physical properties Specific gravity 2.12 Fineness (m2/kg) 36 Min.225 m2/kg Chemical Admixtures Sika Viscocrete 1R is used as high range water reducer (HRWR) SP and Sika Stabilizer 4R is used as VMA. The properties of the chemical admixtures as obtained from the manufacturer are presented in the Table 3.4. Table 3.4 Properties of chemical admixtures Quantity (%) Chemical Specific Admixture gravity ph Solid by content (%) cementitious Main component weight Sika Viscocrete 1R Sika Stabilizer 4R Polycarboxylate Ether

5 Coarse Aggregate Crushed granite stones of size 2 mm and 1 mm are used as coarse aggregate. The bulk specific gravity in oven dry condition (OD) and water absorption of the coarse aggregate 2 mm and 1mm as per IS 2386 (Part III, 1963)79 are 2.6 and.3% respectively. The gradation of the coarse aggregate was determined by sieve analysis as per IS 383 (197)75 and presented in the Tables 3.5 and 3.6. The grading curves of the coarse aggregates as per IS 383 (197)75 are shown in Figs. 3.1 and 3.2. Fineness modulus of coarse aggregate 2 mm and 1 mm are 6.95 and 5.89 respectively. Table 3.5. Sieve analysis of 2 mm coarse aggregate Sieve size Cumulative percent passing 2 mm IS 383 (197) limits 2 mm mm N/A 12.5 mm N/A 1 mm mm -5 Table 3.6. Sieve analysis of 1 mm coarse aggregate Sieve size Cumulative percent passing 1 mm IS 383 (197) limits 1 mm mm mm 2.4-5

6 68 1 2mm Lower Limit (IS 383:197) Percentage Passing 8 Upper Limit (IS 383:197) IS Sieve Size (mm) Fig. 3.1 Grading curve of 2 mm coarse aggregate 1 1mm Lower Limit (IS 383:197) Percentage Passing 8 Upper Limit (IS 383:197) IS Sieve Size (mm) Fig. 3.2 Grading curve of 1 mm coarse aggregate 1

7 69 Dry-rodded unit weight (DRUW) and void ratio of coarse aggregate with relative blending by percentage weight as per IS 2386 (Part III, 1963)79 is shown in Table 3.7 and Fig Table 3.7 Dry-rodded unit weight and void Ratio of a given coarse aggregate blending Coarse aggregate blending by percentage weight ( 2 mm and 1 mm) 17 DRUW (kg/m3) 1: : : : : : : : Bulk density Void Void Dry-rodded bulk density, kg/m3 Void ratio % 2% 4% 6% 8%.35 1% 1mm aggregate to total gravel (by wt.) Fig. 3.3 Dry-rodded unit weight and void ratio of a given coarse aggregate blending

8 7 The DRUW of coarse aggregate 2 mm and 1 mm with relative blending 7:3, 67:33, 6:4 and 4:6 by percentage weight are 1647 kg/m3, 1659 kg/m3, 1646 kg/m3 and 1631 kg/m3 respectively. In this investigation, these four coarse aggregate blending 7:3, 67:33, 6:4 and 4:6 were used to study the effect of coarse aggregate blending on fresh properties of SCC Fine Aggregate Natural river sand is used as fine aggregate. The bulk specific gravity in oven dry condition (OD) and water absorption of the sand as per IS 2386 (Part III, 1963)79 are 2.6 and 1% respectively. The gradation of the sand was determined by sieve analysis as per IS 383 (197)75 and presented in the Table 3.8. The grading curve of the fine aggregate as per IS 383 (197)75 is shown in Fig Fineness modulus of sand is Table 3.8. Sieve analysis of fine aggregate Cumulative percent passing Sieve No. Fine aggregate IS: Zone III requirement 3/8 (1mm) 1 1 No.4 (4.75mm) No.8 (2.36mm) No.16 (1.18mm) No.3 (6μm) No.5 (3μm) No.1 (15μm) 1.9-1

9 71 1 Fine Aggregate Lower Limit (IS 383:197) Percentage Passing 8 Upper Limit (IS 383:197) IS Sieve Size (mm) Fig. 3.4 Grading curve of fine aggregate Water Generally, drinking water is used in concrete. Water from industrial plants, sewage and other contaminated areas should not be used in concrete. If the quality of water is suspected, then that water should be tested before its usage in concrete TEST METHODS This section describes the test methods that are used for testing fresh mortar and SCC properties, hardened, drying shrinkage and microlevel properties of SCC and CC. 1

10 Tests on Fresh Mortar This section describes various tests involved in mortar tests to determine the optimum w/cm and optimum dosage of SP and VMA in mortar Mini Slump Cone and Graduated Glass Plate The test apparatus for measuring the spread and viscosity of mortar comprises a mini frustum (slump) cone and a graduated glass plate. Mini slump cone has top and bottom diameters of 7 cm and 1 cm respectively with a height of 6 cm. The graduated glass plate contains two circular graduations of 1 cm and 2 cm in diameter marked at its center as shown in Fig With this test apparatus, both spread and viscosity of the mortar can be measured from a single test. Fig. 3.5 Mini slump cone and graduated glass plate

11 73 Here in our study, mini slump cone is used to measure the spread of the mortar as described in EFNARC (22)49. To indicate the rate of flow or viscosity of the mortar spread, T2 method (Shekarchi et al., 28)179 is adopted instead of mini V-funnel test. As T2 indicates the intended viscosity of mortar during this test, it is concluded that it is the best replacement of mini V-funnel test. Practically, it is very much feasible to have a single test apparatus to measure both spread and viscosity of mortar so that rigorous mortar tests can be reduced. Determination of spread of mortar In this test, the truncated cone mould is placed exactly on the 1 cm diameter graduated circle marked on the glass plate, filled with mortar (o.56 litre) and lifted upwards. The subsequent diameter of the mortar is measured in two perpendicular directions and the average of the diameters is reported as the spread of the mortar. Determination of T2 (sec) T2 is the time measured in seconds from lifting the cone to the mortar reaching a diameter of 2 cm. The measured T2 indicates the deformation rate or viscosity of the mortar. So, during this test, T2 can be measured first and average of the spread can be measured subsequently. This procedure is similar to slump cone test conducted on SCC. Consistence retention Along with the spread and T2, consistence retention is also an important fresh property of self compacting mortar (SCM). It refers to the period of duration during which SCM or SCC retains its properties, which is important for transportation and placing.

12 74 Consistence retention was evaluated by measuring the spread and T2 of successful SCM at 45 and 6 minutes after adding water. Between measurements, the mortar was stored in the mixing bowl and covered the top to avoid moisture loss. The mortar was remixed for one minute before each test Tests on Fresh Properties of SCC This section describes various tests involved in SCC tests to determine filling ability, passing ability, segregation resistance and consistence retention of SCC Slump Flow Test Slump flow test apparatus is shown in Fig Slump cone has 2 cm bottom diameter, 1 cm top diameter and 3 cm in height. In this test, the slump cone mould is placed exactly on the 2 cm diameter graduated circle marked on the glass plate, filled with concrete (6 litre) and lifted upwards. The subsequent diameter of the concrete spread is measured in two perpendicular directions and the average of the diameters is reported as the spread of the concrete. T5cm is the time measured from lifting the cone to the concrete reaching a diameter of 5 cm. The measured T5cm indicates the deformation rate or viscosity of the concrete V-funnel Test V-funnel test apparatus dimensions are shown in Fig In this test, trap door is closed at the bottom of V-funnel and V-funnel is completely filled with fresh concrete (12 litre). V-funnel time is the time measured from opening the trap door and complete emptying the funnel. Again, the V-funnel is filled with concrete, kept for 5 minutes and trap door is opened. V-funnel time is measured again and this indicates V-funnel time at T5min.

13 L-box Test L-box test apparatus dimensions are shown in Fig In this test, fresh concrete (14 litre) is filled in the vertical section of L-box without any compaction and then the gate is lifted up to allow the concrete to flow into the horizontal section of L-box. The height of the concrete at the end of vertical section represents h1 (mm) and at the horizontal section represents h2 (mm). The ratio h2/h1 represents blocking ratio Segregation Resistance The segregation resistance of SCC was assessed by visual stability index (VSI) from the slump flow test (Daczko and Kurtz, 21)4. A VSI number of, 1, 2 and 3 is given to the spread to characterize the stability of the mixture (refer ). Segregation was considered to be present when a halo of mortar and an uneven distribution or clustering of the aggregates was observed in slump flow. The stability of SCC can easily be examined by evaluating the distribution of the coarse aggregate visually within the concrete mass after the spreading of the concrete has stopped Determination of Consistence Retention Consistence retention is also an important fresh property of SCC in view of workability. It refers to the period of duration during which SCC retains its properties, which is important for transportation and placing. Consistence retention was evaluated by measuring the slump flow spread and T5cm of successful SCC mixes at 6 minutes after adding water. The SCC mix was remixed for one minute before each test.

14 76 Fig. 3.6 Slump flow test

15 77 Fig. 3.7 V-funnel test Fig. 3.8 L-box test

16 Tests on Hardened Properties of SCC and CC This section describes the procedure to determine the mechanical properties of SCC and conventional concrete (CC) and these include compressive strength, modulus of elasticity (MOE) and splitting tensile strength (STS). These properties were tested on cylinder specimens of size 15 mm x 3 mm for all the mixes Compressive Strength Test Compressive strength test was conducted on the cylindrical specimens for all the mixes at different curing periods as per IS 516 (1959)77. Three cylindrical specimens of size 15 mm x 3 mm were cast and tested for each age and each mix. The compressive strength (f c) of the specimen was calculated by dividing the maximum load applied to the specimen by the cross-sectional area of the specimen Modulus of Elasticity Test Modulus of elasticity (MOE) test was conducted on the specimens for all the mixes at different curing periods as per IS 516 (1959)77. Three cylindrical specimens of size 15 mm x 3 mm were cast and tested for each age and each mix. Each specimen was loaded until an average stress of (C+5) kg/cm2 is reached. Here, C is the one-third of the average equivalent cube compressive strength. The equivalent cube strength has been determined by multiplying the cylinder strength by 5/4. Strains at regular interval of loads till the proportional limit, have been measured. Stress-strain curve has been plotted. The secant modulus is calculated from the slope of the straight line drawn from the origin of axes to the stress-strain curve (IS 516, 1959)77 and this secant modulus is the required modulus of elasticity of the concrete (Ec).

17 Splitting Tensile Strength Test Splitting tensile strength (STS) test was conducted on the specimens for all the mixes at different curing periods as per IS 5816 (1999)84. Three cylindrical specimens of size 15 mm x 3 mm were cast and tested for each age and each mix. The load was applied gradually till the failure of the specimen occurs. The maximum load applied was then noted. Length and cross-section of the specimen was measured. The splitting tensile strength (fct) was calculated as follows: fct = 2P/ (Π l d) Where, fct = Splitting tensile strength of concrete (N/mm2) P = Maximum load applied to the specimen (in Newton) l = Length of the specimen (in mm) d = cross-sectional diameter of the specimen (in mm) Drying Shrinkage Test This section describes the procedure to determine the drying shrinkage strain of SCC and CC. Drying shrinkage tests were performed in accordance with ASTM C 157 (28)1. For each concrete mixture, three 15 mm x 3 mm concrete cylinders were cast for drying shrinkage test. These cylinders were cured for 7 days. Initial shrinkage measurements were taken as soon as the cylinders were brought out of the curing. All specimens were kept drying in a controlled environment at 23 ± 2 C and 5% ± 4% relative humidity. The length change of all the specimens was measured by a length comparator having a digital extensometer (-25 cm x.1 mm) as shown in Fig. 3.9 after 1, 7, 14, 28, 56 and 112 days of drying after 7 days of curing. The average value of the measured shrinkage strains of the three specimens for each mix was taken as the drying shrinkage of the concrete.

18 8 Drying shrinkage of concrete at any drying period has been calculated as follows: Drying shrinkage = Ri Rt L Where: Ri = Initial dial gauge reading of the specimen after 7 days of curing Rt = Dial gauge reading of the specimen after t days of drying L = Length of the specimen after 7 days of curing Fig. 3.9 Length comparator

19 Microlevel Properties This section describes the procedure to determine the microlevel properties of SCC and CC after 28, 56 and 112 days of curing. The microlevel properties that were determined are microcrack widths between aggregate and paste by Scanning electron microscope (SEM) analysis; the chemical elements and atomic Calcium-Silica (Ca/Si) ratio of the paste matrix near the ITZ at different ages by Energy dispersive X-ray (EDAX) analysis. To carry out SEM and EDAX, the samples of approximate size 1 mm x 1 mm x 5 mm were collected from the tested cylindrical specimens after the compressive strength tests as per IS 516 (1959)77 on SCC and CC at different ages. All samples chosen for the SEM and EDAX were uncoated and unpolished. The microlevel properties were studied on the samples using the instrument JSM-639 which is equipped with both SEM and EDAX. The SEM scans a focused beam of electrons across the sample and measures any of several signals resulting from the electron beam interaction with the sample (Stutzman, 2)19. The microcracking width images of all samples were acquired by SEM in secondary electron (SE) mode at a magnification setting of X13 and 1μm field width (micron marker). EDAX provided the spectrum of chemical elements with the peaks and quantitative chemical analysis with the relative intensities for all the samples.

20 A SIMPLE TOOL FOR SCC MIX DESIGN This section includes the selection of mix proportions for SCC from the relevant literature, design of SCC mix design tool, calculation of key proportions of constituents for a given SCC scenario. A user-friendly and simple tool has been developed for SCC mix design ( JGJ_SCCMixDesign.xls ) based on the key proportions of the constituent materials of SCC with or without blended cement and with or without coarse aggregate blending. This tool has been used for SCC mix design through out this study Selection of SCC Mix Proportions SCC can be made from any of the constituents that are generally used for structural concrete. In the SCC mix design, it is most common to consider the relative proportions of the key components or constituents by volume rather than by mass (weight) (EFNARC, 22)49. The following key proportions for the mixes listed below (Okamura and Ozawa, 1995; EFNARC, 22; Khayat, 1998; Domone 26b)147, 49, 97, 47: 1. Air content (by volume) 2. Coarse aggregate content (by volume) 3. Paste content (by volume) 4. Binder (cementitious) content (by weight) 5. Replacement of mineral admixture by percentage binder weight 6. Water/ binder ratio (by weight) 7. Volume of fine aggregate/ volume of mortar 8. SP dosage by percentage cementitious (binder) weight 9. VMA dosage by percentage cementitious (binder) weight

21 Design of SCC Mix Design Tool This section describes the material properties required for SCC mix design tool, detailed steps for mix design and output constituent materials for SCC Material Properties for SCC Mix Design Tool The following material properties for the SCC mix design tool are to be determined as shown in Table 3.9 and Table Specific gravity of cement, fly ash, coarse aggregate and fine aggregate. 2. Percentage of water absorption of coarse and fine aggregates. 3. Percentage of moisture content in coarse and fine aggregates. 4. Dry-rodded unit weight (DRUW) of coarse aggregate for the particular coarse aggregate blending. 5. Percentage of dry material in SP and VMA. Table 3.9 Material properties Material Data Material Specific Gravity % Absorption % Moisture Cement 3.15 N/A N/A Additive Fly Ash 2.12 N/A N/A Coarse aggregate (CA1 2mm) Coarse aggregate (CA2 1mm) Fine aggregate (Sand)

22 Detailed Steps for SCC Mix Design Tool The detailed steps for mix design are described as follows: 1. Assume air content by percentage of concrete volume. 2. Input the coarse aggregate blending by percentage weight of total coarse aggregate. 3. Input the percentage of coarse aggregate in DRUW to calculate the coarse aggregate volume in the concrete volume. 4. Adjust the percentage of fine aggregate volume in mortar volume. 5. Obtain the required paste volume. 6. Adopt suitable water/ binder ratio by weight. 7. Input the percentage replacement of fly ash by weight of cementitious material. 8. Input the dosage of SP and VMA (if required) by percentage weight of binder. 9. Adjust the binder (cementitious material) content by weight to obtain the required paste. The coarse aggregate optimization is shown in Table 3.1. The input parameters section is shown in Table Table 3.1 Coarse aggregate optimization or blending Coarse aggregate optimization Material % by weight CA1 2mm 6 CA2 1mm 4

23 85 Table 3.11 Input parameters section Input parameters Dry Rodded Unit Weight(kg/m3) 1646 % of CA in DRUW 44.3 % of Sand in Mortar 46.1 % of Fly ash 35 Wt. Water/Binder.36 Binder (kg/m3) 495 SP (% wt.of binder).9 VMA (% wt. of binder).2 % of Air 2 % of dry material in SP 4 % of dry material in VMA Output Constituent Materials for SCC Once the necessary input data is given, this tool automatically calculates and shows the required output data. Concrete mix proportions by volume and total aggregate by weight are shown in Table Table 3.12 Concrete mix proportions by volume Coarse aggregate (kg/m3) % of CA in concrete volume Concrete Mix proportions by volume (lit/m3) CA Mortar Sand Paste Sand (kg/m3) Total aggregates (kg/m3)

24 86 Paste composition is shown in Table Constituent materials for SCC are shown in Table Constituent materials for self compacting mortar (SCM) are shown in Table This tool also displays the constituent materials for the required volume of SCC or SCM as shown in Table 3.14 and Table Aggregate proportions by volume and by weight are shown in Table Table 3.13 Paste composition Vol. Water/Powder.969 Paste composition Kg/m3 lit/m3 Cement Fly ash Water SP VMA Paste Table 3.14 Constituent materials for SCC Constituent Materials for Concrete Material Required (m3) Initial Adjusted Cement Fly Ash Water CA1 2mm CA2 1mm Sand SP (lit) VMA (lit) Unit Weight Total (kg) Litres (kg/m3).62 g/ml

25 87 Table 3.15 Constituent materials for SCM Constituent Materials for Mortar Material Required (m3) Initial Adjusted Cement Fly Ash Water Sand SP (lit) VMA (lit) Unit Weight Total (kg) Litres.564 (kg/m3).8 g/ml Table 3.16 Aggregate proportions by volume and by weight Aggregate Proportions Material % by Vol % by Weight CA1 2mm CA2 1mm Sand Total Calculation of Key Proportions The detailed steps for calculation of key proportions are presented in Appendix B with an example.

26 OPTIMIZATION OF SUPERPLASTICISER AND VISCOSITY MODIFYING AGENT IN SELF COMPACTING MORTAR This section investigated the use of mini slump cone test along with the graduated glass plate to obtain the optimization of superplasticiser (SP) and viscosity modifying agent (VMA) in self compacting mortar (SCM). As SCC contains lower content of coarse aggregate, mortar employs more effects on the SCC fresh properties than conventional concrete (CC). Mortar not only provides lubrication by wrapping coarse aggregates, it also predominantly influences the fresh properties of SCC with a low yield stress and adequate viscosity so as to ensure the required filling and passing ability without blocking and segregation. Mortar is, thus an intrinsic part of SCC mix design and it has also formed a central part of Jin s research (Jin, 22; Jin and Domone, 22) 88, 89. Hence, self compacting mortar (SCM) is a precondition of the successful production of SCC. Thus, optimization of SP and VMA in SCM must be done to ensure a stable SCC having low yield stress and adequate viscosity for the given w/cm and mix proportion. And this is best done by self compacting mortar tests Experimental Study Our objective was to determine the optimum dosage of SP and VMA in SCM with the available materials. In this respect, 53 grade ordinary Portland cement (OPC 53), class F fly ash as an additive, river sand, SP and VMA were used in preparing SCMs having w/cm.32 and.36. The SCM mixes had 35% replacement of cement with class F fly ash by cementitious weight and 4% and 45% of sand in mortar by volume. The fresh properties that were determined are the mortar spread, T2 and consistence retention.

27 Experimental Procedure Specimen Preparation Mortars were prepared manually in a container in order to observe its behaviour. Mortar volume of.8 m3 (8 x 1-4 m3) (.56 litre) is sufficient for each mortar test using mini slump cone. It is known that mixing procedures have a significant influence on the fresh properties of SCM. Modified Jin s mixing procedure (22)88 was carried out throughout this work to achieve maximum efficiency of SP and VMA. The mixing procedures for SCM with SP only shown in Fig. 3.1 and both with SP and VMA shown in Fig are described as follows. Mixing procedure for mortar with SP only: Cement+W1 W2+SP Fly Ash+Sand Stop Mixing Rest 3 min Remix 1 min before test Minutes W1 = 8% mixing water W2 = 2% mixing water 1 SP = Superplasticiser Fig. 3.1 Mixing procedure for mortar with SP only 1. Cement and 1st part (8%) of water was mixed for two minutes. 2. SP along with the 2nd part (2%) of water was added and mixed for two minutes. 3. Fly Ash and sand was added to the mix and mixed thoroughly for two minutes. 4. The mix was stopped and kept rest for 3 minutes. 5. The mix was remixed for one minute and discharged for mortar test.

28 9 Mixing procedure for mortar with SP and VMA: Cement+W1 W2+SP W3+VMA Fly Ash+Sand Stop Mixing Rest 2 min Remix 1 min before test Minutes W1 = 7% mixing water W2 = 15% mixing water SP = Superplasticiser 1 W3 = 15% mixing water VMA= Viscosity Modifying Agent Fig Mixing procedure for mortar with SP and VMA 1. Cement and 1st part (7%) of water was mixed for two minutes. 2. SP along with the 2nd part (15%) of water was added and mixed for two minutes. 3. VMA along with the 3rd part (15%) of water was added and mixed for one minute. 4. Fly Ash and sand was added to the mix and mixed thoroughly for two minutes. 5. The mix was stopped and kept rest for 2 minutes. 6. The mix was remixed for one minute and discharged for mortar test Tests on Fresh Mortar The test apparatus comprising the mini slump cone and graduated glass plate was used for determining mortar spread, viscosity (T2) and consistence retention (refer ) Mortar Mix Design This research investigated the effect of SP and VMA dosage on the three SCM mixes (Mix 1, Mix 2 and Mix 3) which had 35% replacement of cement with class F fly ash and water/cementitious ratios by weight (w/cm).32 and.36 as shown in Tables 3.17 and In other words, the cementitious proportion is kept same for all the mixes. The volume of paste content was kept at 359 litre/m3 for the two mixes Mix 1 and Mix 2 and 388 litre/m3 for the Mix 3.

29 91 Table 3.17 Mortar Mix Proportions per.8 m3 for 4% of Sand in Mortar w/cm.32 Mix 1.36 Mix 2 %SP %VMA by cementitious weight Cement (g) Fly ash (g) Water (ml) Sand (g) SP (ml) VMA (ml) As per the general purpose mix design method developed by Okamura et al. (1993)142, sand content in mortar is kept at 4% of mortar volume in the first two mixes (Mix 1 and Mix 2) as shown in Table For the third mix (Mix 3), sand is kept at 45% of mortar volume in order to evaluate the optimization of SP and VMA as shown in Table These mortar tests conducted to study the interactions among cement, mineral and chemical admixtures and to determine optimum dosages of SP and VMA for the given w/cm.

30 92 Table 3.18 Mortar Mix Proportions per.8 m3 for 45% of Sand in Mortar %SP w/cm.36 Mix 3 %VMA by cementitious weight Cement (g) Fly ash (g) Water (ml) Sand (g) SP (ml) VMA (ml) Mortar tests started with minimum dosage of SP by percentage weight of cementitious and increased the dosage of SP till the maximum spread of the mortar has reached. When the mortar spread shows halo, minimum dosage of VMA by percentage weight of cementitious was used to avoid the bleeding. For each dosage of SP and VMA, fresh mortars were prepared and tested Results and Discussion Effect of SP and VMA on Spread and T2 for the Mix 1 The influence of SP on mortar spread and T2 (viscosity) for the Mix 1 is shown in Table It is observed that as the SP dosage increases, the spread of mortar increases and T2 decreases. Spread reaches the maximum value and T2 reduces to the minimum at a specific SP dosage. This point is referred as saturation point. For this mix, maximum spread 31 mm was arrived at.9% SP dosage as shown in Fig So, it is the optimum dosage of SP for this mix.

31 93 Table 3.19 Spread and T2 for the Mix 1 Time after water mixing (min) w/cm.32 Mix 1 %SP Initial Spread (mm) T2 (sec) Spread + Halo (mm) Fig Maximum spread of the Mix 1 at.9% SP Beyond this saturation point, adding SP causes decrease in mortar spread and increase in T2. Adding even more SP leads to segregation of mortar. So, it is practically seen that before reaching the saturation point, the addition of SP increases

32 94 the spread and decreases T2. After the saturation point, the addition of SP leads to decrease in the spread and increase in T2. The use of VMA dosage was not used in this mortar Mix 1 as no segregation or bleeding (halo) was observed in the mortar before the saturation point Effect of SP and VMA on Spread and T2 for the Mix 2 But for the Mix 2 with w/cm.36, it is seen that bleeding (halo) was observed at the SP dosage of.7% by cementitious weight itself i.e., before the saturation point as shown in Table 3.2 and Fig As we know that increase of water reduces the yield stress and viscosity and some times leads to segregation. This behaviour is clearly seen in the mixes Mix 1 and Mix 2 when w/cm is increased from.32 to.36. T2 value of Mix 2 is significantly low i.e., rate of flow has been increased as compared to that of Mix 1. Table 3.2 Spread and T2 for the Mix 2 Time after water mixing (min) w/cm %SP Initial %VMA Spread (mm) T2 (sec) Spread + Halo (mm) Mix

33 95 Fig Spread with halo of the Mix 2 at.7% SP At.8% SP dosage, spread does not change, but T2 is decreased from 3.25 sec to 3.3 sec and the thickness of halo is increased as shown in Fig. 3.14(a). It clearly indicates the requirement of VMA in order to resist segregation. Then, fresh mortar mix was prepared both with.8% SP dosage and minimum VMA dosage of.2% by cementitious weight and mortar test was conducted. It is observed that the minimum dosage of VMA stopped the bleeding, but the spread has decreased from 295 mm to 237 mm as shown in Fig. 3.14(b) and T2 increased from 3.3 sec to 6 sec i.e., viscosity is increased. The reason being that VMA can imbibe some free water and increases the viscosity and thus reduces the risk of segregation or bleeding (Khayat, 1998)97. From the result, it indicates that maximum spread has not been arrived at.8% SP dosage and.2% VMA dosage. It leads to increase in SP dosage.

34 96 (a) Spread with halo without VMA (b) Spread without halo with.2% VMA Fig Spread of the Mix 2 at.8% SP

35 97 Hereafter,.2% VMA dosage was maintained for all the mixes of Mix 2 category. Now, fresh mortar mix was prepared with.9% SP dosage and.2% VMA dosage and tested. It is seen that spread was increased from 237 mm to 296 mm without any bleeding as shown in Fig and T2 decreased from 6 sec to 3.15 sec i.e., rate of flow has been increased satisfactorily. Fig Maximum spread of the Mix 2 at.9% SP and.2% VMA After this point, mix prepared with 1% SP and.2% VMA dosage and tested. Mortar spread again decreased from 296 mm to 231 mm which is less than the spread 237 mm that has been arrived at.8% SP dosage and.2% VMA dosage. At this stage, T2 increased from 3.15 sec to 5.78 sec. This result was not satisfactory. Again, a fresh mix with 1.1% SP and.2% VMA was prepared and tested. Spread has increased from 231 mm to 29 mm, but it is less than 296 mm which was maximum spread at.9% SP dosage. When comparing the mixes with.9% and 1% SP dosage, maximum spread 296 mm was observed for the mix with.9% SP dosage.

36 98 So, it is referred as the saturation point. Thereby, it is noted that for the Mix 2 having w/cm.36, the optimum dosages of SP and VMA were.9% and.2% respectively. Interestingly, it is observed that VMA dosage didn t affect the saturation point of SP dosage which is.9%. In our study, saturation point is arrived at the same SP dosage of.9% by cementitious weight for the two mixes Mix 1 and Mix 2 with w/cm.32 and.36 respectively. It is inline with the statement that for mortars with the same powder (binder) proportions, the dosage of SP expressed in terms of percentage by cementitious weight, doesn t change significantly with the variation of w/cm (Nepomuceno et al., 28) Effect of SP and VMA on spread and T2 for the Mix 3 The influence of SP and VMA on mortar spread and T2 (viscosity) for the Mix 3 is shown in Table This mix has 45% of sand in mortar volume. As it can be seen from the Table 3.21, spread has decreased and T2 has increased when compared to that of Mix 2. This is because of increase in percentage of sand in mortar from 4% (Mix 2) to 45% (Mix 3). Table 3.21 Spread and T2 for the Mix 3 Time after water mixing (min) w/cm.36 Mix 3 %SP Initial %VMA Spread (mm) T2 (sec) Spread + Halo (mm)

37 99 But, the behaviour of the Mix 3 is almost similar to that of the Mix 2. Interestingly, maximum spread 293mm was observed at.9% SP dosage which was saturation point. It is to be noted that irrespective of sand content in mortar volume, if cementitious proportions are kept the same for the mixes, the dosage of SP (i.e., saturation point) tends to be the same for those mixes Consistence Retention As it can be seen from Table 3.22, all these three mixes attained good consistence retention in the spread and T2 at 45 and 6 minutes after adding water. So, it can be stated that the used chemical admixtures had good compatibility with the cement and mineral admixture. Thus, polycarboxylate-type superplasticiser can provide higher consistence retention (Hanehara and Yamada, 1999)65. Hence after mixing, SCC should maintain the fresh properties during transportation and placing, generally for 6 to 9 minutes (Kasemchaisiri and Tangermsirikul, 28; RILEM TC 174 SCC, 2; Sonebi and Bartos, 2)95, 165, 186. Table 3.22 Spread and T2 at 45 and 6 minutes after adding water Time after water mixing (min) w/cm %SP %VMA Initial 45 min 6 min Spread (mm) T2 (sec) Spread (mm) T2 (sec) Spread (mm) T2 (sec).32 Mix Mix Mix

38 1 3.4 DESIGN OF M 25 GRADE OF SCC As there is no availability of SCC mix design standards for a particular strength of concrete, this investigation is focused on the design of M 25 grade of SCC satisfying both SCC performance and the desired strength Experimental Study Our objective was to develop M 25 grade of SCC with moderate fines for the use of normal building constructions. To make use of the SCC for normal buildings and to have adequate bond between aggregates and reinforcement in concrete structures, crushed granite stones of size 2 mm and 1 mm were used in this study. In this respect, 53 grade ordinary Portland cement (OPC 53), class F fly ash as an additive, crushed granite stones of size 2 mm and 1 mm, river sand, SP and VMA were used in preparing SCC mixes. The fresh properties that were determined are filling ability, passing ability and segregation resistance and consistence retention. The hardened properties that were determined are compressive strength, modulus of elasticity (MOE) and splitting tensile strength (STS) at different curing periods SCC Mix Design Several methods exist for the mix design of SCC. The general purpose mix design method was first developed by Okamura and Ozawa (1995)147. In this study, the key proportions for the mixes are done by volume as per EFNARC (22)49. As already known that the increased content of powder (fines) and admixture leads to higher sensitivity and cost, combination-type of SCC has been chosen in this study in the SCC mix design so as to use moderate powder and reasonable quantity of chemical admixtures. Three mixes were prepared with different paste contents (36.%, 37.7% and 38.8%) in order to evaluate the SCC fresh properties.

39 11 Keeping in view of the moderate fines, robustness and all SCC properties, w/cm was chosen as.36 (by weight of cementitious) for all mixes. As per EFNARC (22)49, minimum coarse aggregate content of 28% was maintained for all the mixes. Keeping in view of the savings in cost and land fill, greenhouse gas emissions, fresh, mechanical and durability properties of SCC, the replacement level of class F fly ash was kept at 35% as per IS 456 (2)76 for all mixes. In this study, only cementitious material was used as fines and fines (<.125 mm) from the aggregates was not included. 2 mm and 1 mm coarse aggregate particles were blended in 6:4 proportion by percentage weight of total aggregate. The detailed steps for mix design are described as follows: 1. Assume air content as 2% (2 litres) of concrete volume. 2. Determine the dry-rodded unit weight (DRUW) of coarse aggregate for a given coarse aggregate blending. 3. Using DRUW, calculate the coarse aggregate content by volume (28%) of mix volume. 4. Adopt fine aggregate volume of 4 to 5% of the mortar volume. 5. Adopt the required paste volume (litre/m3) in the concrete volume. 6. Keep water/ cementitious ratio by weight (w/cm) as Calculate the binder (cementitious material) content by weight. 8. Replace cement with 35% class fly ash by weight of cementitious material. 9. Optimize the dosages of superplasticiser (SP) and viscosity modifying agent for the given w/cm (.36) using mortar tests by mini slump cone test. 1. Perform SCC tests.

40 SCC Mix Target Typical acceptance criteria and target for SCC are shown in Table Table 3.23 Typical acceptance criteria and target for SCC Property Test method Unit Slump Flow SCC mix target Minimum Maximum Mm T5cm Sec 2 5 V-funnel Sec 6 12 Passing ability L-box h2/h1 (mm/mm).8 1. Segregation V-funnel at resistance T5min Sec 6 15 Filling ability Mixing Procedure for SCC It is known that mixing procedures have a significant influence on the fresh properties of SCC. Liu s (21)118 mixing procedure was carried out throughout this work to achieve maximum efficiency of SP and VMA as shown in Fig Fig Mixing procedure for SCC with SP and VMA

41 13 Mixing procedure for SCC shown in Fig is described as follows: 1. Binder and aggregate are mixed for one minute. 2. The 1st part (7%) of water was added and mixed for two minutes. 3. SP along with the 2nd part (15%) of water was added and mixed for two minutes. 4. VMA along with the 3rd part (15%) of water was added and mixed for two minutes. 5. The mix was stopped and kept rest for 2 minutes. 6. The mix was remixed for one minute and discharged for SCC tests Testing Fresh Properties of SCC The test procedures for slump flow, V-funnel and L-box tests are described in the section to determine slump flow, T5cm, V-funnel time at initial and after 5 minutes (T5min), blocking ratio (h2/h1), segregation resistance and consistence retention Testing Hardened Properties of Concrete The test procedures for compressive strength, modulus of elasticity and splitting tensile strength tests are described in the section These properties were tested on cylinder specimens of size 15 mm x 3 mm for all the mixes Conventional Concrete Mix Design M 25 grade of conventional concrete (CC) has been designed (refer Appendix C) as per IS 1262 (29)85 and IS 456 (2)76 for comparative study Mix Proportions Mix types with percentage relative proportions and mix proportions of constituent materials are shown in Table 3.24 and Table SCC mix design tool ( JGJ_SCCMixDesign.xls ) is used to obtain the SCC mix proportions.

42 14 Table 3.24 Percentage relative proportions of trial mixes Cementitious material OPC+35% fly ash Percentage Coarse aggregate of blending percentage Mix type coarse by weight aggregate (2 mm and 1 mm) w/cm.36 for SCC Percentage Percentage of of Percentage of sand in Mortar paste mortar By volume 28_6:4Aa _6:4Bb _6:4c M a, b c 28_6:4A & 28_6:4B mixes with a paste content of 36.% and 37.7% respectively. 28_6:4: where 28 is the percentage of coarse aggregate volume in a concrete mix and 6:4 is the coarse aggregate blending by percentage weight of 2 mm and 1 mm resp. Table 3.25 Mix proportions of constituent materials of trial mixes Binder Cement Fly ash Water 2mm 1mm Sand SP VMA Kg/m3 kg/m3 kg/m3 l/m3 Kg/m3 kg/m3 kg/m3 l/m3 l/m3 28_6:4A _6:4B _6: M Mix type

43 Results and Discussion SCC Fresh Properties SCC fresh properties i.e., the average values of slump flow, T5cm at initial and at 6 minutes, V-funnel time, V-funnel time at 5 minutes (T5min) and L-box ratio (h2/h1) are presented in the Table 3.26 for all the mixes. Table 3.26 Fresh properties of trial mixes Slump flow (mm) Mix type Initial At 6 min T5cm (sec) Initial At 6 min V-funnel time (sec) L-box ratio (h2/h1) Initial T5min 28_6:4A 657 NDa 5.21 ND Blocked Blocked Blocked 28_6:4B _6: a ND: Not done From the Table 3.26, it is seen that though the mix 28_6:4A got the slump flow spread of 657 mm in 5.21 sec, this mix was failed in V-funnel and L-box tests. The blocking was due to insufficient amount of fines (458 kg/m 3) and paste content (36.%). Whereas, it is seen from the Table 3.26 that the two mixes 28_6:4B and 28_6:4 have met the SCC acceptance criteria. It is clearly observed that the increase in fines (481 kg/m3) and paste content (37.7%) as in the case of 28_6:4B and the increase in fines (495 kg/m3) and paste content (38.8%) as in the case of 28_6:4 resulted in the improvement of SCC fresh properties in these two mixes. Though all the three mixes have the same coarse content (28%) and coarse aggregate blending (6:4), adequate fines and paste content played a key role to achieve successful SCC

44 16 mixes. When comparing the two mixes 28_6:4B and 28_6:4, the mix 28_6:4 performed excellent SCC fresh properties than that of the mix 28_6:4B. This is particularly due to 495 kg/m3 of fines content and 38.8% of paste content of the mix 28_6:4 that resulted in the improvement of the fresh properties. Hence, it is to be concluded that for a given coarse aggregate content and coarse aggregate blending, mixes should have adequate fines and paste content to attain SCC acceptance criteria. It is evidently revealed that for a given coarse aggregate content and its blending, the decrease in fines content decreases the paste content and hence decreases the performance of SCC. From the above results, it is concluded that for the given coarse aggregate content of 28% with the coarse aggregate blending 6:4 (2 mm and 1 mm), the fines content of 495 kg/m3 can be considered as moderate fines and the paste content of 38.8% can be considered as an adequate paste content for a given w/cm (.36) Compressive Strength of CC and SCC From the results obtained in the fresh properties of SCC as shown in Table 3.26, the mixes 28_6:4B and 28_6:4 were considered as successful SCC mixes. Compressive strength results of M 25 and SCC mixes 28_6:4B and 28_6:4 after 7 and 28 days of curing are presented in the Table Table 3.27 Compressive strength of CC and SCC Mechanical property Compressive strength, f c (MPa) Mix type Age (days) M 25 28_6:4B 28_6:

45 17 From the Table 3.27, it is clearly seen that the mix 28_6:4 has attained a compressive strength of 32 MPa after 28 days of curing which was equivalent to 28-day compressive strength of M 25 grade of CC. But the mix 28_6:4B has attained only a compressive strength of MPa after 28 days of curing and was not meeting the requirement of M 25 grade of concrete. This is particularly due to less amount of binder (481 kg/m3) as compared to that (495 kg/m3) of the mix 28_6:4 for a given w/cm (.36) as shown in Table It clearly indicates that for a given w/cm, reduction in fines decreases the paste content and hence decreases the compressive strength of the mix. Moreover, the mix 28_6:4B was compliance with M 2 grade of concrete as per IS 456 (2)76 and IS 1262(29)85. So, for a given w/cm (.36) and 35% replacement level of class F fly ash, the total binder (cementitious) content of 495 kg/m3 i.e., cement of kg/m3 and class F fly ash of kg/m3 has been observed to be the adequate binder content for the mix 28_6:4 to attain 32 MPa. From the results, it is concluded that for a given w/cm and replacement level of class F fly ash, the reduction in the binder content decreases the cement content, fly ash content and paste content and hence decreases the compressive strength of the mix. From the compressive strength results, the mix 28_6:4 has been considered as M 25 grade of SCC Mechanical Properties of CC and SCC Table 3.28 shows the mechanical properties of CC (M 25) and SCC (28_6:4) after 7, 28, 56 and 112 days of curing. As it is seen from the Table 3.28, SCC has attained a lower compressive strength of MPa as compared to that of (23.2 MPa) CC after 7 days of curing due to slower pozzolanic action of class F fly ash at early ages (Siddique, 23)182.

46 18 Table 3.28 Mechanical properties of CC and SCC (28_6:4) Mechanical property Mix type Age (days) M 25 28_6: Compressive strength, f c (MPa) Modulus of elasticity, Ec (GPa) Splitting tensile strength, fct (MPa) Compressive strength (MPa) CC 4 SCC Age (Days) Fig Compressive strength versus Age 112

47 19 After 28 days of curing, SCC and CC have attained a similar compressive strength of 32 MPa. SCC has attained compressive strength of MPa and 48.1 MPa at 56 and 112 days respectively, whereas CC has attained compressive strength of MPa and 39.5 MPa at 56 and 112 days respectively. So, it is clearly seen that the gain of compressive strength was significant in SCC after 28 days as compared to that of CC as shown in Fig This is particularly due to continuous pozzolanic action of class F fly ash at later ages (Siddique, 23)182. For a given compressive strength of 32 MPa at 28 days, MOE of SCC and CC were GPa and GPa respectively as shown in Table It is observed that for a given compressive strength after 28 days of curing, SCC has attained lower MOE than that of CC. It is mainly due to lower coarse aggregate in SCC than that of CC. It is already known that for a given strength higher coarse aggregate content leads to higher unit weight and hence higher MOE of concrete (AASHTO, 26; ACI 318, 1995; Noguchi et al., 29; Tomosawa et al., 199)3, 7, 14, 198. But there was significant improvement observed in the MOE of SCC after 28 days as shown in Fig After 112 days of curing, SCC has attained higher value of MOE than that of CC as shown in Table 3.28 and Fig This is due to continuous improvement in the compressive strength of fly ash blended SCC at later ages due to pozzolanic action of class F fly ash (Siddique, 23)182. For a given compressive strength at 28 days, SCC has attained lower splitting tensile strength (STS) than that of CC as shown Table It is mainly due to the use of 35% of class F fly ash replacement in the cement and attributed to the slower pozzolanic action of fly ash that decreases the STS at early ages (Liu, 21)118. But there was significant improvement observed in the STS of SCC after 28 days as shown in Fig

48 11 35 CC SCC Modulus of elasticity (GPa) Age (Days) Fig Modulus of elasticity versus Age 4.5 Splitting tensile strength (MPa) 4 CC SCC Age (Days) Fig Splitting tensile strength versus Age

49 111 After 112 days of curing, SCC has attained higher value of STS (4.39 MPa) than that of (4.24 MPa) CC as shown in Table 3.28 and Fig Studies already revealed that mechanical properties of fly ash concrete continued to increase with age (Siddique, 23; Siddique, 211; Liu, 21)182, 183, 118. From the above results, it is to be noted that the designed M 25 grade of SCC was performing enhanced mechanical properties at later ages as compared to those M 25 grade of CC. So, the adequate binder content (495 kg/m 3) and 35% replacement level of class F fly ash in the mix 28_6:4 was resulting enhanced fresh and mechanical properties of SCC. 3.5 EFFECT OF COARSE AGGREGATE BLENDING ON FRESH PROPERTIES OF SCC This section is mainly focused on the effect of coarse aggregate blending on fresh properties of SCC and thereby finding the optimization of coarse aggregate blending or proportion in a given coarse aggregate volume of self compacting concrete (SCC). To ensure its high filling ability, flow without blockage and to maintain homogeneity, SCC requires a reduction in coarse aggregate content, high cement content, superplasticiser (SP) and viscosity modifying agent (VMA) (Okamura and Ouchi, 1999)143. According to Okamura (1997)141, blocking depends on the size, shape and content of coarse aggregate. A reduction in the coarse aggregate content and lowering the size are both effective in inhibiting blocking, but it leads to higher drying shrinkage of SCC. According to ACI 237R-7 (27)6, the blending of different sizes of coarse aggregate can often be beneficial to improve the overall properties of the mixture. As a guideline to minimize blocking of SCC, if coarse aggregate size is greater than

50 mm (1/2 in.), then the coarse aggregate content should be chosen between 28% and 32% of the volume of the mix (ACI 237R-7, 27)6. Keeping in view of the drying shrinkage of SCC, attempts have been done in this study to increase the volume of maximum size aggregate and coarse aggregate content. As such, there is no data or specification available to specify the coarse aggregate content that can be suitable for a particular coarse aggregate blending made with 2 mm and 1 mm to make successful SCC. This leads to this research work to determine the suitable (optimum) coarse aggregate blending with 2 mm and 1 mm for the particular coarse aggregate content to obtain successful SCC Experimental Study Our objective was to determine the optimum blending of coarse aggregate with 2 mm and 1 mm for a given coarse aggregate content to make successful SCC. Keeping in view of the M 25 grade of SCC (refer section 6.1), w/cm and paste content was kept at.36 (by weight) and 388 litre/m3 respectively for all mixes. To study the effect of coarse aggregate blending on SCC, 2 mm and 1 mm size aggregates are blended in 7:3, 67:33, 6:4 and 4:6 proportions by percentage of weight of total coarse aggregate. The fresh properties that were determined are filling ability, passing ability and segregation resistance and consistence retention SCC Mix Design SCC mix design procedure was described in the section Coarse aggregate content was chosen between 28 and 35% of concrete volume. The key proportions of constituents of SCC mixes were obtained by using the SCC mix design tool (JGJ_SCCMixDesign.xls).

51 Mix Proportions Mix types with percentage relative proportions and mix proportions of constituent materials are shown in Table 3.29 and Table 3.3. Table 3.29 Percentage relative proportions of mixes Cementitious Material OPC+35% fly ash w/cm.36 Percentage Percentage Coarse aggregate Percentage Percentage of of blending percentage of of Mix type coarse sand in by weight mortar paste aggregate mortar (2 mm and 1 mm) By volume 35_7: _7: _7: _7: _7: _7: _67: _67: _67: _67: _6: _6: _6: _6: _4: _4: _4:

52 114 Table 3.3 Mix proportions of constituent materials Mix type Binder Kg/m 3 Cement Kg/m 3 Fly Ash Kg/m 3 Water l/m 3 2mm 1mm Sand kg/m Kg/m kg/m SP l/m VMA 3 l/m3 35_7: _7: _7: _7: _7: _7: _67: _67: _67: _67: _6: _6: _6: _6: _4: _4: _4:

53 Results and Discussion SCC Fresh Properties SCC fresh properties i.e., the average values of slump flow, T5cm at initial and at 6 minutes, V-funnel time, V-funnel time at 5 minutes (T5min) and L-box ratio (h2/h1) are presented in the Table 3.31 for all the mixes. Table 3.31 Fresh properties of SCC Slump flow (mm) Mix type Initial At 6 min T5cm (sec) Initial At 6 min V-funnel time (sec) Initial T5min L-box ratio (h2/h1) 35_7:3 632 NDa 8.67 ND Blocked ND Blocked 34_7:3 637 ND 8.53 ND Blocked ND Blocked 33_7:3 641 ND 8.47 ND Blocked ND Blocked 32_7:3 645 ND 8.35 ND Blocked ND Blocked 3_7:3 659 ND 7.19 ND Blocked ND Blocked 28_7:3 667 ND 6.36 ND Blocked ND Blocked 35_67:33 64 ND 8.2 ND Blocked ND Blocked 32_67: ND 7.92 ND Blocked ND Blocked 3_67: ND 6.23 ND Blocked ND Blocked 28_67: ND 4.51 ND Blocked 35_6:4 647 ND 7.64 ND Blocked ND Blocked 32_6:4 662 ND 6.86 ND Blocked ND Blocked 3_6:4 687 ND 4.6 ND Blocked 28_6: _4:6 688 ND 5.6 ND Blocked 32_4: _4: a ND: Not done

54 116 From the Table 3.31, it is seen that though the mix 35_7:3 got the slump flow spread of 632 mm in 8.67 sec, this mix was failed in V-funnel and L-box tests. For the given blending 7:3, the reduction in coarse aggregate content from 35% to 28% increased the slump flow spread and the rate of flow, but all these mixes have failed in V-funnel and L-box tests. The blocking was due to 7% of 2 mm aggregate in a coarse aggregate content of 28-35% in the mix that leads to more collision and internal friction within the coarse aggregate particles. As the minimum coarse aggregate content was taken as 28% in this study, effect of the blending 7:3 was not tried for the coarse aggregate content less than 28%. Then, the coarse aggregate blending has been changed from 7:3 to 67:33 and performed the SCC tests on the mixes 35_67:33, 32_67:33, 3_67:33 and 28_67:33. From the results, it is seen that though the mix 35_67:33 got the slump flow spread of 64 mm in 8.2 sec as shown in Fig. 3.2, this mix was failed in V-funnel and L-box tests. The blocking was due to 67% of 2 mm aggregate in 35% of coarse aggregate volume in the mix that leads to more collision and internal friction within the coarse aggregate particles. Similarly, though the mixes 32_67:33 and 3_67:33 got the slump flow spread of 651 mm in 7.92 sec and 672 mm in 6.23 sec respectively, these mixes were also failed in V-funnel and L-box tests due to more internal friction within the coarse aggregate particles. The mix 28_67:33 was successful both in slump flow and V-funnel tests, but this mix was failed in L-box test. Though the coarse aggregate volume was reduced to 28%, the influence of 67% of 2 mm aggregate in the coarse aggregate content also caused the blocking. So, for the coarse aggregate blending 67:33, the mixes 35_67:33, 32_67:33, 3_67:33 and 28_67:33 have not met the SCC acceptance criteria.

55 117 Then, the coarse aggregate blending has been changed from 67:33 to 6:4 and performed the SCC tests on the mixes 35_6:4, 32_6:4, 3_6:4 and 28_6:4. The mixes 35_6:4 and 32_6:4 got the slump flow spread of 647 mm in 7.64 sec and 662 mm in 6.86 sec respectively, but these mixes were failed in V-funnel and L-box tests. From the results, it is observed that the mix 3_6:4 was failed in L-box test as shown in Fig. 3.23, though it was successful in slump flow and V-funnel tests. Among the four mixes 35_6:4, 32_6:4, 3_6:4 and 28_6:4, it is observed that only the mix 28_6:4 was successful and met the SCC acceptance criteria as shown in Table 3.31 and Figs and The reason was that for the given blending 6:4, the increase of coarse aggregate content increased the volume of 2 mm aggregate and leads to blocking of the aggregates. This is inline with the statement that the filling ability and passing ability decreases with an increase in the coarse aggregate content in concrete (Okamura and Ouchi, 23b)145. As compared to the mixes 28_7:3 and 28_67:33, the blocking of coarse aggregate was not observed in L-box test in the mix 28_6:4 as shown in Fig It is to be noted that the 6:4 blending reduced the yield stress or internal friction and increased the deformation rate. Then, the coarse aggregate blending has been changed from 6:4 to 4:6 and performed the SCC tests on the mixes 34_4:6, 32_4:6 and 28_4:6. It is clearly seen from the results, the mix 34_4:6 was failed in L-box test, though it was successful in slump flow and V-funnel tests. From the results, it is observed that for coarse aggregate blending 4:6, both the mixes 28_4:6 and 32_4:6 were met the SCC acceptance criteria. It is practically seen that the influence of 4:6 (2 mm and 1 mm) blending didn t significantly affect the fresh properties of the mix 32_4:6. It can be stated that lower volume of maximum size aggregate (i.e., 2 mm) can lead to increase the coarse aggregate

56 118 content to some reasonable extent. But the increase of coarse aggregate content from 32% to 34% for the same blending 4:6 caused the mix 34_4:6 blocking in L-box test. This is due to more coarse aggregate content as compared to that of the mix 32_4:6. So, it is practically observed from the results that both coarse aggregate maximum size and coarse aggregate volume are influential in obtaining the passing ability of SCC and the same is confirming to ACI 237R-7 (27)6. Interestingly it is seen that both mixes 28_6:4 and 32_4:6 were almost similarly performed the fresh properties. So, it is to be mentioned that if higher volume of maximum size (2 mm) aggregate has to be used, the coarse aggregate content has to be adjusted. Similarly, if the coarse aggregate content has to be increased, the volume of maximum size (2 mm) aggregate has to be adjusted. So, either volume of maximum size (2 mm) aggregate or coarse aggregate volume has to be adjusted for a particular coarse aggregate blending to obtain successful SCC mixes. For the failure mixes, the consistence retention measurements i.e., slump flow and T5cm at 6 minutes were not performed. Segregation resistance of all successful SCC mixes has been assessed by visual stability index (VSI) from the slump flow test. All successful SCC mixes have been considered as stable mixes (VSI 1) as described in Table 2.1. The typical range of coarse aggregate content suitable for a particular coarse aggregate blending is represented as shown in Table Table 3.32 Coarse aggregate content for a particular coarse aggregate blending Coarse aggregate blending (2 mm & 1 mm) Coarse aggregate content 7:3, 67:33 <28% 6:4 28% 4:6 28% to 32%

57 119 Fig. 3.2 Slump flow of 35_67:33 mix Fig Slump flow of 28_6:4 mix

58 12 Fig L-box test of 28_6:4 mix Fig L-box test of 3_6:4 mix

59 EFFECT OF COARSE AGGREGATE BLENDING ON MECHANICAL PROPERTIES OF SCC This section is mainly focused on the effect of coarse aggregate blending on mechanical properties of SCC and these properties were compared to a conventional concrete (CC) of a given compressive strength. The typical mix proportions of SCC are different as compared to those of CC. These could influence the mechanical properties of SCC include unit weight, compressive strength, modulus of elasticity (MOE) and splitting tensile strength (STS). These mechanical properties are crucial to the design and performance of concrete structures Unit Weight As per ACI 237R-7 (27)5, MOE of concrete is related to its compressive strength, aggregate type and content, and unit weight of concrete. AASHTO LRFD (26)3 or ACI 318 (ACI, 1995)7 proposed MOE of concrete as a function of its compressive strength and unit weight. Noguchi et al. (29)14 expressed a conventional equation for MOE of concrete as a function of its compressive strength and unit weight of concrete made with light weight, normal weight and heavy weight aggregates. It is revealed from many conventional equations that coarse aggregate affect the value of MOE of concrete through the value of its unit weight (AASHTO, 26; Tomosawa at al., 199)3, 198. Tomosawa (199)198 considered unit weight of concrete at the time of compression test Experimental Study From the literature, it is revealed that the maximum size of coarse aggregate and coarse aggregate content affects both the fresh and hardened properties of SCC. The objective of this research work is to determine the effect of coarse aggregate blending with 2 mm and 1 mm in a particular coarse aggregate content on

60 122 mechanical properties (unit weight, compressive strength, MOE and STS) of SCC at different curing periods. And these properties were compared to a M 25 grade of CC. From the results of SCC fresh properties shown in Table 3.31, the three successful SCC mixes 28_6:4, 28_4:6 and 32_4:6 were selected for this study. A new parameter called coarse aggregate points (CAP) has been introduced to study the effect of coarse aggregate blending on mechanical properties of SCC. Our objective was to determine the effect coarse aggregate blending (6:4 and 4:6) in a coarse aggregate content (28% and 32%) on mechanical properties of SCC. The hardened properties that were determined are unit weight, compressive strength, modulus of elasticity and splitting tensile strength after 7, 28, 56 and 112 days of curing. The measured MOE of all mixes were compared with ACI 363R (ACI, 1992)8 and AASHTO LRFD (26)3/ ACI 318 (ACI, 1995)7 predicted equations. The measured STS of all mixes were compared with ACI 363R (ACI, 1992)8 and CEB-FIP (199)26 predicted equations Testing Hardened Properties of SCC and CC The test procedures for compressive strength, modulus of elasticity and splitting tensile strength tests are described in These properties were tested on cylinder specimens of size 15 mm x 3 mm for all the mixes. Unit weight or density of hardened concrete (γc) was determined after 7, 28, 56 and 112 days of curing prior to compression test. Weight of cylindrical specimen was measured prior to compression testing and there by unit weight has been calculated by measuring the volume of cylindrical specimen.

61 Coarse Aggregate Points A new parameter, coarse aggregate points (CAP) has been introduced in this study to evaluate the effect of coarse aggregate blending in a particular coarse aggregate content on the mechanical properties of SCC. CAP represents a numerical value based on the size of coarse aggregate, coarse aggregate blending in a given coarse aggregate content. Size of coarse aggregate, coarse aggregate blending and volume of coarse aggregate are used in the calculation of CAP. CAP will be calculated for any size of coarse aggregate, coarse aggregate with or without blending in any coarse aggregate content of SCC mix. CAP can be calculated as below: Let us consider a SCC mix with coarse aggregate content of 28% of concrete volume. Coarse aggregate of sizes 2 mm and 1 mm with coarse aggregate blending 6:4 by percentage weight of total aggregate are used in this mix. Coarse aggregate volume : 28% or 28 litre/m3 2 mm contribution : 6% 1 mm contribution : 4% CAP (coarse aggregate points) : [28*(6/1)*2]+[28*(4/1)*1] = Mix Proportions Mix types with percentage relative proportions along with CAP and mix proportions of constituent materials of SCC and CC are shown in Table 3.33 and Table 3.34.

62 124 Table 3.33 Percentage relative proportions of SCC and CC Cementitious material OPC+35% fly ash Coarse aggregate Mix type blending percentage by weight (2 mm and 1 mm) Percentage of coarse aggregate w/cm.36 for SCC Percentage Percentage of Mortar of sand in mortar Percentage of CAP Paste By volume 28_6: _4: _4: M Table 3.34 Mix proportions of constituent materials of SCC and CC Binder Cement Fly ash Water 2mm 1mm Sand SP VMA kg/m3 kg/m3 Kg/m3 l/m3 Kg/m3 kg/m3 kg/m3 l/m3 l/m3 28_6: _4: _4: M Mix type

63 Results and Discussion Hardened mechanical properties of M 25 grade of CC and SCC mixes are presented in the Table Table 3.35 Mechanical properties of CC and SCC Mechanical property Age (days) Mix type M 25 28_6:4 28_4:6 32_4: Compressive strength, f c (MPa) Modulus of elasticity, Ec (GPa) Splitting tensile strength, fct (MPa) Unit weight, γc (kg/m3)

64 Unit Weight or Density of Hardened Concrete As it can be seen from the Table 3.35, all SCC mixes have attained lower unit weight than that of M 25 (CC) after 7, 28, 56 and 112 days of curing. Because, CC has more coarse aggregate volume (i.e., 43.81%) as compared to that of SCC mixes. Hence for a given concrete strength, higher the coarse aggregate content, higher is the unit weight of concrete. For a given coarse aggregate content (28%), the mix 28_6:4 has got more unit weight than that of the mix 28_4:6 at all ages. It is mainly due to the higher content of 2 mm (6%) in the mix 28_6:4 as compared to that of the mix 28_4:6. So, it is practically seen that for a given coarse aggregate content and concrete strength, higher the maximum size aggregate volume in a coarse aggregate blending, higher is the unit weight of concrete. It is concluded that the change in coarse aggregate blending in a given coarse aggregate content, certainly affects the unit weight of concrete. For a given coarse aggregate blending (4:6), the mix 32_4:6 has got more unit weight than that of the mix 28_4:6 at all ages. It is mainly due to the increase in coarse aggregate content from 28% to 32%. Hence, for a given coarse aggregate blending and concrete strength, higher the coarse aggregate content, higher is the unit weight of concrete. The effect of coarse aggregate blending can also be discussed in the CAP point of view. When comparing the mixes 28_6:4 and 28_4:6, the mix 28_6:4 with a high CAP value (448) attained high unit weight of concrete as compared to that of the mix 28_4:6 having a low CAP value (392). In other words, for a given coarse aggregate content and concrete strength, higher the CAP value, higher is the unit weight of concrete. Similarly, if we compare the mixes 28_4:6 and 32_4:6,

65 127 the mix 32_4:6 with a high CAP value (448) attained high unit weight of concrete as compared to that of the mix 28_4:6 having a low CAP value (392). Hence, for a given coarse aggregate blending and concrete strength, higher the CAP value, higher is the unit weight of concrete. Interestingly, the mixes 28_6:4 and 32_4:6 with the same CAP value (448) have got almost the same value of unit weight of concrete at all ages. This trend can be easily assessed by knowing the CAP value of these two mixes. Hence, it is concluded that for a given concrete strength irrespective of coarse aggregate blending and coarse aggregate content, the mixes with the same CAP value can exhibit almost similar values of unit weight of concrete at all ages. So, the effect of coarse aggregate blending and coarse aggregate content on unit weight of concrete for the given strength can be easily assessed by knowing the CAP value of the mix Compressive Strength All SCC mixes have attained almost the same value of compressive strength at all ages. It is seen that the effect of coarse aggregate blending has not been observed on the compressive strength of SCC mixes at all ages. So, it is agreed that the compressive strength of SCC is mainly controlled by the composition of the binder and w/cm (Domone, 26b)47. After 56 days of curing, all SCC mixes have attained significantly more compressive strength than that of CC. This is mainly due to the influence of class F fly ash in SCC mixes. Hence, it is agreed that fly ash blended concrete mixes attains significantly more compressive strength than that of plain concrete at later ages i.e., after 28 days (Siddique, 23)182.

66 Modulus of Elasticity It is observed that for the given concrete strength after 28 days of curing, all SCC mixes have attained lower MOE than that of CC. It is mainly due to more coarse aggregate content in CC than that of SCC mixes. As it is already seen, higher coarse aggregate content leads to higher unit weight of concrete for the same strength. Due to higher unit weight of concrete, CC has attained high value of MOE as compared to SCC mixes. Hence, it is agreed that the coarse aggregate affects the value of MOE of concrete through its unit weight (AASHTO, 26; Tomosawa et al., 199)3, 198. So it is concluded that, for a given concrete strength, higher the coarse aggregate content, higher the unit weight of concrete and hence higher is the MOE of concrete. After 56 days of curing, though MOE of SCC mixes were less than that of CC, a significant improvement in MOE has been observed in SCC mixes as shown in Fig This is particularly due to increase in strength of all SCC mixes. So, it is well accepted that fly ash blended SCC mixes will continue to increase in strength with age that tends to increase MOE of SCC mixes (Siddique, 23)182. For a given coarse aggregate content (28%), the mix 28_6:4 has attained high MOE than that of the mix 28_4:6 at all ages. It is mainly due to higher value of unit weight of the mix 28_6:4 as compared to that of the mix 28_4:6. The effect coarse aggregate blending has been observed clearly in these two mixes with respect to MOE. So, it is concluded that for a given coarse aggregate content and concrete strength, higher the maximum size aggregate volume in coarse aggregate blending, higher the unit weight and hence higher is the MOE of concrete. For a given coarse aggregate blending (4:6), the mix 32_4:6 has attained high MOE than that of the mix 28_4:6. It is mainly attributed to the higher unit weight of the mix 32_4:6. So, it is concluded that for a given coarse aggregate

67 129 blending and concrete strength, higher the coarse aggregate content, higher the unit weight and hence higher is the MOE of concrete. The effect of coarse aggregate blending and coarse aggregate content can also be discussed in CAP point of view. As we compare the mixes 28_6:4 and 28_4:6, the mix 28_6:4 with a high CAP value (448) attained high unit weight that leads to high MOE of concrete as compared to the mix 28_4:6 having a low CAP value (392). So, it is concluded that for a given coarse aggregate content and concrete strength, higher the CAP value, higher the unit weight and hence higher is the MOE of concrete. Similarly, when we compare the mixes 28_4:6 and 32_4:6, the mix 32_4:6 with a high CAP value (448) attained high unit weight that leads to high MOE of concrete as compared to the mix 28_4:6 having a low CAP value (392). So, it is to be said that for a given coarse aggregate blending and concrete strength, higher the CAP value, higher the unit weight and hence higher is the MOE of concrete. Also, it is observed that both the mixes 28_6:4 and 32_4:6 with the same CAP value (448) attained almost the same value of MOE of concrete at all ages irrespective of coarse aggregate blending and coarse aggregate content as shown in Fig So, it can be concluded that for a given concrete strength irrespective of coarse aggregate blending and coarse aggregate content, the mixes with the same CAP value attains almost the same value of unit weight and MOE of concrete. So, by knowing the CAP value of the mix, the trend of MOE of concrete can be easily assessed for the given concrete strength.

68 M25 28_6:4 2 CAP 448 CAP 392 CAP 448 CAP 71 CAP 448 CAP 392 CAP 448 CAP 71 CAP 448 CAP 392 CAP _4:6 CAP 71 Modulus of Elasticity (GPa) 35 32_4: Age (days) Fig MOE versus age The ACI 363R (ACI, 1992)8 and AASHTO LRFD (26)3 or ACI 318 (ACI, 1995)7 predicted equations for MOE of concrete are presented in the Table Table 3.36 Expressions for MOE Code of practice ACI 363R (ACI 1992) AASHTO LRFD/ ACI 318 Expression for Ec (MPa) 332 f c' (γc) f c' Range of concrete strength No specified maximum strength 21 MPa < f c < 83 MPa

69 131 The measured MOE of all mixes after 28, 56 and 112 days of curing have been compared with the ACI 363R (ACI, 1992)8 and AASHTO LRFD (26)3 or ACI 318 (ACI, 1995)7 predicted equations and presented in the Table Table 3.37 Comparison of measured and predicted MOE of all mixes Mix type Age (days) Modulus of elasticity, Ec (GPa) Experiment ACI 363R ACI _4: M25 (CC) _4: M25 (CC) M25 (CC) 28_6:4 28_4:6 28_6:4 28_4:6 28_6:4 28_4:6 32_4: From the Table 3.37, it is seen that ACI 363R8 equation predicted the lower values of Ec as compared to those of experimental values for all the mixes at 28, 56 and 112 days. Because, ACI 363R8 is not considering the unit weight of concrete in its equation. Whereas, AASHTO LRFD3 or ACI 3187 predicted almost the same values of Ec as compared to those of the experimental values for all the mixes at 28, 56 and 112 days.

70 132 As ACI 3187 included the unit weight of concrete, it is predicting the appropriate values of Ec when compared to ACI 363R8 as shown in Fig Hence, it is more reliable to use Ec predicted models which include both compressive strength and unit weight of concrete. 36 Modulus of Elasticity (GPa) Experiment 28 ACI363R ACI Compressive Strength (MPa) Fig MOE versus compressive strength of SCC mixes Splitting Tensile Strength It is observed from the results that for a given concrete strength, all SCC mixes have attained lower values of STS as compared to those of CC at 28 days. This is primarily due to the use of 35% of fly ash replacement in the cement in SCC mixes that tends to affect the aggregate-paste bond or interfacial transition zone (ITZ) between the aggregate and cement paste at early ages. This is mainly attributed to the slower pozzolanic action of fly ash that decreases the STS at early ages (Liu, 21)118.

71 133 Studies already revealed that though the STS is related to compressive strength, several factors affect the STS such as aggregate type, particle size distribution and mix design (Neville, 1988)136. Parra et al. (211)155 revealed that the use of higher fines or different superplasticisers affect the aggregate-paste bond, which thus have a higher influence on tensile strength than compressive strength. The interfacial transition zone characteristics tend to affect the tensile and flexural strength to a greater degree than compressive strength (Mehta and Monterio, 26)125. The second reason being the lower coarse aggregate content and proportion of SCC as compared to CC. Hence, it can be pointed out that for a given concrete strength, STS primarily depends on the paste composition of the mix and secondarily on the coarse aggregate content and its blending. At 56 days, though the values of STS of all SCC mixes were lower than that of CC, a significant improvement has been observed in the values of STS of all SCC mixes as shown in Fig This is mainly attributed to the use of fly ash in SCC mixes due to which SCC mixes have attained more compressive strength than that of CC at 56 days. This increase in compressive strength caused the increase in STS of SCC mixes. Hence, it is agreed that fly ash blended SCC mixes continued to increase in compressive strength with age and there by increase in splitting tensile strength (Siddique, 23)182. At later ages i.e., after 112 days of curing, both SCC mixes 28_6:4 and 32_4:6 have attained higher values of STS as compared to those of CC. Results shown that for a given coarse aggregate content (28%), the mix 28_6:4 has attained higher STS as compared to that of 28_4:6 for the given strength at all ages as shown in Fig

72 134 Hence, it is pointed out that for the coarse aggregate content and concrete strength, higher the maximum size coarse aggregate volume in coarse aggregate blending, higher is the CAP value and hence higher is the value of STS. If we compare the mixes 28_4:6 and 32_4:6, the mix 32_4:6 exhibited higher value of STS due to more coarse aggregate content as compared to that of the mix 28_4:6. Hence for a given coarse aggregate blending and concrete strength, higher the coarse aggregate content, higher is the CAP value and hence higher is the value of STS as shown in Fig It is agreed that concrete with higher coarse aggregate ratio (content) exhibits slightly higher STS as compared to that of concrete with lower aggregate ratio (content) (Mahdy et al., 22)121. So, it is concluded that coarse aggregate blending in a particular coarse aggregate content has significant effect on unit weight, MOE and STS of SCC mixes M25 28_6: Age (days) Fig STS versus age 112 CAP 448 CAP 392 CAP 448 CAP 71 CAP 448 CAP 392 CAP 448 CAP 71 CAP 448 CAP CAP _4:6 CAP 71 Splitting Tensile Strength (MPa) _4:6

73 135 The ACI 363R (ACI, 1992)8 and CEB-FIP (199)26 predicted equations for STS of concrete are presented in the Table Table 3.38 Expressions for STS Range of concrete Expression for fct (MPa) Code of practice.5 ACI 363R (ACI 1992).59 (f c) CEB-FIP (199) f c' strength 21 MPa < f c < 83 MPa 2 f c < 8 MPa The measured STS of all mixes after 28, 56 and 112 days of curing have been compared with the ACI 363R (ACI, 1992)8 and CEB-FIP (199)26 predicted equations and presented in the Table Table 3.39 Comparison of measured and predicted STS of all mixes Mix type Age (days) Splitting tensile strength, fct (MPa) Experiment ACI 363R CEB-FIP _4: M25 (CC) _4: M25 (CC) M25 (CC) 28_6:4 28_4:6 28_6:4 28_4:6 28_6:4 28_4:6 32_4:

74 136 It is seen from the Table 3.39, ACI 363R (ACI, 1992)8 equation gives a reasonable estimate of splitting tensile strength especially for SCC mixes. Whereas, CEB-FIP model (199)26, under estimates the value of STS as compared to those of experimental values for all mixes. If we consider CEB-FIP26 predicted values as lower bound and ACI 363R (ACI, 1992)8 predicted values as upper bound for SCC mixes, a suitable STS range can obtained for a given concrete strength after 28 and 56 days of curing as shown in Fig Splitting Tensile Strength (MPa) Experiment 3.6 ACI363R CEB-FIP Compressive Strength (MPa) Fig STS versus compressive strength of SCC mixes

75 137 As per the STS range (2.76 MPa 3.35 MPa) of SCC mixes at 28 days for a given strength, it can be concluded that all SCC mixes are within the acceptable range. Similarly, as per the STS range (3.27 MPa 3.68 MPa) of SCC mixes at 56 days for a given strength, it can be concluded that all SCC mixes are within the acceptable range. Also at 112 days, though the SCC mixes 28_6:4 and 32_4:6 have attained higher STS values as compared to those of CC, ACI 363R (ACI, 1992)8 estimated the STS values of SCC mixes reasonably when compared to CEB-FIP26. Hence, it can be concluded that though the coarse aggregate blending and coarse aggregate content have an effect on STS of SCC mixes at all ages for the given concrete strength, ACI 363R (ACI, 1992)8 predicted the STS values of SCC mixes reasonably. 3.7 DRYING SHRINKAGE This section is mainly focused on the determination of drying shrinkage of SCC and CC at different drying periods. Drying shrinkage strains need to be investigated as they can have adverse effects on the serviceability and durability of concrete. Due to higher paste volume and lower coarse aggregate content, SCC leads to higher drying shrinkage than that of CC (ACI 237R-7, 27)6. Ozyildirim and Lane (23)153 recommended a large nominal maximum aggregate size, large amount of coarse aggregate, and low water content to mitigate high drying shrinkage in SCC applications. Studies already revealed that the incorporation of fly ash reduced the drying shrinkage of SCC by densifying the paste matrix and also by serving the unhydrated fly ash particles as fine aggregate to restrain the shrinkage deformation (Poppe and De Schutter, 25; Khatib, 28; Gesoglu et al., 29)162, 96, 57.

76 Experimental Study Our objective was to investigate the drying shrinkage strains of M 25 grade of SCC and CC after 1, 7, 14, 28, 56 and 112 days of drying after 7 days of curing. From the results obtained in the compressive strengths of SCC mixes as shown in Table 3.35, the SCC mixes 28_6:4, 28_4:6 and 32_4:6 have been selected as M 25 grade of SCC mixes to evaluate the effect of coarse aggregate blending and its content on the drying shrinkage of SCC. The test procedure for determining the drying shrinkage of all mixes is described in the section For each concrete mixture, three 15 mm x 3 mm concrete cylinders were cast for drying shrinkage test. These cylinders were cured for 7 days. Drying shrinkage of all mixes was carried out using length comparator after 1, 7, 14, 28, 56 and 112 days of drying after 7 days of curing as shown in Fig Fig Shrinkage measurement of 28_6:4 at 112 days using length comparator

77 139 Mix types with percentage relative proportions along with CAP and mix proportions of constituent materials of M 25 grade SCC mixes and CC are shown in Tables 3.33 and Results and Discussion The average drying shrinkage strains of M 25 grade of SCC and CC after 1, 7, 14, 28, 56 and 112 days of drying after 7 days of curing are presented in the Table 3.4. The drying shrinkage of all mixes as a function of drying period was plotted and shown in Fig Table 3.4 Average drying shrinkage values of M 25 grade of SCC and CC Mix type Drying shrinkage (microstrain) 1-day 7-day 14-day 28-day 56-day 112-day 28_6: _4: _4: M It was observed that CC has attained a lower drying shrinkage strain of 475 x 1-6 (475 microstrain) after 112 days of drying as compared to that of all SCC mixes. This was mainly attributed to the lower paste volume, higher coarse aggregate content and higher CAP value of CC as compared to those of SCC mixes. For a given coarse aggregate content (28%), the mix 28_6:4 has exhibited slightly lower 112-day drying shrinkage than that of the mix 28_4:6. It is mainly due to the higher content of 2 mm (6%) in the mix 28_6:4 as compared to that of the mix 28_4:6. So, it is practically seen that for a given mix proportion and coarse aggregate content, higher the maximum size aggregate volume in a coarse aggregate blending, higher is the CAP value and lower is the drying shrinkage of concrete.

78 14 It is concluded that the change in coarse aggregate blending in a given coarse aggregate content certainly affects the drying shrinkage of concrete. For a given coarse aggregate blending (4:6), the mix 32_4:6 has exhibited slightly lower 112-day drying shrinkage than that of the mix 28_4:6. It is mainly due to the increase in coarse aggregate content from 28% to 32%. Hence, for a given mix proportion and coarse aggregate blending, higher the coarse aggregate content, higher is the CAP value and lower is the drying shrinkage of concrete. Among the three SCC mixes, the two SCC mixes 28_6:4 and 32_4:6 have attained almost the same value of 112-day drying shrinkage strain as shown in Table 3.4. Though these two SCC mixes had different coarse aggregate contents (28% and 32%) with different coarse aggregate blending (6:4 and 4:6), they had the same CAP value as shown in Table From the results, it can be concluded that for a given mix proportion, the mixes with the same CAP value can exhibit almost similar values of drying shrinkage of concrete. 7 Drying shrinkage (Microstrain) 6 5 M _6:4 28_4:6 3 32_4: Drying period (Days) Fig Drying shrinkage of SCC and CC

79 MICROLEVEL PROPERTIES This investigation is mainly focused on the effect of class F fly ash on the microlevel properties of SCC at different curing periods. The microlevel properties studied were the microcrack widths between aggregate and paste; and the chemical elements and atomic Calcium-Silica (Ca/Si) ratio of the paste near the interfacial transition zone (ITZ). It has been suggested that microcracks in the ITZ play an important part in determining not only the mechanical properties but also the permeability and durability of concrete (Mehta and Monteiro, 26)125. Modification of the microstructure in the ITZ has been one of great concern since durability, permeability and strength of concrete are significantly influenced. The addition of mineral admixtures is successful approach in improving the microstructure of concrete (Jing and Stroeven, 24)9. The pozzolanic reaction of fly ash reduces the amount of Ca(OH)2 produced and lowers the Ca/Si ratio of the C-S-H in the cement/fly-ash mix (Wesche, 1991)27. The pozzolanic reaction gives a fly ash concrete its fine pore structure, low permeability, long-term strength gain properties and enhanced durability properties (Sear, 21)175. The Scanning electron microscopy (SEM) has been a powerful tool in the examination of cement and concrete microstructure (Nemati, 1997; Stutzman, 2)133, 19. The energy dispersive x-ray analysis (EDAX) can be used for spectrum analysis to determine the chemical elements and their peaks along with their relative concentrations (Stutzman, 2) Experimental Study In the last few decades, a lot of research has been done regarding the improvement of the concrete performance. Self-compacting concrete with high cementitious content, a lower volume and maximum size of coarse aggregate are

80 142 expected to affect the ITZ in a positive manner, promoting a less porous microstructure and improvement in the mechanical properties. To better understand and correlate the micro and macrolevel (mechanical) properties, this investigation is mainly focused on the effect of class F fly ash on the micro and macrolevel properties of SCC after 28, 56 and 112 days of curing. SEM analysis was carried out to examine the widths of microcracks between aggregate and paste and EDAX analysis was carried out to determine the chemical elements and atomic Ca/Si ratio of the paste matrix near the ITZ at different ages. These properties were also examined on M 25 grade of CC at different days. For this investigation, M 25 grade of SCC mix 28_6:4 and M 25 grade of CC have been considered. The sample preparation and test procedure for determining the microlevel properties of concrete is described in the section To carry out SEM and EDAX, the samples of approximate size 1 mm x 1 mm x 5 mm were collected from the tested cylindrical specimens after the compressive strength tests as per IS 516 (1959)77 on SCC and CC at different ages. Mix types with percentage relative proportions and mix proportions of constituent materials of M 25 grade SCC and CC are shown in Tables 3.33 and Results and Discussion Microlevel Properties This section describes the microlevel properties i.e., the microcracking width and Ca/Si ratio of SCC and CC at different ages Microcracking Width SEM analysis was carried out on SCC and CC samples to observe the microcracking width between coarse aggregate and paste at different ages.

81 143 As the microcracking width is not uniform in concrete microstructure (Mindess et al., 23)126, the microcracking width has been measured at three different places for each sample at each age and the average microcracking width has been calculated and represented in Table Table 3.41 Microcrack widths of CC and SCC (28_6:4) Curing period (days) Microcrack width (μm) CC SCC The microstructure of SCC especially the widths of microcracking after 28, 56 and 112 days of curing have been shown in Figs. 3.3, 3.31 and 3.32 respectively. Fig. 3.3 SEM image of SCC microcrack width at 28 days

82 144 Fig SEM image of SCC microcrack width at 56 days Fig SEM image of SCC microcrack width at 112 days

83 145 It is seen from the Fig. 3.3, the microcracking width of SCC at 28 days was 3.73 μm. After 56 days of curing, the microcracking width was decreased from 3.73 μm to 1.47 μm as shown in Fig After 112 days of curing, the microcracking width was further reduced to.53 μm as shown in Fig It is clearly seen from the results that significant reduction in the microcracking width of SCC was observed with the increasing curing period. This significant reduction in the microcracking width is mainly due to the pozzolanic action of class F fly ash which improves the microstructure of ITZ at later ages. The microcracking widths of CC after 28, 56 and 112 days of curing have been shown in Figs. 3.33, 3.34 and 3.35 respectively. Fig SEM image of CC microcrack width at 28 days

84 146 Fig SEM image of CC microcrack width at 56 days Fig SEM image of CC microcrack width at 112 days

85 147 The microcracking width of CC at 28 days was 4.8 μm as shown in Fig After 56 days of curing, the microcracking width was decreased from 4.8 μm to.93 μm as shown in Fig From the Fig. 3.35, it is to be noted that there was no significant reduction in the microcracking width (.93 μm) after 112 days of curing as compared to the microcracking width (.67 μm) at 56 days. It is clearly seen from the results that there was no significant improvement observed in the microcracking width of CC at later ages as compared to that of fly ash blended SCC. Hence, improvement in the mechanical properties of CC was not significant as compared to that of SCC at later ages. As there was significant reduction in the microcracking width of SCC with the increasing curing period, improvement in the mechanical properties i.e., compressive strength, modulus of elasticity (MOE) and splitting tensile strength (STS) was observed in SCC as shown in the Table Though MOE and STS of SCC were lower than that of CC after 28 and 56 days of curing, SCC has attained higher MOE and STS than that of CC after 112 days of curing as shown in Figs and This is particularly due to significant improvement in the compressive strength of SCC at later ages. Hence, it is concluded that pozzolanic action of fly ash reduces the microcracking width of SCC with age that results in the improvement of bond between coarse aggregate and paste (Kuroda et al., 2; Wong and Buenfeld, 26; Xiong et al., 22)11, 28, 212 and improvement in the SCC mechanical properties Atomic Ca/Si Ratio EDAX analysis was carried out on the paste near the interface at different ages in order to determine the chemical elements with the peaks and their relative intensities.

86 148 Figs represent the spectrum analysis of chemical elements and their relative intensities of SCC after 28, 56 and 112 days of curing. Fig EDAX analysis of SCC at 28 days Fig EDAX analysis of SCC at 56 days

87 149 Fig EDAX analysis of SCC at 112 days Figs represent the spectrum analysis of chemical elements and their relative intensities of CC after 28, 56 and 112 days of curing. Fig EDAX analysis of CC at 28 days

88 15 Fig. 3.4 EDAX analysis of CC at 56 days Fig EDAX analysis of CC at 112 days