Mechanical Properties of Energy Efficient Concretes Made with Binary, Ternary, and Quaternary Cementitious Blends of Fly Ash, Blast Furnace Slag, and Silica Fume

Abstract

When the energy performance of concrete is substantially higher than that of normal type concrete, such concrete is regarded as energy efficient concrete (WBSCSD 2009). An experimental study was conducted to investigate mechanical properties of energy efficient concrete with binary, ternary and quaternary admixture at different curing ages. Slump test for workability and air content test were performed on fresh concretes. Compressive strength, splitting tensile strength were made on hardened concrete specimens. The mechanical properties of concrete were compared with predicted values by ACI 363R-84 Code, NZS 3101-95 Code, CSA A23.3-94 Code, CEB-FIP Model, EN 1991, EC 2-02, AIJ Code, JSCE Code, and KCI Code. The use of silica fume increased the compressive strengths, splitting tensile strengths, modulus of elasticities and Poisson’s ratios. On the other hand, the compressive strength and splitting tensile strength decreased with increasing fly ash.

1 Introduction

The cement industry accounts for approximately 5 % of current anthropogenic carbon dioxide (CO₂) emissions world-wide (WBSCSD 2009). World cement demand and production are increasing; annual world cement production is expected to grow from approximately 2540 million tones (Mt) in 2006 to between 3680 Mt (low estimate) and 4380 Mt (high estimate) in 2050. The largest share of this growth will take place in India, China, and other developing countries on the Asian continent (Liu et al. 2016). This significant increase in cement production is associated with a significant increase in the cement industry’s absolute energy use and CO₂ emissions.

The major difference between normal concrete and energy efficient concrete (EEC) is essentially the use of chemical and mineral admixtures. EEC is the concrete which meets special performance and uniformity requirements that cannot always be achieved by normal materials, normal mixing, placing and curing practices. There have been a number of attempts to develop a method for the proportioning of mixes with cement replacement materials which could be classified as fly ash or blast furnace slag.

Limbachiya et al. (2012) conducted experimental studies of use of recycled concrete aggregate (RCA) in fly-ash concrete. It shows that the use of fly ash in RCA concrete may significantly improve the resistance to chloride ingress. Zain et al. (2002) conducted research work to determine relationship between compressive strength and splitting tensile strength of concrete. Vilanova et al. (2011) evaluated the mechanical properties of self-compacting concrete (SCC) using current estimating models estimating the modulus of elasticity, tensile strength, and modulus of rupture of SCC. It shows that all the models evaluated are suitable for the estimating the modulus of elasticity, tensile strength, and modulus of rupture of SCC.

2 Materials and Mixture Proportions

Commercial Type I Portland cement that complies with the requirements of ASTM C 150 (ASTM Standards 2016) was used as a testing cement. A commercial Class F coal fly ash, blast furnace slag and silica fume was used as a material. The physical properties and chemical composition, as supplied by the fly ash distributor, blast furnace slag and silica fume, are listed in Tables 1 and 2. Coarse and fine aggregate were crushed aggregate and sea sand, respectively. The densities of fine aggregate and coarse aggregate were 2.68 and 2.78, as listed in Table 3. Three levels of energy efficient concrete were designed using the different mixture proportions. Three different mixture group were prepared to achieve the nominal compressive strength. The mix proportions for the three mixture group are listed in the Table 4.

Table 1 Physical properties of cement.

Table 2 Chemical composition.

Table 3 Physical properties of aggregate.

Table 4 The details of mix proportions.

3 Specimen and Test Method

Energy efficient concrete mixtures were mixed in a high-speed shear mixer. For each batch, cylindrical molds of size 100 mm × 200 mm (4 × 8 in.) were cast for the determination of compressive strength, splitting tensile strength and modulus of elasticity of energy efficient concrete. After casting, all the molded specimens were taken to a room at 23 ± 2 °C and humidity and covered with a plastic sheet. When the mixing procedure was completed, tests were conducted on the fresh concrete to determine slump flow and air content. The slump flow tests were performed according to ASTM C 143 (ASTM Standards 2016). The slump flow test measures the horizontal free flow of energy efficient concrete by using a regular slump cone. Air content test method ASTM C 231 (ASTM Standards 2016) was used in this project. Compressive strength and splitting tensile strength tests were performed on cylindrical specimens. Compressive strength of the cylinder specimens was determined in ASTM C 39 (ASTM Standards 2016). Splitting tensile strength testing was conducted in accordance with ASTM C 496 (ASTM Standards 2016). The modulus of elasticity tests was measured following ASTM C 469 (ASTM Standards 2016). These specimens were tested at 7, 28, 56 and 91 days. Three cylinders were tested at each age, and the presented results are an average of the three cylinders.

4 Results and Discussion

4.1 Slump Flow and Air Content

The slump test of concretes was measured for workability of concrete in the fresh state. In addition, the air contents of concretes in its fresh state were measured using pressure gauge method. The test results of slump flow and air content are listed in Table 5. As listed in Table 5, the slump values of for Group I and Group II were 135–210 and 180 to 215 mm, respectively. In addition, the slump flow values of for Group III was 510–600 mm. The air contents for Group I, Group II and Group III, as vindicated values in the pressure gauge, were 3.3–4.4 %, 3.2–4.0 %, and 3.2–3.9 %, respectively.

Table 5 Slump flow and air content.

4.2 Concrete Compressive Strength

Figure 1 shows the variation of the compressive strength of energy efficient concrete with binary and ternary admixture at different curing ages. The compressive strength depends mainly on the water-binder (W/B) ratio. It is also affected by the quality of the constituent materials, mixing and curing methods (Gao et al. 2005).

Fig. 1

Compressive strength.

The compressive strength development of the EEC mixes should be affected because of the different properties and replacement levels of the mineral admixtures. For all curing days, compressive strengths of SF5 specimens were higher than those of other specimens, as shown in Fig. 1. As expected, the compressive strength of concrete containing silica fume are higher than control concrete at all curing ages.

The lower the value of W/B ratio, the higher is the compressive strength of concrete, as listed in Table 6. The average values of 7, 28, 56 and 91 days compressive strength in Group I series was ranged with 15.8–53.0 MPa, 36.3–63.9 MPa, 45.3–71.5 MPa and 48.8–71.9 MPa, respectively. The ratio of compressive strength at 7 days to strength at 28, 56 and 91 days were ranged with 0.44–0.83, 0.38–0.74, 0.31–0.74, respectively. For all the different concrete mixes prepared, 28 days cylindered compressive strengths were found to range between 36.3 and 63.9 MPa. In particular, Specimens BS30 + FA30 and SF 5 concrete mixed in Group I exhibited the lowest and highest 28 days compressive strength, as shown in Fig. 1, respectively. In addition, the compressive strength decreased with increasing fly ash at all ages, compared with Specimens FA 15 and FA 25 in Group I. This is attributed to the fact that pozzolanic reaction depending on contents of fly ash could develop compressive strength slowly.

Table 6 Compressive strength test results.

The average values of 7, 28, 56 and 91 days compressive strength in Group II series was ranged with 21.1–64.5 MPa, 45.8–72.8 MPa, 53.3–84.6 MPa and 60.9–86.1 MPa, respectively. The ratio of compressive strength at 7 days to strength at 28, 56 and 91 days were ranged with 0.46–0.89, 0.37–0.77, 0.35–0.75, respectively. These values in Group II series are similar to those in Group I series. In addition, the trends of 28 days compressive strength for Specimens BS30 + FA30 and SF 5 concrete mixed in Group II series are almost the same to those in Group I series, as shown in Figs. 1a and 1b.

The average values of 7, 28, 56 and 91 days compressive strength for Group III series was ranged with 29.6–73.6 MPa, 57.4–83.6 MPa, 63.7–90.9 MPa and 73.0–93.2 MPa, respectively. The ratio of compressive strength at 7 days to strength at 28, 56 and 91 days were ranged with 0.52–0.88, 0.46–0.81, 0.41–0.81, respectively. These values in Group III series are slightly higher than those in Group I and Group II series. In particular, Specimens BS30 + FA25 + SF5 and SF5 concrete mixed in Group III exhibited the lowest and highest 28 days compressive strength, as shown in Fig. 1.

In all the Group I, II and III, the compressive strength of specimen SF5, binary admixture containing 5 % silica fume, are still much higher than those of specimens FA15, FA 25, BS25 + FA25, BS30 + FA30, BS50 containing fly ash or/and blast furnace slag at all ages and various admixture. This is attributed to the fact that silica fume consisted of ultra fine particles and increased the bond strengths between cement paste and aggregate by making the interfacial zone denser. In addition, the compressive strength of specimens BS30 + FA30, BS30 + FA25 + SF5 decreased with increasing fly ash and blast furnace slag at all ages, compared with Specimens BS25 + FA25, BS25 + FA20 + SF5 in Group I, Group II and Group III. This is attributed to the fact that pozzolanic reaction of fly ash and latent hydraulic activity of blast furnace slag could develop compressive strength slowly.

4.3 The Relationship Between 7 and 28 days Compressive Strength

It may be necessary to predict the compressive strength of concrete not only at an early age but also at later ages. The number of researchers and Codes (Power and Brownyard 1946; Nevile 1997; Hassoun and Choo 2003; Kim et al. 2015) have attempted to predict the relationship between 7 and 28 days compressive strength. The relationship could be expressed by the following equations.

• (1) DIN Code

  In Germany, DIN Code proposed a simple equation to predict the relationship between 7 and 28 days compressive strength

  $$ f_{28} = 1.4f_{7} + 1.0 \sim f_{28} = 1.7f_{7} + 5.9\left( {\text{MPa}} \right) $$

  (1)

  where, f ₇ and f ₂₈ are the compressive strengths at 7 and 28 days.

• (2) Pineiro et al.

  Pineiro et al. proposed equation to predict the relationship of compressive strength between 7 and 28 days.

  $$ f_{28} = k_{2} \left( {f_{7} } \right)^{{k_{1} }} $$

  (2)

  where, f ₇ and f ₂₈ are the compressive strengths at 7 and 28 days. k ₁ and k ₂ are the coefficients, which were ranged from 0.3 to 0.8 and 3 to 6, respectively.

• (3) Hassoun et al.

  Hassoun et al. recommended the following equation for the relationship of compressive strength between 7 and 28 days.

  $$ f_{28} = f_{7} + 2.4\sqrt {f_{ck} } \left( {\text{MPa}} \right) $$

  (3)

  where, ${{f}}_{7}$ and ${{f}}_{28}$ are the compressive strengths at 7 and 28 days, respectively.

• (4) Park et al.

  Park et al. recommended the following equation for the relationship of compressive strength between 7 and 28 days.

  $$ f_{28} = f_{7} + 2.4\sqrt[3]{{f_{ck} }}\left( {\text{MPa}} \right) $$

  (4)

Figure 2 shows the relationship of compressive strength between 7 and 28 days. 7 days compressive strength in Group I were ranged with 15.8–53.0 MPa, which values are 0.44–0.74 times of 28 days compressive strength. In addition, 7 days compressive strength in Group II and Group III were ranged with 21.1–64.5 and 29.6–73.6 MPa, which values are 0.46–0.89 times and 0.52–0.88 times of 28 days compressive strength. Compared with Pineiro’ Eq. (1) and Hassoun’s Eq. (3), the predicted values by DIN code show good agreement with observed values less than 30 MPa and Park’s Eq. (4) over 30 MPa in Group I. Compared with Pineiro’ Eq. (1) and Hassoun’s Eq. (3), the predicted values by DIN code show good agreement with observed values less than 35 MPa and Park’s Eq. (4) over 35 MPa in Group II. This trend is similar to that of Group I. The predicted values by DIN are in a good agreement with measured values less than 50 MPa and Hassoun’s Eq. (3) for 50–70 MPa in Group III. In addition, the predicted values by Park’s Eq. (4) are in a good agreement with measured values over 70 MPa.

Fig. 2

7 days strength versus 28 days strength.

4.4 Splitting Tensile Strength

Figure 3 shows the variation of the splitting tensile strength of energy efficient concrete with binary and ternary admixture at different curing ages. The tensile strength of EEC is much lower than the compressive strength, largely because of the ease with which cracks can propagate under tensile loads.

Fig. 3

Splitting tensile strength.

As listed in Table 7, the average values of 7, 28, 56 and 91 days splitting tensile strengths for Group I were ranged with 1.5–3.7 MPa, 3.2–4.3 MPa, 3.6–4.9 MPa, 3.6–5.0 MPa, respectively. The ratio of tensile strength at 7 days to strength at 28, 56 and 91 days were ranged with 0.45–0.97, 0.38–0.87, 0.33–0.85, respectively. The ratio of tensile strength of FA15 specimens were higher than that of other specimens, while BS30 + FA30 specimens were lower than that of other specimens. In particular, the splitting tensile strength decreased with increasing fly ash at all ages, compared with specimens FA 15 and FA 25 in Group I. Although the trend in splitting tensile strength gain is almost similar with that of compressive strength, the 28 days splitting tensile strength lies in the range of 5–10 % of compressive strength.

Table 7 Splitting tensile strength test results.

The average values of 7, 28, 56 and 91 days splitting tensile strengths for Group II series were ranged with 2.2–3.9 MPa, 4.0–4.7 MPa, 4.1–5.4 MPa, 4.2–5.5 MPa, respectively. The ratio of tensile strength at 7 days to strength at 28, 56 and 91 days were ranged with 0.55–0.87, 0.48–0.85, 0.45–0.81, respectively. The ratio of tensile strength of SF5 specimens were higher than that of other specimens, while BS30 + FA30 specimens were lower than that of other specimens.

Splitting tensile strength of SF5 specimens for Group I and II were higher than that of other specimens at 7 and 28 days. However, 57 and 91 days splitting tensile strength of BS50 specimens were higher than those of SF5. Even though compressive strength of SF5 specimens were higher than those of other specimens, the splitting tensile strength showed different trends.

In addition, the average values of 7, 28, 56 and 91 days splitting tensile strengths for Group III series were ranged with 2.8–4.9 MPa, 4.8–5.3 MPa, 5.0–5.9 MPa, 5.2–6.0 MPa, respectively. The ratio of tensile strength at 7 days to strength at 28, 56 and 91 days were ranged with 0.53–0.96, 0.52–0.94, 0.50–0.83, respectively. The ratio of tensile strength of SF5 specimens were higher than that of other specimens, while BS30 + FA25 + SF5 specimens were lower than that of other specimens.

4.5 The Relationship Between Compressive Strength and Splitting Tensile Strength

The splitting tensile strength generally increases with the compressive strength. The following equation is recommended by KCI Code for prediction of the splitting tensile strength of normal-weight concrete.

• (1) KCI Code

  In Korea, KCI Code (2011) proposed a simple equation to predict the relationship between compressive strength and splitting tensile strength

  $$ f_{\text{sp}} = 0.563\sqrt {f_{ck} } ({\text{MPa}}) $$

  (5)

• (2) ACI 363R-84 Code

  In America, ACI 363R-84 Code (ACI Committee 363 1984) proposed a simple equation to predict the relationship between compressive strength and splitting tensile strength based on a study by Carrasquillo et al. (1981).

  $$ f_{\text{sp}} = 0.59\sqrt {f_{ck} } \left( {\text{MPa}} \right)\quad 2 1\,{\text{MPa}} < f'_{c} < 83\,{\text{MPa}} $$

  (6)

• (3) EC2-02 Code

  In Europe, EC2-02 Code (European committee 2012) recommended the following equation for the relationship between compressive strength and splitting tensile strength (a) $f_{\text{ck}} \le 50\;{\text{MPa}}$

  $$ f_{\text{sp}} = \frac{1}{3}(f_{ck} )^{2/3} ({\text{MPa}}) $$

  (7)

  (b) $f_{\text{ck}} > 50\;{\text{MPa}}$

  $$ f_{\text{sp}} = 2.35{ \ln }\left( {1 + \frac{{f_{cm} }}{10}} \right)({\text{MPa}}) $$

  (8)

• (4) JSCE Code

  In Japan, JSCE Code (Japan Society of Civil Engineers 2008) recommended the following equation for the relationship between compressive strength and splitting tensile strength

  $$ f_{\text{sp}} = 0.23\left( {f_{ck} } \right)^{2/3} ({\text{MPa}}) $$

  (9)

Figure 4 shows the relationship between the compressive strength and the splitting tensile strength of the concrete. The splitting tensile strength can be related to compressive strength, water/binder (W/B) ratio and concrete age. Predicted values by ACI 363R-84, EC 2-02 and KCI Code slightly overestimated observed value, as shown in Fig. 4. Predicted values by Park’s Eq. (4) and JSCE Code show good agreement with observed value in Group I. In addition, predicted values by KCI show good agreement with observed value in Group I and Group II. Predicted values by ACI 363R-84, EC 2-20, KCI and JSCE Code overestimated observed values in Group III.

Fig. 4

Compressive strength versus splitting tensile strength.

4.6 Modulus of Elasticity

The modulus of elasticity of concrete is one of the most important factors to determine the strain distributions and deformation. The modulus of elasticity of concrete, E_c, is an indicator of the resistance to deformation of concrete, which is subjected to compressive load. The modulus of elasticity can be estimated by Equations listed in Table 8. The modulus of elasticity of energy efficient concrete with binary and ternary admixture at different curing ages are listed in Table 9. As listed in Table 9, the average values of 7, 28, 56 and 91 days modulus of elasticities for Group I were ranged with 13.8–27.8 MPa, 21.7–30.2 MPa, 21.0–33.8 MPa, 26.9–34.2 MPa, respectively. The ratio of modulus of elasticity at 7 days to those at 28, 56 and 91 days were ranged with 0.60–0.95, 0.55–0.97, 0.51–0.89, respectively.

Table 8 Estimating equations of the different models.

Table 9 Modulus of elasticity.

The average values of 7, 28, 56 and 91 days modulus of elasticities for Group II were ranged with 18.0–33.0 MPa, 21.7–33.4 MPa, 28.3–35.7 MPa, 30.2–36.0 MPa, respectively. The ratio of modulus of elasticity at 7 days to those at 28, 56 and 91 days were ranged with 0.67–0.99, 0.62–0.92, 0.59–0.92, respectively.

The average values of 7, 28, 56, and 91 days modulus of elasticities for Group III were ranged with 18.7–33.3 MPa, 24.6–35.9 MPa, 31.2–35.9 MPa, 33.2–36.7 MPa, respectively. The ratio of modulus of elasticity at 7 days to those at 28, 56 and 91 days were ranged with 0.70–0.93, 0.59–0.93, 0.56–0.91, respectively. In particular, the modulus of elasticity of SF5 specimens for all the Group I, II and III were higher than that of other specimens. This trend is similar to that of compressive strength.

The relationship between compressive strength and modulus of elasticity is shown in Fig. 5. Predicted values by ACI 318-11 (ACI Committee 2011), CEB-FIP Model (CEB-FIP 1993) and KCI Code, EN 1991 Code (EN 1991 1991) NZS 3101 (New Zealand Standard 1995), CSA A 23.3 (CSA Technical Committee 2004) overestimated observed values in Group I and II. However, predicted values by CEB-FIP Model and KCI Code show good agreement with observed value in Group III.

Fig. 5

Modulus of elasticity.

4.7 Poisson’s Ratio

The Poisson ratio is a basic factor in analyzing, designing and important attribute of the mechanical response of any materials. Poisson’s ratio is defined as the ratio of the transverse extension strain to the longitudinal contraction strain in compression. The Poisson’s ratios are listed in Table 10. Tests were performed on specimens with strain gauge units and tensile strain gauge units. Poisson’s ratio in this study were ranged from 0.101 to 0.236. This values are slightly larger than those of normal concrete. As it can be seen from Fig. 6, the Poisson’s ratio of SF5 specimens were larger than those of other specimens. This trend is similar to that of compressive strength. This is attributed to the fact that silica fume particles are very small, compared with fly ash and blast furnace slag particles. The ultra fine silica fume particles enter the relatively coarse cement inter-particle space. Thus components fineness was effected by particle size of silica fume.

Table 10 Poisson’s ratio.

Fig. 6

Poisson’s ratio.

5 Conclusions

The following conclusions were derived from the experimental results of mechanical properties of EEC with binary and ternary admixture, such as fly-ash, blast furnace slag and silica fume.

• (1) The compressive strengths, splitting tensile strengths, modulus of elasticities and Poisson’s ratios of Specimen SF5 for Group I, II and III showed higher values. This is attributed to the fact that silica fume particles are very small, compared with fly ash and blast furnace slag particles. The ultra fine silica fume particles enter the relatively coarse cement inter-particle space. Thus components fineness was effected by particle size of silica fume.

• (2) The predicted values by DIN code and Park’s Eq. (4) for relationship of compressive strengths between 7 and 28 days show good agreement with observed values in Group I and II. The predicted values by DIN are in a good agreement with measured values less than 50 MPa, Hassoun’s Eq. (3) for 50–70 MPa and Park’s Eq. (4) over 70 MPa in Group III.

• (3) Predicted values by Park’s Eq. (4) and JSCE Code for relationship between compressive strength and splitting tensile strength days show good agreement with observed value in Group I. Predicted values by KCI show good agreement with observed values in Group I and Group II. In addition, predicted values by ACI 363R-84, EC 2-20, KCI and JSCE Code overestimated observed values in Group III.

• (4) The predicted values by ACI 318-11, CEB-FIP Model, EN-1991, NZS 3101, CSA A 23.3 and KCI Code to estimate modulus of elasticity overestimated observed values in Group I and II. However, predicted values by CEB-FIP Model and KCI Code show good agreement with observed value in Group III.

• (5) The Poisson’s ratios in this study were ranged from 0.101 to 0.236. This values are slightly larger than those of normal concrete. In particular, the Poisson’s ratio of SF5 specimens were larger than those of other specimens.