Available online atvwvw.sciencedirect.com

ScienceDirect Procedía

Engineering

Procedía Engineering 145 (2016) 1218 - 1225 -

www.elsevier.com/locate/procedia

International Conference on Sustainable Design, Engineering and Construction

Splitting tensile strength of sustainable self-consolidating concrete

Osama A. Mohamed, P.E., MASCE*, Zubair I. Syed, Omar F. Najm

Department of Civil Engineering, Abu Dhabi University, Abu Dhabi, United Arab Emirates

Abstract

Construction industry is known to be a contributor to environmental pollution in various ways including the production of cement. It is common nowadays to produce concrete in which cement is partially replaced with sustainable minerals that impart favourable properties on the concrete mix, such as fly ash, ground granulated blast furnace slag (GGBS), and silica fume. The properties of high-replacement concrete mixes are still being studied. One of the concrete properties of interest to the designer is the splitting tensile strength of concrete. In this study, sustainable self-consolidating concrete (SCC) mixes were produced in which 80% of the cement was partially replaced with various combinations of minerals. Relationships were developed between the splitting tensile strength and the 28-day compressive strength for the control concrete mix as well as the mixes in which cement was partially replaced with mineral admixtures. The relationships developed in this study were compared to those proposed in selected concrete codes such as ACI 318.

© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review underresponsibility of the organizing committee of ICSDEC 2016

Keywords: self-consolidating-concrete; splitting tensile strength; compressive strength;fly ash; silica fume; slag; regression analysis

J^Ä^TOfil CrossMark

ELSEVIER

1. Introduction

Tensile strength of traditional concrete falls between 8 and 15 percent of the compressive strength. Several factors affect the relationship between tensile strength and compressive strength of concrete, including aggregate type, the presence of compressive stresses transverse to the tensile stresses, and the magnitude of compressive

* Corresponding author. Tel.: +0-971-50-472-2078 E-mail address: osama.mohamed@adu.ac.ae

1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of the organizing committee of ICSDEC 2016

doi: 10.1016/j.proeng.2016.04.157

strength. In addition, different tensile strength tests produce different relationships between tensile strength and compressive strength. The most common tests of tensile strength include: 1) the direct tension test, 2) flexure test, and 3) splitting test. Splitting test is often regarded as simple, reliable, and convenient method for approximating the tensile strength of concrete with test results typically having a low coefficient of variation. ASTM C496 [1] and BS 1881 117-83 [2] prescribe standard procedures for conducting the splitting test. Cylindrical samples are usually used to evaluate the splitting tensile strength of concrete, while BS 1881 117-83 [2] also permits the use of prismatic samples.

This study develops a correlation between the splitting tensile strength and compressive strength for SCC concrete in which 80% was replaced with various individual or combinations of minerals such as fly ash, silica fume, and blast furnace slag. The findings of this study apply SCC mixes made of ASTM C150 type I cement. Mohamed et al. [3] showed that cement type affects both short term and long term properties.

The splitting tensile strength test process involves loading the sample to induce transverse tension. Two strong plane-parallel plates distribute and apply a compressive stress in two diametrically opposite points on the cylinder diameter leading to development of high tensile stresses that cause rupture of the specimen along the vertical plane due to the propagation of cracks through the concrete paste [4].

The splitting fracture stress, fspt, that causes the failure of the specimen is calculated using Eqn. 1:

f- = D (1)

Where:

F: is the applied force

D: is the sample diameter, and

L: is the length of the sample.

Studies show that splitting tensile strength increases with increase in compressive strength. Studies show that Eqn. 2 provides consistent results to relate compressive strength of concrete to the splitting tensile strength.

u=k *(/:)- (2)

Where:

k and n: are non-dimensional coefficients f'c : is the compressive strength of concrete.

Different values of the experimental coefficients k and n were proposed in the literature [5,6], depending on a variety of experimental parameters considered by the investigators, such as, curing effect, curing temperature, and concrete age. Table 1 shows the experimental parameters k and n proposed by various concrete design codes. Some studies suggest that ACI 318 [7] coefficients underestimate the splitting tensile strength for high strength concrete and overestimate it for low strength concrete [5]. Ros and Shima [5] indicated that JCI 2008 [8] coefficients are consistent with their experimental data.

Table 1. Code-based values of non-dimensional coefficients k, and n

Source k n

ACI318-11 [7] 0.56 0.5

JCI 2008 [8] 0.13 0.85

JSCE 2007 [9] 0.23 2/3

CEB-FIB [10] 0.3 2/3

In this study, the non-dimensional coefficients k and n needed for Eqn. 1 are derived using regression analysis, for

SCC mixes in which 80% of the cement was replaced with minerals. In addition, the ability of the coefficient proposed in various codes to predict the splitting tensile strength of sustainable SCC mixes as evaluated using the experimental data collected in this study. The evaluation of the code coefficients and the derived coefficients from regression analysis models were assessed through the use of integral absolute error (IAE) described by Eqn. 3.

Where:

Oi: is the experimental tensile strength

Pi : is the predicted tensile strength from regression analysis.

While there is no consensus on what is an acceptable value IAE, 10% or less is often considered acceptable for prediction of splitting tensile strength [11].

In this study, the 28-day compressive strength was determined experimentally for 150 mm cubes. In order to propose a mathematical relationship for cylindrical samples, and to assess ability of codes to predict the splitting tensile strength, the equivalent cylinder compressive strength was calculated using conversion equations proposed in the literature, as described in the following paragraphs.

The British Standard, BS 1881: part 120:1983 [12] suggests that cube strength may be assumed to be 0.8 times cylinder strength. This is applicable for cylinders with diameter-to-depth ratio of 1:2 and cubes that are 150 mm x 150 mm x 150mm. Murdock [13] suggested the correlation between concrete and cylinder strength depends on concrete strength. Neville [14] reported that the cube strength can be taken as 0.87 times cylinder strength. Dehestani [15] results showed that 100 x 200 mm cylinders and 150 mm cubes gave a correlation factors of 0.87 for high strength concrete and 0.79 for the medium strength concrete, while the low strength concrete gave a 0.62 as a conversion factor. Zabihi [16] experimental results gave similar results to Dehestani [15] except that he proposed a conversion coefficient of 0.74 for medium strength.

This study adopts the study of Mansur [17] to convert cube strengths tested in the laboratory to cylinder strength needed for Eqn. 2. Mansur [17] developed a total of 11 concrete mixes that cover wider compressive strength range from 20 to 100 MPa. He proposed Eqn. 4 to relate the compressive strength of 100 mm x 200 mm cylinders to the compressive strength of 150 mm cubes.

Mansur [17] verified his equation through using data collected from Fong [18] and Chin [19]. 2. Experimental Program

This study includes the development and testing of seven SCC mix and one control mix. The control mix contains 100% Portland cement as cementitious material that complies with ASTM C150, without any other minerals. In each of the remaining seven concrete mixes, 80% of the control mix cement content was replaced with different combinations of pozzolans including fly ash, silica fume, and slag. The control mix design was normally vibrated after being mixed for 80 minutes. A total of 48 concrete samples were tested to obtain the compressive and splitting tensile strength, of which 32 cubes were tested for compressive strength development after 3-, 7-, and 28-days days of moist curing. The remaining 16 cylinders were tested to determine the corresponding tensile strength of the concrete after 28-days of curing.

High Range Water Reducing Admixture (HRWA) based on polycarboxylic ether was used and kept constant at 1.5% for all mixes. The HRWA used in this study is distributed by BASF Corporation under the commercial name Glenium Sky 504.

\(0 - P. )2 P

IAE = X 1 P) 1

In all SCC mixes studied in this paper, the total amount of cementitious materials was approximately 480 kg/m3 and the w/c ratio was maintained at approximately 0.36. Coarse aggregates were crushed aggregate passing 14 mm sieve size with a total amount of 800 kg/m3. Fine aggregates consisted of 582.4 kg/m3 black sand and 313.6 kg/m3 dune sand. Fig. 1 (a) and (b) show the sieve analysis results for course aggregates and fine aggregates, respectively.

120 100 80 60 40 20 0 -20

< y* »

p--- 1 0

Sieve Size Fig. 1. (a) Sieve analysis results for coarse

# SO £1)

Fineness Modulus =3.56

0.01 0.1 1 10 Sieve Size

; (b) Sieve analysis results for fine aggregates

The workability and stability of each mix was evaluated based on the BS1881:105 [20] standards. The superplasticizer dosage was designed to achieve a target slump of 550mm ± 50mm after 80 minutes of continuous mixing. The flowability of concrete was assessed by measuring the average diameter of concrete spread, as shown in Fig.2, and by measuring the T50 time. The stability of concrete mixes was assessed using the visual sight index (VSI).

Fig. 2. Conducting the flow test

The compressive strength of 150 mm x 150 mm x 150 mm concrete cubes was determined in accordance with the testing procedure described in BS1881:116 [21], while the splitting tensile strength of 100 mm x 200 cylinders was determined in accordance with BS 1881-117 [2] code. Prior to conducting the compression test or splitting tensile test, all samples were moist-cured by immersion in water tank until the day of testing. It is generally acknowledged that the curing method affects the compressive strength of SCC mixes, and that moist curing produces best results compared to air and steam curing [22]. Fig. 3 shows a cylindrical sample under splitting tensile test.

Fig. 3. Splitting tensile test setup

3. Test results

The following subsections describe the results of the experimental program including determination of fresh and hardened properties of the sustainable concrete mixes.

3.1. Workability and Stability of Sustainable SCC Mixes

As indicated earlier, seven sustainable SCC mixes were studied where 80% of the ordinary Portland cement was replaced with combinations of fly ash (FA), silica fume (SF), and Granulated Slag (GS). The seven mixes are referred to as Green Mix 1 (GM1) to Green Mix 7 (GM7). After 80 minutes of continuous mixings, all mixes demonstrated good workability as indicated by the final diameter shown in Table 2. Resistance to segregation was measured using the Visual Stability Index (VSI). The most workable concrete mix, as indicated by the large final diameter, was GM7. As shown in Table 2, mixes with silica fume content of 10% to 20% showed greater stability and relatively higher viscosity compared to other mixes. All mixes showed very good resistance to segregation as indicated by VSI of 0 or 1, except for GM3 which exhibited a VSI of 3. The poor stability of GM3 may be attributed to the 60% slag, which was the largest amount of concrete mixes, as shown in Table 2. However, although mixes GM1, GM4, GM5, and GM7 contained 50% and 55% slag, they all showed excellent segregation resistance, as indicated by a VSI of 0. This may be attributed to the presence of silica fume in these mixes that ranged from 10% to 20% of the total cementitious material.

Table 2. Workability results

Mix FA% SF% GS% e o ISA Final Diameter(cm)

GM1 20 10 50 - 0 47

GM2 25 15 40 5 1 53

GM3 15 5 60 5.2 3 50

GM4 15 15 50 9 0 53

GM5 10 15 55 11.3 0 52

GM6 15 20 45 5 1 53

GM7 10 20 50 3.6 0 56

3.2 Compressive strength results

The compressive strength development after 3, 7, and 28-days of curing is shown on Fig. 4(a). Clearly all mixes produced lower strength compared to the control mix at all ages. Nonetheless, despite the high cement replacement ratio of 80%, the 28-day cube compressive strength shown on Fig. (b) is suitable for many practical applications.

Fig. 4(a) and Fig. 4(b) show that GM4 performed the best of all mixes in terms of 28-day compressive strength development. GM4 contained 15% fly ash, 15% silica fume, and 50% slag, and 20% ordinary Portland cement. This mix benefited from favorable optimum dosages of fly ash, silica fume, and slag.

Fig. 4. (a) Compressive strength development after 3-, 7-, and 28-days of curing; (b) Compressive strength after 28-days of curing

3.3. Tensile strength results

Two cylinders were tested for splitting tensile strength for each of the nine sustainable SCC mix, in addition to the control mix, and the average values are shown in Fig. 5.

Fig. 5. Tensile strength after 28-days of curing

4. Assessment of concrete codes and development of correlation between tensile strength and compressive strength

The prediction relationships for four concrete codes are plotted in Fig. 6(a) against the tensile and compressive strength experimental data obtained in this study. The IAE is calculated for each of the code-predicted versus actual experimental data and results are summarized in Table 3. Fig. 6(a) and Table 3 both show that the Japanese Society of Civil Engineers (JSCE) concrete code [9] was the most capable of predicting the experimental data developed in this study. The remaining three codes, namely ACI 318 [7], CEB-FIB [10], and JCI [8] offered conservative estimates of the splitting tensile strength.

Table 3. IAE% for different codes

Source IAE%

ACI318-11 27.9

JCI 13.4

JSCE,2007 4.6

CEB-FIB 29.2

Regression analysis was conducted to develop a relationship between compressive strength and splitting tensile strength. The best fit relationship for ratio of splitting tensile strength to compressive strength, ft/fc is shown in Fig. 6 (b). This relationship corresponds to k and n coefficients of 0.34 and 0.57 respectively. This relationship corresponds to the best attainable accuracy as measured by value of IAE equals to 4.7%. Therefore, the splitting tensile strength of SCC mixes in which 80% of the cement was replaced with fly ash, silica fume, and slag is given by Eqn. (5).

fspt = 0.34 xf')057 (5)

If regression analysis is conducted on the experimental data assuming a value of n = 0.5, the best fit correspond a k value of 0.44 yielding IAE value of 4.9%. Therefore, the splitting tensile strength of SCC mixes in which 80% of the cement was replaced with minerals is given by Eqn. 6.

fspt = 0.44 xf;)0-5 (6)

Compared the ACI318 relation where k = 0.56, it may be concluded that ACI318 relationship overestimates the splitting tensile strength of SCC with high cement replacement ratios.

Fig. 6. (a) Codes' regression analysis; (b) Green mixes regression analysis

5. Conclusion

• Sustainable self-consolidating concrete mixes where 80% of the cement was replaced with various combinations of fly ash, silica fume, and slag. The mix containing 15% fly ash, 15% silica fume, and 50% slag produced the best results in terms of compressive strength development after 3-, 7-, and 28-days of curing. GM4 reached the strength of the control mix after 28-days of curing.

• Sustainable mixes containing 10% to 20% silica fume showed the highest resistance to segregation while mixes with 60% slag showed poor resistance to segregation. However, mixes with 50% and 55% slag showed excellent segregation resistance when 15% silica fume is included in the same mix.

• A regression-based expression was developed for splitting tensile strength as a function of the 28-day compressive strength of concrete for sustainable self-consolidating concrete mixes in which cement was partially replaced with 80% of minerals including fly ash, silica fume, and slag. The reliability of the expression was assessed using the integral absolute error (IAE) which averaged 4.7%.

Acknowledgement

The authors would like to the acknowledge the financial support of the Center on Sustainable Built Environment and

the Office of Research and Sponsored Programs at Abu Dhabi University.

References

[1] ASTM C496 / C496M-11, Standard Test Method for Splitting Tensile Strength of Cylindrical Concrete Specimens, ASTM International, West Conshohocken, PA, 2004

[2] BS-1881:117 (1983). Testing concrete Part 117. Method for determination of tensile splitting strength. London: British Standards Institution.

[3] O. A. Mohamed, K. L. Rens, J. J. Stalnaker, Factors affecting resistance of concrete to freezing and thawing damage, Journal of Materials in Civil Engineering, 12(2000) 26 - 32.

[4] C. Druta, Tensile strength and bonding characteristics of self-compacting concrete, M.A. Thesis, Louisiana State University, Louisiana, 2003.

[5] S. Ros, H. Shima, Relationship between splitting tensile strength and compressive strength of concrete at early age with different types of cements and curing temperatures, Concrete Institute Proceedings. 35(2013) 427-432.

[6] N. Arioglu, Z. C. Girgin, E. Arioglu, Evaluation of ratio between splitting tensile strength and compressive strength for concretes up to 120 MPa and its application in strength criterion, ACI Materials Journals, 103(2006) 18-24.

[7] ACI Committee 318, Building code requirements for structural concrete (ACI 318-11) and commentary American Concrete Institute, Farmington Hills, Mich., 2011, pp111.

[8] JCI 2008, Guidelines for Control of Cracking of Mass Concrete, Japan Concrete Institute, 2008.

[9] JSCE 2007, Standard specifications for concrete structures-2007, Materials and Construction, Japan Society of Civil Engineers, 2007.

[10] CEB-FIB 1991, Model Code for Concrete Structures. Evaluation of the Time Dependent Behavior of Concrete, 199(1991) 201.

[11] A., N., Girgin, A. C., & A., E., Evaluation of ratio between splitting tensile strength and compressive strength for concretes up to 120 MPa and its application in strength criterion., ACI materials journal, (2006) 18-24.

[12] BS-1881:120. (1983). Testing concrete part 120. Method for determination of the compressive strength of concrete cores. London: British Standard Institution.

[13] J. W. Murdock, C. E. Kesler, Effect of length to diameter ratio of specimen on apparent compressive strength of concrete, ASTM bulletin, (1957) 68-73.

[14] A. M. Neville, Properties of Concrete, Harlow: Longman, 1995.

[15] M. Dehestani, I. M. Nikbin, S. Asadollahi, Effect of specimen shape and size on compressive strength of self-consolidating concrete (SCC). Construction and building materials, (2014) 685-691.

[16] N. Zabihi, O. Eren, Compressive strength conversion factors of concrete as affected by specimen shape and size, Research journal of applied sciences, engineering and technology. (2014) 4251-4257.

[17] M. A. Mansur, M. M. Islam, Interpretation of concrete strength for nonstandard specimens, Journal of materials in civil engineering, (2002) 151-155.

[18] Fong, C. S., Effect of specimen shape and size on compressive strength and UPV, BEng. Thesis, National Univ. of Singapore, Singapore, 1998.

[19] Chin, M. S., Deformation and ductility of high-strength concrete, PhD thesis, National Univ. of Singapore, Singapore, 1996.

[20] BS-1881:105 (1984). Testing Concrete Part 105. Method for determination of flow. British Standards Institution.

[21] BS-1881:116 (1983). Testing concrete Part 116. Method for determination of compressive strength of concrete cubes. London: British Standards Institution.

[22] A.F. Bingol, I. Tohumcu, Effects of different curing regimes on the compressive strength properties of self-compacting concrete incorporating fly ash and silica fume, Materials and Design. 51 (2013) 12 - 18.