Supplementary Cementitious Materials to Enhance Bridge Deck Durability

By

Dr. Houssam A. Toutanji and Mr. André D. Danson Department of Civil Engineering

The University of Alabama in Huntsville Huntsville, Alabama

Dr. Nobert Delatte Department of Civil Engineering

The University of Alabama at Birmingham Birmingham, Alabama

Prepared by

UTCA

University Transportation Center for Alabama The University of Alabama, The University of Alabama in Birmingham, and

The University of Alabama at Huntsville

UTCA Report Number 00305

November 30, 2001

ii

Technical Report Documentation Page 1. Report No FHWA/CA/OR-

2. Government Accession No. 3. Recipient Catalog No.

5. Report Date December 2001

4. Title and Subtitle Supplementary Materials to Enhance Bridge Deck Durability

6. Performing Organization Code

7. Authors Houssam A. Toutanji and Norbert Delatte

8. Performing Organization Report No.

10. Work Unit No.

9. Performing Organization Name and Address Department of Civil and Environmental Eng. University of Alabama in Huntsville Huntsville, AL 35899 11. Contract or Grant No.

DTRS98-G-0028 13. Type of Report and Period Covered Final Report / January 2001 – July 2001

12. Sponsoring Agency Name and Address University Transportation Center for Alabama Box 870205, 275 H M Comer Mineral Industries Building Tuscaloosa, Alabama 35487-0205

14. Sponsoring Agency Code

15. Supplementary Notes 16. Abstract Bridge deck durability is an issue of great interest and concern to state highway agencies. The use of supplementary cementitious materials such as fly ash, silica fume, slag, and natural pozzolans has been shown to enhance durability. Since some of these materials are cheaper than Portland cement, there is an economic advantage to wider use. However, clear and conclusive guidelines are lacking as to the optimum use of these materials and the curing practices required to achieve the desired performance and durability. Concrete with silica fume, in particular, has been shown to be more sensitive to curing practices than more conventional mixtures. There is a need for guidelines to help transportation agency personnel specify the materials, proportions, and procedures that will ensure long-term durability of concrete bridge decks. Transferring the latest materials technology to transportation professionals in Alabama will substantially enhance the management and safety of transportation systems both during construction and in service. 17. Key Words Concrete; Durability; Admixtures; Bridges; Cementitious Materials; Silica fume; Fly Ash, Slag, Compressive Strength, Tensile Strength; Dynamic Modulus

18. Distribution Statement

19. Security Classif (of this report) Unclassified

20. Security Classif. (of this page) Unclassified

21. No of Pages 28

22. Price

Form DOT F 1700.7 (8-72)

iii

Contents

Contents………………………………………………………………………………. iii List of Tables…………………………………………………………………………. iv List of Figures………………………………………………………………………… iv Executive Summary…………………………………………………..……………… v 1.0 Introduction…………………………………………………………………..… 1 2.0 Literature Review and Background…………………………………………..… 4 3.0 Testing Procedure……………………...……………………………………….. 8

3.1 Specimens Preparation…………………………………………………….. 8 3.2 Mixing Procedure…………………………...……………………………... 9 3.3 Wet-Dry Test……………………………………………………………… 11 3.4 Freeze-Thaw Test……………………………………………………….… 12 3.5 Compression Test ………………………………………………………… 15 3.6 Tensile Splitting Test …………………………………………………….. 15

4.0 Results………………………………………………………………………….. 16

4.1 Effect of Supplementary Cementitious Materials on Strength………….… 16 4.2 Effect of Wet-Dry Exposure on Strength…………………………………. 19 4.3 Effect of Freeze-Thaw Conditions on Strength………………………..…. 20

5.0 Conclusions…………………………………………………………………….. 23 6.0 Recommendations for Future Research……………………………………….. 25 7.0 References………….………………………………………………………..…. 26

iv

List of Tables

3-1 Mix Proportions………………………………………………………………. 8

List of Figures

Page 3-1 Molds for beams and cylinders………………………………………….. 10

3-2 Cast concrete specimens………………………………………………… 10 3-3 Concrete specimens in curing tank……………………………………… 11 3-4 Schematic of the wet-dry exposure set-up…………………………….… 12 3-5 Freeze-thaw apparatus………………………………………………….. 14 4-1 Compressive strength performance for all mixes……………………….. 17 4-2 Tensile strength performance for all mixes……………………………... 18 4-3 Ratios of compressive strength to tensile strength for all mixes………... 18 4-4 Compressive strength data for both room temperature and wet-dry conditions……………………………………………………………….. 20 4-5 Durability factor for control and mixes with supplementary cementitious materials……………………..……………………….…………………. 21 4-6 Relative dynamic modulus of elasticity of all mixes…………………….. 22

v

Executive Summary

Issues concerning the durability of bridge decks have been of great concern to civil engineers and highway agencies. Many have proposed the use of supplementary cementitious materials to alleviate these concerns. These materials have clearly demonstrated their ability to enhance the durability of concrete. Therefore, it is important to establish how these materials can improve the durability of concrete, while determining their advantages. It should be noted that a standard does not exist for the use of these supplemental materials in concrete. Additionally, prior research has not clearly defined their properties and qualities, nor have the results obtained been consistent. This research focuses on the development of guidelines outlining the correct and safe use of supplementary cementitious materials (silica fume, fly ash, slag, and their combinations) to enhance bridge deck durability. This work primarily deals with the characteristics of the aforementioned materials, including their properties, appropriate curing methods, durability, strength, and resistance to wet-dry and freeze-thaw environments. In order to accomplish these objectives, prior research was reviewed and over 20 mixes were made and compared to a control mix. Each mix was either different in the percentage of additives or was a combination of two or more additives. The freeze-thaw specimens were compared to the control mix by measuring and monitoring weight loss, changes in length, compressive strength, splitting tensile strength and determination of durability factors. In contrast, the wet-dry specimens were compared to the control mix by testing their compressive strength, tensile strength and crack propagation. This research project was a collaborative effort between the University of Alabama in Huntsville (UAH) and the University of Alabama at Birmingham (UAB). The research was completed in 18 months.

1

1.0 Introduction Many applications for concrete have been found in the world of civil engineering ranging from offshore platforms to high rise buildings. The concrete must adapt to the various environmental conditions to which it is exposed. In order to aid in this adaptation process, various chemical and mineral admixtures are added, such as silica fume (SF), fly ash (FA), and ground granulated blast furnace slag (GGBFS). These supplementary cementitious materia ls have been used to improve different characteristics and properties of concrete. These necessary augmentations depend on the specific application and the desired performance of the concrete. Each of these supplementary materials possesses different properties and reacts differently in the presence of water. Silica fume is a relatively recent addition to the list of supplementary cementitious materials. According to ASTM-94A (Ramachandran 1995), a pozzolan is a siliceous or siliceous aluminous material, which in itself possesses little or no cementitious value, but will, in finely divided form and in the presence of moisture, chemically react with calcium hydroxide at ordinary temperatures to form compounds possessing cementitious properties. Silica fume forms as a by-product of the reactions necessary to manufacture silicon and ferrosilicon, which are alloys of quartz and coal. It possesses a bulk density ranging from 200-300 kg/m3 and a specific gravity of approximately 2.20. It was first utilized in the 1970’s as an additive in the concrete used to construct the Fiskaa Smelter (Sabir 1997) in Norway. It should also be noted that only a few million tons of silica fume are produced yearly (Ramachandran 1995). The addition of silica fume increases the early strength of concrete, while reducing its permeability. Silica fume serves a dual role as a filler and pozzolan (Sabir 1997). Due to its small particle size it can enter the spaces that exist between the particles of cement to improve packing. It also contributes to the hydration reaction that takes place between cement and water, due to its high pozzolanicity. Potentially, silica fume could replace small portions of cement, due to its high reactivity with calcium hydroxide during the hydration of Portland cement. It may reduce bleeding and porosity, as a result of its fineness. However, the properties of silica fume are enhanced by the presence of superplasticizers in the mix. The resistance of silica fume to freeze-thaw cycling is unclear, although recent research tends to show that it does resist this environmental cycling well. Fly ash, another pozzolanic material, results from the combustion of coal. According to ASTM C618 (1994), fly ash is defined in two broad categories, Class C and Class F, related to the type of coal burned. Class F fly ash forms from the combustion of anthracite or bituminous coal, whereas Class C fly ash forms from the burning of sub-bituminous coal and lignite. Class F fly ash contains low percentages of CaO, less than 6%, contributing to its pozzolanic properties; however, the low calcium ash does not possess self-hardening properties. In contrast, Class C fly ash contains percentages of CaO greater than 15%. This high calcium fly ash possesses both self-hardening and pozzolanic properties. The specific gravity of fly ash varies between 1.3 to 4.8 depending on the amount of quartz and mullite, which varies proportionally to specific gravity (Ramachandran 1995). The addition of fly ash increases the strength properties of

2

concrete if its carbon content is low. Also fly ash comprised of clean, glassy, spherical particles in finely divided form acts as good void fillers, which may reduce the water required for hydration. Fly ash containing lower percentages of free carbon tends to possess higher pozzolanic activity. In addition, carbon content also affects the freeze-thaw resistance of concrete. The higher the fly ash carbon content, the lower the freeze-thaw resistance of the concrete (Wesche 1991). Slag forms as a waste product from both ferrous and non-ferrous metals. It has been used as an admixture in finely ground form for many years. Slag falls into three categories, blast furnace slag, electric furnace slag, and converter furnace slag. Blast furnace slag is most commonly used due to its cementitious qualities. Ground granulated blast furnace slag (GGBFS) can be further classified into three grades, 80, 100, and 120, depending upon the mortar strengths when mixed with an equal weight of Portland cement. Air-cooled slag has no pozzolanic value when dumped in heaps; however, when the slag is cooled quickly or quenched (granulated) and then ground, it behaves like a pozzolan with some cementitious value (Ramachandran 1995). Slag particles tend to be conchoidal in nature making it more difficult to finish and more difficult to work with. It differs from fly ash in this property; however, striking similarities do exist between fly ash and slag. For instance, there are many similarities in the chemical and mineralogical composition of slag and those of Class C fly ash. They both contain about 75% glass that has similar composition and reactivity. However, unlike fly ash, slag must await the hydration of Portland cement before it can be hydrated itself (Ramachandran 1995). Upon the hydration of cement, alkalis are released, which in turn activate the slag component. The glass present in the slag then hydrates leading to the formation of cementing compounds. In order to increase the reaction activators such as alkalis, gypsum and lime can be used to act as catalysts to speed up the hydration process in the slag. Slag containing one or more of these activators can be characterized by rapid slump loss in spite of the plasticizing effects of alkali solutions. Since alkali activated slag has thixotropic properties, flash setting can be averted by mixing the concrete continuously (Ramachandran 1995). Slag has been used for years, since it improves the durability of concrete placed in marine environments. In combination with the other admixtures, slag can improve the durability of concrete. Concrete that has the ability to resist the destructive forces imposed upon it, by environmental constraints and without requiring excessive maintenance, is termed durable concrete. Factors such as freeze-thaw and wet-dry cycling can cause deterioration of concrete in a relatively short period of time. Therefore, concrete durability is directly correlated to its state of deterioration. The water present in concrete, exposed to freezing conditions, undergoes an expansion process. This raises great concern, primarily due to the 9% expansion of the water in the concrete (Brook and Neville 1987). As the water in the moist concrete freezes, it produces osmotic and hydraulic pressures in the cement paste and the aggregate. As these pressures increase, micro-cracks begin to form and finally rupture. Rupture occurs as the pressure exceeds the tensile strength of the paste or aggregate. Resistance to freezing and thawing depends on many factors: permeability, degree of saturation, amount of freezable water and rate of freezing. Freeze-thaw resistance can be increased with the selection of good quality aggregate, low water/cement ratio, the right amount of cement content and proper curing techniques. Wet-dry cycling conditions can contribute to the deterioration of concrete. This primarily occurs with brackish waters having high salinity. Over time, the salt incorporates into the mix through the voids that naturally occur in the concrete. The salt then attacks the bonds in the concrete, thereby causing the mix to

3

develop cracks and eventually fail. This can also be detrimental to a concrete mix containing steel, since steel corrodes easily. To reduce the effects of wet-dry exposure, recommendations have been made concerning the addition of certain admixtures and air-entrainment mixtures. This reduces the susceptibility of concrete subject to wet-dry cycling.

4

2.0 Literature review and Background Research concerning the use of mineral admixtures to augment the properties of concrete has been going on for many years. Recently, the use of mineral admixtures has dramatically increased, due to an increase in environmental awareness. Higher early strength can be found in some modern cements due to more Ca(OH)2 formation, but this may adversely effect the durability and cost of concrete (Chan and Wu 2000). Since many mineral admixtures are by-products of other necessary reactions, these waste by-products can be used to reduce the amount of cement required, thus, in some cases, reducing the cost of the concrete. Mineral admixtures such as silica fume, fly ash, and ground granulated blast furnace slag should be studied in order to understand their ability to enhance the properties of concrete. Silica fume or microsilica when incorporated into concrete increases its early strength. Perhaps this is the result of the increasing rate of cement hydration in the early hours, due to a release of hydroxide ions (OH)- and alkalis into the pore fluids (Rao 2001). This may be true; however, what effect does this actually have on strength production? The process of hydration is actually quite complicated, requiring the production and dissolution of various compounds at different stages. During these stages, several reactions occur. The combination of these reactions determines the progression of strength development. One belief is that strength can be derived from three mechanisms. First, as silica fume is added, pore size is refined and the resulting matrix is denser. Silica fume causes large pores to become smaller, while the production of certain compounds increases the density of the matrix, thus, increasing the strength of the resulting bonds. Secondly, silica fume reduces the content of Ca(OH)2(CH). Finally, silica fume concrete undergoes a cement paste-aggregate refinement process. This process describes the enhancement of the transition zones between the cement paste and the aggregate (Rao 2001, El-Korchi and Toutanji 1996, Bayasi and Toutanji 1999, Bentur and Goldman 1989, Mohd Zain and Radin 2000).

As alluded to above, the addition of silica fume increases the early strength of concrete. It should be noted that this increase does not occur until after three days. After the three-day time period, hydration begins due to a higher content of calcium silicate hydrate (CSH). CSH is considered to be the main carrier of strength in hardened cement. It is formed through the transformation of weak Ca(OH)2 to strong CSH gel (Rao 2001, Bayasi and Toutanji 1999, Mohd Zain and Radin 2000, Papadakis 1999, Cong, et al. 1992, ACI 226 1987). Although many believe that the above reaction produces CSH gel, it does not mean that strength can be directly obtained through measurement of this parameter. In fact, concrete strength and durability can at best be determined indirectly by knowing CSH, CH, H-content, and porosity (Papadakis 1999). Furthermore, disagreements over which parameters should be considered more important have occurred. For instance, the effects of superplasticizers, or high-range water reducing agents (HRWR), are still being studied. One opinion is that HRWR agents, due to their ability to break up flocs which release water results in a higher degree of hydration and strength development. HRWR provides the necessary water to improve workability and the hydration of cement. For this reason, HRWR agents can, to some extent, enhance the strength of concrete. However, the

5

use of too much HRWR can negatively affect strength through the segregation of particles (Cong, et al. 1992). One of the important processes that develop strength in concrete containing silica fume is the refinement of the transition zone between cement paste and aggregate. Silica fume particles provide nucleation sites to allow the deposition of Ca(OH)2 during the early stages of hydration. These sites become the areas of CSH gel formation, which helps increase the strength of particle-to-particle bonds and particle-to-aggregate bonds (ACI 226 1987, Krstulovic, et al. 2000). The development in strength does not occur until after approximately three days and the trend in strength development may continue until fourteen days whereupon it remains steady (Papadakis 1999). Perhaps in future research, clear and incontrovertible evidence can be used to explain this complicated issue. In the meantime, theory and engineering judgment must be used to describe strength development.

Another important issue that must be resolved is the quantity of silica fume required to produce the optimum strength, while minimizing its potentially adverse effects. The addition of too much silica fume can reduce strength due to the quantities of water and HRWR agents required to construct a workable mixture. One suggestion is that the optimum replacement of cement by silica fume seems to be 15%; however, the highest strengths were observed in mixes containing 20% silica fume (Alexander and Magee 1999, Haque, et al. 1987). Another suggestion was that concrete utilizing 10% SF by mass to the initial binder is the optimum. It should be noted that this conclusion was derived using a range of 5 to 10% silica fume (Alexander and Magee 1999). Other observations were very unclear as to an optimum percentage, and disagreements with the earlier suggestions were expressed (Toutanji 1995). This made it difficult to determine a suitable test group for the current research. As a result a wide range of test samples using varying percentages was used. Literature covering other mineral admixtures had either one suggestion or none concerning the optimum percentage of the additive to be used.

Literature concerning fly ash had no mention of an optimum quantity for use in concrete mixes; however, recommendations have been made suggesting the use of 40% or less. Furthermore, if early strength is not an important factor, fly ash as high as 60% can be used (Naik and Ramme 1989). In order to comprehend the difficulty that exists in determining a single optimum amount, an understanding must be obtained as to the process that determines strength in fly ash mixes. Unlike silica fume mixes, fly ash mixes require longer periods of time to develop strength. It is shown that at 91 days, the strength of mixes incorporating fly ash outperform specimens using no additives (Carette, et al. 1989). Compared to silica fume, fly ash mixes using excessive amounts of HRWR may cause segregation of different materials, resulting in lower strength. Also, a reduction in porosity of concrete incorporating fly ash can be linked to improvements that occur on the interfacial transition zones between the cement matrix and the aggregate. Also, the reaction of CH and fly ash forms CSH gel, which results in increased compressive strengths. On the downside, fly ash tends to have a low reactivity rate when used as an admixture in Portland cement mixes. At 28 days, the degree of fly ash reaction is slightly more than 10% (Lam, et al. 2000, Lam, et al. 2000, Chan and Wu 2000, Fu, et al. 2000). Furthermore, only 20% of the fly ash reacted after 90 days. Despite this, the hydration of the cement was enhanced due to increased effective water to cementitious ratio (w/c) for the paste (Lam, et al. 2000). One explanation for the exceedingly low activation rate in fly ash at room temperature is that the pH of the solution 13 does not meet the requirements of fly ash, 13.3 (Chen, et al. 2000).

6

Another important factor in determining the strength development of fly ash mixes is the presence of carbon. It has been shown that fly ash concrete containing low amounts of carbon tends to have high mechanical strength and high freeze-thaw resistance (Wesche 1991). It is becoming increasingly apparent that determining the development of strength of mixtures containing fly ash is very difficult. One possible method of determining the strength contributed by fly ash in concrete is the sum of the nucleation and pozzolanic factors. For instance, as the fly ash to cementitious ratio increases, the pozzolanic reaction increases. Furthermore, as the amount of fly ash increases, the change in gel space ratio becomes increasingly insignificant (Gopalan 1993). Slag has low strength at early ages when used in large quantities. This is due to the slow initial hydration of slag. Unlike fly ash, slag’s reactivity is dependent more on temperature (Iwaki and Miura 2000, Malhotra and Sivasundaram 1992). Like fly ash, two processes govern slag’s reaction rate during the initial period of hydration. The first, nucleation and the growth rate of hydration phases, is the ability to provide sites for the bonding of Ca(OH)2 compounds. These compounds are then transformed into CSH gel. The second is the phase boundary interactions or the interactions that occur between the old compounds and the newly formed compounds. It has also been found that the initial rate of hydration is dependent on the initial lime content. This can be observed in the sharp decreases in the free lime contents that occur in slag mixes during the first day (Al-Wakeel, et al. 2001). These compounds or hydrates are generally more gel- like in structure than the products of Portland cement. These hydrates add to the density of the cement paste. Also the hydration of slag, in the presence of Portland cement, depends greatly upon the breakdown and dissolution of glassy slag structures by hydroxide anions (OH)-. This ion release occurs during the hydration of cement. ASTM C989 recommends the use of the slag activity index as a basic criterion for evaluating the value of cementitious materials (Iwaki and Miura 2000, ACI 226 1987, Bilodeau, et al. 1992).

Since FA seems to be closely related to slag, it is sometimes substituted for it. Slag has recently increased in cost and as a result FA must be used to maintain a relatively low cost for concrete. When a proper ratio between FA and slag is used, the strength of the combination exceeds that of the slag cement (Chen, et al. 2000). It should be noted that in the case of slag concrete, an optimum does exist. It is found that the best strength and durability is obtained by using a one-to-one ratio of slag to Portland cement. Furthermore, if one exceeds this ratio, the early strength is inversely related to the amount used (Iwaki and Miura 2000, ACI 226 1987). Interestingly enough there are obvious differences between the development of strength between mixes. Many similarities can also be found. For instance, it appears as though the strength in all the mixes using different admixtures is dependent primarily on the same factors. First, proper nucleation sites exist for the bonding of Ca(OH)2 compounds. Second, the bonding that exists between particles and aggregates must be strong. Finally, the weak Ca(OH)2 compounds must be replaced, to some extent, with CSH gel. These similarities perhaps can be linked to their reactions in the presence of each other. Perhaps future research can develop a better understanding of these factors. The most important design parameter for concrete structures is compressive strength. This critical parameter drives the design process and can influence the cost of a project. Through the

7

use of certain mineral admixtures the cost of concrete can be reduced. These admixtures also enhance the properties of mortar or cement mixtures. In some cases, a boost in early strength becomes apparent, while in others an increase in late strength occurs. The partial replacement of cement and mortar with these admixtures leads to improved durability and increased versatility. Also, from an environmental standpoint, the use of these materials is paramount, since the production of cement creates substantial amounts of pollution. As a result, cement used in large construction projects is generally purchased from either Venezuela or Mexico (Fronsdorf and Pasko 1994). Replacing large amounts of cement with mineral admixtures such as slag, fly ash, and silica fume can reduce the high cost. As a result, extensive research has been carried out, or is in the process of being performed on these admixtures in order to determine their usefulness. This research intends to expand the knowledge concerning the proper use of these admixtures, in addition to determining the combined effects of these materials on the properties of concrete.

8

3.0 Testing Procedure 3.1 Specimen Preparation Mixes were prepared using different types of admixtures and various combinations of these admixtures. The materials included: coarse aggregate (limestone), natural sand, manufactured sand, Type I Portland cement, condensed silica fume, Class C fly ash, Grade 120 granulated blast furnace slag, superplasticizer, and water. All mixes were produced using the same mix procedure and the same mixer. The mix proportions are shown in Table 3-1.

Table 3-1 – Mix Proportions

Fine Aggregate Mix

Cement (lb) Natural Sand

(lb) Manufactured

Sand (lb)

Course Aggregate

Water (lb)

Super. (lb)

SF (lb)

SL (lb)

FA (lb)

Slump (in)

Control 37.2 31.52 47.28 109.2 14.88 0.74 6.5 8% SF 34.22 31.52 47.28 109.2 14.76 0.74 2.98 3 10% SF 33.48 31.52 47.28 109.2 14.6 0.94 3.72 3 15% SF 31.62 31.52 47.28 109.2 16.12 0.94 5.58 2 60% S 13.91 18.87 28.3 65.42 8.1 0.334 8.34 4.5 70% S 13.09 18.87 28.3 65.42 8.9 0.334 9.16 4.5 80% S 12.36 18.87 28.3 65.42 8.9 0.334 9.89 5.5

20% FA 18.6 18.93 28.39 65.62 8.48 0.335 3.72 4.5 25% FA 17.86 18.93 28.39 65.62 7.63 0.335 4.46 4

30% FA 17.25 19.02 28.52 65.93 8.64 0.336 5.17 6 Combined A 11.58 19.72 29.6 68.41 9.55 0.459 2.32 5.82 3.49 0 Combined B 11.58 19.72 29.6 68.41 9.65 0.459 1.18 8.16 2.33 0.5 Combined C 11.58 19.72 29.6 68.41 8.71 0.459 1.75 6.98 2.94 3 Combined D 9.3 15.77 22.84 54.6 7.44 0.372 4.65 4.65 3 Combined E 9.3 15.77 22.84 54.6 7.44 0.372 5.58 3.72 2.75 Combined F 9.3 15.77 22.84 54.6 7.44 0.372 7.44 1.86 3.5

The test specimens were cast from 16 separate batches of concrete. In addition to the control and the nine mixes containing different percentages of SF, S, and FA, six mixes made of combinations of the cement, silica fume and fly ash were added. The amount of cement in the combination mixes was kept constant at 50% by mass of the cementitious materials. The combination mixes were designated as A, B, C, D, E and F. The mix proportions of these mixes were as follows: cement:silica fume:fly ash:slag = 50%:10%:15%:25% for Mix A, 50%:5%:10%:35% for Mix B, 50%:7.5%:12.5%:30% for Mix C, 50%:0%:10%:40% for Mix D, 50%:0%,20%:30% for Mix E and 50%:0%:25%:25% for Mix F. For each mix, four to six 3x4x16 inch beam specimens were cast and 12 to 16 4x8 inch cylinders were cast for compression, tension and wet-dry tests. In each test group, specimens were divided into three equal sets, each consisting of four specimens. The final strength values represent the average strength of three to four specimens. Because of project time/limitations, all specimens were moist cured for 14 to 20 days before testing or before they were subjected to the environmental

9

exposures. The specimens were then placed in their respective testing chambers and the required test cycles were run. 3.2 Mixing Procedure In order to acquire uniform results it was important to adhere to a strict mixing procedure. Such a procedure was adopted and followed in this study. The following paragraphs describe the procedure used in the mixing of all samples.

Measure all the mix ingredients in the proportions shown in Table 3-1. Make sure that all the metal instruments and tools are moist (mixing drum, trowels, mixing trays, and rods). This ensures that the moisture content is not drastically reduced by the properties of the metallic instruments. Make sure that there is no standing water in the mixing drum or pans. Place cementitious components (slag, fly ash, silica fume and Portland cement) in the drum mixer. Begin the mixing process by first placing the coarse aggregate into the mixing drum and then add the fine aggregate (natural and manufactured sand). Turn on the mixer and allow it to mix for approximately one minute. Stop the mixer and then place the cementitious materials into the mixing drum followed by half of the liquid. The liquid is prepared by adding the required superplasticizer to the measured water. While the mixing drum is running, pour the liquid deep into the mixer. It is preferable to pour the liquid directly into the mix if possible. After one to two minutes of mixing, add the rest of the liquid and mix for another four to five minutes. When the mixing is done, lower the handle of the concrete mixer and pour the contents out into a moistened pan.

The workability and consistency of the concrete mix are measured by the slump cone test ASTM C143-89a (1994). The slump of the mix should be recorded in inches. Complete the slump test then place the already oiled cylinder molds on a leveled surface to be filled with the concrete mixture (see Figure 3-1). Cast the specimens according to ASTM standard. Level the concrete to the upper surface of the molds. Then cover the specimens using plastic sheets, to prevent the water from evaporating (see Figure 3-2). After 24 hours, strip the specimens from their respective molds and place them in water to cure for 14 to 20 days (see Figure 3-3). Following the casting process, clean all instruments including the mixer thoroughly and prepare for the next mix.

10

Figure 3-1. Molds for beams and cylinders.

Figure 3-2. Cast concrete specimens.

11

Figure 3-3. Concrete specimens in curing tank. 3.3 Wet-Dry Test Wet-dry tests can be used to determine the durability of concrete. The specimens remain in the wet and dry chamber for a required period of time. Following the elapse of this time, they are tested using both the standard compression test and the standard tensile splitting test. The results can then be compared to those obtained for the control specimens (the room temperature specimens). Specimens were placed in a specially constructed environmental chamber, located in the Materials Lab at the Civil and Environmental Engineering Department of UAH, and were exposed to 300 cycles of wetting and drying. The wet-dry environmental chamber is schematically shown in Figure 3-4. The environmental chamber utilizes two pumps and an industrial dryer to cycle the specimens constantly. This constant cycling can be achieved by connecting the pumps and dryer to a series of timers set to activate at specified times. The specimens were subjected to salt water environments in which there were alternating wet and dry cycles (hot air at 35oC average and 90% humidity). Seawater was simulated using 35 grams of salt in a liter of water. This is the approximate content of salt found in the ocean. The duration of the wet cycle was four hours and the dry cycle was eight hours. Using this ratio, the specimens were exposed to a total of 150 days. Following the elapse of the specified exposure period, the specimens underwent two types of tests, a standard compression test and a standard tensile splitting test to determine the strength properties of the specimens after exposure.

12

Test Specimens

Hot air inlet

Water inlet

Water outlet

Figure 3-4. Schematic of the wet-dry exposure set-up. 3.4 Freeze-Thaw Test Although the freeze-thaw test is ideal for determining relative durability, it is not intended for quantitative analysis. The changes in the properties of concrete subjected to the freezing and thawing cycles can be determined. However, the extent of service of a particular type of concrete cannot be determined. Freeze-thaw tests entailed subjecting the specimen to 300 cycles of rapid freezing and thawing conditions. As a specimen underwent the test, it was subjected to many different measurements. For instance, the specimen was weighed, measured, and the fundamental transverse frequency was determined. These factors along with the durability factor are generally assumed to project how well a specimen will perform under repetitive cycles of freezing and thawing. The durability factor reflects the required number of freezing and thawing cycles to yield a particular

13

amount of deterioration. The relative dynamic modulus of elasticity of the test specimens can be determined by measuring the resonant frequency with a sonometer. The apparatus consists of a driver and pick-up circuits. The driver converts electrical power to mechanical vibrations. The mechanical vibrations are imparted to the test specimen to measure the wave velocity through the concrete. Amplitude and frequency of vibrations are controllable in ranges of 0-30 watts and 100-10,000 Hz to a guaranteed accuracy of ± 1% over the entire range. Resonant frequency is determined by observing voltmeter peak readings while varying input frequency. An oscilloscope is used to verify detection of fundamental transverse frequency. Another measurement, the relative dynamic modulus of elasticity uses the ratio of the fundamental transverse frequency at a specified number of cycles to the initial transverse frequency to determine whether a test can continue. According to note 7 of ASTM C666 (1994), it is not recommended that specimens be continued in the test after their relative dynamic modulus of elasticity has fallen below 50%. A major difficulty lies in the fact that a test must be accelerated in comparison with outdoor conditions. Accelerated freezing and thawing tests can result in the destruction of concrete that in practice could be satisfactory. Nevertheless, the ability of concrete to withstand a high number of freezing and thawing cycles can be viewed as a good indication of its durability. The model HM-120 apparatus from Gilson Company, Inc., tests the specimens automatically (see Figure 3-5). The freezing and thawing machine cons ists of a stainless steel cabinet with 18 compartments for 3x4x16 inch beams. The temperature at the center of the control specimens ranges from 40° to 0°F. The cabinet contains three inches of insulation on all sides and a refrigeration unit mounted on the under-frame. Controls and a seven-day temperature recorder can be located at the end of the cabinet. The machine uses several mechanisms to measure the resistance of the concrete specimens to freezing and thawing conditions. It also can be used to evaluate durability of the aggregates and properties of the admixtures. Specimens are subjected to freezing and thawing conditions, while surrounded by 1/8 inch layer of water. The specimens undergo a maximum nine cycles of freezing and thawing per 24-hour period. Simultaneously, a continuous record of the temperature at the center of the control specimen is plotted on a seven-day recording chart. The design of this apparatus corresponds to ASTM C 666 procedure A: Rapid Freezing and Thawing in water (1994). It also follows the provisions found in AASHTO T161 (Procedure A). After the 14-day curing period and before freezing and thawing, the specimens were tested for fundamental transverse frequency as per ASTM C215 (1994). The specimens were weighed and measured for average length and cross sectional dimensions. The specimens were then inserted into thawing water, which began the freezing and thawing test. They were removed in a thawed state at intervals less than or equal to 36 cycles of exposure and were tested for fundamental transverse frequency. The length and weight were measured and the process continued until either 300 cycles of freezing and thawing were completed or until the dynamic modulus of elasticity fell below 50%.

14

Figure 3-5. Freeze-thaw apparatus. The relative dynamic modulus of elasticity, which is used to calculate the durability factor, can be determined as follows:

pc =n1

2

n02

100 (1)

Where:

Pc = relative dynamic modulus of elasticity, after c cycles of freezing and thawing (%) η0 = fundamental transverse frequency at 0 cycles of freezing and thawing, Hz η1 = fundamental transverse frequency after c cycles of freezing and thawing, Hz

The fundamental transverse frequency is determined in accordance with ASTM C 215-Standard Test Method for Fundamental Transverse, Longitudinal and Torsional Resonant Frequencies of Concrete Specimens (1994). The durability factor is found as follows:

DF =PNM

(2)

Where:

DF = the durability factor of the test specimen P = relative dynamic modulus of elasticity at N cycles (%) N = number of cycles at which P reaches the specified minimum value for discontinuing

the test or the specified number of cycles at which the exposure is to be terminated, whichever is less

M = specified number of cycles at which the exposure is to be terminated

15

3.5 Compression Test The compression test was performed using a standard ASTM compression tester, manufactured by Forney. Each of the specimens was allowed to cure for at least 14 days prior to the test date. The cylinders were capped with sulfur to ensure that the specimens were axially loaded concentrically, in order to reduce the compressive strength. The compression-testing machine was first prepared by placing an adjustable load cell in the upper mount. The cylinders were placed in the machine ensuring that they were directly under the center of both the top loading cell and the bottom plates. They were then loaded at a rate of 200 psi per minute until failure. The failure load was recorded and the cylinder removed in preparation for the next test. Following each test the machine was cleaned to ensure that no changes occurred in the loading scenario. The stress of the cylinder is calculated as:

′ f c =pA

(4)

Where:

′ f c = ultimate splitting tensile strength p = maximum applied load indicated by the testing machine A = cross sectional area of the specimen

3.6 Tensile Splitting Test The tensile splitting test, ASTM C 496-86 (1994), is used to determine the tensile properties of concrete specimens. A standard compression tester is used to administer this test. The major diameters of the specimens are first marked using a pencil and a piece of angle iron. The specimen is then placed on its side on top of a bearing strip of plywood, having a nominal thickness of 1/8 inch and an approximate width of 1 inch. The length of the bearing strips has to be the length of the specimen or slightly longer. Another bearing strip is placed on top of the cylinder. The cylinder is then loaded at a constant rate that ranges from 100 to 200 psi per minute. The cylinder tested to failure and the ultimate load is recorded. The splitting tensile strength of a cylinder is calculated as:

T =2pπld

(3)

Where:

T = ultimate splitting tensile strength p = maximum applied load indicated by the testing machine l = length of specimen

d = diameter of specimen

16

4.0 Results

4.1 Effect of Supplementary Cementitious Materials on Strength The addition of silica fume produced the highest increase in strength, when compared to the other supplementary cementitious materials (fly ash, slag and combinations), as shown in Figure 4-1. With the addition of 8, 10 and 15% of silica fume, the compressive strength increased by 33, 32 and 21%, respectively. The addition of 8 to 10% of silica fume by mass seemed to be the optimum content that produced the highest strength. The increase in strength of silica fume concrete can be attributed to an improved bond between the hydrated cement matrix and the sand in the mix. This improved bond was due to the conversion of calcium hydroxide which tended to form on the surface of aggregate particles as calcium silicate hydrate, due to the presence of reactive silica. However, when the dosage of silica fume increased beyond 10% in mass, the compressive strength started to decrease. This reduction in strength may be a result of the lack of water in the mix since self-desiccation of the specimens may have taken place. This limited the pozzolanic effect and thereby reduced the strength of the silica fume concrete. Although the present work did not study the increase of water demand for silica fume concrete, water amount can increase as much as 30% depending upon the amount of silica fume in the mixture (El-Korchi and Toutanji 1995, Darwin and Maher 1977, ACI 226 1987). The addition of slag resulted in a modest increase in strength. Concrete with cementitious materials of 70% slag seemed to produce the optimum strength as compared to mixes with 60% and 80% slag. The addition of 70% slag resulted in an increase of about 15% over plain concrete; whereas, 60% and 80% slag resulted in about 5% increase. The addition of fly ash exhibited a reduction in compressive strength. The compressive strength was reduced by as much as 50% with the addition of 30% fly ash. This reduction in strength is attributed to the fact that specimens were cured for a short period of time, only 14 days. Because of the slow pozzolanic reactions of fly ash, continuous wet curing and favorable curing temperatures are required for proper development of strength. Some fly ashes require 90 days or more to equal or exceed a 28-day control strength (Kostamka and Panarese 2000). It should be noted that fly ash has been used in the production of high-strength concrete when it is cured for a long period of time. Longer curing enables the hydration process to progress. The combination of different supplementary materials increased compressive strength (Figure 4-1). The combination of 10% silica fume, 25% slag, 15% fly ash and 50% cement, Mix A, produced an increase in strength of about 22% over the control mix (concrete without supplementary cementitious materials). However, a combination of 5% silica fume, 35% slag, 10% fly ash and 50% cement (Mix B) produced a reduction in strength of about 15%. The combination of fly ash, slag and cement (Mixes D, E, and F) did not show any significant effect on the strength. The reduction of about 5% is statistically insignificant.

17

2500

3500

4500

5500

6500

7500

8500

9500

10500

Mix Design

Com

pres

sive

Str

engt

h (p

si)

30% FA Comb. B20% FA Comb. FComb. E Comb. D25% FA Control80% S 60% SComb. C 70% S15% SF Comb. A10% SF 8% SF

30% FA

8% SF

10% SF

A

15% SF

70% S

C60% S

80% S

Control

25% FA20% FA

B

SF:SL:FAA = 10, 25, 15B = 5, 35, 10C = 7.5, 30, 12.5D = 0, 40, 10E = 0, 30, 20F = 0, 25, 25

E F D

Figure 4-1. Compressive strength performance for all mixes.

The performance of specimens in tension followed patterns similar to those of compression. It should be noted that this was for mixes with particular admixtures and not necessarily with respect to the dosages of admixtures. Specimens with silica fume exhibited the highest increase in tensile strength, followed by slag and then fly ash, as shown in Figure 4-2. Mix A had the highest strength among all combined mixes and Mix B exhibited the least. The combination of fly ash, slag and cement (Mixes D, E, and F) exhibited reductions in tensile strength of 16, 20 and 23%, respectively, as compared to the control mix. In general, the splitting tensile strength was about 11 to 16% of compressive strength. Specimens with supplementary cementitious materials showed a lower ratio of compressive to tensile strength as compared to control specimens see Figure 4-3. Although, the tensile strength was thoroughly investigated in this study, it was not as critical as the compressive strength, since it is usually is assumed to be zero in design. However, the objective here was to determine how the relationship between compressive and tensile strength changed with the addition of different supplementary cementitious materials. This is depicted in Figure 4-3.

18

0

250

500

750

1000

1250

1500

1750

Mix Design

Ten

sile

Str

engt

h (p

si)

30% FA Comb. B20% FA Comb. FComb. E Comb. D25% FA Control80% S 60% SComb. C 70% S15% SF Comb. A10% SF 8% SF

30% FA

8% SF

10% SF

A

15% SF

70% SC

60% S

80% S

Control

25% FA20% FA

B

SF:SL:FAA = 10, 25, 15B = 5, 35, 10C = 7.5, 30, 12.5D = 0, 40, 10E = 0, 30, 20F = 0, 25, 25

E F D

Figure 4-2. Tensile strength performance for all mixes.

6%

8%

10%

12%

14%

16%

18%

Mix Design

Com

pres

sive

Str

engt

h/T

ensi

le

Stre

ngth

(%)

COL

8 SF

10 SF

15 SF

20 FA

25 FA

30 FA

60 S

70 S

80 S

A

B C

Figure 4-3. Ratios of compressive strength to tensile strength for all mixes.

19

4.2 Effect of Wet-Dry Exposure on Strength The compressive strength of plain concrete specimens decreased due to wet-dry cycling (Figure 4-4). The compressive strength was reduced by as much as 13% due to this exposure. Test results showed that by increasing silica fume content, the durability resistance of specimens seemed to increase compared to the control specimens (specimens without exposure). The compressive strength of specimens containing 8% silica fume was reduced by as much as 10%; however, this reduction was seemingly reversed in specimens with silica fume contents greater than 8%. A moderate increase in strength of about 2 and 6% was exhibited in specimens with 10 and 15% silica fume, respectively. Mixes with fly ash and slag showed significant increases in strength due to wet/dry exposure. There was a positive correlation between the increase in strength and the contents of fly ash and slag in the concrete mixes. For specimens containing 20% fly ash, the increase in strength was about 16%; whereas, specimens with 30% fly ash experienced an increase in strength of over 30%. For specimens containing 60% slag, the increase in strength due to wet-dry exposure was about 18%; whereas, specimens with 80% experienced an inc rease of about 25%. Specimens with different combinations of silica fume, fly ash and slag also showed increases in strength. Specimens with the largest amount of fly ash, Mix A (10% silica fume, 15% fly ash and 25% slag), showed the largest increase in strength, about 15% as shown in Figure 4-4. The reason for this significant increase in strength of specimens containing fly ash and slag was that wet and dry exposure provided these specimens, which contain pozzolanic materials, a longer curing time. Pozzolanic materials require a longer curing time than control specimens; they need more time to cure and to develop strength. It was hard to quantify the effect of salt water on strength; however, it was clear that the long-term exposure, the frequent wet exposure and the high drying temperature contributed to the development of strength of these specimens. This explained why slag has been used in seawater environments for many years, typically for the construction of onshore and offshore rigs (Ludwig and Stark 1997).

20

0

2000

4000

6000

8000

10000

12000

COL SF8

10SF2 SF

1520F

A25F

A30F

ASL

60SL

70SL

80

5-10-3

5

10-15-

25

7.5-12

.5-30

Mix Design

Com

pres

sive

Str

engt

h (P

si)

Room Temperature Wet- Dry

COL

8% SF 10% SF

15% SF

20% FA25% FA

30% FA

60% S70% S

80% S

AC

B

Figure 4-4. Compressive strength data for both room temperature and wet-dry conditions.

4.3 Effect of Freeze-Thaw Conditions on Strength Figure 4-5 represents a qualitative estimate of the durability performance of all specimens. Figure 4-6 shows the relative dynamic modulus of elasticity of all mixes. Results showed that the control samples (COL) possessed higher resistance to freeze-thaw exposure than mixes with supplementary cementitious materials. The closest in durability performance to the control was the mix that contained 8% silica fume. However, increasing content of silica fume beyond 8% seemed to drastically decrease the durability performance. Thus, the durability of concrete containing silica fume was inversely proportional to the amount added. Concrete containing 8% silica fume exhibited a twofold increase in durability performance over specimens containing 10% and almost a 13 fold increase over those with 15%. Specimens with slag performed better than those with fly ash. Results showed no correlation between the contents of slag and the durability factor. Other studies have shown that mixes containing higher than 60% slag may have a higher frost resistance (Ludwig and Stark 1997).

21

It appeared that the greater the amount of fly ash, the lower the resistance to freeze and thaw exposure. With higher contents of fly ash, the durability performance seems to decrease. The difference between the mix containing 20% fly ash and 25% fly ash was small; however, the difference between the mix containing 25% fly ash and 30% fly ash was significant. It is important to note that these mixes were cured for only 14 days before exposure. The results may have been different if the specimens were cured longer. The second highest average durability factors occurred in the combination mixes. On the average, mixes A, B, and C performed better than mixes containing a single admixture. This may be attributed to the fact that the interaction of silica fume, slag, and fly ash resulted in a more stabilized mix. Finding an optimum mix ratio that produces the best durability performance would require further research. In addition, more work should be done to fully understand the reactions and interactions that occur between these various admixtures.

89.7

26.522.5

15.8 13.6 13

4.1 3.7 2.7 0.64

17.618

34.9

0

10

20

30

40

50

60

70

80

90

100

COL8%

SF70%

S A C B60%

S

10%SF

80%S

20%FA

25%FA

15%SF

30%FA

Mix Design

Dur

abili

ty F

acto

r

COL8%SF70%SACB60%S10%SF80%S20%FA25%FA15%SF30%FA

Figure 4-5. Durability factor for control and mixes with supplementary cementitious materials.

22

0

20

40

60

80

100

120

0 50 100 150 200 250 300 350

Number of Cycles

Rel

ativ

e D

ynam

ic M

odul

us (%

)

Control SF 8% SF 10% SF 15% S 60%S 70% S 80% FA 20% FA 25% FA 30%Comb. A Comb. B Comb. C

Figure 4-6. Relative dynamic modulus of elasticity of all mixes.

23

5.0 Conclusions A broad research program was conducted to study the effect of different supplementary cementitious materials on the strength and durability of concrete. This study was a stepping-stone for further studies to address many of the complex issues related to this important topic. This section provides conclusions about the results obtained in this study. 5.1 Effect of the Addition of Admixtures 1. The optimum amount of silica fume to be added to concrete was found to be 8% to 10%.

With higher dosages of silica fume, the compressive strength seems to decrease. 2. A moderate increase occurred in the compressive strength of concrete containing slag. The

optimum content was found to be 70%. 3. Among the single admixture mixes, fly ash contributed to a decrease in strength. Specimens

with the highest amount of fly ash experienced the largest decrease in strength of about 40%. 4. When fly ash was used in combination with the other supplementary materials, the

compressive strength was increased. The combination of 10% silica fume, 25% slag, and 15% fly ash resulted in an increase in strength by as much as 22% over the control mix. On the other hand, a combination with 5% silica fume, 35% slag and 10% fly ash resulted in a decrease in strength by as much as 15%.

5. The tensile strength of the specimens followed the same pattern as the compressive strength with respect to the type of admixtures but not necessarily to the level of admixtures. Results showed that the tensile strength ranged between 11 and 16% of the compressive strength. The control mixes appeared to be on the upper end of this range.

5.2 Effect of Wet-Dry Exposure 1. The compressive strength of plain concrete specimens decreased due to wet-dry exposure. 2. The silica fume specimens exhibited an increase in strength when its content exceeded 8%.

More significant increases were observed in the fly ash and slag specimens over the control specimens. This was due to the fact that slag and fly ash specimens require a longer curing period. In order for the strength to fully develop, the specimens should cure for over 90 days as compared to only 28 days for the control specimens.

3. It was hard to quantify the effect of salt water on strength; however, it was clear that long-term exposure, frequent wet exposure and high drying temperature contributed to the development of strength of these specimens.

4. The combination mixes showed a substantial increase in strength. 5.3 Effect of Freeze-Thaw Exposure The control mix showed excellent durability as compared to the rest of the mixes. Specimens containing fly ash performed quite poorly. On the average, specimens with slag exhibited higher

24

durability characteristics than mixes containing a single admixture. The combination mixes also performed very well when exposed to freeze and thaw conditions. In summary, considering the many mixes that were analyzed in this study, the 8% silica fume mix performed the best overall. In general, the silica fume specimens performed the best out of all of the other admixture and combination specimens. Combined A mix (50% cement, 10% silica fume, 15% fly ash and 25% slag) was the next best in overall strength and durability performance. By far, the least effective mixes were the fly ash series. These mixes experienced the least strength and durability development of all of the other mixes. The control mix performed moderately well in room temperature conditions, poorly in the wet-dry environment, and well in the freeze-thaw condition. It should also be noted that the slag specimens performed moderately well in all tests and should be considered as a possible mix especially in severe environmental conditions.

25

6.0 Recommendations for Future Research

Some researchers believe that the increase in strength due to the addition of admixtures in concrete is attributed to an increase in the bond strength between the cement paste and the aggregate. Others believe that it is the result of an increase in the strength of the cementitious paste matrix. Regardless of these controversies one fact remains : during a process called hydration different reactions occur and promote strength development. A micro- level study should be conducted to characterize strength development when supplementary cementitious materials are added. Because of the limited amount of time available for this study, the specimens were cured for a short period of time, less than 20 days. Such a short curing period was not enough to develop the strength, especially for mixes with slag and fly ash. With respect to silica fume, the curing time allocated in this study was enough since silica fume hydrates much faster than either slag or fly ash. A similar study should be conducted allowing a longer curing time for the specimens, preferably a 90-day period or longer. In addition to unreinforced concrete, specimens reinforced with steel should also be investigated. A theoretical study will be of great benefit to engineers and the scientific community to predict the strength and durability performance of concrete containing different types and dosages of supplementary cementitious materials.

26

7.0 References

ACI Committee 226. “Silica Fume in Concrete.” ACI Materials Journal. Vol. 84, no. 2, pp. 158 166. Mar. - Apr. 1987.

ACI Committee 226. “Granulated Blast-Furnace Slag as a Concrete Constituent.” ACI Materials Journal. Vol. 84, No. 4, pp. 327 - 342. July - Aug. 1987.

ACI Committee 226. “Silica Fume in Concrete.” ACI Journal of Materials. Vol. 84, no. 158. 1987.

ASTM Designation C143-89a. Standard Test Method for Slump of Hydraulic Cement Concrete. 1994.

ASTM Designation C215-97. Standard Test Method for Fundamental Transverse, Longitudinal and Torsional Resonant Frequencies of Concrete Specimens. 1994.

ASTM Designation C496-86. Standard Test Method for Splitting Tensile Strength of Cylindrical Concrete Specimens. 1994.

ASTM Designation C618-89. Standard Specification for Fly Ash and Raw of Calcined Natural Pozzolans for use as Mineral Admixtures in Portland Cement Concrete. 1994.

ASTM Designation C666-97. Standard Test Method for Resistance to Rapid Freezing and Thawing. 1994.

Alexander, M.G., B.J. Magee. “Durability Performance of Concrete Containing Condensed Silica Fume.” Cement and Concrete Research. Vol. 29, no. 6, pp. 917 - 922. June 1999.

Al-Wakeel, E.I., N.Y. Mostafa, S.A.S. El-Hemaly, S.A. El-Korashy, and P.W. Brown. “Hydraulic Activity of Water-Cooled Slag and Air-Cooled Slag at Different Temperatures.” Cement and Concrete Research. Vol. 31, no. 3, pp. 475 - 484. Mar. 2001.

Bayasi, Z., H.A. Toutanji. “Effect of Curing Procedures on Properties of Silica Fume Concrete.” Cement and Concrete Research. Vol. 29, pp. 497 - 501, 1999.

Bentur, A., A. Goldman. “Bond Effects in High-Strength Silica Fume Concretes.” ACI Materials Journal. Vol. 86, no. 5, pp. 440 – 447. Sept. - Oct. 1989.

Bilodeau, A., E. Douglass, V.M. Malholtra. “Properties and Durability of Alkali-Activated Slag Concrete.” ACI Materials Journal. Vol. 89, no. 5, pp. 509 – 516. Sept. - Oct. 1992.

Brook, J.J., A.M. Neville. Concrete Technology. John Wiley and Sons Inc., NY. pp. 438. 1987. Carette, G.G., W.S. Langely, V.M. Malholtra. “High Fly Ash Structural Concrete.” ACI

Materials Journal. Vol. 86, No. 5, pp. 507 - 513. Sept. - Oct. 1989. Chan, W.W.J., C.M.L. Wu. “Durability of Concrete with High Cement Replacement.” Cement

and Concrete Research. Vol. 30, pp. 865 - 879. 2000. Chen, Y., D. Li, J. Shen, J. Su, and X. Wu. “The Influence of Alkalinity on Activation and

Microstructure of Fly Ash.” Cement and Concrete Research. Vol. 30, no. 6, pp. 881 - 886. June 2000.

Chen, Y., D. Li, J. Shen, L. Cheng, and X. Wu. “Study of Properties Fly Ash-Slag Complex Cement.” Cement and Concrete Research. Vol. 30, no. 9, pp. 1381 - 1387. Sept. 2000.

Cong, X., S. Gong, D. Darwin, and S.L. Mcabe. “Role of Silica Fume in Compressive Strength of Cement Paste, Mortar, and Concrete.” ACI Materials Journal. Vol. 89, no. 4, pp. 375-387. July - Aug. 1992.

27

Darwin, A. Maher. “Microscopic Finite Element Model of Concrete.” Presented at the Proceedings, First International Conference on Mathematical Modeling. University of Missouri-Rolla, St. Louis, MO. Vol. III, p.1705. 1977.

El-Korchi, T., H.A. Toutanji. “Effect of Curing Procedures on Properties of Silica Fume Concrete.” Cement and Concrete Research. Vol. 29, pp. 497 - 501. 1999.

El-Korchi, T., H.A. Toutanji. “The Influence of Silica Fume on the Compressive Strength of Cement Paste and Mortar.” Cement and Concrete Research. Vol. 25, no. 7, pp. 1591 -1602. 1995.

Fronsdorff, T.J. Jr. Pasko. “United States Government’s Role in High Performance Materials for Infrastructure.” Presented at the International Workshop on High Performance Concrete. Bangkok, Thailand. Ed. Poud Z, pp. 263 - 265. Nov. 1994.

Fu, X., W. Hou, C. Yang, D. Li, X. Wu. “Studies on High-Strength Slag and Fly Ash Compound Cement.” Cement and Concrete Research. Vol. 30, no. 8, pp. 1239 - 1243. Aug. 2000.

Gopalan, M.K. “Nucleation and Pozzolanic Factors in Strength Development of Class F Fly Ash Concrete.” ACI Materials Journal. Vol. 90, No. 2, pp. 117 - 120. Mar. - Apr. 1993.

Haque, M.N., V. Yogendran, B.W. Langlan, and M.A. Ward. “Silica Fume in High-Strength Concrete.” ACI Materials Journal. Vol. 84, no. 2, pp. 124 - 129. Mar. - Apr. 1987.

Iwaki, I., T. Miura. “Strength Development of Concrete Incorporating High Levels of Ground Granulated Blast-Furnace Slag at Low Temperatures.” ACI Materials Journal. Vol. 97, no. 1, pp. 66 - 70. Jan. - Feb. 2000.

Kosmatka, S., W. Panarese. Design and Control of Concrete Mixtures. Portland Cement Association. Skokie, IL. Thirteenth Edition, 2000.

Krustulovic, R., J. Zelic, D. Rusic, and D. Veza. “The Role of Silica Fume in the Kinetics and Mechanics During the Early Stage of Cement Hydration.” Cement and Concrete Research. Vol. 30, no.10, pp. 1655 - 1662. Oct. 2000.

Lam, L., C.S. Poon, and Y.L. Wong. “A Study on High-Strength Concrete Prepared with Large Volumes of Low Calcium Fly Ash.” Cement and Concrete Research. Vol. 30, no. 3, pp. 447 - 455. Mar. 2000.

Lam, L., Y.L. Wong, C.S. Poon. “Degree of Hydration and Gel/Space Ratio of High-Volume Fly Ash/ Cement Systems.” Cement and Concrete Research. Vol. 30, no. 5, pp. 747 - 756. May 2000.

Ludwig, H.M., J. Stark. “Freeze-Thaw and Freeze-Deicing Salt Resistance of Concretes Containing Cement Rich in Granulated Blast Furnace Slag.” ACI Materials Journal. Vol. 94, No. 1, pp. 47 - 55. Jan - Feb 1997.

Malhotra, V.M., V. Sivasundaram. “Properties of Concrete Incorporating Low Quantity of Cement and High Volumes of Ground Granulated Slag.” ACI Materials Journal. Vol. 89, no. 6, pp. 554 - 563. Nov. - Dec. 1992.

Mohd Zain, M.F., S.S. Radin. “Physical Properties of High-Performance Concrete with Admixtures Exposed to a Medium Temperature Range 20°C to 50°C.” Cement and Concrete Research. Vol. 30, pp. 1283 - 1287. 2000.

Naik, T.R., B.W. Ramme. “High-Strength Concrete Containing Large Quantities of Fly Ash.” ACI Materials Journal. Vol. 86, no. 2, pp. 111 - 116. Mar. - Apr. 1989.

Papadakis, V.G. “Experimental Investigation and Theoretical Modeling of Silica Fume Activity in Concrete.” Cement and Concrete Research. Vol. 29, no. 1, pp. 79 - 86. Jan. 1999.

Ramachandran, V.S. Concrete Admixtures Properties, Science, and, Technology, Parkridge, NJ. 2nd ed., Noyes publications, 1995.

28

Rao, A.G. “Development of Strength with Age of Mortars Containing Silica Fume.” Cement and Concrete Research. Vol. 31, no. 8, pp. 1141 - 1146. Aug 2001.

Sabir, B.B. “Mechanical Properties and Frost Resistance of Silica Fume Concrete.” Cement and Concrete Composites. Vol. 19, no. 4. Aug. 1997.

Toutanji, H.A. “The Influence of Silica Fume on the Compressive Strength of Cement Paste and Mortar.” Cement and Concrete Research. Vol. 25, no. 7, pp. 1591 - 1602. May 1995.

Wesche, K. “Fly Ash in Concrete Properties and Performance.” Report of Technical Committee 67-FAB Use of Fly Ash in Building. NY. RILEM, E & FN Epson, pp. 8 - 19. 1991.