Abstract

Synopsis: A recurring question about use of fly ash in concrete is dealing with setting and hardening of such mixtures with our with out chemical admixtures. This paper presents literature review on the setting and hardening characteristics of cement paste and concrete as influenced by the inclusion of fly ash and chemical admixtures. The paper also reports the work carried out at the University of Wisconsin-Milwaukee (UWM-CBU) on the effects of Class C fly ashes from various sources on the initial-and final-setting times of non-air-entrained and air-entrained concrete; and the effects of Class C fly ash, gypsum, and various types of chemical admixtures (air-entraining admixture (AEA), water-reducing admixture (WRA), superplasticizer, and retarding admixture) on the initial and final setting times of cement paste. Test results indicated that: (1) both the initial-and final-setting times were relatively unaffected at low-percentage replacement of cement with Class C fly ash, although inclusion of fly ash caused large retardation in the times of setting, up to around 60 percent cement replacement; (2) initial-and final-setting times of cement paste remained essentially the same or were slightly delayed with up to 20 percent cement replacement relative to zero percent fly ash content; beyond this range, the setting times of cement paste were accelerated. Increased rate of setting occurred at cement replacement levels of 40 percent and higher irrespective of type of chemical admixtures used. Keywords: Air-entraining admixture (AEA), concrete, fly ash, gypsum, high-range water-reducing admixture (HRWRA), paste, retarder, time of setting, water-reducing admixture. INTRODUCTION Immediately upon mixing of cement and water, various chemical reactions occur leading to formation of numerous types of hydration products. The types and amount of hydration products formed depend upon duration of hydration, water-cementitious materials ratio (W/Cm), properties of constituent materials, temperature, soluble alkalis, and mineral and chemical admixtures. The formation of hydration products causes increase in stiffness of the cementitious matrix. This stiffening behavior of the matrix is determined by the times of initial and final setting. The initial setting of the matrix refers to the beginning of solidification for a given mixture. It is generally accepted that at this stage concrete can neither be properly re-tamped nor handled or placed. The final setting refers to the stage when the mixture attains sufficient hardness to support stress. The subsequent continuing strength gain is called hardening. Setting and hardening of cement mortar mixtures are considerably influenced by inclusion of either mineral or chemical admixtures. Generally, the setting and hardening of mortar are delayed when ASTM Class F (low-lime) fly ash is added to it. Mortar incorporating ASTM Class C (highlime) fly ash, however, has shown either both rapid or delayed setting depending upon the properties and amount of the ash. The setting behavior can be more readily modified when gypsum and chemical admixtures such as water-reducing admixture (WRA), superplasticizer, or retarding and accelerating admixture are used. Even air-entraining admixture is known to slightly modify setting behavior of concrete. A knowledge of setting characteristics of concrete incorporating both mineral and chemical admixtures is needed for efficient scheduling of concrete construction, specifically floor slabs, roadways, pavements, and other flat surfaces. Limited data exist on setting and hardening behavior of paste, mortar, and concrete containing ASTM Class C fly ash and chemical admixtures. 3 LITERATURE REVIEW Many investigators have reported on the effects of fly ash on the times of setting of cement paste and concrete. Dodson 1 investigated the setting characteristics of concretes made with both Class C and Class F fly ashes. He reported that the setting times of concrete are mainly governed by cement content and W/Cm when all other parameters are kept equal. He further added that an increase in cement content caused a decrease in the initial-and final-setting times, whereas an increase in W/Cm increased setting times. However, in general, addition of fly ash increased the setting times. Ramakrishnan et al. 2 reported on the setting characteristics of concretes made with or without fly ash. They used one high-lime fly ash and two types of cement (ASTM Type I and Type II). They concluded that inclusion of fly ash resulted in higher initial-and final-setting times compared to the concrete without fly ash for both types of cement. Lane and Best 3 reported that fly ash generally slows the setting of concrete, although both initial and final times of setting remain within specified limits. Retardation of setting due to the inclusion fly ash may be affected by the amount, fineness, and chemical composition (particularly, carbon content) of the ash. However, the fineness of cement, the water content of the cementitious paste, and the ambient temperature usually have a much greater effect on times of setting than addition of fly ash. Replacement of 60% of cement with high-carbon fly ash by mass resulted in 200% increase in the time of final setting of control concrete mixture. . Gebler and Klieger 6 studied the times of setting of concretes containing Class F and Class C fly ashes from 10 different sources for high content mixtures. They reported that inclusion of the fly ash increased the initial-and finalsetting times of concrete mixtures. Carette and Malhotra 7 reported the setting characteristics of concretes made with fly ashes from different sources. Calcium oxide (CaO) contents of the fly ashes varied between 1 % and 13 %. They concluded that, in general, the fly ashes increased the initial-and final-setting times of concrete. Bilodeau and Malhotra 8 reported properties of concrete incorporating high volumes of Class F fly ashes from three different sources. Cementitious materials content was 300, 370 and 430 kg/m 3 , and three W/Cm (0.39, 0.31 and 0.27) were used. They concluded that for every W/Cm, the initial-and final-setting times of high-volume fly ash concretes were noticeably increased as compared to those of the control concretes (without fly ash). This could possibly be due to the lower cement content of the high-volume fly ash concretes. Carette et al. 9 reported data on the setting time of high-volume (55 % to 60 %) Class F fly ash concretes. Eight sources of fly ashes and two sources of portland cements were used. The initial-and final-setting times varied from 4:50 to 12:51(hr: min), and 6:28 to 13:24 (hr: min), respectively, except for one mixture whose final-setting 4 time exceeded 13:24 (hr: min). Concrete mixtures showed varying setting times depending upon the source of fly ash. In general, for each fly ash source, concrete made with a low-alkali content cement having 6% C 3 A showed longer setting times than concrete made with a high-alkali content cement having 11.9% C 3 A. Malhotra and Ramezanianpour 10 have reported that inclusion of Class F fly ash retards the hydration of C 3 S at very early stages of hydration and then accelerate at later stages. C 3 A contribution from this fly ash increased with increasing its content as a replacement of cement. Thus, fly ash also became a contributor of C 3 A and other reactive components at high fly ash contents. Accelerated setting and hardening occurred due to the reactions of C 3 A present in the fly ash in addition to contributions of reactions associated with cement hydration in presence of fly ash at cement replacements of about 40% and above. Extremely high rate of setting and hardening occurred at 70% fly ash content and beyond due to the presence of relatively higher amount of C 3 A contributed by the fly ash, in addition to that contributed by cement. Hydration of aluminates was very rapid leading to formation of C 3 AH 6 , C 4 AH 19 , and C 2 AH 8 with generation of large amount of heat of hydration 13 . Eren et al. 11 reported the results of setting times of concrete incorporating up to 50 % ground-granulated blast-furnace slag (GGBS) under curing temperatures ranging from 6 to 80 o C. They concluded that: (1) increase in temperature decreased the setting times of concrete; (2) setting times of fly ash concretes were longer than those of Type I cement concretes and GGBS concretes; and (3) at temperatures greater than 20 o C, the setting times of GGBS concretes were shorter than those of Type I cement concretes. Pinto and Hover 12 studied the effects of inclusion of silica fume and superplasticizer on setting behavior of high-strength concrete mixtures. The influence of temperature was also studied by storing mortar specimens at different temperatures. Use of silica fume caused reduction in the initial time of setting. However, an opposite trend was noted when superplasticizer was used. Statistical analysis revealed significant interaction between the two (silica fume and superplasticizer) when the initial time of setting was taken as a response. The effect of temperature was significant on both initial and final times of setting. Samadi et al. 13 studied the influence of phosphogypsum (PG) on the times of setting and soundness of cement pastes. In this study, cement paste mixtures were made using ordinary portland cement (OPC) and pozzolanic portland cement (PPC) at a constant water to cement ratio of 0.6 with PG content varying between 0 and 100 percent. In general both initial and final times of setting increased with increasing PG content. The initial time of setting ranged between 100 to 560 minutes and 120 to 710 minutes for pastes containing OPC (ordinary portland cement) and PPC (pozzolana Portland cement), respectively. The corresponding final time of setting ranged between 250 to 1440 minutes and between 270 to 1440 minutes. The paste expansion also increased with increasing PG content. Brooks 14 investigated the effects of silica fume (SF), metakaolin (MK), fly ash (FA), and ground-granulated blast-furnace slag (GGBS) on the setting times of high-strength concrete using the penetration resistance method (ASTM C 403). He also studied the effects of shrinkage-reducing admixture (SRA) on the setting times of normal and high-strength concretes. Based on the test results, he concluded that: (1) the setting times of the high-strength concrete were generally retarded when the mineral admixtures replaced part of the cement. While the SRA was found to have negligible effect on the setting times of normal strength concrete, it exhibited a rather significant retarding effect when used in combination with a superplasticizer; and (2) the inclusion of GGBS at replacement levels of 40% and greater resulted in significant retardation in setting times. In general, as replacement levels of the mineral admixtures were increased, there was greater 5 retardation in setting times. However, for the concrete containing MK, setting time were only observed up to a replacement level of 10%. Ahmadi 15 studied the initial and final setting times of concrete in hot weather. The effect of field temperature, relative humidity, wind velocity, and admixture on the setting times of concrete were observed. He proposed two equations: (1) the first equation was for determining the initial setting time of concrete with a correlation factor of 0.93 and standard deviation of 5.28%. This equation showed that as the field temperature and field air velocity increased, the initial setting time decreased, and as the field humidity increased, the initial setting time increased; and (2) the second equation for determining the final setting time of concrete with a correlation factor of 0.9 and standard deviation of 5.8% showed similar effects as of initial setting time of concrete. Targan et al. Takemoto and Uchikawa 18 and Uchiwaka and Uchida 19 described a model for hydration reaction process of cement in the presence of pozzolans. The reactions of C 3 A and Class C fly ash resulted in formation of enttringite, monosulphoaluminate hydrate, calcium aluminate hydrates, and calcium silicate hydrate. They reported that presence of pozzolan accelerated hydration of C 3 A due to adsorbing Ca 2+ from the liquid phases and providing precipitation sites for the hydration products. Tay 20 performed a study to investigate properties of mortar and concrete as influenced by inclusion of pulverized sludge ash. The test data exhibited improved workability and increase in initial and final times of setting with increasing sludge ash content. Sawan and Qasrawi 21 concluded that the use of natural pozzolan cause decrease in workability and increase in the times of setting of mortar under normal condition. However, an opposite trend was obtained in hot weather conditions. Uchikawa et al. 22 evaluated the effects of chemical admixtures on the hydration characteristics of cement. They reported that an admixture having a functional group that produces complex salt with decrease in Ca 2+ concentration can cause loss in fluidity and delay in the times of setting of cement pastes. Chen and Older 23 investigated the effect of cement with varying in clinker composition with varying amounts and forms of calcium sulfate on the times of setting of mortars. 6 They indicated that the setting of cement having normal composition was mainly related to hydration of C 3 S content. The formation of enttringite occurred at very high C 3 A contents. Matusinovic and Vrbos 24 and Matusinovic and Curlin 25 reported that setting characteristics of high-alumina cement (HAC) were substantially influenced by inclusion of alkali metal salts. The lithium cation had a greater effect on the times of setting than alkali cations did. The results showed that lithium salt or alkali metal salts could be used as a set accelerator for HAC. Perret et al. 26 investigated the compatibility of six different microfine cements and four different HRWRAs; and the influence of materials and mixture proportions on rheological characteristics and final-setting time of microfine cement-based grouts. Three portland cements and three slag cements, associated with various naphthalene-based and melamine-based HRWRA were investigated. They concluded that: (1) not every microfinecement can be used with every HRWRA; (2) some HRWRAs gave better fluidity, and some gave too long (24 hours) or too short (4 hours) final setting times; and (3) the chemical composition and fineness of cements, as well as the type and chemical characteristics of admixtures lead to different grout properties. INFLUENCE OF FLY ASH ON SETTING TIMES OF NON-AIR-AIR ENTRAINED CONCRETES (Series 1) Experimental Details An experimental program was designed to evaluate the effects of Class C fly ash content and its source on setting times of non air-entrained concrete. Four different Class C fly ashes, obtained from different electric power plants in Wisconsin, were used. The fly ashes corresponding to these power plants are designated as P-4, DPC, Columbia, and Weston. Chemical and physical properties of these fly ashes were determined. Three of the fly ashes (DPC, Columbia, and Weston) exceeded ASTM C 618 requirement for MgO. However, they met all other ASTM C 618 Class C fly ash requirement. Natural sand with 6 mm maximum size was used as a fine aggregate, and a 19 mm maximum size gravel was used as a coarse aggregate throughout this investigation. These aggregates met the ASTM C 33 requirements. Type I cement which met the requirements of ASTM C 150 was used. Concrete mixture proportions were proportioned with all the four Class C fly ashes. Results and Discussion Initial and final setting times of concrete incorporating various sources of Class C fly ash are shown in At high replacements of cement with fly ash (70% or above), the setting of concrete was accelerated. This might be attributed to the fact that at higher cement replacements with fly ash, the concentrations of total C 3 A and gypsum present in the mixture becomes low. This resulted in reduced setting times of the mixtures containing low cement and high fly ash contents. As a result, rapid setting of the concrete mixtures occurred. Therefore, under such conditions, it is desirable to use a set retarding admixtures to allow enough time for proper mixing and placing of concrete. SETTING TIMES OF NON-AIR-ENTRAINED AND AIR-ENTRAINED FLY ASH CONCRETE (Series 2) Experimental Details One source (Pleasant Prairie Power Plant, P-4) of Class C fly ash was used. Three nominal compressive strength levels (21, 28, and 35 MPa) of non-air-entrained and air-entrained concrete mixture proportions, by varying the water-to-cementitious materials ratio (0.45, 0.55, and 0.65) were developed. Cement replacement percentage was 35, 45, and 55%. Replacement was on the basis of Results and Discussion Setting time of non-air-entrained concrete mixtures are given in Setting time data for air-entrained concrete are given in 9 SETTING TIMES OF CEMENT PASTE AS INFLUENCED BY FLY ASH AND CHEMICAL ADMIXTURES Four series of tests were performed: (1) to evaluate only the effects of fly ash addition on the setting times of cement paste; (2) to evaluate the effects of fly ash and two levels of air content on the setting times of cement paste; and (3) to evaluate the influence of fly ash and normal dosage of two types of chemical admixtures (WRA and HRWRA) on the setting times of cement paste; (4) to evaluate the combined effects high dosage of fly ash and three dosage rates of two types of chemical admixtures (retarders and gypsum) on the setting times of cement paste. Experimental Details A portland cement conforming to the requirements of ASTM C 150 was used. An ASTM Class C fly ash, obtained from one source, Pleasant Prairie (P-4), was used. The fly ash met all ASTM C 618 requirements for Class C fly ash. Five chemical admixtures: an air entraining admixture (ASTM C 260), a water-reducer (ASTM C 494, Type A), a retarder (ASTM C 494, Type B), and a HRWRA (ASTM C 494, Type F) were obtained from a local ready-mixed concrete company, the Tews Company, Milwaukee, WI. A total of 82 cement paste mixtures were prepared for evaluating their setting and hardening characteristics. Each mixture was composed of cement, fly ash, and water. Fly ash was used as a replacement of cement ranging from 0 to 100 percent by mass. A ratio of fly ash addition to cement replaced was kept at 1.25. All ingredients were mixed in a laboratory mixer in accordance with ASTM C 305. Normal consistency of pastes containing cement/fly ash was determined in accordance with ASTM C 187. Air content of each paste mixture was determined according to ASTM C 185. Test specimens for each mixture were prepared for measuring the initial and final times of setting using the Vicat apparatus (ASTM C 191). Results and Discussion Effect of fly ash on setting times of pastes without admixtures The initial and final times of setting were essentially the same due to the inclusion of fly ash at 10% compared to the 0% fly ash mixture Effect of air entrainment and content on setting times of paste Effects of air entrainment and content at two dosage levels on setting times of fly ash mixtures are given in Effect of fly ash with normal dosages of chemical admixtures on setting times of paste In this series of tests, fly ash content varied from 0 to 100% with normal dosages of individual chemical admixtures (five different types). Fly ash with a normal dosage of water-reducer Effects of normal dosage of water-reducer on setting characteristics of fly ash mixtures are given in Fly ash with a normal dosage of superplasticizer Effects of normal dosage of superplasticizer on setting characteristics of fly ash mixtures are given in Fly ash with a normal dosage of retarder Effects of normal dosage of retatder on setting characteristics of fly ash mixtures are given in Fly ash with a normal dosage of gypsum Effects of normal dosage of gypsum on setting characteristics of fly ash mixtures are given in Effect of High Fly Ash Contents with High Dosages of Chemical Admixtures on Setting Times of Paste At high fly ash content (above 40%), very rapid rate of setting of mixtures occurred. Use of normal dosage of retarder and gypsum did not cause enough delay to compensate for the rapid rate of setting resulting from the presence of the high-levels of fly ash. Therefore, high dosages of these admixtures were used at fly ash contents of 70, 85, and 100%. The retarder and gypsum were used at their respective double and triple dosages. Fly ash with retarder Effects of high dosage of retatder on setting characteristics of fly ash mixtures are given in Fly ash with gypsum Effects of high dosage of gypsum on setting characteristics of fly ash mixtures are given in CONCLUSIONS Following are the general conclusions from this study: 1. Both the initial-and final-setting times of the concretes were significantly influenced by both the source and amount of fly ash. Both the initial-and final-setting times were relatively unaffected at 10% cement replacement. Although inclusion of fly ash caused large retardation in the setting times, for up to around 60% cement replacement, the rate of strength development were appropriate for most construction applications. Therefore, setting time should not be taken as a sole parameter for selecting a fly ash for a particular 12 application. However, in order to improve construction productivity and efficient construction planning, fly ash content should be reduced and/or chemical admixtures should be added to control the setting times. 2. For non-air-entrained and air-entrained fly ash concretes having compressive strengths of 21, 28, and 35 MPa, the initial-and final-setting time were not significantly different when fly ash replacement for cement was increased up to 55 percent. 3. The water demand of cement paste mixtures decreased with increasing fly ash content. Further decrease in the water demand occurred when were used. Inclusion of fly ash in cement paste mixtures caused small delay in the initial-and final-setting times up to 20% fly ash content depending upon type of chemical admixtures used. Beyond this limit, the setting was accelerated. Fast setting occurred at fly ash content of about 40% and beyond depending upon type of chemical admixtures used. 4. Use of gypsum and water-reducer of normal dosages caused acceleration in the setting of cement paste. However, no appreciable effect of normal dosage of superplasticizer on the setting was observed. ACKNOWLEDGEMENT