INTRODUCTION

Before the 1980s, China was distinguished only in the quantity of dams; now the dam height has increased dramatically. The well-known Three Gorges Project is the largest dam in the world and a milestone of tremendous improvement in dam construction technology in China. Longtan Dam

RCC is often used in the construction of dams and pavements that differ from conventional concrete (CC) not only in regard to the placement method, but also in its consistency and mixture.3 It is a zero slump concrete that is very difficult to compact by the normal methods for workable concrete. To reduce the risk of thermal cracking, these RCC materials are usually characterized by the addition of a low cement dosage and a large volume of mineral by-products, such as fly ash and slag. Although the qualities of all materials and temperature differences in the concrete near the foundation are strictly controlled, superficial cracks were often found in concrete dams, some of which may further develop into inherent and penetrating cracks. The control of cracking in concrete dams has become a problem that attracts the concerns of experts in the world.4

Fly ash provides an economical mineral resource and may improve some properties of concrete; therefore, it has been used widely in the construction of dams (for example, Willow Creek Dam in America, and Arabie Dam and Zaaihoek Dam in South Africa).5 When adding fly ash to RCC, the adiabatic temperature rise and strain may be reduced, but the strength and durability of RCC may be reduced at the same time.6

Due to the aforementioned reasons, the control of the content and quality of the fly ash in the dam concrete is very important. In the Three Gorges Project concrete dam, with approximately 30% fly ash, superficial cracks larger than 2 m (6.6 ft) long have been found. To control the cracks in the RCC Longtan Dam, 50 to 70% fly ash will be used. Due to the limited experience of RCC dams, a number of tests need to be made in the laboratory and in the field for the use of fly ash in dam construction concrete.

RESEARCH SIGNIFICANCE

Cracks, which may reduce the service life and durability of concrete dams, are attracting more attention due to the increasing construction of mass concrete dams. An effective way to reduce cracks is to add fly ash to the concrete. This research studied the effects of fly ash on strength, shrinkage strain, and deformation, and the results provide a better understanding of the cracking mechanism and offer an effective approach to control cracking in concrete dams.

EXPERIMENTAL PROCEDURE

Materials

The mineral composition and strength of the 52.5 MPa (7600 psi) grade, medium heat portland cement (Ce) are listed in Table 1.

Grade I fly ash was used (China National Standard GB1596-91; that is, moisture content =1%, loss on ignition =4%, SO^sub 3^ =3%, water requirement ratio =95%, and residue sieve 45 µm (1.77 ? 10^sup -3^ in.) =12%), and the granite was man-made aggregate with the size grading of: large size stone (10 to 20 mm [0.39 to 0.78 in.]) and small size stone (2.5 to 10 mm [0.098 to 0.39 in.]) equal to 6:4. The fine aggregate (fineness modulus of 2.5) had a well-graded size distribution and a well-rounded shape.

Ex is a novel MgO-bearing expansive agent, which is prepared by calcining two mineral materials at different temperatures. The main chemical compositions of Ex are MgO and CaO. Calcined MgO in dam concrete may produce delayed expansive stress to compensate temperature shrinkage stress, but may also bring an unsoundness problem to concrete. A higher expansive stress caused by an overdose of MgO may change the microstructure of dam concrete, and then decrease the dam durability. CaO may offset the early shrinkage and provide Ca(OH)^sub 2^ to react with excessive fly ash in dam concrete.

The high-range water-reducing admixture was a retarding naphthalene water-reducing agent from the Zhejiang Province. The chemical compositions of these materials and the physical property of the fly ash are shown in Table 2.

Specimens

Concrete cube specimens (100 mm [3.9 in.]) were used in the strength test, and the mixture proportions used for these specimens are shown in Table 3.

The specimens were mixed and molded by vibration with the following conditions: a frequency of 50 Hz and a surcharge mass of 20 kg (44.0 lb). After being placed in a fog room (20 ?C, 90% relative humidity [RH]) for 48 hours, the specimens were demolded and cured in water at 20 ? 2 ?C for 28 days, and then the strength and mass of the specimens were measured.

The concrete specimens for expansion testing were 75 x 75 x 280 mm (2.93 x 2.93 x 10.92 in.). The specimens were demolded and then cured in a fog room (20 ?C, 90% RH) for 48 hours, and the length of the specimens (L^sub 0^) was measured. After placing them in a fog room for 7, 14, 28, 60, 90, 120, 150, and 180 days, the length of specimens (L^sub 1^) were measured. The expansions of RCC may be determined as follows

Expansion (%) = (L^sub 1^ - L^sub 0^) ? 100%/L^sub 0^

EXPERIMENTAL RESULTS AND DISCUSSION

Effects of fly ash content on compressive strength of RCC

The effects of fly ash on the compressive strengths of RCC with 8% expansive agent (by mass of binder) are shown in Fig. 1.

When the content of fly ash varies from 0 to 30 to 50%, the compressive strength of the RCC in the curing time of 7 days is reduced. There is a slight improvement for the compressive strength of RCC at a curing age of 28 days, however, when compared with that of 7 days. At 90 days, the compressive strength of RCC with 30% fly ash is almost equal to that of RCC without fly ash, and with 50% are higher than that of RCC with 30% fly ash and without fly ash. The results showed that appropriate content and good quality fly ash replacing cement does not reduce, but may improve the compressive strength of RCC slightly at later ages.

Effects of fly ash on shrinkage strain

The amounts of fly ash used in the assessment were 0 and 50% (by weight). The mixture proportions of RCC are shown in Table 3. These mixtures were made without the expansive agent. The specimens were cured under adiabatic conditions.

The effects of fly ash contents on shrinkage strain are presented in Fig. 2. When 50% fly ash was used, the shrinkage strains of RCC specimens are less than that of the specimens without fly ash. The change ratio of the shrinkage strains of RCC specimens without fly ash is higher than the specimens with 50% fly ash. Higher change ratios tend to result in superficial cracks in the first 60 days. After 150 days, the shrinkage strains of RCC specimens with 50% fly ash is approximately 33% less than that of specimens without fly ash. There is a logarithmic relationship between the shrinkage strains of RCC specimens with or without fly ash for the same curing time.

Based on these data, the relationship between the shrinkage strain of RCC specimens without fly ash and curing time is as follows

y^sub 0^ = -7.1853ln(x^sub 0^) + 6.3562 (1)

where y^sub 0^ is the shrinkage strains of RCC, in µm/m; and x^sub 0^ is the curing time in days. The correlation coefficient R^sub 0^^sup 2^ was 0.9596.

Similarly, for the RCC specimens with 50% fly ash and curing time, the relationship is as follows

y^sub 50^ = -6.3257ln(x^sub 50^) + 9.7312 (2)

where y^sub 50^ is the shrinkage strain of RCC, in µm/m; and x^sub 50^ is the curing time in days. The correlation coefficient in this case was R^sub 50^^sup 2^ equal to 0.9531.

Effects of fly ash on expansion of RCC

The expansion properties of RCC with 0, 30, or 50% fly ash and 8% MgO-type expansion agents are studied, respectively, and the results are shown in Fig. 3.

The early shrinkage at 28 days and later expansion at 90 days of RCC with expansive agents and without fly ash have been shown in Fig. 3, respectively. When 30 and 50% cement was replaced by fly ash, the shrinkage at early ages was reduced, and the expansion at later ages was controlled. The deformation of RCC with 50% fly ash is lower than that with 30% fly ash. This result showed that the fly ash might decrease the deformation of RCC with 8% expansive agent. Adding 50% fly ash to concrete might be appropriate for RCC dams to control shrinkage cracking under a given environmental condition.

Effects of fly ash on strain and stress

The amounts of fly ash used were 0 and 50% (by weight). The water-binder ratio (w/b) was 0.30. The pastes were cured in a fog room at 20 ? 2 ?C for 180 days, and then the stress and strain of the pastes were measured.

Figure 4 shows the stress-strain curves of pastes with 50% fly ash and without fly ash. The stress-strain curve of the paste with 50% fly ash is different from that without fly ash. The stress-strain curve of paste without fly ash shows a linear relationship before or after reaching the highest stress value, and the increase or decrease process is sharp. Cracks emerge when strain is approximately 0.08%. When 50% cement is replaced by fly ash, the stress-strain curve of paste shows a nonlinear relationship before or after reaching the highest stress value, and the curve increases and decreases slowly. Cracks emerge when strain is approximately 0.15%, with an increase of 88%. The results indicate that the fly ash might decrease the strain and improve the deformation resistance of RCC dam concrete.

Observation of fly ash in RCC

Under the optical microscope, irregularly shaped compaction voids in RCC (Fig. 5) and reacted fly ash were observed.

From Fig. 5, fly ash filled the interface between aggregate and mortar, and decreased the radius and amount of voids in RCC. The radius of voids in RCC with 50% fly ash was almost half of that of RCC without fly ash. The lower the radius and amount of voids in RCC with 50% fly ash, the higher the density of the RCC. As the irregularly shaped compaction voids in RCC are not continuous and close to each other, the permeability of RCC with fly ash may be lower than that of conventional concrete. Due to these reasons, the compressive strength and durability of RCC with fly ash may be improved.

Under the scanning electron microscope, the unreacted fly ash in RCC (Fig. 6) with 50% fly ash can be observed. The unreacted fly ash in RCC may be separate from the paste and may reduce the adherence between mortar and aggregate, thus decreasing the mechanical properties and durability of RCC with fly ash. With the optimum content and quality of fly ash, water reducing agent, and aggregate under the large vibratory rollers that can compact RCC, the number of voids may be reduced to avoid forming continuous void networks, and the durability of RCC dam will be improved.

CONCLUSIONS

1. When the content of fly ash increased from 0 to 30% and to 50%, the compressive strengths of RCC are decreased at early ages of 7 and 28 days, but with 50% fly ash strengths are higher than with 30% fly ash or without fly ash at 90 days;

2. When 50% fly ash was used, the shrinkage strains of RCC specimens are less than that without fly ash. After 150 days, the shrinkage strains of RCC specimens with 50% fly ash is approximately 33% less than that of specimens without fly ash. There is a logarithmic relationship between the shrinkage strains of RCC specimens with or without fly ash and curing time;

3. Fly ash may decrease the deformation of RCC with 8% expansive agent. Adding 50% fly ash to concrete might be a good content for RCC dam in controlling the crack under a given environmental condition;

4. With 50% fly ash, the stress-strain curve of paste increases or decreases slowly before or after reaching the highest stress value and the point when the cracks emerge is approximately 0.15%, higher than 0.08% without fly ash, with an increase of 88%. Fly ash can decrease the strain and improve the deformation resistance of RCC dam concrete; and

5. Fly ash may fill the interface between the aggregate and the mortar and improve the pore structure and density of RCC dam. Unreacted fly ash in RCC may decrease the mechanical properties and durability of RCC. The optimum content and quality of fly ash are two important factors in construction of mass concrete dams.

ACKNOWLEDGMENTS

The authors would like to thank the Natural Sciences Foundation Council of China (2001CB61070503, 50278031), the Bureau of Water Resources & Hydropower Research of China (SPKJ006-13-01-01), and Jiangsu planned projects for postdoctoral research for their financial support of this project.

CONVERSION FACTORS

1 cm = 0.39 in.

1 MPa = 45 psi

1 kg/m^sup 3^ = 1.685 lb/yd^sup 3^

1kg = 2.2 lb