Strength and durability of concretes containing 50% Portland cement ...

2 downloads 0 Views 247KB Size Report
Strength and durability of concretes containing 50% Portland cement replacement by fly ash and other materials. B. W. LANGAN, R. C. JOSHI, AND M. A. WARD.
Strength and durability of concretes containing 50% Portland cement replacement by fly ash and other materials B. W. LANGAN,R. C. JOSHI,AND M. A. WARD Department of Civil Engineering, UniversiQ of Calgary, Calgary, Alta., Canada T2N IN4

Received December 7, 1988

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

Revised manuscript accepted July 27, 1989 Results are presented from an investigation on the compressive strength and durability of concretes containing substitute materials at a 50% replacement level (by mass) of Portland cement. Seven fly ashes (sub-bituminous, bituminous, and lignitic), together with limestone and an inert material (silica flour), were used as replacement materials. Durability studies included freeze-thaw testing (ASTM C666A), scaling resistance (ASTM C672), and abrasion resistance (ASTM C944). The air void system was assessed using the modified point count method of ASTM C457. The results indicate that although concretes with a 50% replacement level of cementitious material did not perform as well as the control concretes with no replacement, such concretes were able to meet minimum durability requirements. As anticipated, air-entrainment is the overriding factor that allows concrete to meet freeze-thaw durability requirements. In the context of this study, compressive strength does not appear to be a significant factor in freeze-thaw durability. Results indicated that concretes with compressive strengths of less than 10 MPa will still pass the freeze-thaw test, provided an adequate air void system is in place. Abrasion resistance tends to increase with compressive strength but not in all the cases. Key words: concrete, fly ash, compressive strength, durability, mineral admixtures. Le prCsent article rend compte des rCsultats d'une analyse de la rCsistance ?i la compression et de la durabilitC de divers Cchantillons de ciment Portland auxquels on a ajoutC des matCriaux de remplacement dans une proportion de 50% (par unit6 de masse). Ces matCriaux de remplacement comprenaient sept types differents de cendres volantes (subbitumineuses, bitumineuses et ligniteuses) ainsi que du calcaire et un matCriau inerte, la poudre de silice. Dans le cadre de l'analyse de durabilitC ont CtC effectuCs des essais de gel-dCgel (ASTM C666A), de rCsistance ?i 1'Ccaillage (ASTM C672) et de rCsistance ?i l'abrasion (ASTM C944). En outre, l'occlusion d'air des Cchantillons a CtC analysCe au moyen de la mCthode ASTM C457 (modified point count method). Les rCsultats dCmontrent que mCme si les bCtons contenant une proportion de 50% de mortiers sont inferieurs aux Cchantillons de contr6le sans remplacement, ils satisfont toutefois aux exigences de durabilitC minimales. Comrne prCvu, l'entralnement d'air constitue le facteur dCterminant qui permet au bCton de satisfaire aux exigences de durabilit6 en condition de gel-dCgel. Dans le contexte de la prCsente Ctude, la rCsistance ?i la compression n'apparait pas comme un facteur important en ce qui ?i trait ?i la rCsistance aux gels et dCgels successifs. Les rCsultats indiquent en effet que les bCtons offrant une rCsistance ?i la compression de moins de 10 MPa rCussissent malgrC tout l'essai de gel-dCgel, ?i condition que leur occlusion d'air soit adCquate. La resistance ?i l'abrasion tend pour sa part ?i augmenter en proportion de la resistance ?i la compression, mais cela ne se produit pas dans tous les cas. Mots clis : biton, cendre volante, resistance ?i la compression, durabilitC, adjuvants minCraux. [Traduit par la revue] Can. J . Civ. Eng. 17, 19-27 (1990)

Introduction Increasing energy costs and depletion of required highquality natural materials for Portland cement and the resultant variability of its properties with respect to durability of concrete has led to the increased use of low-cost mineral admixtures in cement and concrete. During the last decade, in particular, the inclusion of admixtures in both cement and concrete products has increased dramatically. Two commonly used admixtures are pulverized limestone and fly ash. The addition of limestone during the grinding of Portland cement clinker has become commonplace in many industrialized countries. The properties of limestone vary mainly with mineralogy. Generally, high-calcitic limestone is preferred to dolomitic limestone to avoid increasing magnesium compounds in the manufactured Portland cement. In Canada, 5% limestone, by weight of clinker, is currently permitted (Cana-

NOTE:Written discussion of this paper is welcomed and will be received by the Editor until June 30, 1990 (address inside front cover). Printed in Canada / lniprirnd au Canada

dian Standards Association 1983), while in France up to 25% is allowed in some cases (Guyot and Rancl). The properties of fly ashes, on the other hand, are affected mainly by the parent coal mineralogy. Additionally, the degree of pulverization and firing method, the burn temperature as well as additives used in the collection and storage of such ashes have a significant effect (Joshi and Rosauer 1973) on the characteristics of the end product. Concerns have been raised regarding the variability of ash properties, slow early strength gain, and increased demand for chemical admixtures to produce adequate air contents. Introduction of sub-bituminous and lignitic ashes in the last two decades has countered some of these concerns, but at the same time raised other issues regarding the long-term performance of concretes containing such materials. R., and RANC,D. 1983. Controlling the properties of 'GUYOT, concrete through the choice and quality of cements with limestone additions. Presented at the First International Conference on the Use of Fly Ash, Silica Fume, Slag and Other Mineral By-Products, July 31 - August 5, Montebello, Que.

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

20

CAN. J. CIV. ENG. VOL. 17, 1990

The use of fly ash in concrete and limestone in cement has become the rule rather than the exception. Generally, no more than 20-30% of cement is replaced by fly ash in structural concrete. Here, strength development has always been one of the main user concerns. This study of the durability, the other major issue, of concretes containing 50% fly ash, limestone, or silica flour was initiated as part of an ongoing investigation on the compressive strength development of such concretes. Also investigated was the effect of the presence of these replacement materials on the efficiency of air-entraining chemical admixtures in these concretes. The objectives of this study included the evaluation of abrasion resistance, freeze -thaw durability, scaling, and compressive strength of concretes at 50% (by mass) cement replacement with substitute materials. The ultimate goal was to examine the feasibility of replacing up to 50% cement in concrete.

Materials and mixture proportions Type 10 normal Portland cement (Canadian Standards Association 1983) was used throughout this investigation. The chemical analyses of the cement and replacement materials are included in Table 1. Of the seven fly ashes tested, six were from Canada, including three sub-bituminous, two bituminous, and one lignitic ash. The one ash from the United States was also lignitic. Additionally, two other mineral admixtures were investigated, limestone and silica flour. The latter was considered to be an inert replacement material in this study. The limestone is primarily pure calcium carbonate, while the inert material is essentially pure silica. A neutralized vinsol resin air-entraining agent was used in this investigation. The superplasticizer was a modified napthalene sulfonate. The mixture proportions (per m3) used were as follow: cementitious material, 300 kg; coarse aggregate (14 mm max), 1200 kg; fine aggregate, 800 kg; water, 141 kg; and water-cementitious ratio (constant), 0.47. The coarse aggregate used was composed of equal percentages of limestone and quartzite with minor amounts of shale and siltstone present. The sand was a manufactured sand from the same source. Some segregation was noted during preliminary testing and it was deemed necessary to split the fine aggregate into three fractions. These fractions were (1) coarse - retained on 1.25 mm sieve; (2) medium - passing 1.25 mm sieve but retained on the 315 micron sieve; and (3) fine - passing the 315 micron sieve. Prior to mixing, the sand was recombined to meet the requirements of CSA Standard CSA-A23.1-M77 (Canadian Standards Association 1977a). All mixing water came from the City of Calgary water supply. Mixing and casting procedures To maintain accurate control of the water content of these mixes, the coarse aggregate was soaked in water for 24 h and then allowed to drain for 1 h prior to mixing. A blotting and weighing technique was employed to permit an accurate calculation of the mixture water requirements. An Eirich (0.085 m3) pan mixer was used throughout this series. Dry sand was blended with a predetermined amount of water to bring it to the saturated surface dry condition. The coarse aggregate and the cementitious material were then placed in the mixer and blended thoroughly. Next, 2 kg of the

mixture water containing the required amount of air-entraining agent was added, after which mixing continued for a further period of 30 s. The superplasticizer, when required, was added to the mixture at this point. Lastly, the balance of the mixing water was added and mixing continued for 1 min, followed by a minute of rest and a final minute of mixing. Waterto-cement ratios of the labortory mixtures were not corrected for the water in the superplasticizers. Admixture dosages are given in Table 2. Air content of the plastic concrete was determined by the pressure method (ASTM C231 and CSA A23.2-4C). The slump was measured according to ASTM C143 and CSA A23.2-5C and the Vebe time was determined according to British Standard BS 1881 Part 2 (British Standards Institution 1970). Properties of the fresh concrete are included in Table 2. With these tests completed, the concrete was cast into the required moulds. With the low-workability concrete mixes, both vibration and rodding were required to obtain full compaction. On completion of casting, the specimens were transferred to the fog room (20 f 2°C and 95 f 3% relative humidity), covered with burlap to keep the top of the sample moist, and left for 1 or 2 days as required prior to demoulding. It was found necessary to cure the mixes made with Eastern Canadian ashes (Lingan and Nanticoke) for 2 days in the moulds, as the strength development of these concretes at 1 day was insufficient to prevent damage during the demoulding procedure. After demoulding, the specimens were returned to the fog room for curing until testing.

Compressive strength The compressive strength of all concretes was determined at the ages of 7, 14, 28, and 56 days after casting. Additional tests at 112 days were carried out on selected mixes. Three cylinders, 150 mm in diameter x 300 mm in length, were tested at each test age following ASTM C39-84 and CSA A23.2-9C. The averages of the test results are presented in Table 3. Abrasion tests Abrasion testing, using the rotating cutter method (ASTM C944-80), was carried out on samples from nine of the mixes in the test series. These tests were conducted on the moulded end of the cylinders that were subsequently capped and tested for compressive strength. The test ages were 7, 14, 28, and 56 days. Average results of these tests are given in Table 4. Scaling The scaling resistance was measured using ASTM C672-84. One slab (75 x 250 x 250 mm) was cast for each of the 9 mixtures tested. External vibration was used for consolidation. The top surface was leveled with a strike off board, and troweled with a minimum of effort necessary to close the surface. The final surface finish was obtained by brushing. The specimens were cured according to the specification using moist curing until the age of 14 days after demoulding, followed by 14 days at 23 f 1.7"C and 50 f 5% relative humidity. Just prior to the start of testing, a 25 mm high plastic dyke enclosing a surface area in excess of the minimum required (0.046 m2) was fixed to the surface of the specimen using a silicone sealant. A 4 % aqueous solution (by weight) of calcium chloride was used as the deicing agent. After every five cycles of freezing and thawing, the surface was flushed

LANGAN ET AL.

TABLE1. Chemical composition of cement and replacement materials PAC *

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

PC Lime

SiO,

A1203

Fe203

(%)

(%)

(%)

CaO (total%)

MgO

SO,

Lo1

NaO

(%)

(%)

(%)

(%)

K20 (%)

Portland cement (Type 10) Forestburg Sundance Wabamum Laramie Nanticoke Boundary Dam Lingan Inert4 Limestone3 *PAC = Pozzolanic activity index, PC = Portland cement. ?Available alkalies as N+O equivalent. $All other components less than 0.1 %. §C02 content = 43.2%.

with water and a rating of 0 (no scaling) to 5 (severe scaling) was given to the resulting surface. The test results are given in Table 4. Freeze - thaw resistance The freeze-thaw resistance of concrete was determined using ASTM C666 Procedure A, rapid freezing and thawing in water. Only during the initial stages of freeze-thaw testing did all mixtures include specimens cast for this test. Based on the initial freeze-thaw results, it was confirmed that concretes containing no chemical admixture (mixes 9, 11, 13, 15, and 37), either air-entraining agent or superplasticizer, would fail this test. These mixes all failed the test at 50 cycles or less. This was also found to be true for mixes containing only the superplasticizer. Thus, due to a limitation on available space within the freeze - thaw baths, specimens were not cast for the remaining mixtures in these categories, i.e., no air entrainment present (all remaining odd-numbered mixes). Testing for all concretes began at the age of 14 days (i.e., 13 days of moist curing). The average test data for three specimens are given in Table 5. Results and discussions Properties of fresh concrete A majority of the data on the properties of fresh concrete (Table 2) have been presented elsewhere (Joshi et al. 1987); thus, only the main observations are included here. Control mixtures (Nos. 2 and 4) were repeated with the additional mixtures so as to provide a link to the previous work. The results from the repeated mixes compared very favorably with the initial results. In general, the presence of only the superplasticizer resulted in a decreased air content when compared with the companion mixture with no admixture. However, when the superplasticizer was present, a reduced dosage of air-entraining agent was required to achieve the desired air content. The control mixture (Mix 1) was designed without admixtures to have a low slump (slump 10 mm, Vebe time 7.9 s), since it was anticipated that the presence of chemical and mineral admixtures would tend to increase workability. Several of the fly ash mixtures had a zero slump, but exhibited a

lower Vebe time, indicating a more workable mix. In all cases except Mix 33, the Vebe time was reduced when mineral admixtures were added to the mixture. In Mix 33, a higher Vebe time was obtained compared to the control. The inert replacement material (silica flour) was crushed and it is vossible that the particles'were more' angular, which would tend to reduce workability for a given water content. However, this trend was not observed with the limestone (Mix 37) which was also crushed. It is possible the lower hardness and rhombic structure of the limestone did not have the same effect on workability as the silica flour. The presence of chemical admixtures produced significant increases in workability in all cases. Small increases in workability were noted when the airentraining agent alone was added; larger increases when the superplasticizer was added. In general, the highest workability was achieved when both the air-entraining agent and the superplasticizer were added together. The Vebe time is a more sensitive indicator of workability for low-workability mixtures. Properties of hardened concrete Compressive strength The results for the non air-entrained mixtures are shown in Fig. 1. The strength of the concrete containing the high-lime (Laramie) ash exceeded the strength of the control mixture at all ages and was still gaining strength at 56 days, while the rate of gain of compressive strength of the control mixture was much less. The concretes made with Forestberg, Sundance, Wabamum, and Boundary Dam ashes have essentially the same compressive strength as the control at 56 days. At the age of 112 days the control mixture still has essentially the same strength as the concrete made with the Boundary Dam ash. The concretes containing Lingan and Nanticoke ashes show lower initial strength, but exhibit constant strength development throughout their respective testing periods. The compressive strength of the mixture containing silica flour and limestone, as expected, is the lowest with no change in the rate of strength gain after 56 days. As expected, the inclusion of air entrainment (Fig. 2) resulted in lower compressive strengths in all cases. The largest decrease occurred with the Laramie ash. The reduction in compressive strength of concretes made with the same fly ash

CAN. 1. CIV. ENG. VOL. 17, 1990

TABLE 2. Admixture dosages and properties of fresh concrete

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

Mix No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Mineral admixture type

Chemical admixture (mL/kg of cement)

Air Air-entraining Super- Slump Vebe time Density content agent plasticizer (mm) (s) (kg/m3) (%)

A

-

Forestburg Forestburg Forestburg Forestburg Sundance Sundance Sundance Sundance Wabamum Wabamum Wabamum Wabamum Laramie Laramie Laramie Laramie Nanticoke Nanticoke Nanticoke Nanticoke Boundary Dam Boundary Dam Boundary Dam Boundary Dam Lingan Lingan Lingan Lingan Inert Inert Inert Inert Limestone Limestone Limestone Limestone

is essentially constant over all ages tested. The rate of gain of strength is essentially constant from 28 to 112 days for all ashes. The presence of only the superplasticizer (Fig. 3) produced varying results when compared with the non air-entrained mixture. Significant decreases in 56-day compressive strength were observed for mixtures containing Laramie and Wabamum ashes, while the control mixture and the Forestberg ash mixture showed significant increases. The strength of the mixture containing the Wabamum ash and superplasticizer was the same as that of the non air-entrained mix at 112 days. As with the air-entrained mixtures, the rate of gain of strength for the ashes tested to 112 days is essentially constant between 28 and 112 days. The limestone and silica flour mixtures showed a nominal 10% decrease, thus another factor must be contributing to the major strength reduction exhibited by the concretes containing the Laramie and Wabamum ashes. It should be recalled at this point that the water-cementitious ratio was

held constant, thus the ability of the superplasticizer to lower the water requirement (lower the water-cementitious ratio) and thus increase strength did not materialize. Inclusion of the superplasticizer with the air-entraining agent (Fig. 4) in most cases produced little change in compressive strength compared to mixes with only air entrainment present. However, significant increases were noted in the strength of concretes containing the Sundance and Wabamum ashes, and also when these concretes contained the substitute materials. The reason for this is not clear; there does not appear to be any relation with the chemical composition of the cement and (or) the replacement materials. Abrasion Problems were encountered in the use of the rotating cutter method to measure abrasion resistance (ASTM C944). In particular, the presence of a large aggregate particle near the working surface can significantly influence the results

23

LANGAN ET AL.

TABLE3. Compressive strength results (MPa) Mix No.

7d

14d

28d

56d

TABLE 5. Summary of freeze-thaw test results

112d

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

Mix No.

Weight change (%)

Durability factor (%)

Cycles at end of test

*Specimens for 112-day tests not cast. *No specimens cast for freeze-thaw test.

TABLE4. Abrasion and scaling results Abrasion weight loss (g) Mix No.

7d

14d

28d

Scaling 56d

Cycles

Rating

obtained. The weight-loss results presented in Table 4 represent an average of three tests at each age. Owing to the very small weight loss measured (1 1.7 g maximum) compared to the mass of the cylinder (approximately 12.5 kg), considerable

care was required to minimize the effect of moisture loss during the testing period. To minimize this effect, the specimens were removed from the fog room and allowed to stand in the laboratory for 3 h prior to testing. This delay was found to be sufficient to minimize the effect of moisture loss on the measured weight loss. The time required to attain "moisture stability" was established by monitoring the change in weight of a cylinder after removal from the fog room. Moisture stability is defined as being the condition when successive weighings did not vary by more than 1 g within 0.5 h of the previous weighing. This procedure was used in all tests except Mix 2 at the age of 7 days when the effect of varying moisture content was first noticed. The value shown for Mix 2 is the weight loss adjusted for moisture loss as determined by subsequent moisture stability tests. In general, the results (Fig. 5) indicate that the presence of fly ash at high levels of cement replacement increased the weight loss due to abrasion at all ages when compared with the plain concrete. The weight loss tended to decrease as the age

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

CAN. J . CIV. ENG. VOL. 17, 1990

olnert "Limestone

7

14

28

112

56 Age (days)

FIG.

1. Compressive strength vs. age for mixtures without air-entraining agent.

.Boundary

p

Dam

Sundance

-

Limestone

01

'

7 FIG.

I

I

14

28

I

56 Age (days)

112

2. Compressive strength vs. age for mixtures containing air-entraining agent.

at testing increased, although the compressive strength does not seem to have a significant effect on weight loss based on the limited data available. Mixtures 6 and 18 and mixtures 12 and 20 had 56-day compressive strengths equivalent to or greater than the control mixtures (mixes 2 and 4), but still demonstrated a higher weight loss. Gebler and Klieger (1986a), however, noted that abrasion resistance was generally improved with an increase in compressive strength of concrete. This relationship was not confirmed for all mixes tested in this study. Scaling The concern about the ability of fly ash concretes to resist

scaling is well documented. In general, fly ash concretes do not perform well in this test (Gebler and Klieger 1986b). For the control mixtures (Nos. 2 and 4) no scaling was observed even after 100 cycles. Only one fly ash mixtures (No. 6) lasted beyond five cycles; this mixture had a rating of 5 after 10 cycles. All other fly ash mixtures had a rating of 5 after five cycles. It should be noted that normally this test procedure is terminated when a rating of 5 is reached. Owing to the initial lower rate of strength gain of the fly ash concrete, coupled with the relatively high replacement level of 50%, all the fly ash mixtures had a significantly lower compressive strength than the control mixtures at the start of the test, thus one would expect less resistance to scaling. This is

LANGAN ET AL.

Laramie Control

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

Forestberg

Age (doys)

FIG. 3. Compressive strength

VS.

age for mixtures with superplasticizer.

Boundary Dam Sundance Wabamum Lingan

Inert Limestone

0

1

7

14

28

56

112

Age (days)

FIG. 4. Compressive strength vs. age for mixtures with air-entraining agent and superplasticizer.

in agreement with Klieger and Gebler (1987) who observed an increase in scaling resistance with an increase in the compressive strength. It should be noted that concretes in their study contained only 25 % replacement of cement with fly ash. The maximum scaling rating attained at 300 cycles was 3. Clearly, the 50% replacement level used in the present investigation is much too high for concretes subjected to salt scaling effects in service. Freeze - thaw resistance As noted earlier, it was found that mixtures containing no air entrainment or only a superplasticizer failed the freezethaw test before reaching 50 cycles. They are not included in this discussion. All concretes, as anticipated, regardless of the compressive

strength level, passed the severe requirements of this test, provided adequate air entrainment was present. Even those concretes containing the replacement materials, i.e., silica flour and limestone, passed the test when air entrainment was present. In some cases the compressive strength of these mixtures at the start of freeze-thaw testing (14 days) was less than 10 MPa. In general, the addition of the superplasticizing agent with the air-entraining agent produced a decrease in durability factor for the mixtures tested. Nonetheless, the water-to-cement ratio of the mixtures was not corrected for the dosage of superplasticizer, which could be responsible for this observation. Air void analyses were performed on nine of the mixes using the modified point count method of ASTM C457; the results are given in Table 6. All mixes, except Mix 4, met the

CAN. J. C N . ENG. VOL. 17, 1990

0 MIX 2

o MIX 4 A MIX 6 a MIX 8 o MIX14 0 MIX 18 MIX 2 0 t MIX 22 A MIX 2 4

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

*

Control + AEA Control + AEA+ SP Forestberg + AEA Forestberg t AEA t SP Wabarnum t AEA Laramie + AEA Lararnle t AEA + SP Nanticoke t AEA Nantlcoke + AEA + SP

Control

0L

I

I

7

14

56

28 Time (days)

FIG. 5. Weight loss in abrasion test vs. age.

TABLE 6. Air void analyses

Mix No. A (%) P (%)

n

-

1

a (mm-l)

(mm)

NOTES:A_= air content; P = paste content; n = number of g r voids per millimeter; 1 = average chord intercept; a = specific surface; L = spacing factor.

usual maximum allowable spacing factor requirement of 0.20 mrn. In general, the addition of the superplasticizer resulted in an increase in spacing factor, which correlates with reduced durability factors for these concretes. None of the mixtures tested failed the minimum required durability factor of 60% at the termination of the test procedure. All mixtures exhibited some scaling after 150-200 cycles of freezing and thawing. In some cases the weight loss was significant; however, the specimens were still intact at the end of the test. In general, the weight loss was higher when the superplasticizer was present.

Conclusions It should be recalled that these results are for a 50% by mass fly ash replacement level. In general, the lignitic ashes produced the highest strength values from 14 to 112 days. The Laramie ash produced the

maximum strength for non air-entrained concretes, while the Boundary Dam ash produced the best strength results when both chemical admixtures were present. The Laramie ash mixes yielded similar results to the control mixes when only the superplasticizer was used. The mixes containing filler material produced the lowest strengths. The presence of the air-entraining agent affected the magnitude of the compressive strength achieved, but did not alter the rate of gain of strength by itself or in conjunction with the superplasticizer. The presence of 50% fly ash increased the weight loss due to abrasion. The weight loss tended to decrease with an increase in compressive strength, but not in all the cases. However, when both the control mix and the fly ash mixes had the same compressive strength at the test age, the fly ash mixes exhibited a higher weight loss. These test results confirmed that compressive strength is not the most significant parameter in determining the freezethaw resistance of concrete. Even the concretes with compressive strengths as low as 9.4 MPa at the start of testing still passed the severe test requirements, provided an adequate air void system was in place. In general, mixes that contained both chemical admixtures did not perform as well as those that contained only the air-entraining agent. From the limited air void data obtained, there are several instances where the presence of the superplasticizer resulted in significantly increased spacing factors at the same global air content as the control mixes. These mixes also exhibited higher weight losses in the freeze-thaw test. The scaling abrasion results clearly demonstrate that this level of cement replacement, coupled with the relatively high water-cementitious ratios, is too high for concretes in pavements which require high abrasion as well as scaling resistance.

LANGAN ET AL.

Acknowledgements Financial for this was provided through the Strategic Grants Programme Of the Natural Sciences and

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

Engineering Research Council of Canada, Grant No. G1238. The authors wish to thank M r . R. Begert and Mr. P. Gifford for their valuable contributions to this project. ASTM. 1987. Concrete and mineral aggregates. 1987 annual book of standards, Vol. 04.02. ASTM, Philadelphia, PA. BRITISHSTANDARDS INSTITUTION. 1970. Methods of testing fresh concrete. British Standard BS 1881, Part 2. CANADIAN STANDARDS ASSOCIATION. 1977a. Concrete materials and methods of concrete construction. CSA Standard CAN3-A23.1M77, Canadian Standards Association, Rexdale, Ont. 19776. Methods of test for concrete. CSA Standard CAN3-A23.2-M77, Canadian Standards Association, Rexdale, Ont .

27

1983. Portland cements. CSA Standard CAN3-A5-M83, Canadian Standards Association, Rexdale, Ont. GEBLER,S., and KLIEGER, P 1986a. Effect of fly ash on the physical properties of concrete. Special Publication SP-91, American Concrete Institute, Detroit, MI, pp. 1-50. -19866. Effect of fly ash and the durability of air-entrained concrete. Special Publication SP-91, American Concrete Institute, Detroit, MI, pp. 483-519. J o s ~ rR., , and ROSAUER, E. 1973. Pozzolanic activity in synthetic fly ash. American Ceramic Society, Bulletin, 52(5): 456-463. JOSHI,R., DAY,R. L., LANGAN,B. W., and WARD,M. A. 1987. Strength and durability of concrete with high proportions of fly ash and other mineral admixtures. Durability of Building Materials, 4: 253 -270. KLIEGER,P., and GEBLER,S. 1987. Fly ash and concrete durability. Special Publication SP-100, American Concrete Institute, Detroit, MI, pp. 1043 - 1069.

Can. J. Civ. Eng. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 06/20/14 For personal use only.

This article has been cited by: 1. Alaa M. Rashad, Hosam El-Din H. Seleem, Amr F. Shaheen. 2014. Effect of Silica Fume and Slag on Compressive Strength and Abrasion Resistance of HVFA Concrete. International Journal of Concrete Structures and Materials 8:1, 69-81. [CrossRef] 2. Baomin Wang, Yu Han, Ni Tu, Hongmei Ai, Tingting Zhang. 2013. Experimental Study on Abrasion Resistance of Concrete Containing Scrap Rubber Powder. The Journal of Solid Waste Technology and Management 39:3, 214-220. [CrossRef] 3. M. CyrInfluence of supplementary cementitious materials (SCMs) on concrete durability 153-197. [CrossRef] 4. Rafat Siddique, Kushal Kapoor, El-Hadj Kadri, Rachid Bennacer. 2012. Effect of polyester fibres on the compressive strength and abrasion resistance of HVFA concrete. Construction and Building Materials 29, 270-278. [CrossRef] 5. Shadi Riahi, Ali Nazari. 2011. Compressive strength and abrasion resistance of concrete containing SiO2 and CuO nanoparticles in different curing media. Science China Technological Sciences 54:9, 2349-2357. [CrossRef] 6. İlker Bekir Topçu, Tayfun Uygunoğlu. 2010. Effect of aggregate type on properties of hardened self-consolidating lightweight concrete (SCLC). Construction and Building Materials 24:7, 1286-1295. [CrossRef] 7. Rafat Siddique, Jamal M. Khatib. 2010. Abrasion resistance and mechanical properties of high-volume fly ash concrete. Materials and Structures 43:5, 709-718. [CrossRef] 8. Rafat Siddique, Albert Noumowé. 2006. Résistances mécaniques et abrasion de bétons de cendres volantes. Revue Européenne de Génie Civil 10:10, 1217-1233. [CrossRef] 9. Hui Li, Mao-hua Zhang, Jin-ping Ou. 2006. Abrasion resistance of concrete containing nano-particles for pavement. Wear 260:11-12, 1262-1266. [CrossRef] 10. SW Dean, TR Naik, RN Kraus, BW Ramme, Y-M Chun. 2005. Decing Salt-Scaling Resistance: Laboratory and Field Evaluation of Concrete Containing up to 70 % Class C and Class F Fly Ash. Journal of ASTM International 2:7, 11912. [CrossRef] 11. Rafat Siddique. 2003. Effect of fine aggregate replacement with Class F fly ash on the abrasion resistance of concrete. Cement and Concrete Research 33:11, 1877-1881. [CrossRef] 12. Marta Sebastiá, Iñaki Fernández Olmo, Angel Irabien. 2003. Neural network prediction of unconfined compressive strength of coal fly ash–cement mixtures. Cement and Concrete Research 33:8, 1137-1146. [CrossRef] 13. R. Paul Lohtia, Ramesh C. JoshiMineral Admixtures 657-739. [CrossRef] 14. Tarun R. Naik, Shiw S. Singh, Mohammad M. Hossain. 1994. Abrasion resistance of concrete as influenced by inclusion of fly ash. Cement and Concrete Research 24:2, 303-312. [CrossRef]