Effect Of Cement Chemistry On Accelerated Curing Cycle Of Concrete - PCI

Effect of Cement Chemistry on Accelerated Curing Cycle of Concrete Derek Firth, P.E. Regional General Manager Genstar Costain Tie Co. Ltd. Edmonton, Alberta Synopsis This paper presents the results of an investigation into improving the high early compressive strength of concrete containing Type III cement in a prestressed concrete manufacturing plant using accelerated curing techniques. The effects of physical and chemical cement characteristics are discussed with emphasis on C 3S and C3A contents, SO3 optimization, cement fineness, and alkali content. Recommendations are given to the prestressed concrete producer for obtaining the cement quality most likely to suit his needs. Note: This article is based on a paper presented at the 1983 PCI Convention in Las Vegas, Nevada, October 2-5, 1983. PCI JOURNAL/July-August 1984 P recast concrete manufacturers, particularly those who specialize in prestressed concrete, use expensive accelerated curing techniques to achieve high early concrete compressive strengths. In many precasting operations the early target strength of freshly cast products can be as high as 4500 psi (31.0 MPa) after only 14 hours. To attain such strengths, the concrete mix designs usually have a high cement content. (Most prestressed concrete manufacturers use a Type III portland cement, though there are reportedly some so-called "product cements," which are claimed to be of optimum quality for heat-cured products.) Unfortunately, the current prescriptive and performance requirements of ASTM C 150 do not provide specific information to either the cement manufacturer or cement user on the physical and chemical aspects of cement that 45

Table 1. Comparison of objectives of cement manufacturer and prestressed concrete producer in meeting respective specifications and other requirements (Type III cement). Requirements Cement manufacturer's objectives Compressive strength 1800 psi (12.4 MPa) mortar strength (ASTM C 109 minimum) 4200 - 4500 psi (28.9 - 31.0 MPa) concrete strength Time strength required 24 hours 12 - 14 hours Temperature of curing 72 F (22.2 C) 150 - 160 F 65-71C Water added to mix Constant Variable demand Air entrainment None Usually, except in southern United States are best suited in attaining accelerated curing for high early strength concrete. Table 1 provides a comparison of the major objectives of the cement manufacturer and prestressed concrete producer in meeting specifications and other requirements. On the one hand, the cement manufacturer wants to ensure that his cement and mortar meet the ASTM C 150 and C 109 standard specifications, while the prestressed concrete producer is more interested in accelerated curing to attain high early compressive strength for his products. Researchers l 2 have already identified the particular beneficial ranges of chemical constituents of cement which yield best performance in an accelerated curing system. Table 2 shows those ranges that have been identified corresponding to particular chemical constituents. A survey3 of the percentage variations in these chemical consitutents in 46 Prestressed concrete producer's objectives American Type III cement is also included in Table 2. It may be seen from Table 2 that there are many differences between the Type III cements that are presently available in the United States and the type of cement the prestressed concrete producer would likely find most beneficial for accelerated curing purposes. Of particular interest are the sulfur trioxide (SO3) content, the C I S, C2 S, and C I A ranges in addition to the cement fineness and alkali content requirements. Test Program To pursue the objectives of the prestressed concrete producer (listed in Table 1), a program was undertaken over a 3-year period to determine those particular aspects of Alberta Type III cement which would lead to high early strength performance of concrete, using only locally available materials.

Table 2. Chemical requirements of Type III cement. Chemical constituent ASTM C 150 Prestressed concrete manufacturer SO3 3.5 percent max. (if C3 A 8 percent) 4.5 percent max. (if C 3 A 8 percent) Optimized for temperature of curing and time of release C3 S Not specified (usually in range of 34 - 70 percent) 55 - 60 percent Cz S Not specified (usually in range of 0 - 38 percent) 15 - 20 percent C3 A 15 percent max. (usually in range of 7 - 15 percent) 9 - 10 percent 0.80 percent Not specified Alkali content (Na2 0 equivalent) (except 0.60 percent preferable limit for low alkali cement) Blaine Not specified SO 3 Optimization — As a first step in the program the SO3 content of the cement was studied for optimization in an accelerated curing environment [160 to 170 F (71 to 76 C)], where the required period of curing was 13 hours. Since SO3 is present in cement clinker and usually added in the form of gypsum during milling, a study was needed to determine the effect of available sulfate in controlling early compressive strength development. This particular investigation* was conducted at the Portland Cement Association in Skokie, Illinois. The laboratory study was carried out by making paste cubes, first of the cement as produced, then by incremental additions of terra alba — to achieve small increases in SO 3 content. The compressive strength of these test specimens was measured after the cubes had been exposed to an accelerated curing cycle similar to that found in a typical prePCI JOURNAL/July-August 1984 5200 - 5400 cm2/g (520 - 540 m2/kg) stressed concrete production plant. The results of this study (on an Alberta Type III cement) demonstrate the need for the cement SO 3 content (for a particular curing regime) to be some 0.40 percent above the level shown for the more normal moist curing cycle (see Fig. 1). Additionally, the tolerance range of SO3 variations in the cement for peak performance was observed to be much narrower than that shown by the moist curing tests. In perspective, however, it should be pointed out that these findings may only relate to Alberta cement since the types of sulfates incorporated have been found to have a significant effect on the results achieved. Cement Fineness — Another impor*Investigative Report submitted by Construction Technology Laboratories, Portland Cement Association, Skokie, Illinois, December 1981 (Private Communication). 47

6C 0 a- U) a- 5C q 35 I13-HOUR ACCELERATED HEAT CURED z U fr Z 30 w H 40 U) w U U) U p I-DAY MOIST CURED p C 30 r a- U) U) U 20 pp 0 a- 0 U U 15 2.70 2.95 3.20 3.45 3.70 3.95 4.20 4.45 CEMENT S0 3 CONTENT, PERCENT Fig. 1. Cement paste compressive strength versus sulfur trioxide content of cement for 13-hour accelerated heat curing and 1-day moist curing. tant part of the investigation was to Table 3. Advantages and disadvantages of study the fineness of the cement. In increasing fineness of cement (Blaine general, North American Type III ce- method). ments are ground much finer than Advantages equivalent European cements. This is probably due to the fact that the man— Increased early strength ufacturer of cement in North America Disadvantages uses a common clinker for both Types I — Increased water demand and III cements. As a result there is little flexibility in — Increased shrinkage variation of chemical content for Type — Increased reaction with potentially III cement from that of Type I cement reactive aggregates and finer grinds are used in an attempt — Greater proneness to concrete cracking to improve early strength performance. — Needs more SO 3 in cement to retard There are, however, both advantages C3 A and disadvantages of the current trend — C3 A more available to react to increase the Blaine fineness of Type — More air-entraining agent needed to III cements, as shown in Table 3. The achieve desired air content level results of this study (again using Alberta cement) show that a target Blaine fineness of 5250 cm 2/gm (525 m 2/kg) should be sought. 48 — Increased probability of dehydrating gypsum, due to high temperature during grinding

This conclusion was reached after much time and effort by examining some of the negative aspects of higher fineness in a plant producing only one mix design for some 250 cu yds (190 m3) of low-slump concrete each day. It should be noted, however, that production of a non-air-entrained concrete may not be as susceptible to problems in this area as to situations where a consistent air entrainment level is required. Alkali Content — A further aspect of cement which has been receiving special attention recently is the alkali content. Speakers at the 1983 International Conference on Alkalis in Concrete4 pointed out that very few aggregates are totally inert; hence, the effect of alkali content increases in cement could possibly lead to long-term durability problems as a result of alkali-aggregate reaction. Despite this danger, there has been a trend to increase cement alkali content as cement manufacturers have built new production facilities incorporating the new, more cost effective, and energyefficient dry process. Although there are two beneficial aspects of such trends to increase cement alkali content (shown in Table 4), such cost saving moves in cement manufacturing could lead to disastrous consequences if incorporated into concrete with a high volume cement content and with a deleterious aggregate. During the time this study was being carried out, statistical compressive strength control sample data were being collected from the production of accelerated cured concrete including that of mortar cubes (made available by the cement producer). These data are shown in Table 5. In studying the data, it should be recognized that the test specimens were fabricated, cured, and tested as shown in Table 1 by both the cement manufacturer and the prestressed concrete producer. In addition, it should be recognized PCI JOURNAL/July-August 1984 Table 4. Effect of alkali cement content. Effect of alkali content Remarks 1. Surveyed range of alkali 0.2 - 1.3 content of cement 2. Current trends (To achieve energy percent To increase earlier levels cost savings) 3. Effect of alkali content increase on the user: (a) Harmful if aggregates are alkali reactive (b) Lower air-entraining agent addition for specified levels (c) Higher early strengths Result: Very serious Beneficial Beneficial that the mortar specimens were produced and cured under well-controlled conditions. Yet the same cannot be assumed for the concrete specimens. There appears, therefore, to be a good indication from the data that the concrete producer has to put a significant amount of quality control effort into preventing interaction of other quality variations in aggregates, admixtures, and cement constituents. Failure to minimize such interaction could result in concrete variations totally unacceptable to customers. CONCLUSIONS The work carried out on one particular, locally available Type III cement, together with reference to work already published, allows the following conclusions to be made: 1. The prestressed concrete producer who uses an accelerated curing cycle is likely to become aware that: (a) Type III cement is not normally manufactured to suit his curing regime. (b) The cement producer's quality control data are not likely to be indicative of the cement's 49