Effect of Mix Design on Restrained Shrinkage of Concrete

Metin Husem

Effect of Mix Design on Restrained Shrinkage of Concrete

Metin Husem

2013

visibility

…

description

9 pages

link

1 file

The impetus for this research came from a project that involved the reconstruction of Twelve bridge decks in Tabriz, East Azerbaijan, IRAN that was completed in the summer of 2009. Shortly after construction was completed, shrinkage cracks were observed on 8 of the 12 bridge decks. Such cracking can cause long term durability problems by facilitating chemical ingress and moisture penetration that may aggravate problems such as alkali-silica reaction and corrosion of the reinforcing steel . Shrinkage cracks can also increase deterioration caused by cyclic loadings. Any of these problems can decrease the service life of a bridge deck . The mixture used in the bridge decks had a w/c+p ratio of 0.33 and contained 26.2% fly ash (Class F). Although the mixture was not intended to be a high performance concrete (HPC) mixture, it did have characteristics similar to HPC mixtures. As the use of HPC becomes more common, studies on the effect of pozzolanic materials and the ability of admixture...

Sign up for access to the world's latest research

checkGet notified about relevant papers

checkSave papers to use in your research

checkJoin the discussion with peers

checkTrack your impact

Figures (11)

Tab. 1. The granulometry of aggregates used in production of concretes 2.1.Mixture Proportions ——oer vere Type I/II Portland cement, expansive (Type K) Portland cement, and Class F fly ash were used as cementations materials in this study. The aggregates used in this study were obtained from the alluvial source used to obtain aggregates from Tabriz. Physical properties for these aggregates are provided in Tab. 1. Drinking water is used in the mixing and cure. A shrinkage- reducing admixture was also used for two sets of mixtures. Polypropylene fibers, 19 mm in length, were used in one series. es rs Six sets of concrete mixtures were investigated. The mixture proportions were obtained by varying the proportions used for the. The mixture proportions used in construction are presented as Mixture A4 in Tab. 2. These proportions included 26.2% fly ash by mass of cement. The fly ash was used as an addition to the cement, not as a replacement. The first series of mixtures (Series A) included four mixtures with constant water and cement contents, but decreasing fly ash content. Mixtures A3-A1 had fly ash contents of 20, 10, and 0%, respectively. Since the water content remained constant, w/c+p ratios for the mixtures in Series A were 0.33, 0.35, 0.38, and 0.42. Each of the other five sets of mixtures contained four mixtures with these fly ash contents. In the second series of mixtures (Series B), the cementations materials content and w/c+p ratio were kept constant. Consequently, the cement content increased as the fly ash content decreased. The w/c+p ratio for these mixtures was 0.38. The third series of mixtures was designated as Series C. The mixtures in Series C had the same mixture proportions as the mixtures in Series B, but the Type Il cement was replaced with a shrinkage compensating cement (Type K). The fourth series (Series D) repeated variations in Series B, but also contained the manufacturer's minimum recommended dosage of SRA. The fifth series (Series E) also repeated the variations in Series B, but contained the

manufacturer's recommended dosage of the SRA for optimal shrinkage reduction. The sixth serie (Series F) repeated variations in Series A, but 1% polypropylene fibers by volume were included i each mixture. The mixture proportions for all mixtures are presented in Tab. 2. Preparation of Test Soecimens For each mixture, six prisms (76.2 x 102 x 406 mm) and three cylinders (102 by 204 mm) were prepared. The cylinders and three of the prisms were used for measurements of compressive and flexural strength, respectively. The remaining three prisms had 19.1 mm gage studs embedded in their ends to facilitate length change measurements. After placement, the specimens were left in the molds and covered in plastic for 24 + 4 hours. After removal from the molds, the prisms with gage studs were moist cured for seven days, and then air-cured for the remainder of testing. The prisms and cylinders used for strength testing were moist cured for 28 days. The prisms with gage studs were monitored for length changes to an age of 56 days, and then oven-dried to obtain total shrinkage.

Fig. 1. Flexural strength with varying fly ash contents. generally observed at ages later than 28 days [3,4]. _ ee eee EI EE EE EEE t/a Compressive strengths for all mixtures are illustrated in Fig. 2. The most important trend in Fig. 2 is that compressive s decreased as the compressive strengt rength increased as fly ash content increased in Series A. However, w/c+¢ ly ash content increased in Series A. A similar trend is observed in the hs from Series F, which has mixture proportions similar to Series A. It appears that the expected trend of increasing strength with decreasing w/c+p ratio is primarily responsible fo. this trend because f remaining mixtures. y ash content does not have a significant effect on compressive strength for the As with flexural strength, fly ash was not expected to have a strong effect or compressive strenat h at 28 days.

~The effects of fly / ash content on shrinkage can be seen in Fig. 4. Again, the mixture with 20% fly ash had the least shrinkage. 3. The water and cement content was held constant in these mixtures. Shrinkage strains for the various fly ash contents are similar. Mixture A20, with w/c+p = 0.35, exhibited approximately 120 microstrain less shrinkage than the other mixtures in Series A. Slump results for this mixture were similar to the other mixtures in this series, so an erroneous water content does not appear to be a factor in the lower shrinkage strains. The mixture with the lowest w/c+p, 0.33, exhibited shrinkage strains similar to the mixture with the highest w/c+p. In very low w/c+p concrete, capillary shrinkage due to self-desiccation tends to be greater. However, shrinkage also increased as the w/c+p increased from 0.35. It is important to note that the increasing w/c+p and shrinkage were not accompanied by an increasing water content that would result in more cement paste. Because the fly ash was treated as an addition in this series, the paste content was decreasing as w/c+p increased. Consequently, it appears that the 20% fly ash content produced an optimal effect in terms of reducing capillary tension or limiting moisture loss. af Fife nmnte af Fl., Aek Mawmiant

-ig. 4. Shrinkage strains for mixtures from Series B.

~ Shrinkage comparisons of specimens containing Type K cement are shown in Fig. 5. shrinkage ¢ 6 days was observed in the mixtures with Type K cement compared to mixtures in Series B. Initi xpansion of the specimens made with Type K cement was greater in the specimens with 20% an % fly ash content and similar to the specimens with Type I/Il cement for the other two fly as contents. By 56 days, the Type K specimens had experienced about 50 microstrain more shrinkag han comparable specimens in Series B. Type K mixtures require higher w/c+p ratios to ensure th yroper formation of ettringite, sometimes as high as 0.5 [1]. The w/c+p ratio used in this study for th nixtures with Type K cement was 0.44. As a result, greater shrinkage was observed in Set yecause the expansive sulfoaluminates in the Type K cement did not have sufficient water 1 naximize expansion. The effects of fly ash content on shrinkage of Series C mixtures was similar 1 he effects observed for Series A and Series B. Excluding mixture C3, shrinkage strains at 56 day vere similar for all the Type K specimens. Mixture C3, with 20% fly ash, had approximately 10 nicrostrain less shrinkage at 56 days than the Type K mixtures with other fly ash contents. Th upports the observation that moderate amounts of fly ash are helpful in controlling shrinkage. 3.5.Effects of SRA The effects of two different dosages of SRA are shown in Figg. 6-7. Fig. 6 presents the shrinkage results for mixtures containing the manufacturer’s minimum recommended dosage of the SRA. At this dosage, the SRA produced a minor reduction in shrinkage. Results for specimens with the minimum dosage of SRA initially run along the same curves as the results for specimens without SRA. Between 14 and 24 days, specimens with SRA began to show less shrinkage than the control specimens. Shrinkage strains in Series D, with a minimum dosage of SRA, were 50-90 microstrain less than shrinkage strains for similar mixtures in Series B at 56 days. Shrinkage results for mixtures containing the manufacturer's recommended dosage of SRA for optimal performance are presented in Fig. 7. At this dosage, there is a substantial difference in the shrinkage strains of the specimens with SRA compared to those without SRA. While the specimens were moist-cured, the length changes that occurred in two sets of SRA mixtures were similar to those in non-SRA mixtures. Differences in shrinkage strains became apparent in the first reading after the specimens began the dry-curing phase. Initial differences in shrinkage strains are small. However, differences in shrinkage strain increase significantly by 56 days for all mixtures except E4. 4 exhibited 60 micro strain less shrinkage than B26. The remaining mixtures in Series E experienced reductions in shrinkage strain of 230-260 microstrain. This is a reduction of approximately 50% when the manufacturer’s optimal dosage of SRA was added to the mixture. Close inspection of Figg. 6-7 reveals that Mixtures MS20

weer ee ee 8 ee ae The effects of using fibers in the concrete mixture are shown in Fig. 8. The polypropylene fibers used in this study had little effect on shrinkage. Fly ash content in the fiber mixtures also had little effect on shrinkage. The lack of a trend for the Series F mixtures was not surprising since fibers are generally used to control cracking by limiting crack widths, not by reducing shrinkage[6].

J.f.l All Olal ONMMMNKage COMPAarsons specimens were oven dried for 48 hours following the 56-day reading. After oven drying, a final length measurement was performed and total shrinkage strain was calculated. Total shrinkage resul s are shown in Fig. 9. The optimum dosage of SRA resulted in the lowest total shrinkage strain for each of the different fly ash contents, excluding 30% fly ash. Although a minimum dosage of SRA expe shrin rienced more total shrinkage than Series E, these specimens generally exhibited less total kage than any of the other sets. Specimens made with Type K cement exhibited similar or greater total shrinkage when compared to similar specimens made with Type I/Il cement. There is no distinct trend in the total shrinkage strain of the specimens containing fibers. Mixtures with 20% fly ash had less shrinkage than comparable mixtures with other fly ash contents in four of the six seauences.

David Darwin

Structural Engineering and …, 2007

The effects of paste volume, water-cement ratio, aggregate type, cement type, curing period, and the use of mineral admixtures and superplasticizers on the free shrinkage of concrete are evaluated with the goal of establishing guidelines to reduce cracking in reinforced concrete bridge decks. Three concrete prisms were cast and tested in accordance with ASTM C 157 for each mixture up to an age of 365 days under controlled conditions of 23 ± 2°C (73 ± 3°F) and 50 ± 4 percent relative humidity. The work was organized in five test programs. The first program included mixes with water-cement ratios of 0.40, 0.45, and 0.50, and aggregate contents of 60, 70, and 80 percent, with Type I/II cement and Type II coarse-ground cement. The second program included the mixes with one of three coarse aggregate types, granite, limestone, and quartzite. The third program evaluated the effects of Class C fly ash, ground granulated blast-furnace slag, and silica fume as partial volume replacements for portland cement. The fourth and fifth programs were used, respectively, to evaluate the effect of curing period (3, 7, 14, or 28 days) and the use of different superplasticizer types and dosages. The results indicate that concrete shrinkage decreases with an increase in the aggregate content (and a decrease in the paste content) of the mix. For a given aggregate content, no clear effect of water-cement ratio on the shrinkage is observed. In general, granite coarse aggregates result in lower shrinkage than limestone coarse aggregates. A similar conclusion cannot be made with quartzite coarse aggregate, although in some cases shrinkage of concrete containing quartzite coarse aggregate ii was lower than that of concrete containing limestone. The use of partial volume replacement of portland cement by Class C fly ash without changing the water or aggregate content generally leads to increased shrinkage. The use of partial volume replacement of portland cement by blast furnace slag without changing the water or aggregate content can lead to increased early-age shrinkage, although the ultimate shrinkage is not significantly affected. An increase in the curing period helps to reduce shrinkage. The use of Type II coarse ground cement results in significantly less shrinkage compared to Type I/II cement. The use of superplasticizers in concrete appears to increase in shrinkage to a certain degree. The results, however, do not present a clear picture of the effect of superplasticizer dosage on shrinkage.

Log In

Effect of Mix Design on Restrained Shrinkage of Concrete

Sign up for access to the world's latest research

Figures (11)

Related papers

Related papers

Related topics