Click for next page ( 36


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 35
35 0.6 0.5 Soil Penetrometer (MPa) 0.4 0.3 0.2 0.1 0 0 2 4 6 8 10 12 ASTM C403 Penetration (MPa) Figure 3.5. Comparison between ASTM C 403 and soil penetrometer values. and 22r, which were identical except the use of an accelerat- ing the effects of cylinder size, capping material, and load rate. ing admixture in 26, indicated that the use of an accelerating Based on these original mixtures, some useful predictive mod- admixture reduced bleeding and slightly reduced subsidence. els were developed to predict the strength gain (short and long The results of subsidence testing are shown in Table 3.21. term) of CLSM. The original study led to several follow-up studies, each of which focused in more detail on issues involv- ing testing parameters. Detailed investigations on load rate, Hardened CLSM Properties curing and conditioning of cylinders, effects of drainage on strength, and the use of alternative capping materials were per- Unconfined Compressive Strength formed. The findings of the initial broad study and the later de- A great deal of emphasis was placed in assessing the uncon- tailed studies were used to refine and improve existing methods fined compressive strength of CLSM. This section first summa- of measuring the unconfined compressive strength of CLSM. rizes the findings from the initial mixtures (38 in all), in which The unconfined compressive strengths of the originally pro- several aspects of compression testing were examined, includ- posed mixtures at 3, 7, 28, and 91 days are shown in Table 3.22. 0.5 0.4 Vane Shear (MPa) 0.3 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 Soil Penetrometer (MPa) Figure 3.6. Comparison between soil pocket penetration and vane shear values.

OCR for page 35
36 Table 3.21. Subsidence results of six selected CLSM mixtures. Mixture 4 Mixture 24 Time (h) Subsidence (mm) Time (h) Subsidence (mm) 1.25 3.0 3.0 5.3 2.30 3.4 5.5 6.8 3.57 3.4 6.0 6.8 4.57 3.5 5.17 3.4 Mixture 23 Mixture 6 Time (h) Subsidence (mm) Time (h) Subsidence (mm) 3 0.3 1.00 9.4 4 0.3 2.17 10.5 3.33 12.7 4.33 14.9 6.50 15.9 Mixture 22r Mixture 26 Time (h) Subsidence (mm) Time (h) Subsidence (mm) 3.17 1.2 3.67 1.5 5.17 2.2 5.67 1.5 6.33 2.2 Note: The total specimen height was 600 mm. Table 3.22. Unconfined compressive strength of original 38 CLSM mixtures. 3-day f c C.O.V. 7-day f c C.O.V. 28-day f c C.O.V. 91-day f c C.O.V. Mixture (MPa) (%) (MPa) (%) (MPa) (%) (MPa) (%) 1 0.12 8.2 0.21 6.8 1.09 4.9 1.87 2.8 2 0.29 2.4 1.76 10.1 3.69 4.0 6.26 13.5 1r 0.13 14.4 0.24 1.2 1.35 7.7 2.34 0.3 15 0.07 8.2 0.11 5.7 0.18 4.0 0.25 1.4 3 0.33 10.5 0.57 2.4 1.36 6.8 2.02 2.9 8 0.09 9.8 0.11 8.3 0.25 4.7 0.33 10.6 10 0.12 2.3 0.16 15.1 0.22 7.9 0.26 1.3 9 0.09 9.7 0.13 12.4 0.22 5.0 0.25 2.9 5 0.14 13.2 0.18 8.2 0.46 16.1 0.57 5.3 12 0.30 16.2 0.27 6.2 0.57 4.7 0.86 3.4 4 0.34 4.9 0.48 6.2 0.79 11.0 1.08 13.6 7 0.09 3.2 0.11 3.4 0.12 9.7 0.16 5.8 3r 0.46 13.1 0.58 4.4 1.49 5.8 1.97 8.3 4r 0.41 13.6 0.57 5.8 0.94 4.0 1.03 6.9 24 0.34 4.8 0.22 1.4 0.44 0.1 0.58 4.6 23a 0.04 6.4 0.14 9.5 0.18 7.6 18a 0.33 6.8 0.70 1.1 0.79 4.0 14 0.58 6.6 1.07 13.5 2.15 8.1 3.49 16.8 2r 0.42 9.3 1.58 2.9 4.90 2.4 6.87 0.3 29 0.18 0.6 0.31 0.2 0.63 1.9 0.98 6.1 30 0.09 7.1 0.14 3.1 0.26 8.6 0.28 2.8 17a 0.01 31.8 0.07 18.9 0.13 16.8 11 0.33 1.7 0.42 4.3 0.75 3.5 0.94 4.7 6 0.40 10.2 0.47 0.7 0.83 4.9 1.09 4.7 16a 0.06 11.9 0.13 12.0 0.16 8.5 21a 0.09 10.6 0.16 11.8 0.18 11.7 22a 0.43 9.0 0.73 4.3 1.01 4.8 22r 0.32 4.2 0.50 9.7 0.96 17.5 0.93 7.0 5r 0.17 12.5 0.28 10.3 0.55 10.1 0.78 16.3 26 0.43 7.1 0.76 8.2 1.14 15.8 1.53 2.6 16ra 0.07 9.7 0.15 23.6 0.17 8.3 13 0.28 0.5 0.35 3.3 0.74 3.2 1.12 4.4 25 0.17 4.4 0.30 5.7 0.40 30.9 0.50 9.0 19a 0.02 0.71 0.06 45.0 0.06 13.2 20a 0.04 36.4 0.21 1.0 0.29 26.0 27 0.22 4.6 0.29 1.7 0.36 3.8 0.55 6.6 20ra 0.04 49.9 0.15 32.8 0.24 10.4 28 0.28 3.0 0.47 0.9 0.70 1.95 0.94 0.2 a Mixtures were too weak to be tested at 3 days.

OCR for page 35
37 3000 found to work well for the entire range of materials and mix- 28 days 2500 ture proportions; however, predictive models for subsets of the mixtures were found to be quite accurate. For instance, sepa- 2000 rate models were developed for air-entrained (both for "high" Load (N) 7 days 1500 and "moderate" air contents) and nonair-entrained CLSM. The results obtained from mixtures containing bottom ash 1000 3 days or foundry sand were not included in the data that were used 500 to develop the predictive compressive strength model of air- 0 entrained CLSM (up to 91 days). These data were excluded 0 1 2 3 4 5 because air entrainment was found to be too difficult in mix- Deformation (mm) tures containing these aggregates. The model predicting the Figure 3.7. Load-deformation response of mixture compressive strength of air-entrained CLSM mixtures is shown 12 at 3, 7, and 28 days. in Equation 3.2 (Du et al. 2002): f c = a e b(w c ) (3.2) The data shown in this table were for small, sulfur-capped cylinders (75 150 mm) tested on a smaller capacity machine at a loading rate of 0.38 mm/min, as previously described. where The results were found to be repeatable, with quite low values f c = compressive strength (MPa) of coefficient of variation. a = 0.3074 ln(t) + 0.2289 An interesting observation that illustrates the uniqueness of b = 0.0086 ln(t) - 0.272 CLSM is that most mixtures show a drastic change in the load- t = age (days) deflection curve as the curing time is increased. Figure 3.7 w/c = water-cement ratio illustrates this behavior for Mixture 12, which was typical of most CLSM mixtures. At early ages, CLSM acts more like a The measured and predicted compressive strengths using soil, with more ductile behavior, but as time progresses, CLSM the model shown in Equation 3.2 for air-entrained mixtures begins to act more like concrete, with higher strength and are plotted in Figure 3.8. There was very good correlation, with lower ductility. an R2 value of 0.97. This model was also found to be effective Efforts were made in this project, based on the strength re- in predicting long-term strength gain (i.e., beyond 91 days). sults for the initial 38 mixtures, to develop predictive models For example, cylinders from mixture 22r that were tested for for the compressive strength of CLSM. Various models and compressive strength after 256 days had an average strength of statistical approaches were considered. No single model was 1.0 MPa, compared to the predicted value of 1.1 MPa. 1.2 1.0 Predicted Strength (MPa) 0.8 0.6 0.4 0.2 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Measured Strength (MPa) Figure 3.8. Measured vs. predicted strengths of air-entrained CLSM.

OCR for page 35
38 A predictive model was also developed for nonair- b4(t) = 0.75 - 0.018 t when t 30 days entrained CLSM mixtures. If the water-cement ratio was the b4(t) = 0.22 when t > 30 days only variable used to predict the compressive strength, the model yielded an R2 value of 0.8. To improve this accuracy, a Although the models presented previously are valid for only model was developed that included the water-cement ratio, the materials used in this project, they provide important in- aggregate type, fly ash type, and the fly ash content as strength- sights into the strength development of CLSM mixtures. Most predicting variables. A critical aspect to this approach was to of the significant effects were related to the influence of water assign numerical values to the non-numerical variables used demand on compressive strength. Mixtures containing ma- in the model. Through an iterative process, the constant, k, terials that increased the required water content for the target was selected for the materials used in this investigation. Con- spread (such as foundry sand and high-carbon fly ash) gener- crete or river sand (kriver sand) was assigned a value of 1.0, ally yielded lower compressive strengths compared to mix- foundry sand (kfoundry sand) a value of 0.2, bottom ash (kbottom ash) tures with lower water contents. The chemical reactivity of fly a value of 1.0, Class C fly ash (kC ash) a value of 2.2, Class F fly ash was found to be critical, because the strength of CLSM ash (kF ash) a value of 1.0, and high-carbon fly ash (kHC ash) a mixtures containing Class C fly ash was higher than similar value of 0.75. The equation for predicting the compressive mixtures containing Class F or high-carbon fly ash. Class C fly strength, S(t), is shown in Equation 3.3 (Du et al. 2002), and a ash has a higher CaO content than Class F fly ash (and the comparison between predicted and actual compressive high-carbon fly ash used in this study) and it increases the strengths is shown in Figure 3.9. early and final strengths of the mixtures. Because the devel- oped models are valid for only the specific materials used in this study, the researchers recommend preparing and testing b (t ) S ( t ) = b0 ( t ) ( kagg . type ) 1 (k fly ashtype )b2(t ) a series of trial mixtures to predict the strength of CLSM mix- b3 ( t ) b4 ( t ) tures containing different materials. (w c ) ( k fly ashcontent ) (3.3) As stated earlier, various modifications to the unconfined compression test were studied in the initial investigation, where some of which were later addressed in more detailed research. S( t) = compressive strength (MPa) The following paragraphs briefly discuss the findings from t= age (days) this initial investigation that focused on cylinder capping b0(t) = 0.0007 t2 + 0.13 t - 0.76 methods, cylinder size, and testing machine capacity. b1(t) = 0.0001 t2 + 0.013 t + 0.42 Table 3.23 shows a comparison between sulfur-capped b2(t) = 0.00008 t2 + 0.015 t + 0.094 cylinders and neoprene-capped cylinders for the selected six b3(t) = 0.003 t - 1.03 CLSM mixtures. In general, sulfur-capped cylinders yielded 1.2 1.0 Predicted Strength (MPa) 0.8 0.6 0.4 0.2 0.0 0.0 0.2 0.3 0.5 0.7 0.8 1.0 Measured Strength (MPa) Source: Du et al. (2002) Figure 3.9. Measured vs. predicted strengths of non air-entrained CLSM mixtures.

OCR for page 35
39 Table 3.23. Comparison of compressive strengths with different capping materials and methods. 3 days 7 days 28 days Sulfur Neoprene Sulfur Neoprene Sulfur Neoprene Mixture (C.O.V.) (C.O.V.) (C.O.V.) (C.O.V.) (C.O.V.) (C.O.V.) MPa (%) MPa (%) MPa (%) MPa (%) MPa (%) MPa (%) 4 0.34 (4.9) 0.25 (15.1) 0.48 (6.2) 0.34 (6.0) 0.79(11.0) 0.55 (14.4) 6 0.45 (6.8) 0.25 (14.7) 0.47 (0.7) 0.36 (6.3) 0.81 (4.9) 0.61 (3.2) 23a 0.04 (6.42) 0.03 (12.92) 0.14 (9.52) 0.12 (3.14) 24 0.34 (4.8) 0.15 (0.3) 0.22 (1.4) 0.19 (20.5) 0.44 (0.1) 0.30 (18.8) 22r 0.32 (4.2) 0.22 (8.1) 0.50 (9.7) 0.37 (10.6) 0.93 (17.5) 0.54 (4.4) 26 0.43 (7.1) 0.29 (9.1) 0.76 (8.2) 0.42 (11.2) 1.11 (15.8) 0.73 (7.2) a Mixture was too weak to be tested at 3 days. higher strengths than cylinders using neoprene pads. The the researchers placed additional emphasis on assessing the ef- neoprene pads used in this initial study had a durometer fects of loading rate on the compressive strength of CLSM. value of 50, which is a typical durometer value for conven- The loading rate is an important parameter when considering tional concrete cylinder testing. A more comprehensive study compression testing. The testing of CLSM cylinders should on capping materials, including neoprene pads with signifi- first of all be accurate. If strength values are found to be cantly lower durometer values, was subsequently performed, strongly influenced by loading rate, then a finite range must as described later in this chapter. be defined and required for accurate testing. The loading rate In a study focusing on the effects of cylinder size (75 also determines the length of time needed to test a given cylin- 100 mm vs. 150 300 mm), cylinder size across this range der. This required length of time must be sufficient to ensure was found to have little impact on strength values. This study accuracy (i.e., not a sudden cylinder failure) and to complete used deflection-controlled testing with all cylinders tested at a given test in a reasonable amount of time. In a laboratory or the same effective strain rate. Thus, like conventional con- testing facility, the time required to test cylinders may be crit- crete, different cylinder sizes of CLSM can be used provided ical, especially if many tests are performed in a single day. that the length-diameter ratio remains at 2:1, that the cylin- A deflection-controlled machine, often used to test soils, was der size is sufficiently large for the maximum aggregate size, used for this investigation with the following rates of deflection: and that the load capacity of the machine is adequate to ac- 0.13 mm/min, 0.25 mm/min, 0.38 mm/min, 0.51 mm/min, curately measure the peak load. and 0.89 mm/min. The deformation at peak load was used to Most of the compression testing in this project used a smaller determine if changes in loading rate resulted in changes in capacity testing machine under deflection control. However, modes of failure, such as a change from relatively ductile to many laboratories that typically test conventional concrete are brittle failure. currently using larger capacity concrete compression machines The effects of load rate on compressive strength were found under load rate control. To assess the relative difference in to vary, depending upon the type of mixture and the age of strength values, a limited study was performed to compare the testing. Interestingly, some mixtures (e.g., mixture A-2) were results of a small-capacity (100 kN) machine under deflection relatively insensitive to load rate, whereas others were quite control to a larger capacity (1780 kN) compression machine sensitive. No consistent trend among all mixtures suggested using load control. It should be noted that in order to meet the that compressive strength was either directly or indirectly time-to-failure limits described in ASTM D 4832 (not less than proportional to load rate (based on the range of rates eval- 2 minutes), a load rate of 6.9 kPa/s was used for the large ma- uated). In general, the range of loading rates from 0.25 to chine. The results indicated significantly lower strength values 0.64 mm/min produced the most consistent results. Also, and higher variations for the large compression machine, com- within this range of loading rates, there was little impact on pared to the results from the smaller, deflection-controlled the deformation at peak load, suggesting that the mode of fail- machine. Thus, caution should be taken when using a large- ure (e.g., ductile vs. brittle) was not greatly affected by loading capacity (Tinius Olson, 1780 kN capacity) machine under load rate modifications. Therefore, a range of loading of 0.25 to 0.64 control. When using a large-capacity machine for testing mm/min is recommended (as detailed in Appendix B) based CLSM, one must ensure the machine is properly calibrated in on the findings that these rates generated accurate and repeat- the lower range of load values typically encountered for CLSM. able strength values in a reasonable amount of time. Effects of Loading Rate. Because of the general lack of Effects of Cylinder Curing and Conditioning. The over- guidance provided in ASTM D 4832 regarding loading rate, all objective of this study was to determine the most efficient

OCR for page 35
40 and accurate method(s) of curing and conditioning CLSM test tures containing Class C fly ash. Thus, using laboratory-cured cylinders. CLSM samples were exposed to the four different cylinders to assess the field performance of CLSM containing curing conditions (A, B, C, and D), as previously described in fly ash (especially high-calcium fly ash) in hot environments Table 3.18. All cylinders were capped with sulfur capping must be done with caution. The effects of temperature on compound and tested at a loading rate of 0.38 mm/min at CLSM hydration are especially important when large amounts 28 and 91 days. The curing regime recommended by ASTM of fly ash are used and when the fly ashcement ratio is high. D 4832 is labeled as curing condition C in this study. This temperature-driven impact on strength triggered a more The compressive strengths measured at 28 and 91 days are comprehensive study on the effects of temperature on strength shown in Tables 3.24 and 3.25, respectively. The remainder of gain, as described later in this chapter. this section discusses these results and also describes some of For high air-content mixtures G-3 and G-6, the effects of the nuances observed when testing different materials and curing methods varied with cement contents (and strength lev- mixture proportions. Some interesting observations were els). Mixture G-3 contained 30 kg/m3 of cement and exhibited made that illustrate that the strength of CLSM is significantly relatively low strengths. In fact, mixture G-3 suffered such a re- affected by variations in curing conditions, and further, that duction in strength when stored outdoors (compared to fog these variations are a function of specific materials and mix- room curing) that cylinders could not be tested at an age of ture proportions. 28 days. At 91 days, cylinders could be tested, but the resultant In general, there were substantial differences between the strengths were significantly lower than fog roomcured cylin- strength of CLSM cylinders stored outdoors (in hot Austin, ders. Mixture G-6, which contained 45 kg/m3 of cement, ex- Texas, weather) and cylinders cured in the fog room. How- hibited less difference in strength (comparing outdoor curing ever, there was not a consistent trend for all the mixtures to fog room curing) than the lower cement-content mixture studied, illustrating that the effects of temperature and cur- (G-3). For mixture G-6, there was still a reduction in strength ing conditions are sensitive to material type and proportions. for outdoor-stored cylinders compared to fog-room cured Some mixtures lost strength when stored outdoors, whereas cylinders at 28 days, but this difference became negligible at others showed significant increases in strength when stored 91 days. These findings suggest that high air-content CLSM outdoors. mixtures (without fly ash) benefit more from moist curing The largest increase in strength for mixtures stored outdoors than they do from high-temperature exposures. was observed for mixtures G-2, G-8, and G-10, which con- Overall, these findings regarding curing temperature and tained Class C fly ash, where the high temperatures helped to cylinder storage led the research team to initiate a final inves- activate the fly ash. Mixture G-2, a rapid-setting mixture with tigation on the effects of curing temperature and humidity on 275 kg/m3 of Class C fly ash and no portland cement, showed compressive strength, as discussed in the next section. a 40 percent increase in strength when stored outside rather The effect of drying time on the compressive strength of than in the fog room. Mixtures containing Class F fly ash also CLSM cylinders was evaluated; the results are shown in exhibited higher strengths for cylinders cured outdoors in a hot Table 3.26. CLSM cylinders were first cured in the fog room climate, but the differences were more pronounced for Class C for various time periods (7, 28, and 91 days) and then allowed fly ash. In fact, the difference between fog roomcured and to dry at room temperature for various time periods before outdoor-cured cylinders was as high as 250 percent for mix- compression testing. Table 3.24. Compressive strength at 28 days using different curing conditions. Curing Condition A Curing Condition B Curing Condition C Curing Condition D (Normal) (Mold) (Cap) (Outside) Mixture Average C.O.V Average C.O.V Average C.O.V Average C.O.V (kPa) (%) (kPa) (%) (kPa) (%) (kPa) (%) G-1 559.1 3.2 365.1 2.3 344.3 6.0 536.0 5.9 G-2 246.8 16.5 267.9 4.7 269.2 14.3 380.1 5.2 G-3 89.5 24.1 93.6 24.6 59.4 6.5 a b G-4 247.8 11.6 167.0 16.1 164.9 4.8 b b G-5 893.5 1.7 877.0 3.3 991.3 4.4 1259.0 4.8 G-6 369.5 7.8 326.1 10.2 306.6 7.3 295.5 8.2 G-7 150.8 11.5 145.4 4.8 161.5 3.7 280.0 8.3 G-8 170.3 9.2 137.3 6.4 160.9 16.1 600.1 8.3 G-9 317.9 12.9 328.6 13.3 279.0 18.7 496.0 14.5 G-10 486.3 3.6 412.2 9.6 411.6 6.1 986.9 4.4 a Mixture was too weak to be tested. b Not enough specimens were available for testing at this age.

OCR for page 35
41 Table 3.25. Compressive strength at 91 days using different curing conditions. Curing Condition A Curing Condition B Curing Condition C Curing Condition D (Normal) (Mold) (Cap) (Outside) Mixture Average C.O.V Average C.O.V Average C.O.V Average C.O.V (kPa) (%) (kPa) (%) (kPa) (%) (kPa) (%) G-1 754.6 7.7 500.7 5.0 479.4 6.0 808.8 2.3 G-2 346.1 7.8 308.6 11.1 266.5 15.3 422.7 4.5 G-3 101.9 7.3 97.4 2.0 96.3 15.5 49.6 6.1 G-4 305.8 11.4 272.9 7.9 239.0 8.1 408.5 7.1 a G-5 1248.8 19.2 1342.3 14.2 1244.3 7.9 a G-6 378.0 13.9 367.7 3.7 366.6 7.1 378.0 13.9 G-7 175.8 10.7 179.3 4.1 199.6 7.6 271.9 8.3 G-8 218.0 2.0 210.5 8.8 201.5 12.6 943.0 2.7 G-9 408.8 6.3 347.3 14.0 353.4 24.1 501.1 13.7 G-10 785.9 17.8 688.0 6.2 617.7 5.4 1080.1 6.3 a Not enough specimens were available for testing at this age. The standard test for compressive strength of concrete the cylinders were immediately transported to the appro- cylinders, ASTM C 39, requires samples to be tested in a priate temperature-controlled chambers. After 3 days of moist state. Neville (1996) pointed out that testing samples storage in the chambers, half the cylinders from each mix- when moist would provide more reproducible results because ture were stripped or removed from the molds and re- the moisture conditions (especially near the surface) of dried turned to the same chamber until the time of testing. This samples may vary. For air-dried CLSM samples compressive regime is referred to as "dry" curing in subsequent discus- strength first increased with increasing drying periods and sions. The other half of the cylinders from each mixture then dropped. The largest compressive strength values were were kept inside the molds with the caps firmly in place obtained from samples that were dried for 2 hours. Re- until the day of testing (designated as "wet" curing). These searchers observed that, similar to drying concrete, drying cylinders were placed directly next to the cylinders that had CLSM samples could increase the measured compressive already been stripped. strength values as much as 17 percent. This observation indi- The results of compression testing at 7, 28, and 91 days are cates that it is not necessary to air-dry CLSM cylinders for shown in Table 3.27. Also included in this table are the mois- 4 to 8 hours before capping as required by the ASTM D 4832. ture contents of cylinders that were just tested in compression, This procedure eliminates any potential differences in strength which were measured to assess the effects of curing conditions due to variable moisture conditions of cylinders and increases on the moisture content (or evaporable water content) and the number of cylinders that can be tested in a given day at a strength of CLSM. testing facility. Mixtures containing fly ash exhibited significant strength gain at 38C, compared to lower curing temperatures. The in- Effects of Curing Temperature and Humidity on Com- crease in strength was more pronounced for Class C fly ash, pressive Strength. This study was a follow-up to previ- compared to Class F fly ash, mainly because of the differ- ous testing that suggested that temperature plays a major ence in their reactivity. For example, at 38C curing temper- role in CLSM strength development. Three curing tem- ature, the increase in compressive strength of Class C fly ash peratures (10C, 21C, and 38C) and six CLSM mixtures containing mixture H-5 was 160 percent. However the increase (H-1 to H-6 in Table 3.14) were selected to study the in compressive strength of Class F fly ashcontaining mixture strength gain of CLSM across a range of practical construc- H-2 was only 40 percent. The CaO content of fly ash is gen- tion conditions. After casting the mixtures into plastic molds, erally the most important factor affecting the compressive Table 3.26. Influence of air drying on compressive strength (mixture B-2). Drying Time 0-0.5 h 2h 4h 8h Age Strength C.O.V. Strength C.O.V. Strength C.O.V. Strength C.O.V. (kPa) (%) (kPa) (%) (kPa) (%) (kPa) (%) 7 days 310.1 5.8 329.2 11.3 357.5 2.7 363.2 2.0 28 days 1575.1 1.9 1536.5 7.8 1447.2 5.7 1649.8 3.3 91 days 3289.4 2.0 3226.9 4.6 3430.4 3.9 3536.1 4.6

OCR for page 35
42 Table 3.27. Effects of temperature and relative humidity on CLSM compressive strength. Age Temp. a 7 days 28 days 91 days Mixture (C) MCb CSc C.O.V.d MCb CSc C.O.V. d MCb CSc C.O.V. d (%) (kPa) (%) (%) (kPa) (%) (%) (kPa) (%) 10 D 36.9 328.6 3.5 17.8 548.5 7.4 2.0 258.5 2.7 10 W 38.0 210.5 6.0 37.1 240.9 7.8 27.0 323.4 5.5 21 D 18.1 367.5 2.8 1.6 607.3 6.5 1.1 480.2 6.5 H-1 21 W 38.0 299.2 23.0 37.8 266.2 11.8 26.8 632.3 3.3 38 D 19.5 828.5 2.2 1.8 722.3 4.4 1.1 756.7 12.2 38 W 22.6 440.4 13.5 35.3 802.2 2.4 26.0 917.3 0.4 10 D 7.4 188.3 10.2 1.2 314.1 7.0 2.7 118.9 13.6 10 W 10.3 151.9 8.6 9.6 194.2 12.3 5.5 260.5 3.6 21 D 1.8 214.1 2.3 0.3 171.4 10.1 6.5 142.7 7.5 H-2 21 W 9.8 172.4 2.1 9.1 233.1 6.5 3.3 345.3 10.2 38 D 0.3 256.6 12.2 0.2 211.9 4.9 12.2 263.6 3.2 38 W 8.6 220.9 11.7 8.9 458.2 7.3 0.4 634.6 2.9 10 D 6.0 1384.3 1.5 1.6 2315.4 4.9 1.1 1395.4 13.0 10 W 8.7 937.7 16.2 9.1 1141.2 8.9 7.9 1367.9 3.2 21 D 1.6 1213.4 5.1 0.6 963.6 9.6 0.5 852.9 8.9 H-3 21 W 8.7 695.7 8.8 8.6 786.2 3.2 7.4 919.8 6.7 38 D 0.3 1222.5 6.8 0.0 944.5 3.8 0.4 1042.0 13.5 38 W 6.3 864.0 2.1 6.2 3844.8 7.4 2.3 3880.9 20.4 10 D 7.4 314.0 2.8 1.4 1486.2 2.8 1.1 895.8 14.8 10 W 10.0 185.9 9.4 37.7 893.2 9.6 8.1 1670.3 4.9 21 D 1.2 669.6 3.0 0.3 628.6 9.8 0.4 458.6 9.9 H-4 21 W 8.8 501.7 4.1 7.7 1570.7 7.3 3.8 3743.6 4.9 38 D 0.4 2615.0 8.5 0.9 2041.2 3.5 0.3 2060.6 5.6 38 W 5.9 2098.8 9.2 3.7 12116.8 11.1 1.4 11512.6 7.0 10 D 8.7 273.1 6.9 1.4 711.4 3.2 0.8 421.9 5.8 10 W 10.0 232.6 4.5 9.7 544.5 9.6 8.6 1362.6 6.3 21 D 1.4 420.8 3.9 0.3 411.2 5.3 0.3 330.7 12.1 H-5 21 W 10.7 316.8 6.7 10.2 815.2 2.9 8.8 1497.7 4.6 38 D 0.3 1524.7 9.1 0.2 1423.7 4.6 0.1 1339.4 11.9 38 W 7.4 1472.5 7.3 7.7 2282.0 8.6 3.5 2638.2 12.0 10 D 6.3 281.9 16.7 1.6 740.5 8.7 1.1 669.3 3.4 10 W 8.7 210.6 6.9 8.1 470.5 1.2 7 922.2 5.6 21 D 1.0 480.4 16.9 0.3 434.5 15.0 0.3 372.9 18.7 H-6 21 W 7.5 371.4 15.8 6.2 744.7 7.2 6.2 929.6 9.4 38 D 0.4 816.8 11.9 0.2 828.0 28.1 0.2 782.0 4.1 38 W 6.1 562.5 10.7 4.8 786.3 12.6 0.4 991.3 7.7 a D = cylinders stripped after 3 days, W = cylinders kept in mold until time of testing. b MC = moisture content c CS = compressive strength d C.O.V. = coefficient of variation strength of CLSM mixtures, especially at high temperatures. such, CLSM mixtures that are produced with locally available In general, the compressive strength values of CLSM mix- materials for specific field applications should be tested in field tures without fly ash were less sensitive to curing temperature conditions. Issues such as the long-term strength gain of CLSM than mixtures containing fly ash. mixtures in the field conditions should be addressed prior to the Air drying of CLSM cylinders from the third day of curing use of CLSM mixtures. An assessment of the on-site strength of generally increased their 7-day strength, compared to the CLSM should take into account laboratory-obtained test samples that were kept continuously in molds for 7 days. results, but it should also take into account climatic conditions. However, the 91-day compressive strength of air-dried cylin- An understanding of material reactivity is helpful in extrapolat- ders was generally lower compared to the samples that were ing laboratory results to field performance. kept in molds. At 28 days, air-dried cylinders and the samples that were kept in molds gave mixed results. Effects of Drainage Conditions on Compressive Strength. This study reinforced the need to recognize that field in- To evaluate the effect of different drainage conditions on the stallations of CLSM may possess vastly different strengths than compressive strength of CLSM a specially designed "curing one might predict from laboratory-cured tests, especially box" was constructed, as described earlier in this chapter. Re- when CLSM contains fly ash and is used in hot climates. As searchers evaluated four different storage conditions, which

OCR for page 35
43 ranged from "normal" curing in a fog room to curing in cylin- eral, sulfur capping generated the lowest variations com- ders that allowed seepage from the bottom and/or sides and pared to the other capping methods. Lower strength cylin- evaporation from the top (to mimic field conditions in a trench, ders tested with higher hardness value neoprene pads ex- for example). The "curing box" was kept at a higher tempera- hibited higher variations in the results. Compressive strength ture than fog roomcured cylinders, and therefore, this study results obtained using durometer 20 neoprene pads per- also was intended to assess temperature-related effects. formed better, especially with weaker cylinders, and exhib- The main finding from this study was that the effects of ited only slightly larger variations than the results obtained water seepage to adjacent sand and loss of water by evapo- using sulfur capping. ration did not significantly impact the strength of CLSM. ASTM D 4832 states that capping systems are acceptable The study also confirmed that temperature plays a key role when the average strength obtained is not less than 80 percent in many CLSM mixtures and suggested that drainage and of the average strength of companion cylinders capped with evaporation may not be as critical as temperature-induced sulfur capping compound. According to this criterion, only the effects. use of gypsum and neoprene pads with a durometer value of 50 could be qualified using the ASTM C 1231 method. How- Alternative Capping Materials for Compression Testing. ever, the qualification method described in ASTM C 1231 is The capping materials evaluated in this study included neo- developed for concrete samples; if the ASTM C 1231 process is prene pads, sulfur caps, and gypsum caps (or "hydro- slightly modified to recognize the unique properties of CLSM stone"). Neoprene pads with Shore A durometer values (see Folliard et al. [2001] for more details), neoprene pads with of 20, 40, 50, 60, and 70 were evaluated. CLSM cylinders a durometer value of 20 could be qualified as an acceptable were capped and tested after 7, 28, and 91 days of curing capping material. using a load rate of 0.38 mm/min. Gypsum paste prepared In an additional study, samples of four CLSM mixtures for capping had a gypsum-water ratio of 0.3 and required were tested after 7 days of curing using sulfur capping com- approximately 40 minutes to harden. Because gypsum pound, neoprene pads with a Shore A durometer hardness capping was a time-consuming process, it was only used for of 50, polyurethane pads, and unbonded polyurethane- 28-day compression testing. Table 3.28 summarizes the neoprene pads. strength data for the various capping methods and materi- Figure 3.10 shows the ratios of compressive strength values als. Table 3.29 shows the corresponding coefficients of vari- obtained using different capping methods to the compressive ation of measured compressive strength using the various strength values obtained using sulfur capping for similar capping methods. samples. The abscissa of the plot is the mean compressive For almost all cases, sulfur capping yielded the highest strength of the samples capped with sulfur compound. Re- strength values for all eight mixtures tested. Also, in gen- sults indicated that, for compressive strength values lower Table 3.28. Compressive strength results using different capping materials. Capping Mixture Age Material E-1 E-2 E-3 E-4 E-5 E-6 E-7 E-8 7 days 0.31 0.11 0.08 0.13 0.20 0.34 0.43 0.19 Sulfur 28 days 0.53 0.14 0.29 0.33 1.03 1.19 0.66 91 days 0.85 0.14 0.45 0.48 1.58 1.71 1.24 Gypsum 28 days 0.52 0.11 0.24 0.36 0.95 1.11 0.59 Neoprene 7 days 0.28 0.09 0.06 0.12 0.20 0.28 0.36 0.17 Pad 28 days 0.42 0.11 0.24 0.34 0.92 1.05 0.42 D70 91 days 0.10 0.34 0.46 1.12 1.39 0.91 Neoprene 7 days 0.29 0.09 0.05 0.11 0.16 0.33 0.36 0.18 Pad 28 days 0.57 0.12 0.26 0.27 0.66 1.02 0.42 D60 91 days 0.10 0.37 0.31 1.29 1.50 0.89 Neoprene 7 days 0.26 0.08 0.05 0.13 0.21 0.34 0.40 0.19 Pad 28 days 0.51 0.10 0.25 0.28 0.87 0.96 0.47 D50 91 days 0.10 0.35 0.32 1.23 1.44 1.08 Neoprene 7 days 0.27 0.07 0.04 0.13 0.22 0.33 0.34 0.21 Pad 28 days 0.51 0.10 0.22 0.25 0.91 1.01 0.50 D40 91 days 0.79 0.14 0.33 0.45 1.36 1.37 0.80 Neoprene 7 days 0.25 0.09 0.04 0.11 0.20 0.32 0.40 0.17 Pad 28 days 0.71 0.09 0.24 0.31 0.90 0.96 0.57 D20 91 days 0.13 0.40 0.36 1.30 1.43 1.05 "" = Not enough specimens were available for testing at this age.

OCR for page 35
44 Table 3.29. Coefficients of variation (%) for compressive strengths using different capping materials. Capping Mixture Age Material E-1 E-2 E-3 E-4 E-5 E-6 E-7 E-8 7 days 1.6 5.8 5.0 7.1 15.7 17.6 8.4 6.9 Sulfur 28 days 16.3 14.2 5.7 20.7 13.3 4.0 10.3 91 days 3.6 15.3 10.6 10.5 6.9 5.3 8.1 Gypsum 28 days 18.0 7.7 6.1 5.5 5.3 7.8 14.9 Neoprene 7 days 5.3 12.4 20.3 5.9 3.7 12.0 9.7 9.5 Pad 28 days 45.9 19.3 22.0 44.5 16.1 14.9 20.8 D70 91 days 22.8 6.8 24.9 8.3 3.0 3.7 Neoprene 7 days 18.0 6.5 15.5 7.2 17.3 12.0 13.2 22.1 Pad 28 days 12.9 2.9 2.5 4.0 23.9 7.6 22.0 D60 91 days 30.4 12.9 16.4 12.5 6.6 21.4 Neoprene 7 days 17.5 5.7 28.7 15.9 7.4 0.7 4.6 30.0 Pad 28 days 18.1 15.5 5.0 20.9 15.1 6.4 6.5 D50 91 days 17.0 22.2 12.6 4.2 4.0 11.5 Neoprene 7 days 17.9 27.6 23.3 21.4 10.1 13.0 6.8 15.2 Pad 28 days 15.4 14.0 9.5 20.5 19.5 10.8 11.2 D40 91 days 12.7 2.7 4.6 27.8 6.6 14.6 9.4 Neoprene 7 days 16.5 1.9 20.5 15.8 10.3 7.3 9.6 16.6 Pad 28 days 7.1 1.1 14.5 10.1 19.2 6.8 5.9 D20 91 days 6.1 7.4 22.2 9.1 6.0 23.5 "" = Not enough specimens were available for testing at this age. than approximately 200 kPa, the non-sulfur capping meth- CLSM with compressive strength lower than 1.0 MPa ods generally underestimate the compressive strength. How- should be tested using unbonded polyurethane pads ever, for compressive strength values greater than 200 kPa, (Shore OO 50, equal to Shore A durometer 5) the use of non-sulfur capping methods provided results that CLSM with compressive strength between 1.0 and 2.0 MPa were acceptable following the criteria given in ASTM D 4832. should be tested using either polyurethane pads (Shore OO As noted, for different capping methods to be acceptable, the 50) or neoprene pads (Shore A durometer 50) ASTM D 4832 standard requires the obtained compressive CLSMwithcompressivestrengthgreaterthan2.0MPa should strength values to be not less than 80 percent of the corre- be tested using neoprene pads (Shore A durometer 50) sponding values obtained using sulfur caps. Based on these results, the following recommendations can The selection of durometer 50 neoprene pads for higher be made with regard to generating acceptable strength data strength CLSM mixtures was due to the general availability of using unbonded pads: these pads in concrete laboratories and because the pads can be 1.1 Ratio of Strength Measured Using Sulfur Caps 1.0 0.90 0.80 Polyurethane pads 0.70 Unbonded PN Bonded PN D 50 0.60 0.50 0 200 400 600 800 1000 1200 Compressive Strength by Sulfur Caps (kPa) Figure 3.10. Comparison of strength values from different capping methods.

OCR for page 35
45 qualified in accordance with ASTM D 4832. Polyurethane pads good predictor of excavatability. Another example of lack of were found to be too weak to use under high compression loads. correlation was the fact that mixture 24 had a higher stiff- ness than mixture 22r, yet it was much easier to excavate. The findings of this initial study led the researchers to per- Excavatability form more comprehensive research on excavatability, in- This section summarizes the results of tests that are directly cluding the assessment of other test methods and indirect or indirectly related to the excavatability of CLSM. Included indices, as described next. are the initial findings from tests conducted on selected The researchers performed a comprehensive follow-up CLSM mixtures (six from the original mixture series) and the study to the initial excavatability investigation. A wide range results of subsequent, more comprehensive testing on exca- of CLSM mixtures (C-series) was included in the investiga- vatability and related indices. The splitting tensile test was tion, and the following methods or approaches were assessed also evaluated as a potential index of excavatability, and as as possible indices (direct or indirect): such, the tensile results are provided in this section. As previously described in this chapter, the excavatability Unconfined compressive strength (field-cured cylinders) of CLSM was assessed for six of the original thirty-eight mix- Field penetrometer (field version of ASTM C 403 needle tures by casting CLSM into 450 450 300 mm plywood penetrometer) boxes. The early strength or stiffness of CLSM was assessed DCP using a soil penetrometer, and these values were correlated CBR (estimated from DCP) with "walkability" or the time at which an average person can Stiffness gauge (GeoGauge) walk on the material. Soil penetrometer values in the range of RE 4.32 to 7.35 kPa were found to correlate with initial walkabil- Splitting tensile strength ity. Long-term excavatability was assessed for the six CLSM mixtures at an age of approximately 9 months using typical Table 3.31 summarizes additional results from the exca- hand tools, including a shovel and a pick. Just prior to assess- vatability study, including DCP values, stiffness values (using ing the excavatability, the "stiffness" of the samples was mea- GeoGauge), and calculated RE values. The table also shows sured using the GeoGauge instrument (as described earlier). the compressive strength for laboratory-cured cylinders (at Compressive strengths of laboratory-cured cylinders were 28 days) and field-cured cylinders, which were cured adjacent also measured at the time of excavation. to the excavatability boxes and tested at the time of excava- As shown in Table 3.30, there was no clear correlation be- tion (240 days). The densities of the laboratory-cured and tween compressive strength, excavatability, and stiffness (as field-cured cylinders were measured before testing them in measured by the GeoGauge). For example, the laboratory- compression, and these values were used in RE calculations. cured compressive strength of mixture 23 was quite low, but The relative ease of excavation was assessed using a hand the field-cured excavation box was not excavatable. Previ- shovel. ous testing has shown that laboratory-cured cylinders may The GeoGauge was used to assess the relative stiffness of not be accurate indicators of in-situ strength or stiffness, es- the CLSM specimens. As CLSM mixtures were quite strong pecially when CLSM is exposed to higher temperatures in (relative to soil), a thin layer of wet fine sand was placed on the field (as was the case for these samples). Also, the results the surface prior to testing, as per the recommendations of the suggest that compressive strength, by itself, may not be a manufacturer. Three readings were taken for each mixture. Table 3.30. Results of initial excavation study. Relative Ease of Compressive Strength Stiffness Stiffnessa Excavation Mixture Strength C.O.V. C.O.V. (MN/m) (with shovel (MPa) (%) (%) and/or pick)b 24 0.31 4.86 11.62 10.27 1 22r 1.01 5.49 9.76 5.40 7 6 0.92 9.66 13.20 6.72 9 4 0.70 5.24 30.72 9.10 10 26 1.61 5.91 34.57 2.09 10 23 0.12 22.31 17.15 7.51 9 a Stiffness was measured using the GeoGauge device. b Each mixture was assigned an ease of excavation value from 1 to 10, where 1 is easiest (able to excavate with minimal pressure applied to shovel and/or pick) and 10 is most difficult (not able to excavate with shovel and/or pick, even under heavy pressure).

OCR for page 35
46 Table 3.31. Results of follow-up excavatability study. Mixture Tests C-1 C-2 C-3 C-4 C-5 C-6 C-7 C-8 C-9 28-day compressive strength (lab-cured) 533 807 144 292 1027 332 1192 658 417 (kPa) Density (kg/m3) 1512 1946 1724 1858 2094 2252 2143 2171 1660 240-day compressive strength (field-cured) 1134 493 63 447 3290 362 1397 1615 199 (kPa) RE (using 28-day lab- 0.84 1.51 0.53 0.85 1.9 1.2 2.12 1.6 0.85 cured cylinders) RE (using 240-day 1.22 1.18 0.35 1.05 3.39 1.25 2.29 2.51 0.59 field-cured cylinders) Modified RE (using 28-day lab- 0.79 1.50 0.51 0.83 1.91 1.23 2.14 1.63 0.82 cured cylinders) Modified RE (using 240-day field- 1.16 1.17 0.34 1.03 3.42 1.29 2.32 2.55 0.57 cured cylinders) DCP index (mm) 6.4 6.5 29 5.2 0.5 5 1.1 0.6 10 CBR (%) 37 36 7 46 100 48 100 100 22 Stiffness (using GeoGauge) 19.37 40.56 19.03 28.97 30.87 18.54 11.96 23.89 23.03 (MN/m) Relative ease of excavation (with 3 7 1 6 9 8 8 10 4 shovel)a a Each mixture was assigned an ease of excavation value ranging from 1 to 10, where 1 is easiest (able to excavate with minimal pressure applied to shovel) and 10 is most difficult (not able to excavate with shovel, even under heavy pressure). The variations were quite high for the device, with coeffi- Another parameter that may potentially be used as an cients of variation as high as 40 percent for some specimens. index for excavatability is the splitting tensile strength of There was no clear trend between stiffness values and DCP, CLSM. Some preliminary trials found that tensile strength nor was there a clear trend between stiffness values and actual may, in fact, be more suitable than compressive strength in excavatability (by shovel). In general, the GeoGauge was not assessing excavatability. Although splitting tensile tests were found to be an effective means of assessing the properties of not performed on the C-series mixtures, some tests were per- CLSM, both because of poor reproducibility and inability to formed on other mixture series. The results are provided in predict excavatability. this section because of the potential of applying tensile data The DCP index value, which indicates the penetration depth to excavatability predictions. per blow, was measured for each of the excavation boxes. The The splitting tensile strengths of a range of CLSM mixtures minimum value for a recordable blow corresponded to a pen- (E-series) were measured, as shown in Table 3.32. A split cylin- etration of at least 25 mm. DCP values were found to decrease der from mixture E-1 is shown in Figure 3.12. For the E-series until the specimens ultimately suffered large cracks. After the CLSM mixtures, the splitting tensile strength to compressive large cracks appeared, the DCP values progressively decreased. strength ratio ranged from 9 percent to 17 percent, which is Thus, the lowest index value was taken for each mixture and higher than those typically observed for conventional concrete. used in Table 3.31 because it represented the most difficult por- Unlike concrete, this ratio did not substantially decrease with tion to excavate, thus providing a conservative index. The correlation between DCP index and the RE values (based an increase in compressive strength. on 240-day field-cured cylinders) is shown in Figure 3.11. As Additional splitting tensile tests were performed using the E- shown in the figure, a DCP index of 5 mm per blow correlated series mixtures to assess the effects of drying on tensile strength well with an RE value of 1.0. This correlation suggests that the and the tensilecompressive strength ratio. This testing was ini- DCP may be an effective, user-friendly method of assessing tiated because drying generally has a more profound effect on excavatability in the field. This approach was further investi- tensile strength than compressive strength, at least in the case of gated in the field testing component of this project (Chapter 4), conventional concrete. This behavior is generally attributed to where excavatability will be assessed not only using hand tools, the effects of microcracks. The results, shown in Table 3.33, but also using typical, commercial excavation equipment (i.e., confirmed that drying had a similar effect on CLSM, signifi- backhoe). cantly lowering the tensilecompressive strength ratio.

OCR for page 35
47 40.0 35.0 30.0 DCP Index (mm per blow) y = 5.3212 x -1.7953 25.0 R2 = 0.928 20.0 15.0 10.0 5.0 0.0 0 0.5 1 1.5 2 2.5 3 3.5 Modified RE (using field cured cylinders) Figure 3.11. Correlation between DCP index and RE calculated using 240-day compressive strength (field-cured cylinders). California Bearing Ratio and Resilient Modulus (measured after 28 days of moist curing) were in the range of silty clays, silty or clayey fine sands, silts, clayey silts, and clays. CBR and resilient modulus of six CLSM mixtures were Results indicate that water-cement ratio was an important fac- measured following modified AASHTO T 193 and T 292, re- tor affecting the coefficient of permeability. Generally, per- spectively. Table 3.34 shows the measured CBR and resilient meability decreased with decreasing water-cement ratio. In- modulus values. With the exception of some mixtures that terestingly, the high air content of mixture I-5 did not increase contained fly ash or high air content, observed CBR values its permeability significantly, indicating that the entrained air were high, indicating that the tested mixtures would func- bubbles were not well connected. The water permeability of tion as a suitable base or subbase material. More important, the CLSM samples was easily measured using equipment the results and experience confirm that it is feasible to deter- commonly used to characterize soils. Additional information mine CBR and resilient modulus values for CLSM using on the effect of freeze-thaw damages on water permeability equipment commonly used to evaluate soils in typical test- of CLSM samples is provided in the section "Freezing and ing laboratories. Thawing." Water Permeability Triaxial Shear Strength The water permeability (or hydraulic conductivity) test re- Using the same materials and mixture proportions as the sults of six CLSM mixtures (I-series) are shown in Table 3.35. water permeability study, the triaxial shear strength of sev- According to Bowles (1984), all of these permeability values eral CLSM mixtures was measured. The results, shown in Table 3.32. Compressive and splitting tensile strengths at 7, 28, and 91 days. 7 days 28 days 91 days Mixture Average C.O.V. f st /f c Average C.O.V. f st /f c Average C.O.V. f st /f c (kPa) (%) (%) (kPa) (%) (%) (kPa) (%) (%) E-1 57.0 4.0 18.5 62.5 9.8 11.7 161.5 8.2 19.0 E-2 21.1 5.3 17.5 28.4 19.9 14.5 31.2 19.1 11.9 E-3 14.4 8.2 17.1 20.6 17.5 14.3 27.2 11.7 18.9 E-4 25.0 14.5 19.6 29.5 8.3 10.1 57.6 6.7 12.8 E-5 46.9 12.9 23.6 101.1 23.1 9.8 150.7 23.0 9.5 E-6 34.2 19.3 10.0 57.3 14.7 17.3 61.9 8.3 13.0 E-7 50.5 6.9 11.8 164.2 30.3 13.8 188.5 18.5 11.0 E-8 33.2 16.3 17.3 75.4 8.1 11.5 146.1 1.8 11.8 f'st = splitting tensile strength f'c = compressive strength

OCR for page 35
48 Figure 3.12. A cylinder from mixture E-1, before and after being tested for splitting tensile strength. Table 3.36, confirmed the observation of other researchers For mixtures I-1, I-2, and I-3, the internal friction angles (Bhat and Lovell 1996) that the strength of CLSM is com- and cohesion both increased with time (between 7 and 28 days). posed of both chemical bonding and internal frictional re- Their friction angles at 28 days were in the range of very dense sistance. For the mixtures investigated in the present study, granular soil, and the mixtures behaved like dense sand, with the behavior was found to be a function of specific material lower residual strengths than ultimate strengths. For mixture and mixture proportions, and the effects changed with in- I-4, the strength development was manifested mainly as an creased curing time. increase in internal friction angle, whereas for mixture I-5, an Table 3.33. Effects of temperature and drying conditions on splitting tensile strength of CLSM. Mixture H-1 Mixture H-2 Condition Average C.O.V. f st /f c Average C.O.V. f st /f c (kPa) (%) (%) (kPa) (%) (%) 10C, dry 16.7 23.1 6.5 6.0 20.2 5.1 10C, wet 25.6 19.3 7.9 21C, dry 23.8 8.6 4.9 8.8 39.6 6.1 21C, wet 89.0 27.9 14.1 38C, dry 55.0 18.8 7.3 18.9 11.4 7.2 38C, wet 74.5 6.1 8.1 53.4 15.5 8.4 Mixture H-3 Mixture H-4 Condition Average C.O.V. f st /f c Average C.O.V. f st /f c (kPa) (%) (%) (kPa) (%) (%) 10C, dry 100.4 14.1 7.2 65.5 33.0 7.3 10C, wet 166.3 32.1 12.2 95.8 101.4 5.7 21C, dry 49.2 8.5 5.8 32.2 5.0 7.0 21C, wet 114.5 15.5 12.4 455.9 36.0 12.2 38C, dry 87.8 7.5 8.4 158.6 6.3 7.7 38C, wet 525.4 30.9 13.5 1791.7 12.5 15.6 Mixture H-5 Mixture H-6 Condition Average C.O.V. f st /f c Average C.O.V. f st /f c (kPa) (%) (%) (kPa) (%) (%) 10C, dry 28.8 12.3 6.8 56.1 67.5 8.4 10C, wet 133.0 7.5 9.8 84.5 22.8 9.2 21C, dry 25.6 16.6 7.7 64.1 5.3 17.2 21C, wet 127.8 8.6 8.5 142.8 7.3 15.4 38C, dry 106.8 9.1 8.0 114.1 16.6 14.6 38C, wet 198.2 6.0 7.5 125.3 12.5 12.6 f'st = splitting tensile strength f'c = compressive strength "" = Not enough specimens were available for testing at this age.