Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
49 Conclusions Results of the literature review and a survey of SHAs sug- gested the need for improvements in tests and specifications for CFA used in highway concrete to better identify those properties affecting the concrete performance. Areas need- ing improvement include better characterization of (1) the strength development associated with the use of CFA, (2) the carbon fraction of CFA and its influences on air entrainment, and (3) the level of cement substitution with a specific CFA to mitigate ASR. Improved specifications would generally require a new clas- sification approach that better characterizes CFA performance in concrete mixtures. The purpose of classification is to group CFAs that are similar without excessive testing. The purpose of characterization is to measure and report properties that are known to affect performance, and those properties serve as a basis for the classification system. Therefore, the method used for CFA characterization may not necessarily be appro- priate for use as a classification method. For CFA characterization, 30 fly ash sources were evaluated using the test methods and specification limits stipulated in AASHTO M 295-07 and new test methods to examine the important CFA properties. As a result, refinements to existing CFA characterization protocols were proposed, and changes to AASHTO M 295-11 were made to provide better predic- tion of CFA performance within the framework of the existing classification method. Changes to Chemical Requirements Chemical Classification The primary distinction between Class F and Class C CFA is the bulk composition based on the sum of the oxides (i.e., %SiO2 + %Al2O3 + %Fe2O3). However, this classification omits consideration of calcium content. This research has shown a clear, linear relationship between the sum of the oxides and the calcium content; therefore, specifying either characteristic would provide the same result. Although there was no unique relationship between calcium oxide content and CFA performance to serve as a basis for clas- sification, above a threshold of approximately 15%, there was evidence of a decreasing effectiveness for mitigating sulfate attack or ASR (Dhole et al., 2011; Thomas, 2011). Therefore, determining the calcium content is important and is proposed for inclusion in Table 1 of AASHTO M 295. Reporting the cal- cium oxide content and the magnesium oxide, sodium oxide, potassium oxide, and the equivalent alkali (Na2Oe) content is also proposed. It should be noted that AASHTO PP 65 requires the CaO and Na2Oe values for determining strategies for ASR mitigation. Effects on Air Entrainment To assess the effect of a CFA on air entrainment, AASHTO M 295 calls only for determining the LOI content of the ash. However, this single measure has been shown to not be suffi- cient. In this research, an experimental investigation was con- ducted to develop new tests for predicting the AEA demand of CFA. This investigation included evaluation of the foam index test and development of two new tests: the CFA iodine number and direct adsorption isotherm tests. After review and conduct of a wide range of foam index tests, a modified version of the test method by Harris et al. (2008a) was proposed as a standard test. The modifications included using a range of standard solution concentrations to determine the solution that achieves a stable foam in a consis- tent time (15 ± 3 min) for that fly ash. The solution strength is determined by iteration and a procedure is provided in Attachment B. The proposed test time allows for the CFA to achieve a consistent degree of contact with the AEA solution, although not necessarily long enough to achieve equilibrium. The modified Harris test specifies incremental addition of AEA to minimize the error associated with adding AEA to the C H A P T E R 5 Conclusions and Suggested Research
50 slurry and requires use of a mechanical agitator to minimize operator-induced variability. Overall, the foam index test is suitable for assessing CFA and AEA interactions, but it is not necessarily suitable for assessing the influence of CFA on air entrainment. Even with the pro- posed improvements, the test has a high level of subjectivity and variability. To enhance applicability, there was a need for developing a standard approach for conducting the test. The CFA iodine number and the direct adsorption iso- therm tests are recommended to be included in AASHTO M 295 as a Supplementary Optional Chemical Requirement for assessing the adsorption potential of a specific CFA source. The research has shown that CFA sources with a CFA iodine number ⤠0.1 mg iodine/gram CFA have little impact on air entrainment, and a CFA with an iodine number > 0.1 mg iodine/gram CFA may have some effect on air entrainment. However, the CFA iodine number characterizes only the CFA, and it is necessary to evaluate both the CFA and AEA as a sys- tem. Therefore, establishing an upper threshold for the CFA iodine number is not possible without defining the AEA to be used. From a classification perspective, it is recommended to retain the LOI measure as a means of limiting the maximum carbon content, rather than establishing a maximum CFA iodine number. Instead, the CFA iodine number and direct adsorption isotherm test can be used as a means for iden- tifying those CFA sources that have a potential to adversely influence air entrainment. The CFA iodine number is recom- mended for use as a screening test. If the test results in a value > 0.1 mg iodine/gram CFA, a direct adsorption isotherm is required with a specified AEA and the capacity determined for the CFA-AEA combination is reported to provide the user with a quantitative measure of the effect of the CFA on air entrainment. Another consideration is the use of powdered activated carbon (PAC) for pollution control in power plants, which significantly increases the adsorption capacity of the CFA with potentially only a small change in the LOI. PAC has an adsorption potential much higher than carbon in CFA result- ing from combustion. Both the CFA iodine number and direct adsorption isotherm tests are extremely sensitive to the inclu- sion of PAC and can quantify its effect on air entrainment. This research has shown the CFA iodine number and the direct adsorption isotherm test can be used together by per- forming both tests on a suite of fly ash sources that covers the expected range of carbon content. The direct adsorption isotherm test is performed using the AEA of interest. When the CFA iodine number is plotted versus the direct adsorp- tion isotherm test, iodine capacity can be converted directly to AEA capacity. The user then only needs to measure the iodine number for any new CFA source being used with the same AEA. To implement this approach, the CFA iodine number versus direct adsorption isotherm relationships for the AEAs will need to be determined. In this case, the specifi- cation would only require reporting the CFA iodine number. The approach of correlating the CFA iodine number and the direct adsorption isotherm test depends on the consistency of the AEA adsorption characteristics. If the adsorption capacity of an AEA varies with production, the correlation would vary and would need to be repeated. The CFA iodine number test uses standard chemicals making it less subjec- tive. An alternative application of the direct adsorption iso- therm test would be to evaluate specific AEAs with one CFA source to allow users to select the AEA that performs best with this CFA source. Given that a concrete producer tends to use the same source of CFA, but numerous AEAs are avail- able, the approach of matching materials would have practi- cal advantages. Available Alkali Limit The available alkali test currently specified in AASHTO M 295 takes over 1 month to complete, and its precision is questionable. A preferable alternative is to base the classifica- tion on the determined potassium oxide and sodium oxide contents (i.e., total alkali). ASTM C441, ASTM C1567, and ASTM C1293 failed to provide acceptable correlations with available alkali test results. Given the limitations of the avail- able alkali test, and total alkali limits in AASHTO PP 65 that greatly exceed the current available alkali limits, the require- ment for the available alkali limit would be better replaced with the total alkali contents to establish required levels of ASR mitigation in AASHTO PP 65. Changes to Physical Requirements Strength Activity Testing The current strength activity index (SAI) has a minimum strength requirement that can be met by inert materials (i.e., finely ground quartz). This research showed that increasing the 7-day limit from 75% to 85% would be appropriate for rejecting inert material. An alternative to the SAI test method is the pozzolanic activ- ity index (PAI) using 35% by mass replacement of cement with CFA, together with accelerated curing. This test is similar to ASTM C1240, Standard Specification for Silica Fume Used in Cementitious Mixtures, where mortar cubes are moist cured for 7 days at 65°C. However, developing an accelerated PAI test method would need considerable refinement and an inter- laboratory study to provide a 7-day test criterion for Class F fly ashes (the old AASHTO M 295 PAI test includes only a 28-day criterion). The Keil hydraulic index (KHI) was investigated as a poten- tial replacement for the current SAI test. The results of that
51 analysis showed that KHI values are not significantly different from those obtained by the SAI test, but the test better defines the effect of a CFA on strength contribution at different ages and replacement levels. Also, specimens prepared using inert fillers did not pass the KHI test using existing SAI limits. The KHI test showed an influence of the type of portland cement used (a similar influence was seen in the SAI test). This research also indicated the type of inert filler used had negli- gible effect on 28-day strengths. Additional refinement and an inter-laboratory study are required to provide appropriate test criteria for the KHI test. The KHI and PAI tests require further development. There- fore, the SAI test should remain part of the AASHTO M 295 specification. To address the issue of inert fillers meeting the SAI test requirement, the minimum SAI needs to be raised to 85% at 7, 28, or 56 days of age. The current AASHTO specifica- tion requires a CFA to meet the strength index requirement at 7 or 28 days, or at 56 days only if specified by the purchaser. This research showed that Class F ashes not meeting the 7-day limit of 85% would meet the 85% limit at 28 or 56 days. ASR Mitigation The AASHTO M 295 standard test for ASR mitigation uses Pyrex glass as a synthetic alkali-silica reactive aggregate, which is readily available and can be used to compare the perfor- mance of CFAs. However, the ability of a CFA to mitigate del- eterious ASR expansions is influenced by the type of reactive aggregate used in concrete. Also, the performance criterion for the test method is vaguely stated and may not provide a reli- able level of performance. ASTM C1567 uses the proposed job aggregate to determine an effective level of cement replace- ment using a particular CFA. Both tests can be completed in about 2 weeks after preparing the mortar bars. Because AAS- HTO PP 65 makes use of the ASTM C1567 test, it is recom- mended the test be incorporated into AASHTO M 295 with a requirement for a maximum expansion limit of 0.10% after 14 days of exposure in NaOH solution. The research also indicated that the expansion limit of 0.10% after 28 days for evaluating ASTM C1567 results was not obtained by three of the four Class F ashes with 40% replacement of cement, and the results differed from those obtained from the 2-year concrete prism tests (ASTM C1293) and the performance of Class F ash with aggregates of similar type and level of reactivity. These findings indicate that this evaluation criterion is inappropriate. The ASTM C1293 concrete prism test allows testing of both the coarse and fine aggregates but requires an excessive time period to complete (2 years is recommended in ASTM C1293 and AASHTO PP 65). Changes to Testing Requirements Moisture content and LOI tests are currently limited to a sample mass of 1 g. The small sample mass can affect the precision of the results, particularly for low values. Increas- ing the sample mass would eliminate this concern. Increased sample mass can be accomplished by modifying the moisture content and LOI test methods in ASTM C311 by removing the reference to the procedure in ASTM C114, which involves a 1 g sample requirement and a reheat requirement (i.e., con- stant mass requirement). Laboratories may select the length of the ignition time and sample mass using reference materi- als to document both the precision and bias of the procedure. This research showed some inconsistency in sulfur (SO3) determinations caused by a loss of SO3 in the LOI determina- tion. Therefore, it is proposed that the potential for SO3 loss during the LOI procedure be noted in AASHTO M 295. Suggested Research The following suggestions for future research are made to improve specifications and test methods for CFA: 1. Develop a precision statement for the foam index, CFA iodine number, and direct adsorption isotherm tests 2. Establish the relationship between the CFA iodine number and the direct adsorption isotherm tests for a range of AEAs 3. Develop a modified PAI test method using 35% CFA by mass replacement of cement and accelerated curing (e.g., 55°C) with consideration for using a fixed w/cm 4. Establish appropriate specification limits for the mortar air content test and the shrinkage test (with the CFA being treated as a cement replacement) 5. Develop a water-soluble alkali test for CFA, precision information, and specification limits.
