United States Environmental Protection Agency Research and Development Risk Reduction Engineering Laboratory Cincinnati, OH 45268 EPA/600/SR-93/167 October 1993 or EPA Project Summary Evaluation of Solidification/ Stabilization Treatment Processes for Municipal Waste Combustion Residues David S. Kosson, Teresa T. Kosson, and Hans van der Sloot The investigations described in this report were carried out to compare and evaluate the effectiveness of solidifica- tion/stabilization (S/S) processes as treatment technologies for residues from municipal waste combustion (MWC). A full, factorial, experimental design was used to evaluate five S/S processes. The two experimental fac- tors were the residue type to be treated and the S/S process. The three experi- mental levels within the residue type factor were (1) bottom ash, (2) air pol- lution control (APC) residue, and (3) combined ash. The six experimental lev- els within the S/S process factor were the untreated residue, a Portland-ce- ment-only control process, and four selected vendor processes. Thus, 2 ex- perimental factors at 3 and 6 experi- mental levels, respectively, resulted in the evaluation of 18 experimental cases. Evaluation of each experimental case included analysis of chemical compo- sition, physical properties, durability, and leaching characteristics. The test- ing included moisture content, loss on ignition, bulk density, modified Proctor density, particle size distribution, per- meability, specific surface area, po- rosity, cone penetrometer, unconfined compressive strength, pozzolanic ac- tivity, unconfined compressive strength after immersion, wet/dry, freeze/thaw, toxicity characteristic leaching proce- dure (TCLP), availability leach test (ALT), distilled water leach test (DWLT), acid neutralization capacity (ANC), and the monolithic leach test (MLT). Based on comparison of untreated residues with treated residues, the S/S processes evaluated generally did not decrease the potential for the release of target contaminants. A phosphate process did, however, reduce the po- tential for lead to be released. Whether treated or not, the typical release po- tential for metals was a small fraction of the total metal concentration present in the residues. This Project Summary was developed by EPA's Risk Reduction Engineering Laboratory, Cincinnati, OH, to announce key findings of the research project that is fully documented in a separate report of the same title (see Project Report ordering information at back). Introduction The proper management of MWC resi- dues is necessary to ensure that the use of combustion as a solid waste manage- ment method protects human health and the environment. Although recent Federal regulations for municipal solid waste (MSW) and landfills (Federal Register, Oct. 9, 1991) address the disposal of MWC residues, regulations specific to ash dis- posal generally remain the responsibility of individual states. Recently, the U.S. Congress has considered legislation that would require the U.S. Environmental Pro- tection Agency (EPA) to develop compre- hensive national ash disposal, treatment, and utilization standards. To have the sci- entific data available to support possible future legislative requirements to promul- gate technical guidance and/or regulations Printed on Recycled Paper ------- for the proper management of MWC resi- dues, the EPA initiated studies on MWC ash characteristics, ash disposal facilities, and ash management practices. This study adds to that database. Specific Objectives and Approach The investigations described in this re- port were carried out to compare and evaluate the effectiveness of S/S pro- cesses as treatment technologies for bot- tom ash, APC residue, and combined ash. Rather than determine how ash charac- teristics are affected by municipal waste combustor designs, operating conditions, and waste input, the program emphasized the evaluation of S/S treatment technolo- gies. Therefore, the residues included in the study were limited to those from a single, modern, waste-to-energy facility. The specific objectives of this study were to: 1. define residue sampling, preparation, and characterization protocols to permit bench and pilot-scale demonstrations of S/S treatment processes with representative residues; 2. carry out MWC residue S/S treatment process demonstrations under carefully controlled and monitored conditions; 3. determine how the S/S treatment process affects the fundamental physical and chemical properties of the MWC residues; 4. compare the effects of the S/S processes on leaching properties of MWC residues through laboratory procedures, which include the TCLP and other tests that can estimate contaminant release potential and release rate over a prolonged period of time and under diverse environmental conditions; and, 5. evaluate the physical durability of the treated MWC residues during aggressive environmental cycling tests. A full factorial design was used to evalu- ate five S/S processes for treating MWC residues. The two experimental factors were the residue type to be treated and the S/S process. Residue types tested were (1) bottom ash, (2) APC residue, and (3) combined ash. The six experi- mental levels within the S/S process fac- tors were the untreated residue, the Wa- terways Experiment Station (WES) Con- trol (Portland cement only) S/S process, and four selected vendor S/S processes. Thus, 18 experimental cases were evalu- ated in triplicate. The three residue types were obtained during a single composite sampling taken from a typical, state-of-the-art, mass-burn municipal waste combustor. The facility has a lime-slurry spray drier (wet-dry), an acid gas scrubber, and a fabric filter par- ticulate removal system. Bulk residue samples were dried, size reduced, screened, and homogenized before use in this program. Thus all process demon- strations, testing, and evaluations were carried out on preprocessed residues to facilitate laboratory-scale testing and the comparisons of treatment effects. Five S/S processes were evaluated. Four of the processes were proprietary modifications of four different generic S/S processes. The four proprietary processes evaluated were: • Portland cement and polymeric additives or other proprietary additives (Process 1); • Portland cement and soluble silicates (Process 2); • Cement kiln dust and proprietary additives (Process 3); • Addition of soluble phosphates (Process 4). The fifth process (WES) used Type 1 Portland cement only (control process). The Type 1 Portland-cement-only process was selected to provide a baseline com- parison of the treatment effects of Port- land cement without vendor additives. All vendors demonstrated their specifi- cally designed S/S processes on each residue type. The vendors were provided approximately 50 Ib of each residue type to develop optimum formulas before the process demonstrations. They were also provided a list of test and program objec- tives that would be used to evaluate each process. The vendors were not provided specific performance criteria to which the residue should be treated; performance criteria were left to the vendor's discre- tion. Based on analysis of results, vendor process optimization may have focused on minimizing contaminant release based on TCLP, concurrent with minimizing cost, and not on maximizing the physical prop- erties of the treated residue. Demonstrations were conducted under the observation of EPA representatives and U.S. Army Corps of Engineers per- sonnel. Each demonstration, done in trip- licate, consisted of the vendor carrying out the specified process to produce ap- proximately 100 Ib of treated residue. Each experimental case was analyzed for chemi- cal composition and tested for physical properties, physical durability, and leach- ing characteristics with the use of the fol- lowing tests: moisture content, loss on ignition, bulk density, modified Proctor den- sity, particle size distribution, permeability, specific surface area and porosity, cone penetrometer, unconfined compressive strength, pozzolanic activity, unconfined compressive strength after immersion, wet/ dry, freeze/thaw, TCLP, ALT, DWLT, ANC test, and MLT. Summary of Leaching Tests The leaching tests were selected to pro- vide a broad understanding of contami- nant release under a variety of potential environmental conditions. The primary ob- jective was to evaluate fundamental leach- ing properties rather than to simulate spe- cific environmental exposure scenarios. These fundamental leaching properties were (1) release potential, (2) elemental solubility as a function of pH, and (3) release rate under diffusion-controlled con- ditions. This approach permits the use of leaching data to estimate contaminant re- lease for a variety of environmental condi- tions instead of only the particular expo- sure scenario tested. TCLP also was car- ried out on untreated and treated residues for comparison purposes. Untreated resi- due samples were tested after mechani- cal processing steps, which included screening, size reduction, and homogeni- zation. Therefore, results from testing the untreated residues may not be indicative of the behavior of "as disposed" residues, which have not been mechanically pro- cessed. The leaching tests and the basis for selecting each are discussed in the following paragraphs. The TCLP was selected to allow these results to be compared with a broad data- base of results from tests of other materi- als. The TCLP was carried out in accor- dance with the method outlined in the U.S. Code of Federal Regulations, Title 40, Part 268, Appendix 1.1988. In this test, a solid sample is crushed to a sample size less than 9.5 mm and is extracted with dilute acetic acid at a 20:1 liquid-to- solid ratio. The extraction solution is either buffered or unbuffered depending on the alkalinity of the material to be tested. Only a fixed quantity of acid is used for the extraction, and therefore, the final pH of the extract may vary widely. Thus, metals concentrations observed in the extract of- ten reflect the pH-dependent solubility con- straints of the specific element. The con- taminant concentrations in the test leachate are compared with a published list of limits that apply to hazardous wastes (not MWC residues). The ALT was selected to assess the maximum amount of specific elements or species that could be released under an ------- assumed "worst case" environmental sce- nario. This test, originally developed by the Netherlands Energy Research Center, is carried out on a sample size reduced to less than 300 nm. Two serial extractions are carried out, each at a 100:1 liquid-to- solid ratio, with the use of distilled water. The pH is controlled to pH 7 during the first extraction and to pH 4 during the second extraction by using an automatic pH controller that delivers dilute nitric acid. Thus, the final extraction pH is controlled rather than the amount of acid used. The first and second extracts are combined for analysis. The very large liquid-to-solid ra- tio ensures that the contaminant release is not constrained by its solubility at the final pH and that the amount of contami- nant extracted is the maximum amount that would be available at that pH. This test generally extracts all species not tightly bound in a mineral or glassy matrix. The test does not provide information on the rate of contaminant release. The DWLT was selected to assess the amount of specific elements or species that might be released under continued exposure to rainfall or to clean water per- colation. Synthetic acid rain solutions were not selected because the very high natu- ral alkalinity of the residues would signifi- cantly limit the effects of acid rain acidity. The DWLT was carried out in accordance with procedures of the sequential batch leaching test used at the Environmental Laboratory, U.S. Army Corps of Engineers. A sample, crushed in size to less than 2.0 mm, is extracted four times in succession, each at a 10:1 liquid-to-solid ratio with the use of distilled water as the extractant. Acid is not added nor is pH controlled. Thus, the natural buffering capacity of the material controls the final extract pH, which was typically between pH 10 and 12 for the materials tested. The first and second extracts were combined for analysis, as were the third and fourth extracts. Results are used to estimate the amount of con- taminant released over prolonged expo- sure to the leachant and to provide limited information on the rate of contaminant re- lease. The ANC test assesses the solubility of specific metals over a broad pH range. The test was carried out on a residue sample crushed and size reduced to less than 300 urn. Eleven separate extractions are performed using separate samples at a liquid-to-solid ratio of 5:1. The low liq- uid-to-solid ratio results in the extraction being solubility-constrained for some analytes. Each extraction receives a dif- ferent amount of dilute nitric acid, varying from 0 to 12 meq/g dry residue, resulting in a range of final pHs. A titration curve also is obtained for each material tested. The MLT was used to assess the re- lease rate of specific elements and spe- cies from the untreated and treated resi- dues under diffusion controlled conditions. This would be the case under field condi- tions where the flow of infiltration or con- tacting water is predominantly around monolithic structures (e.g., blocks, other forms, or low permeability, compacted fill). The MLT was carried out based on a modification of the American Nuclear So- ciety (ANS), American National Standard Measurement of the Leachability of Solidi- fied Low-Level Radioactive Wastes by a Short-Term Test Procedure (ANSI- 16.1,1986). A 4-cm diameter by 4-cm long cylindrical, monolithic sample was tested instead of the specified size test speci- men. The monolithic samples were ex- tracted by contacting them with 8.47 L distilled water for up to 64 days. Contact- ing water was replaced with fresh distilled water at 1, 2, 4, 8, 16, 32, and 64 days and analyzed for metals and other spe- cies. A new test method was developed for evaluating compacted granular materials. Release rate data were obtained for un- treated bottom ash and combined ash by compacting each ash at optimum mois- ture content, using modified Proctor compactive effort, in 4" diameter by 4" long cylindrical polyethylene molds. Speci- mens were cured in the mold for 28 days at 24°C and 98% relative humidity. The exposed face of the specimen in each mold was covered with a 22-mm-thick layer of 3-mm-diameter glass beads and con- tacted with 8.47 L distilled water. Contact- ing water was replaced with fresh distilled water at 6 hr and at 1, 2, 4, 8, 16, and 32 days. Modeling of the release data in con- junction with the results of the availability leach test was used to determine effective diffusion coefficients, tortuosity, and chemi- cal retention factors for estimating long- term release rates for selected species. The full report provides the results of each of these determinations for each element assayed and for each S/S process and residue type evaluated. Test Results When the extracts from laboratory leach tests are chemically analyzed, the con- centration in the aqueous phase can be determined. However, transforming the concentration data into element or spe- cies release (e.g., the mass of an element or species emitted from the solid matrix into the extract per unit mass of solid extracted) permits normalization and the comparison of data obtained from differ- ent leach tests. The TCLP, DWLT, and ALT are all intended to assess the poten- tial for (or the maximum extent of) species release under different extremes of leach- ing conditions. Each test, however, em- ploys different liquid-to-solid ratios and extraction conditions. Only the TCLP has a defined concentration basis for interpre- tation of resulting extract concentrations. Thus, data interpretation on an extract concentration basis is of limited value. Using additives for each treatment pro- cess and varying treated residue moisture contents may result in dilution effects, fur- ther confounding direct comparison of ex- tract concentration data. To provide more uniform data interpretation from these leaching tests, extract concentration data were transformed to a basis of release per mass of residue extracted. Results are based on a comparison of treated residues with the untreated results. The typical release potential for metals was a small fraction of the total metal concentra- tion present in the residues. Detailed re- sults for each element assayed for each process and residue type are provided in the full report. The TCLP, DWLT, and the ALT leach- ing data are interpreted on a release ba- sis because these test results most fre- quently are viewed as the potential for release under the extreme conditions rep- resented by the testing procedures. The ANC test is the principal exception to leaching data interpretation on a release basis. The ANC test was performed at a low liquid-to-solid ratio (5:1) to facilitate determining pH titration curves and satu- rated solution concentrations of a variety of elements as a function of pH. There- fore, it was most useful to present ANC data on a concentration basis. In addition, extract data also have been presented on a concentration basis for cadmium, cop- per, lead, and zinc from TCLP extractions. Table 1 summarizes the levels of im- mobilization achieved by each process for each of the target elements in all three residues, as obtained by the most aggres- sive of all leaching procedures used, the ALT. These levels ranged from total im- mobilization of aluminum in APC residue by Process 4 to no immobilization at all of cadmium in the same residue by Pro- cesses 1, 2, 3, and WES control. No one ------- Table 1. Fractions of Elements Released in the Availability Leach Test after Treatment A PC Residue Aluminum Cadmium Calcium Chloride Copper Lead Sodium Potassium Zinc Bottom Ash Aluminum Cadmium Calcium Chloride Copper Lead Sodium Potassium Zinc Combined Ash Aluminum Cadmium Calcium Chloride Copper Lead Sodium Potassium Zinc Total" Concentration (mg/kg ash) 25,586 137 290,725 90,325 515 2,969 20,467 15,598 17,453 51,749 35 1 13,087 24,301 1,477 1,563 19,777 9,510 6,793 56,083 32 123,357 28,922 1,734 1,054 21,678 13,245 6,172 Process 1 (%) 16 >100f >100f >100f 30 40 >100f >100f 38 5 23 53 61 5 9 17 21 10 11 >100f >wot 55 14 24 32 52 27 Process 2 (%) 31 >100f 95 >100f 55 72 >wot >100f 62 10 29 >100f >100f 18 10 >100f 48 >100f 27 63 >100f >wot 23 47 >100f 57 21 Process 3 (%) 33 >100f >100f >100f 76 76 >100f >100f 69 11 46 >100t >100f 13 20 38 >100f 24 21 63 >100f >100f 22 >100f 34 >100f 32 Process 4 (%) 1 87 85 >100f 25 0.1 94 >100f 38 1 49 78 71 15 3 17 24 27 1 81 85 >100f 14 4 24 36 35 WES Control (%) >100f >100f >100f >wot >100f >100t >100f 9 51 92 >100f 10 21 23 47 31 9 75 >100f >100f 19 32 24 43 34 * All total element concentrations are based on NAA results except lead, which is based on SW-846 results. t Values nominally greater than 100% were calculated because of either contributions from process additives or correction for process dilution. process demonstrated superiority over other processes for all elements. Note, however, that Process 4 retained lead in all the treated residue types significantly better than did the other processes. Data from the other leaching tests, al- though not given in this Project Summary, are presented in detail in the full report. No clear pattern could be noted other than that no single process demonstrated clear superiority over the other processes for all target elements as was noted with the ALT, except for lead in Process 4. Another measure of immobilization used in this study is the level of total dissolved solids (IDS) released during the DWLT. Data obtained from this test showed that all processes were comparable to one an- other with releases ranging from 4% to 32%. Significantly more TDS were re- leased from the APC residue compared to the other ash types. Many of the treat- ments resulted in increased release of TDS when compared to the untreated resi- dues. Table 2 shows the releases from all three residues by each of the processes as well as the control and the untreated samples. Conclusions Based on the results presented in the full report of all the testing conducted on the untreated and treated residues, the key findings and conclusions are as fol- lows: • The S/S processes evaluated generally did not decrease the potential for release of target contaminants based on comparison of untreated residues with treated residues. The phosphate process, however, did reduce the potential for lead to be released. • Whether the MWC residues were treated or not, the typical release potential for metals (lead, cadmium, zinc, etc.) was a small fraction of the total concentration present in the residues. Release rates of the elements were very low for compacted, granular, untreated bottom ash and combined ash. Release rates also were very low for bottom ash and combined ash treated by processes that produced physically durable specimens. The S/S processes evaluated did not successfully treat the residues to reduce the potential for release of TDS and soluble salts. Whether the MWC residues were treated or not, the release potential and release rates were high for TDS and the salts of calcium, sodium, potassium, chloride, and sulfate. The total amounts of these constituents released typically approached the total concentration in the MWC residues. In the case of the APC residues, the treatment ------- Table 2. Comparison of Total Dissolved Solids Released for the Distilled Water Leach Test (g release/kg ash, dry solid), and the Weight % of the Material Released *n parenthesis). Sample source Untreated Process 1 Process 2 Process 3 Process 4 WES control Bottom Ash 58 (6%) 53 (4%) 187 (12%) 126 (7%) 47 (4%) 59 (5%) A PC Residue 289 (29%) 640 (32%) 565 (26%) 578 (24%) 194 (15%) 671 (30%) processes increased the release potential of the salts. • The high concentration and ultimate fate of soluble salts in MWC residues should be carefully considered in the design of treatment processes and the use and disposal of the residues. • Based on results from the program, ARC residues have the least potential for use in applications requiring structurally durable products. The physical retention values for the treated ARC residues indicated limited or no physical retention. The major contaminant release from the ARC residue was salts. In excess of 30% of the total mass was released in the form of sodium, calcium, chloride, arid sulfate salts. • The use of proprietary additives in the evaluated S/S processes did not enhance the strength of the treated residues. The Portland-cement-only (WES control) process produced test specimens with unconfined compressive strengths greater than or equal to those of all the processes with proprietary additives. It should be noted, however, that high strength probably was not an objective of the vendor's processes. Process 4 developed a granular product. • Evaluation of S/S process design, performance, and treatment efficiency should be based on a matrix of several testing protocols. No single test, such as TCLP, can provide all the information required to evaluate contaminant release potential, contaminant release rate, and physical durability. An appropriate test matrix to evaluate S/S processes shoulc include tests that will address these factors. Formulas for most processes evaluated in this study were probably developed on a limited number of testing procedures. Variations in Portland-cement-based and other S/S technologies will influence the Combined Ash 60 (6%) 54 (4%) 208 (13%) 144 (8%) 56 (5%) 79 (6%) degree of durability and chemical leaching potential. Therefore, substantial improvements in S/S process optimization may be obtainable by optimizing process design based on results of multiple test criteria. • The Portland-cement-based proc- esses can be formulated to produce S/S test specimens of MWC bottom and combined residues with high structural integrity and increased resistance to weathering. These types of processes, if properly designed, are likely to be successful in producing monolithic products with physical properties acceptable for various uses. This does not mean, however, that the chemical characteristics would also be acceptable. Physical durability or possessing a monolithic structure does not ensure acceptable performance with respect to contaminant release. • The release rate of most potentially toxic metals will be very slow to negligible for S/S-treated MWC residues, at least those that have minimal leachability in the raw state. • The unconsolidated, granular nature of the ash material required that a method for estimating diffusion- controlled release from compacted granular materials be developed. Such a method was developed for this evaluation, and the application of a modified MLT to determine intrinsic leaching properties for granular materials has proven very consistent. The data are comparable with results from other types of diffusion measurements. The tortuosity data obtained in the experimental setup are consistent with diffusion measurements using radiotracers. TCLP was not a good indicator of release from untreated and treated residues for several reasons. Variable end-point pH for the extraction resulted in wide variation in estimated metals release because of pH- dependent solubility constraints. The low liquid-to-solid ratio for the TCLP (20:1) also may have resulted in solubility limitations for many elements of concern. Finally, TCLP does not provide for determination of the total release of soluble salts and anions. • The most durable test specimens to the cyclic weathering tests and the immersion tests were those with the highest unconfined compression strength (UCS). Thus, UCS may be useful as a preliminary indicator of physical durability. • The MLT for construction materials and stabilized products provides intrinsic information on long-term leaching effects and usefulness in relation to product quality. The MLT also provides useful information for improving product quality. By focusing on the controlling parameter requiring adjustment, initial estimates of release rates and fluxes for varied application scenarios can be obtained. The distinction between physical retention and chemical retention and the release mechanisms (dissolution, wash-off, and diffusion) can be made.' Existing regulatory tests do not provide such useful information. • Physical retention was directly correlated with the compacted, dry densities of the material for the bottom and combined ashes. The test specimens with the greater densities had more physical retention. • The EPA-recommended methods of chemical analysis (SW-846) were not comparable in many cases to the neutron activation chemical analysis (NAA) for total elemental concentrations in the raw and treated residues. The EPA method results indicated significantly lower elemental concentrations than did the NAA methods; this suggests that only partial analytical recoveries occurred. This discrepancy warrants further- investigation into the chemical analysis methods to investigate and develop more applicable methods for similar type solid matrices. The full report was submitted in partial fulfillment of CR 818178-01-0 by Rutgers The State University of New Jersey CR 813198-01-0 by the New Jersey Institute of Technology and an interagency agree- ment between the U.S. Environmental Pro- tection Agency and the U.S. Army Corps of Engineers, Waterways Experiment Sta- tion, under the sponsorship of the U S Environmental Protection Agency GOVERNMENT PRINTING OFFICE: W3 - 750-07I/JJOW3 ------- ------- ------- David S. Kosson is with Rutgers, The State University of New Jersey, Piscataway, NJ 08855; Teresa T. Kosson is with the U.S. Army Corps of Engineers, Vicksburg, MS 39180; and Hans van der Sloot is with the Netherlands Energy Research Foundation (ECN), Petten, The Netherlands 1755ZG. Car/ton Wiles is the EPA Project Officer (see below). The complete report, entitled "Evaluation of Solidification/Stabilization Treat- ment Processes for Municipal Waste Combustion Residues," (Order No. PB93-229 870/AS; Cost: $61.00, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Risk Reduction Engineering Laboratory U.S. Environmental Protection Agetncy Cincinnati, OH 45268 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati, OH 45268 Official Business Penalty for Private Use $300 BULK RATE POSTAGE & FEES PAID EPA PERMIT No. G-35 EPA/600/SR-93/167 ------- |