United States Environmental Protection Agency Industrial Environmental Research Laboratory Cincinnati OH 45268 .Si."/ Research and Development EP A-600/S2-84-132 Nov. 1984 &ERA Project Summary Evaluation of Hazardous Waste Incineration in a Lime Kiln: Rockwell Lime Company D. R. Day, L. A. Cox, and R. E. Mournighan During a one-week test burn, hazard- ous waste was used as supplemental fuel and co-fired with petroleum coke in a lime kiln in eastern Wisconsin. Detailed sampling and analysis was conducted on the stack gas for principal organic hazardous constituents (POHCs). particulates, paniculate metals, HCI, SO2, NOx, CO, and THC and on process streams for metals and chlorine. POHCs were also analyzed in the waste fuel. Sampling was conducted during three baseline and five waste fuel test burn days. Results show average destruction and removal efficiencies (DREs) greater than 99.99% for each POHC and little change in pollutant emissions from baseline to waste fuel test conditions. In addition, material balance results show that 95% of chlorine enters the process from the limestone feed and the chlorine exits the kiln in the baghouse dust and lime product at 61 % and 38%, respectively. This Project Summary was developed by EPA's Industrial Environmental Research 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 Cofiring hazardous wastes in high- temperature industrial processes is an attractive alternative to incineration, because it makes use of the heat content of the waste. Many cofiring devices, which include cement and dolomite kilns, glass furnaces, steel furnaces, and some industrial boilers, provide temperatures and residence times similar to those required for incinerators dedicated to hazardous wastes. In addition to the savings derived from the heat value, using existent equipment saves the capital required to build a separate incinerator and may thus provide an environmentally acceptable alternative to conventional hazardous waste disposal. Because of their high energy use, lime kilns are an excellent example of this concept. Such kilns typically operate at temperatures over 1093°C (2000°F), have gas residence times exceeding 1.5 seconds, and have a highly turbulent combustion zone. However, the need exists for data that show the effect of cofiring hazardous waste on the emissions from the lime process. The State of Wisconsin Department of Natural Resources (DNR) and the U.S. Environmental Protection Agency (EPA), Region V, issued a temporary permit to Rockwell Lime Company to conduct a hazardous waste test burn. This test would allow the burning of hazardous liquid waste as supplemental fuel along with petroleum coke. The waste fuel would replace natural gas as a fuel component. Through a contract with the Industrial Environmental Research Laboratory, Cincinnati, OH (lERL-Ci), Monsanto Re- search Corporation (MRC) performed the sampling and analysis of stack gases and process samples during the test burn conducted at the Rockwell Lime Company in Rockwood, Wisconsin. The primary objectives of the sampling and analysis were to (1) determine the effects of cofiring petroleum coke and hazardous waste OP the emissions from ------- the kiln, (2) determine the fate of the principal organic hazardous constituents (POHCs) and determine destruction and removal efficiencies (DREs), (3) determine the fate of chlorine and trace metals in the kiln process, (4) determine the concentration of SOa, NOX, particulates, HCI, metals, total hydrocarbons, and carbon monoxide in the stack gas at baseline and waste fuel test burn conditions, and (5) evaluate kiln operation during hazardous waste fuel burning conditions. This testing provides the Wisconsin DNR and EPA Region V with the data necessary to determine whether a permit can be issued to Rockwell Lime Company to burn hazardous waste. The testing also will provide the EPA-ORD with additional data in their research on the incineration of hazardous waste and the environmental problems associated with incineration. Facility and Process Description The Rockwell Lime Company's lime kiln in Rockwood, Wisconsin, approximately 10 miles north of Manitowoc, produces lime at approximately 1.3x106kg(1,430 tons) per week, which varies based upon product demand. The process involves heating limestone to approximately 1,100°C (2,000°F) in a horizontal rotary kiln. Calcining is achieved by interfacing the hot gases with the limestone, which drives off the C02 from the limestone, leaving the lime product (CaO). The kiln, with refractory linings, is 2.4 m (8 ft) in diameter and 67.1 m (220 ft) long. The kiln rotates at approximately Natural Gas m Supply Petroleum Coke—^*.. Supply | Primary Air Screw Conveyor one revolution per minute and has a gentle slope to allow material to pass through by gravity. It also has a counter- current flow pattern, that is, solids travel in one direction and hot gases and dust emissions travel in the opposite direction, as shown in Figure 1. Limestone is fed into the upper end of the kiln at approxi- mately 15,440 kg/hr (34,000 Ib/hr). At the opposite end of the kiln, a mixture of coal and natural gas is burned at approximately 1,450 kg/hr (3,200 Ib/hr) and 142 mVhr (5,000 ftVhr) to provide a heat input of approximately 14,700 kw (50 million Btu/hr) or approximately 6.5 million Btu/ton of lime product. As the limestone feed travels down the inclined rotating kiln, it passes through various temperature zones, and the hot gases calcine the limestone into the lime product. The product is produced at approximately 7,720 kg/hr (17,000 Ib/hr). After transformation in the kiln, the lime product is air cooled and either directly stored in silos or hydrated prior to storage. Primary air mixed with the gas, coke, and secondary (heated) air from the lime product cooler is fed to the kiln to provide oxygen for the combustion of the coke and natural gas (or waste fuel). The kiln exhaust gases pass through a series of large radiator coolers that cool the gases before they enter the baghouse; this removes particulates and SOa from the gas stream. The gases then pass through the induced draft fans and out the stack at approximately 200°C (392°F) and 5.5 m/s (18 ft/s). The collected dust is stored in a silo and mixed with water to granulate. Some of the dust is sold and Limestone Fe.ed \Exhaust Stack Klln ^°' Dust Air the remainder is disposed of in the quarry. No dust is reinjected into the kiln. During baseline conditions, a blended combination of petroleum coke and natural gas was used to fire the kiln. During the waste fuel runs, a temporary 1-inch diameter stainless steel pipe was placed on the burner pipe with its nozzle pointed into the flame, and the waste fuel and petroleum coke were fed unblended to fire the kiln. The hazardous waste fuel was trucked to the site and stored in a 5,000-gallon tanker between kilns 1 and 2 near the burner end. The diaphram-type waste fuel pump, located next to the tanker, pumped fuel through the stainless steel pipe to the flame. Air was added to the pipe to supply oxygen for combustion and to cool the pipe. The waste fuel consisted primarily of lacquer thinner solvents, alcohols, still bottoms, paint wastes, and a small fraction of chlorinated hydrocarbons (0.4%). Tetrachloroethylene and tri- chloroethylene were spiked to the waste fuel before the test to bring the total chlorine content to approximately 3.0%, which would allow easier evaluation of the destruction of the chlorinated species. During test conditions, the waste fuel ranged from 8% to 36% of the Btu input to the kiln, and petroleum coke ranged from 64% to 90%. Comparatively, under baseline conditions, the petroleum coke averaged 90% and the natural gas averaged 10% of the Btu input to the kiln. Experimental Program Table 1 summarizes the test program. Measured pollutants in the stack gas Radiators Exhaust Gases I.D. Fans \ ~\ •*• Baghouse (8 Modules) NXXXXXX/ Dust to Storage Silo Lime Product Figure 1. Schematic diagram of lime kiln process. 2 ------- include POHCs (tetrachloroethylene (PERC), trichloroethylene (TCE), methylene chloride (MeCI2), 1,1,1-trichloroethane (CH3CCI3), methyl ethyl ketone (MEK), and toluene), paniculate matter, particulate trace metals, hydrogen chloride, sulfur dioxide, nitrogen oxides, carbon monoxide, total hydrocarbons, and oxygen. In addition, the distribution of metals and chlorine was measured in all of the process input and output streams. Waste fuel, coke, baghouse dust, and lime product samples were submitted for sulfur analyses. Coke samples were analyzed for ash and Btu content. Waste fuel samples also were analyzed for POHCs and Btu content. Sampling was conducted under baseline conditions (i.e., no waste fuel burned) on April 15, 29, and 30. Sampling at waste fuel conditions (i.e., waste fuel burned) was conducted from May 2 to May 6. A Quality Assurance (QA/QC) Project Plan was reviewed and approved prior to the test program. A full description of the QA/QC results involving replicates, blanks, spikes, and standards is provided in the full report. Results and Discussion Waste Fuel A detailed summary of the waste fuel composition for two waste fuel samples collected is shown in Table 2. Tables 3 and 4 show the concentration of each POHC and other properties for the five waste fuel samples (one sample per day, Runs 4-8). POHC Destruction and Removal Efficiencies The complex combustion chemistry for organic materials is perplexing when a Table 1. Summary Parameter of Rockwell Lime Kiln Sampling and Analytical Program Sampling method Analytical method Stack Gas POHCs* Particulate matter Metals on particulate Hydrogen chloride Carbon dioxide and oxygen Nitrogen oxides Sulfur dioxide Carbon monoxide Volatile organic sampling train (VOST) EPA Method 5 EPA Method 5 Impinger absorption in 0.5 M NaoAc (back half of EPA Method 5) EPA Method 3 Continuous Continuous Continuous CC/MS, thermal desorption and SIM EPA Methods ICP Specific ion electrode Fyrite Chemiluminescence photo- metric analyzer Pulsed fluorescence TECO analyzer Infrared-EPA Method 10 Total hydrocarbons Oxygen Waste fuel POHCs Metals Chlorine, sulfur Btu content Baghouse dust Metals Chlorine, sulfur Lime product Metals Chlorine, sulfur Dry limestone feed Metals Chlorine Primary fuel coke Metals Chlorine, sulfur Btu content Continuous Continuous Grab Grab Grab Grab Grab Grab Grab Grab Grab Grab Grab Grab Grab — composite — composite — composite -~ composite -~ composite — composite — composite — composite — composite — • composite — composite — composite — composite Flame ionization detector Teledyne's micro-fuel cell GC/MS ICP ASTM D240-64 ASTM D482-IP4 ICP XRF ICP XRF ICP XRF ICP XRF ASTM D240-64 broad range of organic compounds in a liquid waste are burned. On a weight basis, most of the organic carbon in the waste is oxidized to COa in the combustion process, but trace amounts of organic chemicals survive oxidation and are only partially reacted. Accordingly, the test burn investigated the amount of destruc- tion of the organic compounds in the hazardous waste. The ORE for an incineration/air pollu- tion control system is defined by the following equation: DRF= Win'w°u'(inn) W,n (1) where DRE=destruction and removal efficiency, % W,n = mass feed rate of principal organic hazardous constit- uent(s) (POHCs) fed to the incinerator Wout = mass emission rate of prin- cipal organic hazardous constituent(s) (POHCs) to the atmosphere (as mea- sured in stack prior to dis- charge). ORE calculations are based on combined efficiencies of the destruction of the POHC in the incinerator or the lime kiln and the removal of the POHC from the gas stream in the air pollution control system. The presence of POHCs in solid discharges from the air pollution control devices is not accounted for in the ORE calculation as currently defined by EPA. RCRA, Part 264, Subpart 0 regulations for hazardous waste incinerators require a ORE of 99.99% for all principal organic hazardous constituents of a waste during trial burns unless it can be demonstrated that a higher or lower ORE is more appropriate based on human health criteria. Specifi- cation of the POHCs in a waste is subject to best engineering judgment, considering the toxicity, thermal stability, and quantity of each organic waste constituent. ORE requirements in the Subpart 0 regulations do not apply to metals or other noncom- bustible materials. Toluene, MEK. Pare, and TCE were present in high concentration for organic compounds (see Table 2). Spikes of Perc and TCE were added to the waste fuel prior to the test burn to obtain the higher concentrations. Perc, TCE, methylene chloride, and 1,1,1-trichloroethane were selected because the chlorinated hydro- carbons are, in general, difficult to destroy thermally. All six of the com- *Tetrachloroethylene. trichloroethylene, methylene chloride, ketone, and toluene. 1,1,1-triehloroethane. methyl ethyl ------- 30 10- Methylene Chloride 30 8 20< 3 10- Trichloroethylene 30- 20 10 1234 Number of Nines ORE Methylethyl Ketone 30- 10- 12345 Number of Nines ORE Tetrachloroethylene / 2 3 4 5 Number of Nines ORE 30 4 1.1,1 Trichloroethane 12345 Number of Nines ORE 10 I 12345 Number of Nines ORE Figure 2. Destruction and removal efficiencies. 30- | 20- "5 "6 tj 10- Toluene 1 2 3. 4 5 Number of Nines ORE pounds except methylene chloride were found in the top twenty constituents of the waste fuel. All six compounds are listed as hazardous in RCRA Part 261, Appendix VIII. Approximately six VOST sampling runs were made each day (Runs 1 -8). Destruc- tion and removal efficiencies, calculated for waste fuel runs 4-8, are summarized in Figure 2. In general, DREs ranged from 99.60% to >99.999% for all compounds and averaged 99.9989%. Only four runs had DREs less than 99.99%; three of these were for methylene chloride, the fourth for 1,1,1-trichloroethane. DREs for methylene chloride (MeCI2) ranged from 99.60% to <99.999% and averaged 99.983% ± 0.15% (95% con- fidence limits). Methyl ethyl ketone (MEK) had an average ORE of 99.999% ± 0.0002% (95% confidence limit) and ranged from 99.998% to greater than 99.999%. These high destruction efficiencies were consistent throughout the test runs. DREs for 1,1,1-trichloroethane (CH3 CCI3) ranged from 99.989% to 99.999% and averaged 99.997% ± 0.004% (95% confidence limits). Only Run No. 4E had a ORE less than 99.99%. DREs for trichloroethylene (TCE) were greater than 99.999% for all runs. TCE was spiked to the waste fuel prior to testing to increase its concentration and allow easier detection of TCE in the stack gas. Spiking of TCE to concentrations greater than approximately 1.3% by volume was not possible due to permit requirements which specified a maximum of 3.0% by volume for chlorine. DREs for tetrachloroethylene (Perc) also were greater than 99.99% for all runs. Like TCE, Perc was spiked to the waste fuel to the maximum allowable concentration described in the test burn permit prior to the test. Toluene was the POHC of highest concentration in the waste fuel (average 11.6% by weight). DREs for toluene were above 99.999% for all runs. Data for toluene was very consistent during all waste fuel test runs. Stack Samples Results for stack conditions and paniculate, hydrogen chloride, sulfur dioxide, nitrogen oxides, carbon monoxide, and total hydrocarbon emissions for baseline and waste fuel runs are sum- marized in Table 5. The overall stack rate averaged 917 mVmin (32,420 ftVmin) and the dry stack rate averaged 487 dscm/min (17,210 dscf/min). As evi- denced by the high standard deviations the CO and, to a lesser degree, the S02 fluctuated. Minor kiln upsets (i.e., coke feed chute cleaning, clumps of coke falling to kiln, change in process condi- tions) created high CO excursions. An increase in SOz by —200 ppm, followed by a reduction in NOX by ~50 ppm and a subsequent increase in CO by ~500 ppm occurred quite often over a 15 minute period. These trends are expected when a lower intensity flame occurs (or kiln upset). However, as revealed by Figure 2, the kiln upsets had little or no effect on the ORE results. Chlorine, Sulfur and Metals Balance Chlorine and sulfur material balances are summarized for baseline and waste fuel conditions in Table 6. The majority of chlorine (for either baseline or waste fuel conditions) enters the kiln in the limestone feed and exits the kiln in the lime product and baghouse dust. Sulfur (for either baseline or waste fuel conditions) enters the kiln in the petroleum coke and exits the kiln distributed in the lime product (—9%), baghouse dust (—27%), and stack gas (—64%). ------- Typical metals material balance is shown in Table 7. There was no difference for baseline and waste fuel conditions for distribution of metals in the kiln process. As shown in Table 7, the majority of mass entering the kiln is contributed by the limestone feed, except for zinc. The mass exiting the kiln is distributed between the lime product and baghouse dust. Baseline vs. Waste Fuel and Kiln Operation Emissions were evaluated under baseline and waste fuel conditions. For Table 2. Results of Capillary GC/MS Analysis of Major Components of Waste Fuels Concentration, wt% Waste fuel component Acetone Methyl ethyl ketone (MEK) (POHC) 1, 1 ,1 -Trichloroethane CH3 CCI3 (POHCi 1 -Butanol Trichloroethylene TCE (POHC) 2-Ethoxyethanol Methyl isobutyl ketone Toluene (POHC) Tetrachloroethylene (PercJ (POHC) Butyl acetone Ethylbenzene Xylene (isomer No. 1) Xylene (isomer No. 2) 2-Butoxyethyanol 2-Ethoxyethyl acetate Ca-Benzene (isomer No. 1) Cs-Benzene (isomer No. 2) C 10- Alkane Alkane >Ca Alkane Cn- Alkane 2-Cyclohexen-1 -one or 3,5,5-Trimethyl (isomer) Alkane >C6 Number 4' 0.23 2.48 0.24 0.32 1.73 0.85 1.06 11.0 2.17 0.27 1.42 4.92 1.43 1.99 5.91 0.28 0.46 0.80 0.24 0.14 1.26 0.15 0.27 Number 7" 0.22 3.17 0.22 0.37 2.16 0.92 1.16 12.5 2.49 0.32 1.58 5.58 1.60 2.07 6.37 0.32 0.57 0.94 0.28 0.18 1.48 0.18 0.24 aAver age of split sample. Table 3. Concentration of POHCs Run No. 4 5 6 7b 7° 8 POHCs concentration, wt %* MeClz 0.101 0.097 0.106 0.120 0.120 0.116 MEK 2.48 2.75 2.48 3.17 3.17 2.59 CH3 CC/3 0.238 0.239 0.228 0.216 0.216 0.282 TCE 1.73 1.64 1.78 2.16 2.16 1.89 Perc 2.17 2.02 2.05 2.49 2.49 2.56 Toluene 10.97 10.55 10.95 12.50 12.50 12.90 *No waste fuel burned on baseline runs 1 -3. "Runs 7A-7C. cRuns 7D-7F. the pollutants listed in Table 5, HCI, NO* and THC showed a significant difference in stack emissions under baseline and waste fuel conditions. For the POHCs, only methylene chloride and toluene showed an increase from baseline to waste fuel conditions. All remaining POHCs showed no significant difference in baseline vs. waste fuel emissions. As described previously, the kiln operation fluctuated as indicated by CO and SO2 emission variations during waste fuel burning. Kiln fluctuations were caused by several factors, including non-constant fuel rates, product rushes, clumps of coke fed to kiln accidentally, and operator inexperience with burning waste fuel. The fluctuation resulted in occasional kiln 02 increases and stack gas SOa decreases that caused a poorer quality lime product most likely due to excess sulfur. The following items were identified as ways to improve kiln operation under waste fuel conditions: • Change waste fuel burner configura- tion such that at low waste fuel rates the waste fuel is mixed with the coke to maintain a flame. • Decrease the fan speed (i.e., reduce the draft) to lower the 02 in the kiln, thus lowering the sulfur in the product and increasing the sulfur in the stack. Conclusions Constant achievement of at least 99.99% ORE was demonstrated for each POHC (MeCI2, MEK, CH3 CCI3, TCE, Perc, and toluene) in the lime kiln process. Emissions of pollutants were deter- mined and ranged as follows: particulates 0.7-1.4 kg/hr; HCI 0.04-0.26 kg/hr; S02 123-2,100 ppm; NO, 280-550 ppm; THC 1.5-10 ppm; and CO 10-5,000 ppm. Except for HCI, NOX, THC, MeCI2, and toluene, emissions for pollutants were statistically not different for baseline and waste fuel conditions. Typically, sulfur enters the kiln in the petroleum coke and exits the kiln distri- Table4. Waste Fuel Conditions Run number* 4 5 6 7 8 Chlorine content, % (vol.) 3.14 2.66 3.04 3.05 3.51 Sulfur content, % 0.10 0.08 0.065 0.06 0.11 PCB concentration, ppm 1.0 1.0 1.0 1.0 1.0 Heat value, Btu/lb 12,301 12,084 12,267 13,612 14.064 Specific gravity, g/cc 1.031 1.042 1.035 0.986 0.971 Feed rate, gal/min 0.76 1.21 2.05 0.78" 2.90" 2.88 Mass rate. g/min 2,990 4,770 8,020 2,910 10,820 10,590 *No ivsste fuel burned during Run Nos. 1, 2, and 3. *Runs 7A-7C. cRuns 7D-7F. ------- TableS. Average Stack Emissions Baseline Waste fuel Standard Parameter and unit Stack rate, rrf/min Stack velocity, m/sec Particulates mg/dscm kg/hr HCI, ppm SOi, ppm NOx, ppm CO. ppm THC, ppm Range Average 805 - 975 5.0 - 6.0 24.0 - 35.0 0.66- 1.1 0. 74 - 3.9 123 - 730 306 - 460 10 - 4,900 6.7 - 12.7 917 5.7 28.7 0.9 2.0 553 386 477 8.2 deviation Range 76 0.4 4.7 0.2 1.4 110 49 966 1.9 791 - 938 4.9 - 5.8 24.9-48.7 0.68 - 1.4 2.5 - 6.0 183 - 1.924 288 - 552 10 - 4,540 1.5 - 10.0 Average 847 5.2 35.3 1.0 4.4 596 446 599 3.5 Standard deviation 52 0.3 8.0 0.3 1.2 240 64 1.409 1.1 Table 6. Chlorine and Sulfur Material Balance Run number Coke Chlorine Baseline * 4 Waste fuef 5 Sulfur Baseline* 100 Waste fuef 99 * Average values of baseline Percent to kiln Waste Limestone fuel feed 0 96 2 93 0 NA 1 NA Runs 1-3. Total mass in. kg/hr 23 21 59 54 Lime product 47 34 8 10 Percent from kiln Baghouse Stack dust gas 52 1 65 1 29 63 25 65 Total mass out. kg/hr 13 20 91 70 Material balance closure. % 55 105 73 77 b Average values of waste fuel Runs 4-8. Table 7. Typical Rockwell Lime Metals Material Balance* Metal Coke Be 10 Ca 1 Cr 1 Fe 1 Mg <1 Ni 20 Pb 14 In 5 Percent to kiln (by wtj Waste Limestone fuel feed 0 90 <1 99 <1 98 <1 99 0 99 <1 80 6 80 53 42 Total mass in. 9/hr 25 3x106 811 13,160 2 x 106 660 <500 170 Lime product 75 93 75 83 90 74 55 41 Percent from kiln (bywt) Baghouse Stack dust gas 25 0 7 <1 25 <7 77 <7 70 <7 26 <1 45 <1 58 1 Total mass out. g/hr 16 3x 70s 700 73,300 2x 70s 460 <450 100 Percent closure 64 1OO 86 1O1 700 70 90 58 * Average values for Runs 1-8. buted in the lime product, baghouse dust, and stack gas. Typically, a metal enters the kiln in the limestone feed and exits the kiln in the lime product and baghouse dust. The kiln operation fluctuated resulting in an occasionally lower-quality lime product. Improving the burner system, reducing draft (and % 02), and allowing operators sufficient time to run the system may minimize fluctuations and improve product quality. •&U. S. GOVERNMENT PRINTING OFFICE: 1984/559-11 I/I 0732 ------- D. R. Day and L A. Cox are with Monsanto Research Corporation, Dayton, OH 45407; ft. E. Mournighan (also the EPA Project Officer, see below) is with the Industrial Environmental Research Laboratory, Cincinnati, OH 45268. The complete report, entitled "Evaluation of Hazardous Waste Incineration in a Lime Kiln: Rockwell Lime Company," (Order No. PB 84-230 044; Cost: $ 16.00, 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: Industrial Environmental Research Laboratory U.S. Environmental Protection Agency Cincinnati, OH 45268 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 BULK RATE POSTAGE & FEES PAID EPA PERMIT No. G-35 Official Business Penalty for Private Use $300 ------- |