Back to Publications

Gossman Consulting, Inc.


Michael von Seebach, Ph.D., F.M. Miller, Ph.D., Charles Lamb, Ph.D., and R. Simmons Southdown, Inc., Houston, TX


David Gossman, Gossman Consulting, Inc.

Submitted For Presentation At Air & Waste Management Association Specialty Conference On New RCRA Regulations

Orlando, FL March 17-19, 1992


Extensive research and development efforts have been devoted over the past ten years to the task of developing a list of suitable Principal Organic Hazardous Constituents (POHCs) or POHC surrogates to determine Destruction Efficiencies (DEs) in hazardous waste combustion processes. There have been proponents of various types of scales for POHC destruction difficulty -- those based on the heat of combustion, on the auto-ignition temperature, or on various thermochemical kinetic and/or thermodynamic parameters. However, many workers agree that one of the best methods to select a POHC that will challenge the combustion system's destructive capability is to recognize the likely "failure mode" of the organic under the conditions to which it will be subjected. Thus, for example, a compound such as sulfur hexafluoride or perchloroethylene might be selected if the principal concern were adequate temperature, while a compound like monochlorobenzene (MCB) might be selected if the system operated at modest excess oxygen levels, but at very high temperatures (such as, for example, cement kilns). After exhaustive work, Dellinger and co-workers (1,2,3) have developed a listing of thermal stability for 320 organic compounds. Both pyrolytic and oxidative modes of failure of these compounds have been considered in preparing this list.

Monochlorobenzene (MCB) appears as POHC number 15 on this list (3). Most of the compounds above MCB in the list are highly toxic (such as HCN, cyanogen, and carcinogenic polyaromatic hydrocarbons), difficult to measure (such as acetonitrile), or notorious products of incomplete combustion (PICs) such as benzene. SF6 is an attractive possibility, since it is #4 on the list, very easy to measure in low concentrations, and non-toxic (4,5). However, if the failure mode of concern is one of low oxygen, rather than low temperature, its selection is subject to criticism, since SF6 will decompose given adequate temperature, irrespective of O2 concentration. Therefore, while it may be of use for generating supporting data, SF6 should probably not be selected as the only POHC for a trial burn in cement kilns.

MCB thus seems an attractive choice for regulatory agencies, as it is quite stable under high excess O2 conditions, and even more stable at lower O2. Furthermore, MCB is easily obtained in quantities great enough to run an extended test, is not prohibitively expensive, and is of only moderate toxicity. However, as recent data demonstrate, it is generally not an appropriate choice for use in cement kilns. The purpose of this paper it to provide the data to support this conclusion.


Southdown's cement kiln in Fairborn, OH is depicted schematically in Fig. 1. It is a four-stage dry-process suspension preheater kiln, without precalciner, with a production rate of 85 short tons of clinker per hour. As usual for this type of kiln system, the raw meal is fed to the uppermost of the four cyclone stages of the preheater, while the hot combustion gases from the kiln enter the preheater from the gas duct to the lowest stage. Thus, the raw meal is preheated to approximately 1550F when it enters the rotary kiln while the combustion gases entering the lower part of the preheater at 1900F-2000F are cooled to 700F at the preheater outlet. The preheater off-gases are further cooled in the raw material drying and grinding system, prior to entering the baghouse air pollution control device. From the baghouse, the gases are discharged into the stack. The plant typically uses 7.5 tons/hour of pulverized coal as primary fuel, supplemented with about 2.8 tons/hour of liquid hazardous waste fuel, which is atomized with compressed air concentrically at the tip of the pulverized coal burner. Also, 0.5 ton/hour of tire derived fuel is added to the kiln system in the transition housing between the rotary kiln and preheater.

The kiln system is equipped with a bypass above the transition housing. The bypass allows 10%-15% of the combustion gases from the rotary kiln to be removed from the system. The gases at 1900F-2000F are quenched with ambient air and water before they enter a separate baghouse air pollution control device. From this baghouse, the gases are discharged into the bypass stack.

The Regional Air Pollution Control Authority in Ohio, together with the Ohio EPA, requested that a very comprehensive stack test be carried out at this facility, to include the determination of destruction efficiencies on a number of POHCs. These included carbon tetrachloride, perchloroethylene, and sulfur hexafluoride (which represent materials undergoing destruction mostly as a result of high temperature -- the kinetic mechanism of failure is thought to be largely unimolecular for these materials). Also included were MCB, 1,2,4-trichlorobenzene, and tetrachlorobenzene, whose kinetic mechanism for failure probably is largely bimolecular and requires the intimate presence of oxidizing agent (1,2).


Both the main and the bypass stacks were subjected to comprehensive testing in April of 1991, when emissions of particulate, HCl, metals, total hydrocarbons, carbon monoxide, sulfur dioxide, oxides of nitrogen, polychlorinated dibenzo-p-dioxins and dibenzofurans, and DEs were determined. This paper will discuss only the DE results, but will reference total hydrocarbons and carbon monoxide results. A follow-up test was requested by the regulatory agencies to be carried out in December, 1991, when the DE for MCB was again determined, along with HCl and chlorine emissions, and other parameters not discussed here. In the April test, the normal fuel mix was used, but in the December test, no tire-derived fuel was added. In a four-stage preheater system, regardless of the fuel mix, all combustion of fuel occurs only in the rotary kiln itself; no fuel is introduced into the preheater. In the April test, the concentrations of the POHCs were measured in the main stack, and also in the bypass stack. Reference to Fig. 1 reveals that most of the gases exiting the rotary kiln travel through the preheater and eventually to the main stack, while a small volume (about 12-15% of the total) actually is withdrawn at the kiln exit.

The results of the DE measurements for both April and December tests are shown in Table 1. Actual emissions in pounds per hour are given for all POHCs, both in the main stack and in the bypass stack. From these measurements, it is possible to calculate what the total emissions would be if the measurements from the bypass stack only were considered, and were scaled up on the basis of gas volumes to represent all kiln exit gas. These results are given in Table 2, with a comparison to the results determined using all emissions for comparison.

It can readily be seen from Table 2 that only for MCB is there a significant difference between the destruction efficiencies calculated in the two different manners. Whereas the DE for MCB determined on the basis of the bypass stack is the same order of magnitude as those for the other POHCs, the DE based on total emissions is not in compliance with the requirements of RCRA, which dictate that 99.99% of the POHC must be destroyed.


The test results for MCB were not expected. It was unclear why the MCB emission concentrations were considerably higher in the main stack than in the bypass stack. Since combustion of hazardous waste fuel occurs only in the kiln, it was obvious that MCB was being generated in the preheater, in a mechanism that had nothing to do with combustion in the rotary kiln. Reference to the literature revealed that a similar observation had been made before. Trenholm, Hlustick and Hansen (6) reported that MCB emissions from the preheater at Ash Grove's precalciner kiln in Louisville, NE were considerably higher than those at the bypass; the difference was that the Louisville kiln is equipped with a precalciner; the MCB emissons were at the time attributed to PIC formation in the precalciner. (6) This explanation does not apply for Fairborn, as no burning takes places in the preheater. Therefore, the question remained: what was the source of the organic precursors to MCB in the preheater?

Some clarification can be obtained by considering the THC and CO data from the main and the bypass stacks during the April test. As shown in Table 3, both THC and CO are higher in the main stack. Again, where were these hydrocarbons coming from? Studies have shown that almost every cement plant kiln feed contains at least a low level of organics; reference to this fact also appeared in the Trenholm paper (6). As the kiln feed in a preheater kiln passes down through the cyclone stages and is preheated, at some point the temperature of volatilization, or pyrolysis, of these organics will be reached. If they are then combusted immediately (which is likely only for certain hydrocarbons), they will not report to the stack, and the THC readings from the preheater will be expected to be as low as those from the bypass stack. However, the higher THC readings at the main stack evidence that volatilization or pyrolysis without complete combustion does occur in the preheater.

The most stable pyrolysis products have been shown to be simple aliphatic and simple aromatic hydrocarbons. Benzene is especially stable, as indicated also by its high position on the Dellinger incinerability index (3). This would then provide part of the answer to the question. The remaining part of the question is even more perplexing: How was the benzene chlorinated, since benzene is more stable than chlorobenzene?

EPA Method 26, which has been adopted as part of the "Methods Manual for Compliance with the BIF Regulations" (7), is designed to measure hydrogen chloride and chlorine gas in stack emissions. While there is evidence (8) that the HCl determination in cement kilns will also measure ammonium chloride and very finely divided alkali chloride salts, Method 26, obviously also measures chlorine or oxidizing chlorine compounds. For the December test, chlorine as well as "HCl" was determined in the main stack at Fairborn. The results of the three chlorine determinations and the three corresponding MCB measurements are given in Table 4 and plotted in Fig. 2. The correlation is convincing: the results clearly demonstrate that MCB in the main stack results from the reaction of chlorine with benzene from the raw materials. The MCB in the main stack is therefore neither an undestroyed POHC, nor is it a PIC, but it results from chemical reactions in the preheater. This reasoning has been supported by others (9, 10), and Dellinger has mentioned similar problems with MCB forming as a PIC in incinerators (2,11).

Support for this hypothesis is provided in a study which was performed at Southdown's Knoxville, TN facility. The tests were carried out burning only pulverized coal, with no waste fuel at all. Calcium chloride was added, as the only chlorine source to the kiln system. As shown in Table 5, monochlorobenzene was measured exiting the stack, in quantity roughly proportional to the input calcium chloride (12). These results clearly show a correlation similar to that measured at the Fairborn facility: with chlorine as measured by Method 26, the MCB in the stack gases increases.


Evidence from two cement kilns is presented showing convincingly that monochlorobenzene is an unsuitable POHC for DE testing on cement kilns. However, where the kiln is a preheater or precalciner kiln equipped with bypass, the DE can be determined at the bypass. The bypass gas composition is representative of the kiln exit gas composition, and therefore will accurately reflect the capability of the kiln system to efficiently destroy MCB or any other POHC selected. The DE determination from the bypass will be conservative, as the scale up is based upon gas volumes. The DE determination in the bypass also avoids confusion arising from the effect of any reactions subsequent to the kiln in the preheater.


1. Dellinger, Barry et al (University of Dayton), "The Theory of Thermal Zone Chemistry And Its Influence on Hazardous Waste Incinerations", Presented at the 79th Annual Meeting of the Air Pollution Control Association, June 22-27, 1986

2. Dellinger, Barry and Philip H. Taylor (University of Dayton) and C. C. Lee (Hazardous Waste Environmental Research Laboratory, USEPA), "Development of Hazardous Waste Incinerability Surrogate Mixtures", Presented at the Second Annual National Symposium on Incineration of Industrial Wastes, San Diego, CA, March 9-11, 1988

3. Taylor, Philip H., Barry Dellinger, and C. C. Lee (University of Dayton and USEPA), "Development Of A Thermal Stability Based Ranking Of Hazardous Organic Compound Incinerability", Environmental Science and Technology. Vol. 24; Pg. 316, March, 1990

4. England , Walter G., Lynn H. Teuscher, and Steven L. Quon (Tracer Technologies), "The Correlation of SF6 Destruction With Principal Organic Hydrocarbon Destruction In Incineration Processes", Presented at the 80th Annual Meeting of the Air Pollution Control Association, New York, NY, June 21-26, 1987

5. Waterland, Larry L. (Acures Corporation) and Laurel J. Staley (USEPA), "Pilot Scale Testing Of SF6 As A Hazardous Waste Incinerator Surrogate", Presented at the 82nd Annual Meeting and Exhibition of the Air and Waste Management Association, June 25-30, 1989

6. Trenholm, Andrew (Midwest Research Institute), Dwight Hlustick (USEPA), and Eric Hansen (Ash Grove Cement), "Emissions From A Cement Kiln Burning Hazardous Waste", Presented at the 83rd Annual Meeting of the Air & Waste Management Association, Pittsburgh, PA, June 24-29, 1990

7. United State Environmental Protection Agency, "Methods Manual for Compliance with the BIF Regulations", Office of Solid Waste, Washington, DC, December 1990

8. von Seebach, Michael (Southdown, Inc.) and Gossman, David (Gossman Consulting, Inc.), "Cement Kilns, Sources of Chlorides Not HCl Emissions", Presented at the Waste Combustion in Boilers and Industrial Furnaces Conference, Kansas City, MO, April 17-20, 1990

9. Trenholm, Andrew (Midwest Research Institute), Private Communication

10. Gossman, David (Gossman Consulting, Inc.), Private Communication

11. Tiery, Debra A., Richard C. Striebich and Barry Dellinger (University of Dayton), and Harry E. Bostian (USEPA), "Comparison Of Organic Emissions From Laboratory And Full-Scale Thermal Degradation Of Sewage Sludge", Hazardous Waste and Hazardous Materials, Vol. 8; No. 3, 1991

12. Simmons, R., (Southdown, Inc.), "Dixie Chlorobenzene Testing", Internal Report, December 16, 1991




APRIL 1991 ------------------------- Repetition ------------------------

1 2 3 Average
Carbon Tetrachloride

Mass Feed Rate, lb/hr 276.1 276.1 276.1 276.1
Mass Emission Rate, lb/hr

Main Stack 0.000108 0.000178 0.000133 0.000139
Bypass Stack 0.000074 0.000052 0.000127 0.000084
Main + Bypass 0.000182 0.00023 0.00026 0.000224


Mass Feed Rate, lb/hr 68.73 68.73 68.73 68.73
Mass Emission Rate, lb/hr

Main Stack 0.0122 0.0145 0.0124 0.013033
Bypass Stack 0.000034 0.000079 0.000038 0.000050
Main + Bypass 0.012234 0.014579 0.012438 0.013084


Mass Feed Rate, lb/hr 46.53 46.53 46.53 46.53
Mass Emission Rate, lb/hr

Main Stack 0.000147 0.000217 0.000191 0.000185
Bypass Stack 0.000080 0.000071 0.000159 0.000103
Main + Bypass 2.27E-04 2.89E-04 3.50E-04 0.000288

Sulfur Hexafluoride

Mass Feed Rate, lb/hr 5 5 5 5
Mass Emission Rate, lb/hr

Main Stack 4.83E-06 4.42E-06 4.76E-06 0.000004
Bypass Stack 3.81E-06 3.87E-06 3.85E-06 0.000003
Main + Bypass 8.64E-06 8.29E-06 8.61E-06 0.000008


Mass Feed Rate, lb/hr 21.85 21.85 21.85 21.85
Mass Emission Rate, lb/hr

Main Stack 0.000054 0.000059 0.000055 0.000056
Bypass Stack 0.000027 0.000026 0.000025 0.000026
Main + Bypass 8.21E-05 8.60E-05 8.17E-05 0.000083


Mass Feed Rate, lb/hr 36.25 36.25 36.25 36.25
Mass Emission Rate, lb/hr

Main Stack 0.000076 0.000132 0.000115 0.000107
Bypass Stack 0.000065 0.000081 0.000084 0.000077
Main + Bypass 1.42E-04 2.14E-04 1.99E-04 0.000185



Mass Feed Rate, lb/hr 349.3 402.1 352.6 368
Mass Emission Rate, lb/hr

Main Stack 0.107 0.0496 0.0722 0.076266
Bypass Stack 0.000162 0.000232 0.00039 0.000261
Main + Bypass 0.107162 0.049832 0.07259 0.076528




APRIL 1991 ------------------------- Repetition ------------------------

1 2 3 Average
% Bypass 12.62 13.17 12.23 12.67

Carbon Tetrachloride

DRE, %(based on bypass) >99.99978 >99.99985 >99.99962 >99.99975
DRE based on total emitted >99.99993 >99.99991 >99.9999 >99.99991


DRE, %(based on bypass) >99.9996 >99.99912 >99.99954 >99.99942
DRE based on total emitted >99.98219 >99.97878 >99.9819 >99.98096


DRE, %(based on bypass) >99.99863 >99.99883 >99.9972 >99.99822
DRE based on total emitted >99.99951 >99.99937 >99.9924 >99.99937

% Bypass 11.92 12.64 12.73 12.43
Sulfur Hexafluoride

DRE, %(based on bypass) >99.99936 >99.99938 >99.99939 >99.99938
DRE, %(total emissions) >99.99982 >99.99983 >99.99982 >99.99982


DRE, %(based on bypass) >99.99893 >99.99905 >99.99907 >99.99902
DRE, %(total emissions) >99.99962 >99.9996 >99.99962 >99.99961


DRE, %(based on bypass) >99.99847 >99.99821 >99.99817 >99.99828
DRE, %(total emissions) >99.9996 >99.99941 >99.99944 >99.99948

DECEMBER TEST ------------------------------------Repetition--------------------------------

1 2 3 Average
% Bypass 13.9 13.9 13.9 13.9


DRE, %(based on bypass) >99.99966 >99.99958 >99.9992 >99.99948
DRE, %(total emissions) >99.96932 >99.9876 >99.97941 >99.9792





MAIN 1 209 119.2
MAIN 2 254 111.5
MAIN 3 233 96.4
BYPASS 1 3.5 2.5
BYPASS 2 23.2 1.3
BYPASS 3 13.8 2.3

MAIN 1 198 69.4
MAIN 2 210 98.8
MAIN 3 198 75.5
BYPASS 1 3.5 1.5
BYPASS 2 10 1.3
BYPASS 3 20 2.2






2 0.0732 0.678
3 0.0496 0.0365

0 0
















0.00894 0.01 0.0152


0.147 0.15 0.132