U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
PB-265 541
DESTROYING CHEMICAL WASTES IN COMMERCIAL
SCALE INCINERATORS FACILITY REPORT NO. 1 -
THE MARQUARDT COMPANY
TRW Defense and Space Systems Group
Redondo Beach, California
Prepared for
Environmental Protection Agency, Washington, D. C,
1977
-------
BIBLIOGRAPHIC DATA
SHEET
1. Report No.
2.
PB 265 541
4. Tide and Subtitle
DESTROYING CHEMICAL WASTES IN COMMERCIAL SCALE INCINERATORS
Facility Report No. 1 - The Marquardt Company
5. Report Date
April 1977
7. Authors)
J. F. Clausen, R. J. Johnson, C. A. Zee
8. Performing Organization Kept.
No.
9. Performing Organization Name and Address
TRW Defense and Space Systems Group
One Space Park, Redondo Beach, California 90278
10. Project/Task/Work Unit No.
II. Contract/Grant No.
68-01-2966
12. Sponsoring Organization Name and Address
U.S. Environmental Protection Agency
Office of Solid Waste
Washington, D. C. 20460
13. Type of Report Si Period
Covered
Facility test report
14.
15. Supplementary Notes
16. Abstracts
Incineration tests were conducted at The Marquardt Company, Van Nuys, California, to
determine the effectiveness of thermally destroying two selected industrial liquid
wastes: ethylene manufacturing waste and nexachlorocylopentadiene (C-5,6). Each
waste was burned at three different test conditions to determine the effects of
normal operating and equipment variables. Analysis of combustion gas samples indi-
cated destruction efficiencies of over 99.999 percent for each waste constituent.
Standard EPA Method 5 tests were performed on stack emissions to determine particulate
loading and composition. Analysis of scrubber water samples indicated no increase in
organic content compared to fresh scrubber water. Burner head residue formed during
incineration of each waste contained no chlorinated organics. Results of these tests
indicated that either waste can be effectively destructed in a liquid injection
incinerator; however, because of tarry residual, the C-5,6 waste may be better
matched to rotary kiln incineration. Estimated cost of destroying an assumed annual
17o. Descriptors
17. Key Words and Document Analysis.
Waste Disposal
Industrial Wastes
Ethylene
Incinerators
Field Tests
Gas Sampling
Chemical Analysis
Cost Analysis
17b. Identifieis/Open-Ended Terms
The Marquardt Company
Liquid Injection Incinerator
Hexachlorocylopentadi ene
17c. COSATI Field/Group 07-01, 14-02, 14-01
production of 15 million kilograms of ethylene
waste was $1.8 million capital investment and
$1.0 million annual operating costs ($69/metric
ton). Cost of incinerating 4.5 million kilogram
of C-5,6 per year was estimated to be $1.6
million capital investment and $2.2 million
annual operating costs ($488/metric ton).
18. Availability Statement
19. Security Class (This
Report)
UNCLASSIFIED
20. Security Class (This
"^CLASSIFIED
21. N> -'•>-*•
22. Price pcfiota
FORM NTIS-BB (REV. 1O-73) ENDORSED BY ANSI AND UNESCO.
THIS FORM MAY BE REPRODUCED
USCOMM-DC B2B3-P74
-------
DESTROYING CHEMICAL WASTES IN
COMMERCIAL-SCALE INCINERATORS
Facility Report No. 1
This final report (SW-122c) describes work performed
for the Federal solid waste management program
under contract no. 68-01-2966
and is reproduced as received from the contractor
U.S. ENVIRONMENTAL PROTECTION AGENCY
1977
-------
This report has been reviewed by the U.S. Environmental Protection
Agency and approved for publication. Publication does not signify
that the contents necessarily reflect the views and policies of the
Environmental Protection Agencyt nor does mention of commercial
products constitute endorsement or recommendation by the U.S. Government.
An environmental protection publication (SW-122c) in the solid waste
management series.
ii
-------
PREFACE
The purpose of this study is to evaluate the environmental, technical,
and economic feasibility of disposing of industrial wastes via incinera-
tion. This objective is being pursued through a series of test burns con-
ducted at commercial incinerators and with real-world industrial wastes.
Approximately eight incineration facilities and seventeen different indus-
trial wastes will be tested under this program. The incineration facil-
ities were selected to represent the various design categories which
appear most promising for industrial waste disposal. The wastes were
selected on the basis of their suitability for disposal by incineration
and their environmental priority.
This report describes the test conducted at the Marquardt Company,
which was the first facility of the series. Marquardt manufactures incin-
eration equipment and has a liquid injection test unit which was used for
this project. Wastes resulting from the manufacture of ethylene and from
hexachlorocyclopentadiene were tested and found to be successfully
destroyed by the Marquardt incinerator.
The content of this report is primarily of an objective nature pre-
senting the equipment description, waste analysis, operational procedures,
sampling techniques, analytical methods, emission data and cost informa-
tion. Facility reports similar to this one will be published as soon as
possible after the testing has been completed at a facility so that the
raw data and basic results will be available to the public quickly.
This report and the other Facility Reports will contain a few subjec-
tive conclusions. However, the detailed subjective analysis will be
reserved for the Final Report, which will be prepared after all testing is
complete.
ACKNOWLEDGEMENTS
TRW wishes to express its sincere gratitude to the Marquardt Company
personnel, particularly Messrs. Joel Hudson, Robert Babbitt and Kenneth
Scheurn, for their cooperation in conducting these facility tests. Thanks
are also due to Dr. Philip Levins and Mr. Jeffrey Adams of Arthur D. Little,
Inc., for their advice in modifying the combustion gas sampling probe to
improve cooling capability. The project is deeply indebted to Messrs.
Alfred Lindsey and John Schaum of the Office of Solid Waste Management
Programs, U.S. Environmental Protection Agency, for their advice and tech-
nical direction.
iii
-------
CONTENTS
Page
1. Summary ]
2. Introduction 6
3. Process Description 8
3.1 Facility Process 8
3.1.1 Incinerator and Reaction Tailpipe 8
3.1.2 Air Supply System 14
3.1.3 Waste Fuel System 14
3.1.4 Auxiliary Fuel System 14
3.2 Instrumentation 14
3.2.1 Process Parameters 14
3.2.2 Gas Analysis 17
3.3 Emission Controls 17
3.3.1 Venturi Scrubber and Separator Tank 17
3.3.2 Caustic Solution and Water Supply System 17
3.3.3 Scrubber Liquid Collection System 19
4. Test Description 20
4.1 Wastes Tested
4.1.1 Ethylene Manufacturing Waste 20
4.1.2 Hexachlorocyclopentadiene Manufacturing Waste 22
4.2 Operational Procedures 24
4.2.1 Test Procedures 25
4.2.2 Safety Procedures 26
4.2.3 Test Commentary 27
4.2.4 Disposal of Waste Residues 36
4.3 Sampling Methods 36
4.3.1 On-Line Gas Monitoring 36
4.3.2 Sampling of Combustion Products 43
4.3.3 Sampling Emissions and Waste Products 48
iv
-------
CONTENTS (CONTINUED)
Page
4.4 Analysis Techniques 50
4.4.1 Extractions and Sample Preparation 50
4.4.2 Analytical Methods 52
4.5 Problems Encountered 54
4.5.1 Field Test Problems 55
4.5.2 Laboratory Analysis Problems 56
5. Test Results 59
5.1 Combustion Gas Data Summary 59
5.2 Waste Destruction Analysis Summary 59
5.2.1 Organic Composition 61
5.2.2 Inorganic Characterization 66
5.3 Final Emissions 69
5.3.1 Stack Gases 69
5.3.2 Scrubber Waters 71
5.3.3 Solid Residues 76
6. Waste Incineration Cost 81
6.1 Capital Investment 81
6.2 Annual Operating Costs 84
Appendices
Appendix A - Assessment of Environmental Impact 87
of Destructlng Chemical Wastes at
the Marquardt Company
Appendix B - Sample Volume Data 91
Appendix C - Calculation of Incinerator 97
Performance
Appendix D - Analytical Chemistry Details 101
-------
FIGURES
Page
3-1 TMC Test Facility Process Flow Diagram 9
3-2 Overall Test Installation 10
3-3 Incineration Unit 11
3-4 Waste Fuel Supply System 12
3-5 Basic Burner System Geometry 13
3-6 Process Controls and Instrumentation 16
3-7 TMC On-Line Gas Analysis Equipment 18
4-1 MA-II Clinker 30
4-2 MA-III Clinker 32
4-3 MA-IV Clinker 33
4-4 MA-V Clinker 34
4-5 MA-VI Clinker 35
4-6 MA-VII Clinker 37
4-7 Sampling System For On-Line Monitors 3.8
4-8 Gas Conditioning and Analysis System For On-Line Instruments 39
4-9 Instrument Racks 41
4-10 Calibration Gases and Gas Conditioner 42
4-11 Work Area Behind Instrument Racks 42
4-12 Combustion Zone Sampling Train Schematic 44
4-13 Combustion Zone Sampling Train 45
4-14 Water Cooled Probe Schematic 46
4-15 Sampling System For Combustion Zone 47
4-16 Sampling System For Stack 49
4-17 Sorbent Trap Extractor 58
vi
-------
FIGURES (CONTINUED)
Page
5-] Filters From Combustion Zone Gas Sampling After Solvent 67
Extraction
5-6 Toxic Trace Metals at Level of Interest (mg/m3) by AAS 68
5-2 Filters From Stack Gas Sampling 70
vii
-------
TABLES
Page
1-1 Results Summary 4
3-1 Instrumentation for Waste Destruction Testing - Marquardt 15
4-1 Composition of Ethylene Manufacturing Waste Representative 21
Sample
4-2 Trace Metals in the Ethylene Representative Waste Sample 22
4-3 Composition of Hexachlorocyclopentadiene Waste Survey 23
Sample
4-4 Composition of C-5,6 Waste Representative Sample 24
4-5 Trace Metals in the C-5,6 Waste 25
4-6 Incinerator System Parameters Data Summary 28
4-7 Description of On-Line Instruments 40
4-8 Summary of Analytical Methods 53
5-1 Total Gas Composition in the Combustion Zone by Volume 60
5-2 Results of Gas Chromatographic Analysis 62
5-3 Organic Material Extracted From Combustion Zone Filters 63
5-4 Organic Material Extracted From Sorbent Traps, Survey 64
Analysis
5-5 Analysis of Organics in Grab Gas Samples 65
5-6 Toxic Trace Metals at Level of Interest (mg/m3) by AAS 68
5-7 Particulate Mass Loadings 71
5-8 Percent Sodium Salts as Nad on Stack Filters 72
5-9 Gastec^Tube Results 72
5-10 Results of Gas Chromatographic Analysis of Scrubber 73
Water Samples
5-11 Results of Analysis of Concentrated Organic Extracts of 74
Scrubber Water Samples
5-12 Elemental Analysis of Scrubber Water by AAS (ppm) 75
vi ii
-------
TABLES (CONTINUED)
Page
5-13 C and H Analysis of Clinkers 76
5-14 Results of Analysis of Clinker Extracts by Gas 78
Chromatography
5-15 Residue Extracted From Clinker 78
5-16 Inorganic Composition of Clinkers (ppm) 80
5-17 OES Detection Limits 80
6-1 Capital Investment - 15 Million kg./Yr. Ethylene Manu- 82
facturing Waste Incineration Plant
6-2 Capital Investment - 4.5 Million kg./Yr. Hexachlorocyclo- 83
pentadiene Waste Incineration Plant
6-3 Annual Operating Cost - 15 Million kg./Yr. Ethylene 85
Manufacturing Waste Incineration Plant
6-4 Annual Operating Cost - 4.5 Million kg./Yr. Hexachloro- 86
cyclopentadiene Waste Incineration Plant
B-1 Gas Volume Data 93
B-2 Final Corrected Gas Volumes 94
B-3 Liquid Impinger Volume Data 95
D-l Extracted Material From Control Samples 103
D-2 Filter Weights Through Analyses (mg) 112
D-3 Particulate Mass Loadings 113
D-4 Elemental Survey of Selected Samples (mg/m3) 115
D-5 Caustic Impinger Analyses 116
ix
-------
1. SUMMARY
Numerous organic wastes from industrial operations are generated in
liquid form. From a physical point of view, they are often suited for dis-
posal in commercial liquid injection incinerators. If a given liquid waste
possesses a significant heating value, it may be a candidate for heat re-
covery by utilizing it as a supplementary fuel or by combustion-for example-
in a water wall incinerator. Ideally, such a high heating value waste
would have a relatively low toxicity level, little ash content, minimal
suspended solids, be pumpable at ambient temperatures, and annually generated
in significant volumes. At the other end of the spectrum are those liquid
wastes which are quite hazardous and have low heating values. Organochlorine
wastes represent an excellent example of this class of waste. Annual waste
quantities generated tend to be large, hence the disposal problem is signi-
ficant. Solids loadings and ashing characteristics will vary. Because of
the hazardous nature of these liquid wastes, however, thermal destruction
appears to be a desirable disposal method relative to other alternatives.
Two representatives of liquid organic industrial wastes were selected
for incineration at The Marquardt Company facility. In great measure, they
exemplify extremes of this generic waste form. The two wastes selected for
incineration were those resulting from the manufacture of ethylene and hexa-
chlorocyclopentadiene (C-5,6). On the one hand, ethyl ene manufacturing
waste is a relatively nonhazardous. high heating value waste. It is gen-
erated at a rate of approximately 15 million kilograms per year
This waste was a clear light brown solution with no apparent sediment
or particulate at room temperature. Indeed, it possessed physical and chem-
ical properties not unlike distillate oil. For example, this waste had a
specific gravity of about 0.9 and a heating value of around 10,000 kcal/kg.
Its chemical composition was approximatelv 89 percent carbon and 9 percent
hydrogen. Halogens-as chlorine-were about 0.004 percent. Sulfur and nitrogen
contents were relatively low; namely 1.32 and 0.13 percent, respectively.
Ash content (nonvolatile inorganics) in this waste was less than 0.01 per-
cent. A viscosity of 1.28 centistokes (at 22°C) allowed this material to be
pumped and injected into the incinerator without preheating or other condi-
tioning. In reality, this waste appears to be a good candidate for use as
a supplementary fuel.
Conversely, hexachlorocyclopentadiene manufacturing waste (4.5 million
kilograms annually) was significantly different in comparison with the
ethylene waste. This highly chlorinated waste would not sustain combustion
without auxiliary fuel. Its heating value was only 2,400 kcal/kg. Physi-
cally, the C-5,6 waste was a dark brown dense and viscous liquid with sus-
pended, wax-like particulate matter. Specific gravity of the C-5,6 was in
excess of 1.7, while its viscosity at room temperature was 11.3 centistokes.
Fortunately, this waste was miscible in No. 2 fuel oil which facilitated
injection and combustion in a liquid incinerator.
Selection of hexachlorocyclopentadiene waste is important since it
represents a class of heavily chlorinated wastes. Safe destruction of
this hazardous material is indicative of what may be anticipated by
1
-------
Incinerating similar organochlorine wastes. Further* C-5,6 is the princi-
intermediate for synthesis of important polychlorocyclodiene pesticides,
including chlordane, haptachlor, aldrin, dieldrln, isodrin. endrin. mirex
and kepone. In this regard, prolonged storage of such pesticides can re-
sult in product degradation in which C-5,6 may be formed. Further, incin-
eration of chlorinated pesticides may result in the formation of C-5,6 as
an intermediate reaction product. Therefore, adequate thermal destruction
of C-5,6 waste suggests that similar wastes as well as unused C-5,6 based
pesticide formulations could be safely incinerated with a significant
degree of confidence.
Incineration tests with these two wastes were conducted at The
Marquardt Company facility. Van Nuys, California. The purpose of the tests
were to determine the effectiveness of thermally destroying these two dif-
ferent wastes from a combustion standpoint! and to insure that incineration
could be accomplished in an environmentally safe manner. Marquardt1s
incineration facility was chosen for a number of reasons. Basically, it
is a liquid injection system and thus compatible with the waste farm,
Further, the incinerator was capable of combustion temperatures in the
800°C to 1400°C (1500°F to 2500°F) temperature range desired for adequate
waste destruction. In addition, a caustic venturi scrubber was available
to neutralize HC1 formation derived from combustion of the heavily
chlorinated C-5,6 waste.
Marquardt's liquid injection incinerator consisted of a SUE® (for
SUdden Expansion) burner; air, waste, and auxiliary fuel feed systems; and
aThigh energy venturi scrubber. The incinerator was a well-instrumented
test unit but similar in operation to their commercial configuration.
Marquardt has numerous commercial incinerator installations of this basic
system in field operation. The incinerator is scaled up to process larger
waste throughputs by simply adding more basic units with proper manifolding,
controls and an appropriately sized scrubber. In this regard, the test
incinerator was very representative of typical Marquardt commercial instal-
lations. Further, the incinerator is setup to use No. 2 fuel oil as an
auxiliary fuel; a necessity with low heating value fuels such as C-5,6
waste.
Each waste was burned at three different conditions to ascertain the
effects of normal operating and equipment variables as well as attempting
to define minimum destruction requirements where feasible. Ethylene waste
incineration proved relatively straightforward and trouble free. In
reality, its combustion characteristics were quite similar to No. 2 oil,
although some solid residue formation was apparent following long duration
runs.
The C-5,6 waste presented obvious problems associated with a liquid
injection and feed system. Waste suspended solids tended to clog filters
rapidly. Smaller particulate passing through a filtration system will,
after prolonged operation, abrasively erode injection orifices. This can
be minimized by premixing the C-5,6 with fuel oil in holding tanks and
allowing the particulate to settle. In any event, a steam or electrically
heated and insulated waste feed system would be necessary where premixing,
settling or other processing takes place at ambient temperatures less
than 20°C.
2
-------
Elemental chemical analysis of the C-5,6 showed the following composi-
tion: 20.76 percent carbon, 0.67 percent hydrogen, 0.37 percent N, 0.02
percent S, and 76.47 percent halogens as chlorine. The chlorinated species
were predominantly C5C16 (hexachlorocyclopentadiene) and C5C18 (octachloro-
cyclopentane). As a result, incineration of this waste necessitates use of
a caustic wet scrubber to remove the large quantities of the HC1 combustion
product.
As indicated previously, the test program was conducted using three
operating conditions for each waste. The ethylene wastes were burned at
three different combustion temperatures to evaluate the impact on destruc-
tion efficiency. The combustion of C-5,6 however, required a different
approach due to the fact it would not support combustion without the pres-
sence of an auxiliary fuel. Because of this, there was concern that
chlorinated species (principally CsCle or CsCls) would be emitted to the
atmosphere as a result of incomplete destruction at lowered combustion tem-
peratures. For this reason, it was arbitrarily decided that the fuel/air
ratio would be held approximately constant at the highest combustion tem-
perature attainable. Thus to evaluate destruction of C-5,6 under different
operating conditions, this waste was burned using three blends of fuel oil
and waste.
A summary of the test results is presented in Table 1-1, beginning
with the range of test conditions for each waste, including combustion tem-
perature, residence time, and waste feed rate. Standard EPA Method 5 tests
were performed on the stack emissions to determine particulate loading and
the nature of the particulate. Grain loadings of 20-25 mg/mg3 were obtained
for the No. 2 oil and ethylene waste tests. Particulate loadings of 36 to 113
mg/m3 were obtained for the C-5,6 tests, but 90 percent of the particulate
matter from these tests was found to be sodium salts generated by the
reaction of the caustic scrubber with HC1 and 0)3. The particulate was
analyzed for toxic metals and none were found to exist above 0.1 mg/m3 in
the stack gases.
The stack gases were sampled for hydrocarbons and SOX on all tests,
and for HC1, Cl2» and phosgene on the C-5,6 tests. The only specie
detected was SOX at levels consistent with the fact that the ethylene waste
contained more than 1 percent sulfur. The limits of detection for the
other species are discussed in Section 4.4.
Benzene and mesityl oxide were found in the test filter and sorbent
trap samples at levels which ranged from <0.2 to 2.4 mg/m3. These levels
are about the same as those obtained for the appropriate control samples.
Nothing of the less volatile materials, e.g., heavily chlorinated organics,
POM, etc., were seen for any of the tests above the lower limits of detection
for the analyses. These detection limits were estimated to be 0.1 mg/m3.
Samples of scrubber water were also analyzed for organic content.
Organic material was found in the scrubber water but at the same approxi-
mate levels as in the control samples of fresh scrubber water. Gas
chromatographic retention data indicates that the same compounds are pre-
sent in all samples.
-------
TABLE 1-1. Results Summary
Combustion Temperature (°C)
Residence Time (sec)
Feed Rate ( kg/ml n)
Quality of Stack Emissions:
Parti art ate (mg/m )
Trace Metals (mg/m )
Quality of Combustion Gas:
2
Total Organlcs (mg/m }
2
Waste Content (mg/m )
Trace Metals (mg/m )
Quality of Scrubber Water:
Total Organlcs (mg/1)
Trace Metals (mg/1)
Quality of Solid Residue (Clinker):
Waste Content (mg/g)
Destruction Efficiency:
Total Organics (percent)
Waste Constituents (percent)
Capital Cost ($)
Operating Cost ( I/metric ton)
Plant Size (Kg/yr)
Ethyl ene Waste
1349-1752
0.14-0.19
1.4-2.0
20-25
0.001-0.003 Pb
13-22
Not Detected
(<0.02)
0.002-0.003 Pb
Not Significant^3^
0.02 Pb
1.8-2.4
99.96-99.98
>99.999
1 ,318,000
69.31
15 million
C-5,6 Waste
1348-1378
0.17-0.18
2.0-2.9*15
36-113<2>
0.003-0.006 Pb
0.034-0.065 Hn
21-27
Not Detected
(<0.02)
0.55-1,40 Pb
0.077-0.19 Mn
0.090-0.33 Co
Not Sign if leant (3)
0.21-0.54 Pb
Not Detected
(<.026)
99.97
> 99. 999
1,630,000
487.59
4.5 million
(^Mixture of C-5,6 and No. 2 oil; feed rate of C-5,6 alone is 0.66-1.5
(2'85-90 percent of stack particulate content was calculated (based upon
sodium analysis) to be NaCl produced by the reaction of the caustic
scrubber solution with HC1 and C02.
^ 'Levels not significantly higher than background water samples.
4
-------
A burner head residue was found to form during incineration of both
waste materials. These residues were typically 95 to 99 percent carbon
and had some remnants of the fuel feed contained in them at <0.5 percent.
Constituents of ethylene waste were easily detectable in the clinker from
those runs, but the organics found in the C-5,6 test clinker were not
chlorinated.
The waste destruction performance is also summarized in Table 1-1 for
both total organics and the specific waste constituents. Emissions and
efficiencies are measured at the combustion zone exit prior to the scrubber
system. A sample destruction efficiency calculation is presented in
Appendix C.
The results of these tests indicate that the C-5,6 waste can be
effectively destructed in a liquid injection incinerator. No evidence of
C-5,6 or any other chlorinated organic was found in any of the samples
including the "coke like" burner head residue referred to as clinker. How-
ever, because of the tarry residual in the waste, the C-5,6 may be better
matched to rotary kiln incineration, which can accommodate tars and sludges,
Chemical composition, combustion characteristics, and emissions of the
ethylene waste are so nearly identical to No. 2 oil that the use of this
waste as a primary or secondary boiler fuel should be seriously considered.
With the possible exception of the formation of clinker, which appears to
have small impact, the incineration of this type of ethylene waste is
clean and efficient.
Both capital investment and annual operating costs were estimated for
SUE®incineration systems to handle the annual source plant productions of
the ethylene and C-5,6 wastes. Annual waste production of the ethylene
waste was assumed to be 15 million kilograms (33 million pounds), result-
ing in an estimated total capital investment of $1,818,000 and annual oper-
ating costs of $1,037,500 ($69.31/metric ton). An annual waste production
of 4.5 million kilograms (10 million pounds) was used to estimate the
C-5,6 waste incineration costs. Total capital investment was estimated to
be $1,630,000 with annual operating costs of $2,211,700 ($487.59/metric
ton).
-------
2. INTRODUCTION
The objective of this facility test program was to evaluate the ef-
fectiveness of thermally destructing specific industrial chemical wastes
In an existing commercial scale processing facility. These facility tests
are part of an overall U.S. Environmental Protection Agency sponsored
program involved with selective testing of seventeen different wastes at
eight generic types of thermal destructing facilities. The purpose of
this test program is to acquire useful disposal technology as well as
economic information. This report describes test operations and results
of incinerating two different liquid wastes in a commercially available
liquid injection incinerator operated by The Marquardt Company (TMC),
Van Nuys, California. TMC is a manufacturer of incinerators, not a waste
disposal company,and operates a test facility for evaluation of their
incinerator with different wastes.
TMC has been involved in combustion research and production of com-
bustion devices for over 30 years, including ramjets, rocket-ramjet hybrids,
and rocket engines. Concurrent with the evolution of this advanced tech-
nology, TMC developed a burner for in-house use to generate high pressure,
high temperature gases for the testfng of supersonic ramjets and materials~
as well as related aerodynamic testing. The TMC burner is called the SUE®
(for SUdden Expansion burner) and was ultimately integrated into a full-
scale, commercially available incineration system.
The TMC incinerator was selected for use on this program for several
reasons. It utilizes a unique liquid injection configuration which has
good air/fuel mixing properties. Development and application of the TMC
incinerator system has been ongoing for over 12 years, it is commercially
available and currently in use at various industrial facilities. As a
result, this incinerator concept has been found to be effective in de-
stroying waste propel!ants and solvents as well as chlorinated hydrocarbons
including herbicides and pesticides. TMC's incinerator is a well-instru-
mented installation equipped with a high energy venturi scrubber for control
of atmospheric emissions.
The two materials tested at TMC, ethylene waste and hexachlorocyclo-
pentadiene (C-5,6) waste, were selected based upon physical compatibility
with the incinerator, annual waste generation quantities and the hazardous
nature of the wastes. Both wastes are liquid and thus suitable for a
liquid injection system. Similarly, both wastes are generated in relatively
large quantities. Further, they are representative of even larger classes
of similar Industrial wastes. The ethylene waste has a high heating value
and is therefore ultimately a candidate for heat recovery although the
present TMC incinerator did not utilize a water wall combustion chamber.
Conversely, the C-5,6 material has a very low heating value (requiring
auxiliary fuel to sustain combustion), but is heavily chlorinated and quite
hazardous. For these reasons, the two wastes were burned in the TMC
facility to demonstrate incineration as a viable alternative to landfill
disposal.
-------
Although the TMC incinerator tested on this program had a maximum
input feed rate of 3.5 liters/min (55 GPH), it is truly representative
of commercial scale applications. This is because applications requiring
greater feed rates are achieved simply by using multiples of the existing
design. A 12.6 liters/min system, therefore, would only necessitate
manifolding four TMC incinerators onto a common foundation. Controls and
safety devices would be essentially the same as with a single incinerator.
In most applications, exhaust gases would be manifolded to a single ade-
quately sized wet scrubber. For these reasons, data generated at the
Marquardt test facility are deemed representative of larger systems. In
addition, the single TMC unit was both convenient and economic insofar as
the EPA program objective of demonstrating waste destruction efficiency.
The fact that TMC's facility was very well instrumented facilitated the
conduct of a well controlled and environmentally sound test program.
The following report sections describe in detail the incinerator
process equipment (Section 3), and the waste destroyed and test and
sampling procedures followed (Section 4). Test results are presented and
discussed (Section 5), including effectiveness of destruction of the wastes.
An estimate of the capital investment and operating costs of disposing of
wastes using this type of incinerator equipment is also included in the
report (Section 6).
-------
3. PROCESS DESCRIPTION
3.1 FACILITY PROCESS
The Marquardt liquid incineration test facility process is shown
schematically in Figure 3-1. Basic system components, shown in Fig-
ures 3-2 through 3-4, include:
t Bidden .Expansion (SUE®) incinerator
• Radiation cooled reaction tailpipe
• Air supply system
• Waste fuel system
t Auxiliary fuel system
• Instrumentation
• Emission control system
Following is a description of the incinerator and feed system. Facil-
ity instrumentation and emission controls are discussed in subsequent
report Sections 3.2 and 3.3, respectively.
3.1.1 Incinerator and Reaction Tailpipe
The configuration of the SUE® incinerator and radiation cooled reac-
tion tailpipe is shown in Figure 3-5. The burner consists of an inlet
pipe connected to a large diameter combustion chamber by means of a flat
plate. Fuel nozzles protruding through the plate spray the fuel radially
into the inlet oxidizer stream. Burning occurs in the recirculation zone
formed by the flat plate and the combustion chamber wall. An overall view
of the combustion chamber and tailpipe is presented in Figure 3-3.
The combustion chamber wall is constructed of 310 stainless steel and
is actively cooled by process air prior to its entry into the combustion
zone, thus preheating the air to between 150° to 370°C. The reaction
tailpipe, also of 310 stainless steel, is cooled by radiation to the envi-
ronment and provides a hot walled chamber for completion of the incinera-
tion process. The mating flanges of the reaction tailpipe are internally
water cooled to prevent warping or leakage and are sealed with high-
temperature asbestos fiber material. Ports were provided as shown on Fig-
ure 3-5 for instrumentation and gas sampling. This unit operates at
velocities of 30 to 45 m/sec, depending on mass flow and temperatures,
with resulting "stay time" to the end of the reaction tailpipe (and gas
sampling ports) of approximately 0.12 to 0.2 seconds. The incinerator
was designed for a continuous feed rate of 190 to 230 liters per hour and
a maximum wall temperature of 1650°C, and has successfully destructed
chlorinated wastes such as herbicides and pesticides.
8
-------
_^_ -. ©
g> YY
^*—
43b—
S^i.-j€ "^IT^ IIMMI
Figure 3-1. TMC Test Facility Process Flow Diagram (Figure Courtesy of TMC, Van Nuys, California)
-------
STACK SAMPLING
'' PLATFORM
,V
EXHAUST
STACK
SEPARATOR
TANK
I
VENTURI
SCRUBBER
CONTROL
ROOM
REACTION
TAILPIPE
SUE INCINERATOR
- COMBUSTION CHAMBER
AUXILIARY
. FUEL INLET
WASTE SUPPLY
INLET
'igure 3-2. Overall Test Installation (Courtesy of The Marquardt Company)
10
-------
SPENT SCRUBBER WATER CATCH TANKS
.REACTION TAILPIPE
CHAMBER EXIT
THERMOCOUPLES
SUE* INCINERATOR
COMBUSTION CHAMBER.!
S AIR SUPPLY SYSTEM
PREHEAT
THERMOCOUPLf
Figure 3-3. Incineration Unit (Courtesy of The Marquardt Company)
-------
r •
WASTE FUEL FEED TANK
RECIRCULATION PUMP
DRUM TRANSFER
STATION
TRANSFER PUMP
Figure 3-4. Waste Fuel Supply System (Courtesy of The Marquardt Company)
-------
TM MMMIUMT CMFOtATMM
Mn /*-**- y v
8MS/C
rzsr sysre^t
7C4.
wttrf
_Lf
/A/
i.zsjr.o.
.-r_tf
r
^_ 4-
LL
_/x«7 rt>
**
f*## SfOfJ
US. 01TKMr "O. 3,074-. 44,9
Figure 3-5. Baste Burner System Geometry
-------
3.1.2 Air Supply System
Combustion air was supplied from the 4,140-kilopascal (600 psi) facil-
ity storage tanks. The Marquardt facility normally maintains a high pres-
sure air supply for ramjet engine testing. Combustion air flow rate is
regulated by a remotely controlled valve and metered with a sonic venturi.
Air feed lines to the combustion chamber are shown in Figure 3-3. This
system is used by TMC due to its availability. Commercial installations
utilize conventional forced draft fans for combustion air.
3.1.3 Waste Fuel System
Waste fuel was transferred from 208-liter (55 gallon) drums into a
1,140-liter, 3,450-kilopascel (500 psi) feed tank (Figure 3-4). The tank
was pressurized with nitrogen, and fuel was fed to the incinerator through
either of two parallel 5 micron filters, a remote control valve, and a
turbine flowmeter. The fuel line was purged with N2 after use. A recir-
culation system was used to mix the tank contents. In commercial practice,
the waste feed would be pumped from a run tank into the incinerator.
3.1.4 Auxiliary Fuel System
The auxiliary fuel inlet to the incinerator is shown in Figure 3-2.
Propane gas was used to preheat the incinerator system to an equilibrium
temperature of approximately 815°C before introduction of the waste liquid.
Once combustion was established with the waste fuel, the propane was
turned off and a small air flow introduced into the propane system for
burner nozzle cooling. For controlled test purposes, a gaseous nitrogen
(GN2) purge was included to clean the system during shutdowns. All flows
were controlled remotely and metered with sonic venturi.
3.2 INSTRUMENTATION
Instrumentation capability provided at the Marquardt facility for this
test program is shown on the Figure 3-1 schematic. All instrumentation was
calibrated and certified by the Marquardt Standards Laboratory prior to
initiation of testing. Measurements were made of all process parameters,
including pressures, temperatures, and mass flow rates. In addition, on-
line gas analysis of combustion products was conducted.
3.2.1 Process Parameters
Continuous measurements were made of all pressures, temperatures, and
mass flow rates required to control and monitor the incineration process,
including the scrubbing and sampling systems. Process instrumentation is
listed in Table 3-1. Figure 3-6 shows the process control panel and
instrumentation readouts. Direct pressure gauges were located outside of
the viewing windows. Temperatures were recorded on 8 or 16 point recorders.
Turbine flowmeter readouts were read directly in Ibs/sec. Estimated waste
flow measurement accuracy was ±2 to 4 percent, depending on whether the
flow rate was in the mid range of the calibration or not.
14
-------
Table 3-1. Instrumentation for Waste Destruction Testing - Marquardt
(Refer to Figure 3-1)
System
Air
Natural
Gas or
Nozzle
Cooling
A1r/GN2
Waste
Fuel
SUE
Burner
Scrubber
Systems
Sampling
Systems
Symbol
d*
PT1
TC-1
d*
n
T2
£
Ps2
TC-2
PT5
Wf
TC-3
PT3
pTg
Py
Li
TC-8
PT3
*P1
TC-4
TC-5,6,7
Wr
c
Ww
TC-12
TC-10
TC-9
PT6
"T7
Tc-n
Function 1 Note
I
Air flow Venturl
Upstream total pressure
Inlet total temperature
Gas flow Venturl
Upstream total pressure
Throat static pressure
Inlet total temperature
Manifold pressure
Turbine flowmeter
Inlet fuel temperature
Manifold pressure
Supply tank pressure
Supply pressure downstream
filters
Fuel tank liquid level
Fuel tank temperature
Burner Inlet pressure
Burner pressure drop
Burner air Inlet temperature
Exhaust gas temperature
Caustic Solution
Water flowmeter
Scrubbed effluent gas tem-
perature
Scrubber water exit tem-
perature
Caustic solution Inlet tem-
perature
Beckman probe cooling air
TRW probe purge air
Beckman sample gas temperature
N/A
2
1
N/A
2
2
1
2
2
1
2
2
2
3
2
2
2
1
1,2
2
2
2
1
1
2
2
2
Size or Range
0.80" d1a.
0-200 psig
0-100°F
0.227" dia.
0-50 psig
0-50 psig
0-100°F
0-50 psig
.05-. 20 pps
0-200°F
0-500 psig
0-500 psig
0-500 psig
Sight Gauge
0-200° F
0-10 psig
0-25" H20
0-1000°F
0-2400° F
0.5-1.6 pps
0.5 - 3 pps
0-200°F
0-200°F
0-100°F
0-100 psig
0-100 psig
0-300°F
NOTES
1. Continuously measured and recorded parameters.
2. Continuously measured but manually read/recorded every 30 minutes or
whenever deemed necessary by operational changes.
3. Manually measured/checked and recorded whenever deemed necessary.
15
-------
o
DIRECT PRESSURE
INSTRUMENTS
r PW-
REMOTE PRESSURE!
INSTRUMENTS
Irrmtrm
mt •
'''.'//.•' .-••'"•'*'
FLOW
READOUTS;
»•••••
TEMPERATURE
RECORDERS
SYSTEM CONTROLS
•w. ,'^..,-
,
•
t '
Figure 3-6. Process Controls and Instrumentation (Courtesy of The Marquardt Company)
-------
3.2.2 Gas Analysis
On-line continuous monitoring of combustion gas products was performed
for all tests using the following instruments (shown in Figure 3-7).
CO - Beckman 315A (Infrared), Span 0-5000 ppm
NO - Beckman 315A (Infrared), Span 0-200 ppm
CHX - Beckman 109A (Flame lonization), Span 0-30,000 ppm (Variable)
Valves were provided to allow sampling from either the combustion
zone or the stack exit, as shown in Figure 3-1. Data from the combustion
zone gas analysis was used to indicate relative system combustion effici-
ency and to verify attainment of steady state operating conditions prior
to initiating gas sampling runs.
3.3 EMISSION CONTROLS
Atmospheric emissions from the combustion of liquid wastes during the
Marquardt incineration tests were controlled by a high energy venturi
scrubber system. The major components of the scrubber system, shown in
Figures 3-2 and 3-3, are described in the following paragraphs. Disposal
of spent scrubber liquids is also discussed.
3.3.1 Venturi Scrubber and Separator Tank
Combustion gases leaving the reaction tailpipe passed through the
high energy venturi scrubber into the separator tank (Figure 3-2). Quench-
ing and scrubbing water with caustic solution (10-12% NaOH/water) were
injected at the venturi inlet and mixed with the combustion gas at veloci-
ties up to 125 m/sec in the venturi throat. The separator tank is equipped
with a metex screen demistor. Spent scrubber water was collected in the
separator tank and transferred to holding tanks (Figure 3-3) by a cyclic
pumping system. The water saturated, scrubbed effluent gases were dis-
charged up the stack.
3.3.2 Caustic Solution and Water Supply System
Fresh water to the venturi scrubber was supplied from the
965-kilopascal (140 psi) facility water system at the rate required to
quench the gases and scrub at a rate of approximately 19 liters/min per
28 actual cubic meters per minute. When burning the C-5,6 waste, caustic
solution was also injected to neutralize the HC1 and Clg in the combustion
products . Caustic flow (approximately 12 percent NaOH in water) was
injected at a rate about three times the amount required to neutralize the
theoretically expected amounts of HC1. Both flows were regulated by
remote control valves and metered by turbine flowmeters.
17
-------
SAMPLE FLOW «
CONTROLS
HYDROCARBON
ANALYZER
Figure 3-7.
TMC On-Line Gas Analysis Equipment (Courtsey of
The Marquardt Company)
18
-------
3.3.3 Scrubber Liquid Collection System
Spent scrubber water was collected in any of three 21-kiloliter
(5,500-gallon) catch tanks (Figure 3-3). The pump cycle was monitored to
allow taking of samples during actual separator tank pumpout. The col-
lected liquid was discharged to the facility's 5.3 megaliter reservoir.
Reservoir water is analyzed and neutralized, if required, before release
to the municipal sewer system.
19
-------
4. TEST DESCRIPTION
This section presents the manner in which the tests were carried out.
It is divided into the following subsections, listed in order of
discussion:
• Physical and chemical description of the wastes that were
tested
• Operational procedures used and a test-by-test commentary
• Sampling methods
• Analysis techniques
• Description of the problems encountered related to the facility
and sampling.
4.1 WASTES TESTED
The two wastes selected for testing at TMC were from the manufacture
of ethylene and hexachlorocyclopentadiene (C-5,6). Survey samples were
received as early as possible before the tests and representative samples
were obtained by compositing another sample at the time the waste material
to be burned was drummed. These samples were then analyzed for both organic
and inorganic composition to determine the expected compounds of interest
in the test burn samples.
4.1,1 Ethylene Manufacturing Waste
The ethylene waste was a clear light brown solution with no apparent
sediment or particulate at room temperature. It had a strong pungent odor
but did not fume when exposed to the atmosphere. The loss on ignition at
800°C (LOI) for this material was greater than 99.99 percent, indicating a
very low concentration of non-volatile inorganics. Other physical charac-
teristics of the waste sample were:
• Heat content of 10,050 kcal/Kg (18,100 Btu/lb)
9 Specific gravity of 0.91
• Kinematic viscosity of 1.28 centistokes at 22°C
Elemental analyses performed on the ethylene waste showed the composi-
tion to be:
carbon - 88.7 percent
hydrogen - 8.6 percent
nitrogen - 0.13 percent
sulfur - 1.3 percent
halogens as chlorine - 40 ppm
20
-------
4.1.1.1 Organic Composition
Analytical techniques used to determine the organic composition
included infrared spectrophotometry (IR), low resolution mass spectroscopy
(LRMS), and combined gas chromatography/mass spectroscopy (GC/MS). The IR
spectrum revealed the presence of aromatic and aliphatic hydrocarbon struc-
tures. All significant peaks correlated to major constituents found in
the GC/MS analysis discussed below. No evidence of -OH, -NH, -COOH, -C=0,
or any other peaks indicating the presence of other functional groups were
observed. Analysis by LRMS indicated alkyl substituted benzenes, indenes,
and napthalenes as the primary species. The spectra indicated a rather
complex mixture of species, but no appreciable signal was detected for
compounds of molecular weight over 170.
The combined GC/MS analysis of the survey sample found eleven species
whose chromatographic response was 0.1 percent or more of the total nor-
malized area. The results, summarized in Table 4-1, showed the waste to
consist primarily of unsaturated cyclic and aromatic compounds. No organic
sulfur compound could be identified which would relate to the 1.32 percent
S found by elemental analysis. Thus, the sulfur is most likely present
in an inorganic form. These compositional values are based on equal
response of all these hydrocarbons to the flame ionization detector (FID).
This assumption is adequate for the purpose of identifying major compounds
and/or partial decomposition products in the combustion test samples. The
representative sample of the ethylene wastes fed to the incinerator was
analyzed by GC under the same conditions used to analyze the survey sample.
The chromatograms were compared and the results indicated that the two
materials were identical except for small differences in the relative peak
areas of a few of the smaller peaks. The relative concentrations of the
major constituents of the waste remain essentially the same as shown in
Table 4-1.
Table 4-1. Composition of Ethylene Manufacturing
Waste Representative Sample
Compound
Approximate Concentration
(percent)
Cyclopentadiene
1,3, 5-Hexatriene
Methylcyclopentadi ene
Benzene
Cyclohexadiene
Toluene
Xylene
Styrene
Methyl Styrene
Indene
Molecular Weight 131 or 132
Methylindene and Divinyl Benzene
Naphthalene
Ethyl Naphthalene
30.
20.
20.
3.
20.
0.8
5.0
<0.1
1.
3.
0.2
21
-------
4.1.1.2 Trace Elements
Trace elemental analysis was performed by spark source mass spectro-
scopy (SSMS). This analysis showed no metals present at greater than 1 ppm
levels. A few potentially toxic metals were found at low ppb levels.
These elements and their predicted combustion concentrations from the SSMS
data, using Run IV as an example for waste feed rate and other operating
conditions, are shown in Table 4-2. Of these elements, only Pb and As are
susceptible to loss during the dry ashing sample preparation step for SSMS
and may be somewhat low. The mercury determination was performed by a
hignly quantitative atomic fluorescence technique in order to be sure of
an accurate measurement. The extremely low concentrations of these ele-
ments in the ethylene waste indicate an expectation of correspondingly low
emissions.
Table 4-2. Trace Metals in the Ethylene
Representative Waste Sample
Element
Cr
Cu
Mn
Pb
As
Hg
Ba
Measured
Concentration
in Waste (ppb)
14
41
2
16
8
70*
1
Predicted
Concentration Produced
by Combustion (yq/m3)
1.0
3.0
0.2
1.2
0.6
4.9
0.1
*Determined by atomic fluorescence
4.1.2 Hexachlorocyclopentadiene Manufacturing Waste
The C-5,6 waste was a dark brown liquid with darker, suspended, semi-
solid, wax-like particulate matter. The material had a slight odor but did
not fume when exposed to the atmosphere. The material lost 99.96 percent
of its mass upon ignition at 800°C, indicating a low inorganic content.
Heat content for the survey sample was determined to be 2,400 Kcal/Kg
(4,300 Btu/lb). Its specific gravity was 1.74, and its viscosity at 22°C
was 11.34 centistokes.
22
-------
Elemental analyses performed on the C-5,6 waste showed the composition
to be:
carbon
hydrogen
nitrogen
sulfur
halogens as chlorine
20.8 percent
0.67 percent
0.37 percent
0.016 percent
76.5 percent
4.1.2.1 Organic Composition
Analytical techniques used to determine the organic composition in-
cluded IR, LRMS, and GC/MS. The infrared spectrophotometric analysis did
not reveal the presence of -CH, -OH, -NH, -C=0, -COOH, or any other func-
tional groups other than those attributable to halogenated organics. Most
of the major peaks in the spectrum matched those of hexachlorocyclopenta-
diene, Cede- Other peaks matched that of its eight chlorine analogue,
octachlorocyclopentene, CsClg. Low resolution mass spectrometry confirmed
the two major species to be €5015 and CsClg- Heating the waste sample and
pumping removed the most volatile compounds of the mixture. The residue
yielded a mass spectral pattern of higher molecular weight analogues of
CcCls including the dimer of pentachlorocyclopentadiene. Analogues with
molecular weights as high as 470 were noted.
Combined gas chromatographic-mass spectrometric analysis of the survey
sample found five compounds whose chromatographic response was 0.1 percent
or more of the total normalized area. The results of this analysis are
presented in Table 4-3. As mentioned previously, these values are based
on assuming an equal response to all hydrocarbons by the FID.
Table 4-3. Composition of Hexachlorocyclopentadiene
Waste Survey Sample
Compound
Approximate Concentration
(percent)
hexachlorocyclopentadiene
C5HCl7 heptachlorocyclopentene
Chromatogram had doublet peak but
mass spectrum shows only
octachlorocyclopentene, C5Clg
MW 308, chlorinated organic
MM 281, chlorinated organic
MW 376, C5HC19 or C8C18
50.
5.
30.
15.
0.2
0.1
23
-------
The representative sample of the C-5,6 waste was analyzed by GC using
an OV-225 column and compared to the chromatogram of the survey sample
obtained under the same conditions. The chromatograms were very similar
including even the minor constituents. There was, however, a change in the
relative composition of the major constituents. The areas were computed
and normalized under the assumption of equal response to equal weights of
the major constituents. The composition of the C-5,6 representative sample
is shown in Table 4-4. The remaining minor constituents shown in Table 4-2,
did not appear to significantly change relative to each other. However
these minor constituents increased in the normalization to account for the
drop in the level of 05^7 and CsClg from those measured in the survey
sample. These changes do not impact the conduct of the test or the analyt-
ical methods for the test samples.
4.1.2.2 Trace Elements
Elemental inorganics were determined by SSMS on the representative
sample. The highest ppm concentrations found were for innocuous metals,
such as Fe @ 82, Si @ 27, Al @ 38, Na @ 34, and Ca @ 25 ppm. Lower and
trace ppb levels were found for numerous other non-hazardous elements.
The potentially toxic elements, their concentrations in the waste and their
predicted combustion concentrations using Run VI for operating condition
data (including the dilution with No. 2 oil) are shown in Table 4-5. Of
these elements, all have some tendancy to volatilize as chloride salts due
to the conditions (high Cl concentration, dry ashing) of their preparation
for SSMS analysis. Thus, all of the predicted combustion values could be
low, with the exception of mercury which was determined by a highly quanti-
tative atomic fluorescence technique in order to be sure of an accurate
measurement.
4.2 OPERATIONAL PROCEDURES
Detailed operating procedures, including both a test plan and a safety
plan, were reviewed and approved prior to arrival of the TRW sampling team
on site. Procedures and operating conditions were also recorded during the
field tests. Following are brief summaries of both plans, a test-by-test
commentary on events that took place in the field, and information on the
disposal of waste residues.
Table 4-4. Composition of C-5,6 Waste
Representative Sample
Compound
Approximate Concentration
(percent)
Hexachlorocyclopentadiene
Heptachlorocyclopentene
CeClg Octachlorocyclopentene
Other Minor Constituents
66
3
29
2
24
-------
Table 4-5. Trace Metals in the C-5,6 Waste
Element
Cr
Cu
Mn
Pb
As
Hg
Ba
Co
Se
V
Measured
Concentration
In Waste (ppb)
12,000
1,900
330
7
1
10*
36
42
3
4
Predicted
Cone. Produced by
Combustion (wg/rn^)
400
64
11
0.2
0.03
0.3
1.2
1.4
0.1
0.1
*Determined by atomic fluorescence
4.2.1 Test Procedures
Tests at TMC were run with two wastes, ethylene manufacturing wastes
and C-5,6, as previously described in Section 4.1. The basic procedure for
each waste test was:
e Fill waste tank, measure specific gravity
• Verify instrumentation and sampling systems ready
• Ignite on auxiliary fuel (propane) and stabilize
temperatures
• Activate on-line analyzer system
• Transfer to waste fuel combustion and observe effluent
• Stabilize system thermally
• Extended burn duration
- Process data acquisition
- Combustion gas composition data acquisition
- Combustion zone and stack gas sampling
- Scrubber liquid sampling
25
-------
• Transfer to auxiliary fuel combustion
• Shutdown and secure, inspect for and collect samples of
deposits in combustion chamber.
The test series for each waste consisted of three burn periods during
which a three hour combustion gas sample was acquired at steady state oper-
ating conditions. A three hour sample run with No. 2 oil was also required
to obtain background data prior to the C-5,6/No. 2 oil mixture runs.
Target test conditions for the ethylene waste were:
• Air flow rate - 680 g/sec
• Fuel flow rate - 27 to 41 g/sec
• Theoretical combustion temperature - 1430° to 1815°C
Initial test conditions for the C-5,6 wastes were:
• Air flow rate - 680 g/sec
• Fuel flow rate - 39 to 66 g/sec
• Theoretical combustion temperatures - 1430° to 1870°C
• C-5,6/No. 2 oil mixture ratio - 1/1 and 1/2 by weight
4.2.2 Safety Procedures
TMC standard safety procedures for handling and incinerating industrial
chemicals were observed during this test program, including the following:
• Only authorized personnel with prior approval were permitted
in the test area during operations.
• Chemicals were handled only by personnel wearing suitable
protective clothing and trained in handling such materials.
• Safety shower and eyewash was available in immediate area.
• All leaks or spills were to be flushed with JP4 and incinerated
in the same manner as test wastes.
• Visual observation of the test system was maintained at all
times during operation.
• Safety ropes were used to isolate the test area during actual
operation.
26
-------
o Canister gas masks were available to all personnel
- Masks were placed in easy access in control room
- Personnel outside control room carried masks
• Emergency medical treatment was available both onsite and
at defined outside medical facilities.
4.2.3 Test Commentary
All recorded data for the test burn incinerator operating conditions
were provided by TMC and are summarized in Table 4-6. Each test was num-
bered consecutively by TMC Run Nos. 1 through 15. TRW Run Nos. I through
VII were assigned only to runs during which combustion zone and stack gas
samples were acquired. The ratios shown in Column 3 for the C-5,6/No. 2
oil are by weight. Certain of the thermocouples (i.e., TCs and TCe, TCy
and TC-|3) are redundant and are meant to provide an average temperature
at the same point.
After checkout and background data tests with No. 2 oil were accom-
plished, three test conditions were evaluated with both ethylene and C-5,6
wastes. Initial test conditions included calculated combustion temperature
variations from 1315° to 1760°C. Corresponding measured temperatures at
the end of the reaction tailpipe where combustion gas samples were taken
were 982° to 1150°C.
4.2.3.1 Checkout and Background Tests
A series of tests were conducted with No. 2 oil as fuel to checkout
the incineration and sampling systems. A full duration sampling run was
then performed to obtain background emission data burning No. 2 fuel oil
only.
TMC Run No. 1: An initial checkout run with No. 2 oil was made to
verify operation of the TRW trailer on-line instrumentation as well
as to check out the incinerator and scrubber systems. Sampling
trains were operated at the combustion zone and exhaust stack, but
no actual acquisition of samples was intended. This run was 70 min-
utes in duration at a reaction tailpipe temperature of 1132°C. Com-
bustion sample qas temperatures were 105° to 110°C at the filter
inlet. All'systems operated satisfactorily, and preparations were made
for full duration sample tests.
TMC Run No. 2: This test was conducted to measure stack gas velocities
after a second stack sampling train monorail system was added. Capa-
bility was thus provided to make sampling traverses 90 degrees apart.
Stack velocity measurements were utilized to select a probe nozzle
diameter for isokinetic sampling.
TMC Run No. 3; A baseline sampling test with No. 2 oil was begun at
the same operating conditions as TMC Tests 1 and 2. This test,
intended as a 3-hour sample run, was terminated after one hour when
27
-------
Table 4-6 Incinerator System Parameters Data Summary
1
TMC
Test
NO
1
2
3
4-1
4-2
4-3
4-4
5
6-1
6-2
6-3
6-4
7
6
9
2
TRW
Run
No
-
-
-
-
-
-
-
3
Fuel
and Waste
To Fuel Oil
Ratio
No. 2 oil
No 2 oil
No 2 oil
No 2 Oil
1 , No 2 oil
~
-
II
-
ill
IV
Ethyl ene
Waste
Ethyl ene
Waste
Ethyl ene
waste
Ethyl ene
Waste
in i No 2 Oil
10 ~ (Rinse)
11-1 1- CcC!,/oil
56
11-2 1 2
1?
13
14
15
V !CcjCl6/oil
VI
VII
-
i i
C5C1 6/oil
C.C It/oil
11
JP-7
4
5
6
7
Process Flow Rates
Air
Avg.
Kg/Sec
0 694
0 689
0.689
0 694
0 694
0 694
0 694
0 694
0 694
0.694
0 694
OE94
0 572
0 567
0 558
0 685
0 572
0 572
0.567
0 572
0 572
0 504
Fuel
Avg
Kg/Sec
0 0324
0.0312
0.0315
0.0182
0.0227
0.0285
0 0317
0 0301
0.0227
0.0273
0 0349
0.0330
0.0316
0 0229
0.0317
0 0295
0 0356
0 0295
0 0340
0 0393
0 0490
0 0322
Scrub.
H20
Avg.
Kg/ Sec
1.42
1 37
1.36
1 39
1.39
1 41
1.41
1 37
1 36
1.36
! 36
1 37
1 20
1 19
1.23
1 39
1 13
1 07
1 13
1 07
0 88
1 31
Scrub.
NAOH
Sol.
Avg
Kg/Sec
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0 34
0 34
0 27
0 37
0 57
0 33
8
lhamb
Press
"3
Kilo-
Pascals
9 65
9.65
8 96
9 31
9 31
8.96
8.96
10 0
10.34
8.96
11.03
9 65
8.96
8 96
8 96
11 03
12.41
11 72
11 38
11 38
13 44
12 41
9
10
11
12
13
14
15
Temperatures
Fuel
TC3
Hvg.
°C
23
27
19
14
13
13
13
13
14
14
14
16
21
26
19
22
14
14
24
18
14
10
Pre-
Heat
TC^
Avg
216
216
210
99
143
182
202
202
135
182
254
246
321
238
381
199
257
232
271
25b
266
230
Chamb
Exit
TC5
Avg
1235
1249
1218
860
1029
1174
1229
1198
893
1054
1166
1137
1241
1031
1229
1168
1166
1060
1185
1219
1244
1204
Chamb
Exit
TCg
Avg
-
1254
1252
849
1029
-
-
1184
918
1049
1216
1169
1256
1052
1244
1168
1181
1088
1183
1184
-
1196
Smpl.
Point
TC?
Avg.
1132
1121
1132
860
971
1089
1138
1119
885
993
1143
1114
1151
960
Smpl.
Point
TC]3
Avg
-
1088
1104
843
949
1060
Stack
Exit
TC,2
Avg
80
79
79
69
74
78
1113 80
1073 ( 79
857
960
1093
1048
1046
968
- ,1144
-
-
-
1117
1108
1142
1110
1099
1077
933
1102
1121
1154
1107
71
75
79
79
79
74
79
79
79
79
78
79
81
78
16
17
18
19
20
21
22 23
24
Calculated Process Parameters
Fuel/
Air
Ratio
Kg/Kg
0.0470
0.0450
0 OlbO
0 0262
0 0327
0 0411
0.0456
0.0433
0 0327
0 0393
0 0503
0 0476
0 0553
0 0404
0 0566
0 0430
0 0622
0 0515
0 0596
0 0687
0 0861
0 0540
TC
Theo
°C
-
-
-
-
-
-
-
1646
1131
1318
1602
1535
1726
1349
1752
-
1393
1201
1348
1370
1378
-
TC
Avg
°C
-
-
-
-
-
-
-
1371
1001
1147
1360
1308
1412
1156
1448
-
1235
1097
1228
1242
1263
-
Chamb
Vel
m/Sec
-
-
-
-
-
-
44 1
32 9
36 9
42.7
41 7
36 9
30 8
36 9
-
Stay
Time
sec
-
-
-
-
-
-
-
0 131
0.176
0 157
0 135
0 139
0 157
0 188
0 157
NAOH
Applied
Avg
kg/Sec
-
-
-
-
-
-
-
-
NAOH/
NA04 Excess
Theo • Air
Avg
- ; -
- ' -
- ' -
-
- ; -
- ' -
; -
- 56 8
130
-
- ' 91 3
49 3
- ' - ' 58 0
Stoic
Fuel/
Air
Ratio
Kg/ Kg
-
-
-
-
-
-
-
0 0679
0 0752
0.0752
0 0752
0 0752
35 5 ,0 0752
- ;- lto.i
0 0752
- 1 - 32.9 |00752
, 1
33 1 '0 1741 0 040 38 54 7 ;0 0962
29 7 0 195
32 5 0 178
0 040 4 6 86 8
0 033 33 61 4
33 2
33 G
-
0 174 | 0 044 ;3 1 54 3
0 172
-
0 068 3 2 41 1
0 0962
0 0962
0106
0 122
" j " , i "
25
26 !
Sectarian
Stack
HC
'PHCHj
3 5
-
-
18
7
7
12
8
-
9
-
1.5
2
2 5
2
-
4
5
5
! * 5
O 0
-1 0
Snip' •
Point
HC
PPIICH^
3 5
""
\ * '
20
7
! 9
14
9
;
-
-
1 5 ,
3 ,
i
2 5
1 5 !
1 _
1
~ t
1
1
—
\
1 -
rs>
CO
-------
combustion zone gas temperatures at the filter inlet suddenly rose
from the normal 110° to 177°C. This higher temperature would have
been detrimental to the filter seals, so the test was terminated.
Cooling of the combustion gases by an air stream around the quartz
probe liner had been satisfactory in prior tests, but appeared to be
marginal in cooling capability. Water as well as air was added to
the liner coolant flow (Figure 4-14) for subsequent tests.
TMC Run No. 4: This test was conducted to verify the effectiveness
of the modified combustion gas cooling system prior to repeating the
sampling run. A temperature excursion from 860° to 1140°C as meas-
ured at the reaction tailpipe was made by varying the fuel/air ratio
of the incinerator. The water/air combustion gas cooling system
enabled filter inlet temperatures of 80° to 95°C to be maintained,
allowing a considerable safety margin in assuring temperatures of less
than 115°C at the filter inlet. Preparations were again made for a
full duration sample run.
TMC Run No. 5. TRW Run I: A background run with No. 2 oil for a full
3 hour combustion zone sample time was completed. Stack sampling was
conducted for two hours. Temperature at the reaction tailpipe aver-
aged 1095°C, and calculated theoretical combustion temperature was
1645°C (Table 4-6). Residence time in the reaction tailpipe was
0.131 seconds.
4.2.3.2 Ethylene Waste Tests
After completion of the baseline sample tests, No. 2 oil was drained
from the feed tank and ethylene waste was loaded.
TMC Run No. 6. TRW Run II: The first test with ethylene waste was
performed. Prior to sampling, a 3 point traverse of reaction tailpipe
temperatures was made at nominally 870°, 980°, and 1150°C, with cor-
responding combustion temperatures of 1130°, 1320°, and 1600°C. A 3-
hour sample test was made at 1090°C reaction tailpipe temperature and
1535°C calculated combustion temperature. Residence time was 0.139
seconds. Deposits in>the combustion chamber after the test were
photographed (Figure 4-1) and clinker samples were taken.
TMC Run No. 7; This run was intended as the second ethylene sample
test at an increased reaction tailpipe temperature of 1150°C (combus-
tion temperature of 1725°C), and residence time of 0.157 seconds.
After one hour of sampling, filter inlet temperature on the combustion
zone sample train suddenly dropped from a normal 88° to 93°C, to 16°C
and the quartz probe liner was observed to have moved, probably re-
tracting the probe tip through the seal. The test was discontinued,
as combustion gases were not being acquired by the sampling system.
The probe liner was safety-wired to prevent axial movement in sub-
sequent tests, thus precluding repetition of this leak.
TMC Run No. 8. TRW Run III; The second ethylene run was successfully
completed. Three hours of combustion zone sampling were conducted at
a reaction tailpipe temperature of 970°C and a calculated combustion
29
-------
LO
O
Figure 4-1 MA-II Clinker (Courtesy of the Marquardt Company)
-------
temperature of 1350°C. Residence time was 0.188 seconds. Deposits
in the combustion chamber were again photographed (Figure 4-2) and
samples taken.
TMC Run No. 9. TRW Run IV: The third ethylene test was conducted at
increased operating temperatures. Reaction tailpipe temperature was
nominally 1150°C and calculated combustion temperature was 1750°C.
Residence time was 0.157 seconds. A 3-hour sample run was completed,
although accumulation of deposits in the combustion zone resulted in
some excursions in system temperatures during the run. Deposits were
photographed (Figure 4-3), sampled, and removed prior to the next run.
TMC Run No. 10: This run was made to flush the system prior to C-5,6
runs by rinsing the fuel tank with No. 2 oil and operating the in-
cinerator on No. 2 oil for over one hour.
4.2.3.3 Hexachlorocyclopentadiene Tests
The heating value of the C-5,6 waste was not sufficient to maintain
combustion, and therefore was mixed with No. 2 oil for these tests. Dif-
ferent mixture ratios of C-5,6 to No. 2 oil were incinerated, as noted in
Table 4-6 and mentioned in the following test descriptions.
TMC Run No. 11: A checkout run was made with C-5,6 waste. A 15-
minute sample was taken with a TMC benzene train for analysis before
full duration testina. Laboratory analysis of the samole indiratpri
less than 0.22 ppm of C-5,6 in stack exhaust. A TRW gas chromatograpn
read hydrocarbon values of 20 ppm, indicative of satisfactory com-
bustion efficiency.
TMC Run No. 12, TRW Run V: The first sample run with C-5,6 was made
at a reaction tailpipe temperature of 1107°C and calculated combustion
temperature of 1363°C. Residence time was 0.178 seconds. The C-5,6
waste was mixed with No. 2 oil at a weight ratio of 1:2 (1 kg C5.6
to 2 kg No. 2 oil). The test was terminated at 2 hours 15 minutes
when caustic solution in the large impinger in the combustion zone
sample train appeared to be neutralized by HC1; however, a sufficient
sample had been obtai-ned for analyses. Deposits were photographed,
(Figure 4-4) sampled, "and removed. TMC changed the fuel manifold dia-
meter from 23 to 20 cm to improve combustion in an attempt to reduce
deposits. A decision was made to maintain high combustion tempera-
tures of nominally 1370°C, and vary the ratio of C-5,6 to No. 2 oil
for subsequent tests.
TMC Run No. 13. TRW Run VI: The second C-5,6 sample run was made at
a 2:3 weight ratio of C-5,6 to No. 2 oil. Reaction tailpipe nominal
temperature again was 1107°C, and calculated combustion temperature
was 1370°C. Residence time was 0.174 seconds. Some fluctuation in
system pressures and temperatures occurred due to deposit formations.
A full 3-hour sample duration was attained. Deposits were photo-
graphed (Figure 4-5) and removed from combustion chamber.
31
-------
-'A
•
Figure 4-2 MA-III Clinker (Courtesy of the Marquardt Company)
-------
Figure 4-3 MA-IV Clinker (Courtesy of the Marquardt Company)
-------
• .
Figure 4-4 MA-V Clinker CCourtesy of the Marquardt Company)
-------
00
en
Figure 4-5 MA-VI Clinker (Courtesy of the Marquardt Company)
-------
TMC Run No. 14. TRW Run VII: The third C-5,6 run was made at a
1150°C reaction tailpipe temperature, and a 1378°C calculated com-
bustion temperature. Residence time was 0.172 seconds. Equal weights
of C-5,6 and No. 2 oil were mixed for this test. Deposit formations
again caused some system fluctuations. Shutdown was premature at
about 2-1/2 hours due to depletion of waste fuel in tank. Deposits
were photographed (Figure 4-6) and samples taken.
TMC Run No. 15: The fuel tank was rinsed with JP-7, which was then
burned for over 1-1/2 hours to flush the system.
4.2.4 Disposal of Waste Residues
The residual ethylene and C-5,6 waste materials along with the empty
drums left from the test burns were disposed of by a contract disposal
firm in accordance with all applicable regulations.
4.3 SAMPLING METHODS
Sampling methods used in the tests at TMC were chosen to cover three
basic areas. These were:
1) Continuous, on-line monitoring of gas composition to deter-
mine and follow steady state conditions.
2) Collection and concentration of hot zone combustion products
to identify and quantify the trace organic and inorganic
species formed.
3) Collection of final emission and waste products to evaluate
the environmental safety of the tests.
Following is a brief summary of the methods for each of these areas.
More detailed discussions can be found in the Marquardt Analytical Plan.*
4.3.1 On-Line Gas Monitoring
Gases were drawn continuously from the hot zone through a ceramic
probe and then through a heated Teflon sample line to the trailer. The
location of this probe at the Marquardt test stand is shown in Figure 4-7.
The gas then entered the system shown in Figure 4-8. The gas conditioner
supplied a cool, dry, particulate free sample to all of the analyzers with
the exception of the hydrocarbon (HC) monitor which uses an untreated sam-
ple. A heated Teflon line carries the HC gas sample from a tee in the
unconditioned sample line to the HC analyzer. Because it uses an untreated
gas sample, the HC monitor is the only instrument exposed to the moist acid
(HC1 coming from the C-5,6 waste burns) stream and is thus the most
*TRW Document #27033-6001-RU-00, "Analytical Plan for Facility No. 1 Tests
to be Conducted at Marquardt Corporation", by J. F. Clausen and C. A. Zee.
36
-------
• •
•
Figure 4-6 MA-VII Clinker (Courtesy of the Marquardt Company)
-------
!0
00
CONTROL BOX
FOR STACK
TRAIN
CONTROL BOX
FORCOMBUSTION._
ZONE TRAIN
HEAT
SHIELD
FOIL WRAPPED
MULLITE PROBE
FORON-LI&JE
INSTRUMENTS
HEATED SAMPLE LIN
O TRAILER
WATER-COOLED
PROBE FOR
SAMPLING TRAIN
Figure 4-7 Sampling System For On-Line Monitors (Courtesy of the Marquardt Company)
-------
to
to
2B,
rl-3
HCF
'1-2
COMPRESSED
AIR SOURCE
VENT
2 WAY BALL VALVE
WHITEY SS-4234
3 WAY BALL VALVE
WHITEY SS-43XS4
SS-45X38
5 WAY BALL VALVE
WHITEY SS-432
FLOWMETER WITH INTEGRAL
CONTROL/SHUTOFF VALVE
HYDROCARBON REMOVAL
FILTER
PRESSURE REGULATOR WITH
GAGE (2 STAGE)
DRYER/FILTER
CONDENSATE TRAP
1/4" TUBE
I/S" TUBE
Figure 4-8 Gas Conditioning and Analysis System For On-Line Instruments-
-------
susceptible to corrosion. To avoid potential problems, a Perkin-Elmer 881
gas chromatograph with an open tubular column and a flame ionization detec-
tor was substituted as a total hydrocarbon analyzer during tests with the
C-5,6 waste.
The monitoring instruments used are listed with their operating ranges
in Table 4-7. Data was recorded on Hewlett-Packard 680M strip chart rec-
orders. Figure 4-9 shows the instrument racks mounted in the sampling
trailer. The analyzers, recorders, and manifold valves were all located
in racks to provide ease of operation and accessibility. The location of
calibration gases and gas conditioning system across from the racks is
shown in Figure 4-10, and the work area behind the racks is shown in
Figure 4-11.
Table 4-7. Description of On-Line Instruments
Species Analyzed
Manufacturer
and Model
Range*
Total hydrocarbons (HC)
Carbon monoxide (CO)
Carbon dioxide (C02)
Oxygen (02)
Oxides of nitrogen (NO )
A
Beckman
model 402
Beckman
model 865
Beckman
model 864
Taylor
OA 273
Thermo
Electron
model 10-A
0.05 ppm - 10% with
eight ranges
2-200 ppm
10-100 ppm
0.05 -5%
0.02 - 20%
0.05 - 5%
0.25 -25%
1 - 100*
0.05 - 10,000 ppm
with eight ranges
*A11 of these manufacturers report an accuracy of ±1 percent of full scale
for their instruments.
40
-------
Hgure 4-9 Instrument Racks
-
-------
i
.
CALIBRATION GAS CYLINDERS
Figure 4-10 Calibration Gases and Gas Conditioner
^
Figure 4-11 Work Area Behind Instrument Racks
42
-------
4.3.2 Sampling of Combustion Products
The sampling train used to collect hot zone gases, vapors, and parti-
culate is shown schematically in Figure 4-12 and as set up for a test run
in Figure 4-13. It consisted of a standard EPA Method 5 train with the
following important modifications.
• There was a stainless steel jacketed, water-cooled probe
(shown schematically in Figure 4-14} with a quartz liner.
The liner provides an inert surface for the sample gas and
the cooled, stainless steel jacket shields this gas from
extreme hot zone temperatures in order to quench any further
reactions of the sample constitutents. Further cooling of
the gas can be modulated by aspirating an air/water mixture
into the space between the steel jacket and quartz liner.
• Special fittings were fabricated to allow a back purge of the
probe with purified compressed air while the sampling train
was not in operation. This eliminated the possibility of con-
tamination from the relatively high amounts of organic, partial
combustion products produced during start-up and shut-down of
the incinerator. The back purge connection was made at the
point where the dogleg from the probe liner mates to the filter.
• A chromel/alumel thermocouple was potted into the dogleg going
from the quartz probe liner to the filter housing to check
the temperature of the gas stream at that point.
• An ultra high-purity glass fiber filter was used, Gelman Spec-
troquality Type A. The filters were muffled to remove organ-
ics and have extremely low background levels of inorganics.
They were tared by desiccating and weighing on consecutive
days to a constant weight { ±0.1 mg), and were then stored and
handled throughout the tests and analyses in glass petri
dishes.
• A solid sorbent trap, designed to adsorb the organic consti-
tuents in the sample gas stream, is located downstream of the
heated filter and upstream of the first impinger. The sorbent
trap, with overall dimensions of 170 x 45 mm, contained ^40 g
of XAD-2, an Amberlite resin of the type commonly used as a
chromatographic support.
• A Teflon valve was added to the glass connector between the
sorbent trap and the first impinger through which glass bulbs
were filled with the sample gas to be analyzed for any volatile
or gaseous components not collected by the sorbent trap. This
valve is also shown in Figure 4-13.
• For the C-5,6 test burns where large quantities of HC1 were
produced by combustion, a large, two-liter impinger (shown in
Figure 4-13) was added between the solid sorbent trap and the
first modified Greenburg-Smith (G-S) impinger. One liter of
43
-------
4 INCH
FILTER
HOLDER
INCINERATOR
WALL
WATER
COOLED
PROBE
SOLID
SORBENT
TRAP
GAS
SAMPLE
VALVE
QUARTZ
LINER
HEATED AREA-
THERMOMETERS
ORIFICE
THERMOMETER
CHECK
VALVE
BVALVES IMPINGERS VACUUM
(MAXIMUM SIX) GAUGE
7
VACUUM LINE
MAIN
VALVE
AIR-TIGHT
PUMP
DRY TEST METER
Figure 4-12 Combustion Zone Sampling Train Schematic
-------
:
P ARTICULATE
FILTER
£1
SOLID SORBENT
TRAP
NTT*
ATER-COOLED
PROBE
SAMPLE VALVE
LARGE CAUSTIC
IMPINGFR FOR
TESTS V THRU VII
BACK-PURGE
LINE
ASPIRATED AIR/WATER
COOLANT LINES
Figure 4-13 Combustion Zone Sampling Train (Courtesy of the Marquardt Company)
-------
PROBE
LIIIER
AIR/WATER COOLANT
FOR SAMPLE GAS
INLET
Figure 4-14 Water Cooled Probe Schematic
15 to 20 percent NaOH was placed in this large impinger in
addition to 100 ml of 5 percent NaOH in each of the first
two G-S impingers. The third impinger was empty and the fourth
held silica gel. For the background No. 2 oil and ethylene
waste tests, only distilled water was used in the first two
G-S impingers.
This hot zone train was operated at a flow rate of approximately
30 liters/min for 3 hours during each test, thereby sampling an average of
4 to 5 cubic meters. Gas volumes were measured to 0.03 liter, with a leak
rate of less than 0.6 liter/min. Operating parameters for the train and
sample volume data are tabulated in Appendix B.
The following hot zone samples were obtained from each test:
• Solvent probe wash
• 10 cm diameter particulate filter
• Solid sorbent trap
e Grab gas
• Combined impinger solutions
• Acidified split of combined liquid impingers
• Spent silica gel
The location of the combustion zone sampling train at the test site is
shown in Figure 4-15.
46
-------
UMBILICAL TO
CONTROL BOX
;» •
SAMPLING TRAIN
CONTROL BOXES
VENTURI SCRUBBER
:,'
, itti^#4M
WATER-COO LEDfES
PROBE
COMBUSTION GAS
SAMPLING TRAIN
HEAT SHIELD
Figure 4-15 Sampling System For Combustion Zone (Courtesy of the Marquardt Company)
-------
4.3.3 Sampling Emissions and Waste Products
Samples of the stack effluent, spent scrubber water, and solid com-
bustor residue (clinker), were taken during and after each test to evaluate
the environmental safety of the final emissions. An EPA Method 5 test was
performed at the stack for particulate mass loading and composition deter-
minations. Location of the sampling train at the test site is shown in
Figure 4-16. The standard train was used and two right angle, six-point
traverses of the one-meter diameter stack were made, sampling for one hour
at approximately 20 liters/min. It was decided at the start of the Mar-
quardt tests that owing to the low velocity in the stack (6 to 8 ft/sec),
the sample would be drawn superisokinetically in order to have an adequate
particulate sample for compositional analyses. Gas volumes were measured
to 0.03 liter, with a leak rate of less than 0.6 liter/min. Operating
parameters for the train and sample volume data are tabulated in Appen-
dix B.
The following samples were obtained for each test from the stack sam-
pling train:
• Aqueous probe wash
• 10-cm diameter particulate filter
• Impinger solutions
* Acidified split of impinger solutions
• Spent silica gel
Gaseous stack effluents were determined by the use of Gastec®
indicating absorbtion tubes. Toxic gases in ppm levels such as NO , S02,
CL2, HC1, and COC12 were looked for at the stack.
All of the spent scrubber water samples were obtained from a tap down-
stream of the tank which holds the water after it has passed through the
venturi scrubber. Periodically during the test, as this water level rises,
a pump removes the scrubber water from the tank and transfers it to a stor-
age tank. It was during these pumping periods that composite samples were
taken which make up the entire sample. The city water control sample was
taken from a line which supplies water to the venturi scrubber, whenver
caustic scrubbing solutions are required in the scrubber, the caustic con-
centrate is blended with city water in a line leading to the venturi scrub-
ber. No tap or sampling point was available in the line between the blend-
ing point and the venturi scrubber, hence, it was necessary to prepare the
fresh caustic scrubber water control sample by blending the correct portions
of concentrated caustic with city water. The mixing was performed within
the one-gallon sample jug. Aliquots of these scrubber samples were acid-
ified for subsequent metal analyses.
Samples of solid residue (or clinker) which accumulated in the combus-
tor samples were collected when the burner assembly was removed after each
test burn. The residue was then scraped out from the burner as completely
as possible and wrapped in aluminum foil.
48
-------
TRAVERSE
MONORAILS
EFFLUENT GAS METHODS
SAMPLING TRAIN
SCRUBBED EFFLUENT
EXHAUST STACK
UMBILICAL TO
CONTROL BOX
Figure 4-16 Sampling System For
Stack (Courtesy of the Marquardt Company)
49
-------
4.4 ANALYSIS TECHNIQUES
Samples taken as described In Section 4.3, were analyzed for both
organic and inorganic constituents. When necessary, extractions were per-
formed first to concentrate the sample in a suitable form for analysis.
Techniques used for these extractions and analyses will be briefly summa-
rized here. For more detailed discussions, see the Marquardt Analytical
Plan.
4.4.1 Extractions and Sample Preparation
Both solvents and acids were used to extract organics and inorganics
respectively, from the appropriate samples. These procedures and the basic
sample preparation steps are listed by sample type:
• Probe Mashes
Combustion Zone.. The methylene chloride rinse was added
directly to the filters after they were folded and placed in
the Soxhlets for organic extraction (see filters below).
Stack. The acetone probe rinse was evaporated and the residue
weighed. This weight was added to the weight of the particu-
late on the filter for total mass loading calculations, in
accordance with EPA Method 5 procedures.
• Filters
Combustion Zone. The tared sample filters plus two controls
were desiccated and weighed on consecutive days to a con-
stant weight ±0.1 mg, and then extracted in a Soxhlet appa-
ratus for 24 hours with methylene chloride. Solvent extracts
were evaporated to 10 ml for analysis. Filters were then
cut in half and one of the halves was weighed, low tempera-
ture plasma ashed, reweighed, and extracted with constant
boiling aqua regia for two hours. This acid extract was
reduced to 50 ml for analysis.
The tared sample filters were weighed and cut in
One of the halves was weighed, low temperature plasma
reweighed, and extracted with constant boiling aqua
regia for two hours. The acid extracts were reduced to
50 ml for analysis.
Solid Sorbent Traps
Combustion Zone. The XAD-2 resin was transferred from the
traps to pre-extracted paper thimbles and then extracted in
a Soxhlet apparatus with pentane and methanol for 24 hours
with each solvent. These extracts were evaporated to
10 ml for analysis. Two unused traps were also extracted
for background values and a blank (empty thimble) control
was also run.
50
-------
The original analysis plan had called for extraction of the
sorbent traps with only one solvent, pentane. However, as
discussed in Section 4.5.2, a second extraction with meth-
anol was found to be necessary to remove the majority of
material adsorped on the resin. In addition, the first set
of traps to be extracted had even a third extraction with
benzene. The benzene removed very little additional material
and this third extraction was not continued through the
remainder of the traps.
Stack. No solid sorbent traps were used in the stack sampling
train.
• Grab Gas
Combustion Zone,. No special preparation was required.
Stack. No gas.samples were taken at the stack.
t Impingers
Combustion 7nnp and Stack. The volume of liquid in the
impingers was measured and the spent silica gel was weighed
in the field after each test burn to determine the amount
of water collected. The impinger liquids from the com-
bustion zone were also combined and a 100 to 250 ml aliquot
acidified in the field to stabilize the metals for anal-
ysis. No extractions or other special preparation steps
were performed on any of the impinger samples.
• Scrubber Waters
500 to 1000 ml aliquots of the scrubber water samples were
extracted for organics according to the procedure for the
separatory funnel extraction with Freon for oil and
grease from water recommended by the EPA Handbook on
Methods for Chemical Analyses of Water and Wastes (National
Environmental Research Center, Cincinnati, Ohio, 45268,
EPA-626-/6-74-003). However, instead of evaporating the
material to the dried residue, the Freon extracts were
concentrated to a 10 milliliter sample by use of a Kuderna-
Danish concentrating evaporator. Aliquots of this 10 mil-
liliter sample were then used for the survey analysis
(IR and LRMS) and for gravimetric determination of residual
material after evaporation at ambient conditions and im-
mediate weighing.
0 Solid Combustion Residues
The solid residues (clinkers) were first weighed, then
ground and blended to provide a homogeneous sample. Ten
to 20 g portions were extracted in a Soxhlet apparatus
for 24 hours with methylene chloride. The solvent ex-
tracts were then evaporated to 10 ml for analysis.
51
-------
4.4.2 Analytical Methods
After extraction of the samples for organic material and other
preparation for inorganic material, the concentrated extracts, and aqueous
solutions were analyzed by several methods which are summarized in
Table 4-8. A general treatment of the sample preparation and analytical
procedures is discussed below.
4.4.2.1 Organic Analyses
With the exception of the gas grab samples in glass bulbs, all of the
samples for organic analysis were in 10 ml solvent extracts. Gas samples
were analyzed on the Hitachi-Perkin-Elmer mass spectrometer. The samples
were contained in double ended glass sampling vessels which have a septum
through which syringe samples can be drawn. A gas tight syringe was used
to remove part of the sample from the sample vessel and inject it into a
heated stainless steel inlet system. The pressure of the sample in a con-
stant volume portion of the inlet system was measured with a quartz man-
ometer. Known pressures of helium and a 10.6 ppm propane standard were also
introduced to the mass spectrometer (MS) for the purpose of background and
calibration of instrument response. Tests samples were introduced to the
MS with interspersed background samples to ensure that the instrument was
not slowly accumulating a "memory" in the m/e peaks of interest.
The concentrated solvent extracts of the filters, sorbent traps,
scrubber waters, and sand bed samples were analyzed by gravimetry, IR, LRMS
and GC techniques. An aliquot of each extract was evaporated at ambient
conditions to remove the solvent. The residue was weighed and analyzed
by IR and LRMS.
The IR and LRMS analyses yield qualitative information about the
classes or types of compounds (e.g., hydrocarbons, phenols, ROMs, etc.)
present as well as an idea of the complexity of the concentrated sample.
Knowledge of the classes of compounds present provides a measure of the
toxicity, if any, of the residue. The detection limits for these analytical
techniques vary somewhat with the type of compound (see Table 4-8).
Two gas chromatograph systems were employed to cover the boiling point
range of compounds which were expected in the samples if less than complete
combustion were to occur. The first system described is for what are being
called low boilers. This includes the very volatile solvents including
methylene chloride, pentane and includes all of the compounds found in the
ethylene waste with boiling points increasing up to but not including
napthalene and ethylnaphthalene. These two compounds do in fact elute
from this column system but with such long retention times that they are
better analyzed with the second column system. The column system used for
low boilers is as follows:
Column Chromasorb 102,80/100 mesh, 6 ft x 1/8 inch
Temperature 200°C isothermal
Carrier Gas helium at 30 cc/min
Detector FID
52
-------
Table 4-8 Summary of Analytical Methods
en
CO
Method
Organic Analyses
Gravimetry
Infrared
Spectrophotometry
(IR)
Low Resolution Mass
Spectrometry
(LRMS)
Gas
Chroma tography
(GC)
Combined Gas
Chromatography/Mass
Spectrometry
(GC/MS)
Inorganic Analyses
Inductively Coupled
Plasma Optical Emission
Spectrophotometry (ICPOES)
Atomic Absorption
Spectrophotometry
(AAS)
Spark Source
Mass Spectrophotography
(SSMS)
Instrument Manufacturer
and Model
Mettler, microbalance
Perkin Elmer, 521
Hitachi-Perkin Elmer,
RMU-6 Mass
Spec tome ter
Varian, 1860 dual
FID
Varian, 1860 GC and
Hitachi-Perkin Elmer,
RMU-6 MS
or
Finnigan, 9500 GC and
Finnigan, 3100D Quadrapole
Mass Spectrometer
Applied Research
Laboratories, QA-137
Jarrell Ash, 810
AEI Scientific
Apparatus Ltd., MS 702R
Detectability for a Compound or Element
Being Searched For
1 jig
- 3-5% of the sample being
examined
-10 jig
(IX of a 1 mg sample)
per jjl of sample
~1 jig per jil of sample
-100 ng per jil of sample
-0.5-2000 ppb
-1-0.001 ppm
-50-100 ppb
-------
The second column system used in the GC analysis was for what are
called high boilers. These materials are defined as those which have the
higher boiling points which begin around 200°C and increase from there.
Napthalene is one of the first eluting compounds this system and the larger,
higher boiling materials, such as POM, hydrocarbon oils etc., elute after
napthalene. The operating parameters used for the GC separation of the
high boilers follows. These two column systems combined were able to elute
all the constituents found in both the wastes tested at TMC as well as
possible incomplete combustion products or species formed in combustion.
Column OV-225
Temperature 1800C isothermal
Carrier Gas helium at 30 cc/min
Detector FID
4.4.2.2 Inorganic Analyses
Inorganic analyses were performed using atomic absorption spectros-
copy (AAS), optical emission spectroscopy (OES), and inductively coupled
plasma optical emission spectroscopy (ICPOES). Selected samples of the
acid extractions of the particulate filters and the acidified splits of
the scrubber waters and the impingers were surveyed for trace metals by
ICPOES.
The ICPOES analysis determines 32 elements, including most of the toxic
elements of interest to the program, down to ppb levels with an accuracy
of 100 to 200 percent. The purpose of this survey is primarily to check
that the metals in these test samples are in approximately the same amounts
relative to each other as they were in the waste material. Those elements
which from the results either of the ICPOES survey or of the analysis of
the waste material seemed to be present at potentially toxic levels, were
determined quantitatively by AAS. The sensitivity of this method varies
from approximately 1.0 to 0.001 ppm for the elements which were determined,
with an accuracy between 10 to 50 percent.
In addition to the AAS and ICPOES analyses performed on the test
samples, spark source mass spectrophotography (SSMS) was used to analyze
the waste material for trace elements.
Some wet chemical methods were also performed on the combustion zone
impinger solutions. Total available chlorine was determined by a ferrous
ammonium sulfate titration, chloride by a modified Volhard tritration, and
carbonate by evolution and gravimetry.
4.5 PROBLEMS ENCOUNTERED
Problems which occurred during either the field testing or laboratory
analysis phases of the Marquardt test program are described in the following
sections. Corrective actions taken and final disposition, when applicable,
are also discussed.
54
-------
4.5.1 Field Test Problems
In spite of careful selection of equipment and checkout at actual or
simulated test conditions, a few incidents occurred during field testing
that had not been anticipated. Corrective actions were immediately taken
in each case, and testing was completed as scheduled. All necessary
samples were acquired.
4.5.1.1 Inadequate Cooling of Combustion Gas Sample
Combustion gases at 1090°C or higher had to be cooled to 115°C or
less at the entrance to the sampling train to avoid damage to the filter
seals. A thermocouple in the gas stream at the entrance to the filter was
used to monitor this temperature. Cooling of the sample gas was accomplished
both by radiation from the quartz probe liner to the water-cooled probe
(Figure 4-14) and by convective cooling with a stream of compressed air
directed around the probe liner. Checkouts of this cooling method both in
a boiler at TRW and during initial tests at TMC (TMC Run #1) were successful
in maintaining a sample gas temperature of 220-230°F at the filter inlet.
However, during the first sampling test (TMC Run #3), the sample gas
temperature suddenly rose to 1770C at the filter inlet even after over an
hour of steady-state operation had been completed with the temperature at
this point well controlled at 105 to 11QOC.
This problem was corrected by adding water as well as compressed air
to cool the combustion gas. Facility air and water flow, controlled by
metering valves, were mixed in a tee fitting and fed into a tube exiting
along the quartz probe liner (Figure 4-14). Both air and water flows could
then be precisely metered to control the sample gas temperature. Using
this cooling method, sample gas temperature was maintained at safe levels
of 80 to 105°C at the filter inlet for the remainder of the tests.
4.5.1.2 RAC^ Train Modifications
The standard EPA Method 5 sampling trains ordered from Research
Appliance Corporation required certain modifications before they could be
used on this program. These modifications were completed before the start
of actual sampling tests, so no schedule delays were involved.
First, the heated oven box enclosing the particulate filter was poorly
insulated and sealed, making control of the internal temperature erratic.
Next, the oven box fan took in cold outside air instead of recirculating the
warm air within the box. This resulted in large thermal fluctuations and
gradients.
Modifications to the sample train included re-insulating the heater
box by removing the loose filter fiber glass matting and installing fire
resistant 1.3-cm thick acoustic tiles to the interior of the box. Although
the oven box blower motor had to remain outside of the heater box (because it
could not withstand the internal temperature of nominally 120°C), the
squirrel-cage blower itself was re-installed inside the box to recirculate
only the heated air. These modifications greatly improved the control of
oven box temperature by reducing thermal gradients and eliminating most of
the heat losses.
55
-------
4.5.1.3 HC1 Corrosion of Hydrocarbon Analyzer
Examination of the analyzer components prior to testing made it
apparent that the HC analyzer could not withstand continuous, long-duration
exposures to moist, hot, HC1 gas streams without significant damage to its
brass and copper plumbing. The choice was then made to substitute a Perkin-
Elmer 881 gas chromatograph with an open tubular column and FID for the
Beckman instrument for monitoring hydrocarbon levels during the C-5,6 waste
burns. The chromatograph's primarily stainless steel plumbing was less
susceptible to HC1 attack and could be operated intermittenly to further
minimize its exposure to the corrosive sample stream.
4.5.1.4 Caustic Scrubbing of Acid Gases
The acid gas stream produced in the C-5,6 runs would also corrode the
pump and gas meter in the sampling train control box. To protect these
components and also to measure quantitatively the volume of HC1 produced,
a large, 2-liter impinger (shown in Figure 4-13) was added to the train.
This impinger was filled with 100 to 1200 ml of 13 to 22 percent sodium
hydroxide solution. A small amount of phenolphthalein was added to this
impinger before each test run in order to visually be able to observe
neutralization of the caustic solution.
Early neutralization did occur on the first test burn (TRW Run #V),
causing sampling to be stopped after 2-1/4 hours. The premature exhaustion
of the caustic solution was attributed to underestimating the effect of (XL
being scrubbed out as sodium carbonate. On subsequent test runs, allowance
was made for this factor and no further problems with the caustic impingers
were encountered.
4.5.2 Laboratory Analysis Problems
4.5.2.1 Trace Metal Losses from Chlorinated Organic wastes
It is not totally unexpected that it is proving difficult to obtain
accurate and precise determinations, with a minimum detectability of
1 ppm, of the trace metals in the chlorinated, organic matrix of our waste
materials. Of the techniques which can be performed on the neat sample,
the detection limits of X-ray fluorescence range from 5 to 500 ppm; and
chlorine is a major interferent with neutron activation analysis. Other
techniques with good accuracy and sensitivity, such as SSHS, AAS, and
ICPOES, require an ashed sample. During ashing, volatile metals and metal
halides are lost which negates the quantitative capability of the analysis.
A set of comparative experiments is planned in order to evaluate
special ashing and oxidizing procedures for trace metal retention. A
commercially prepared standard containing 20 elements in fuel oil will
be mixed with chlorinated organics to obtain an appropriate matrix. Tech-
niques for analyzing the ash will in turn be compared using the optimum
ashing and preparation methods selected.
56
-------
4.5.2.2 Extraction of the Solid Sorbent Traps
Extraction of the collected samples from the solid sorbent traps was
complicated by problems with the designed extractors. The extractor, shown
in Figure 4-17 was fabricated so that the sorbent trap would fit between
the upper and lower portions and distilled solvent would be continuously
passed through the resin bed. The only difficulty was in obtaining a leak-
free seal around the ball and socket joints. The best solution has turned
out to be Teflon sleeves. Some were ordered, but were not received until
late in the analysis schedule. With the Teflon sleeves, the designed
extractors work well, and will be used for future test samples.
In the meantime, it was decided to use a standard Soxhlet apparatus.
This technique involved pre-extracting paper thimbles, transfering the
XAD-2 resin to the thimble, and then extracting the resin in the Soxhlet.
The principle behind the two methods is the same. However, the Soxhlet
method presents the question of background organic contributions from the
thimbles.
4.5.2.3 Use of Multiple Solvents for Extractions
After completion of the initial pentane extraction of the first set
of sorbent traps, the resin was observed to have remained considerably dis-
colored. This discoloration was assumed to represent organic compounds
still absorbed on the resin. Subsequent extraction of the traps with a
polar solvent, methanol, removed five to six times more material by weight.
These results showed that a single solvent extraction, even over a 24-hour
period, would not remove those compounds which are only slightly soluble in
the solvent. Proceeding with this new information, a third extraction was
performed using benzene since this would be the best solvent for aromatics,
chlorinated aromatics, and ROMs. However these extracts contained insigni-
ficant amounts of material, showed no trace of any of the above compounds,
and had high backgrounds due to the resin not having been precleaned with
benzene. Consequently, the benzene extractions were not continued, and
only pentane and methanol extracts were made of the remaining sorbent traps,
57
-------
H20
-CONDENSER
24/40 JOINTS
--FLEXIBLE TEFLON
-I COUPLING
FRIT "
24/40 JOINTS
250 ml FLASK
Figure 4-17. Sorbent Trap Extractor
58
-------
5. TEST RESULTS
Test burns at The Marquardt Company (TMC) consisted of a background
test with No. 2 fuel oil (MA-I), three tests with ethylene waste (MA-II,
III, IV) and three tests with C-5,6/fuel oil blends (MA-V. VI. VII). This
section presents results of analyses performed on these test samples. The
section is divided into results for: 1) on-line combustion gas analysis,
2) waste destruction analysis, and 3) final emissions analysis which in-
clude stack gases, scrubber waters and burner head residues. Methods and
procedures for the preparation and analysis of test samples can be found
in Section 4.4.
5.1 COMBUSTION GAS DATA SUMMARY
Gas concentrations (by volume) as measured primarily by the on-line
instruments are presented in Table 5-1. Other methods were used to quantify
those species which were not measured by continuous monitors. Percent water*
was calculated according to EPA Method 4 using the volume of water col-
lected in the combustion zone sampling train. To determine the mole
percent of HC1 and Cl2 gases, chlorides and total available chlorine were
titrated in the combustion zone sampling train caustic impinger solutions.
In addition, SO gases were determined by Gastec® indicating tubes.
X
The data generally follow the operating parameters of the respective
tests. The ethylene tests consisted of varying the waste and air feeds to
achieve different combustion temperatures and therefore the C02i H20, and
NOx values rise along with increasing feed/temperature conditions. Due to
the high sulfur content of the ethylene waste (1.32%) the resulting SOx
emissions could only be.«roughly estimated. SOX levels were difficult to
measure with the Gastec® tubes because the tubes were calibrated for 1 to
100 ppm.
The C-5,6 tests consisted of holding nominally constant feed rates
while increasing the waste/auxiliary fuel ratio. This produced corres-
ponding increases in the levels of HC1 and Cl2 while C02. NOX, and SOX
levels remained essentially the same. The fairly constant level of CO
throughout all seven tests indicates that good combustion was achieved at
all tested conditions. The HC values are not believed to indicate signif-
icant differences betwen tests since: 1) readings were at the low range
of the Beckman analyzer where its accuracy is somewhat poor, and 2) mea-
surements were made non-redundantly with two different analyzers (Tests I to
IV with the Beckman 402, and Tests V to VII with the gas chromatograph)
so that the two sets of values cannot be directly compared.
5.2 WASTE DESTRUCTION ANALYSIS SUMMARY
Sampling of the combustion zone from the SUE®burner reaction tail-
pipe was accomplished just upstream of the venturi scrubber with the
sampling train described in Section 4.3.2. The samples from the train
consisted of a probe wash, filter, sorbent trap, grab gas sample and
impinger liquids. These samples were then separated into their organic
and inorganic constituents and analyzed by appropriate techniques, as
described in Section 4.4. Analysis of the combustion products is aimed
59
-------
Table 5-1. Total Gas Composition In the Combustion Zone by Volume
Run No. and
Fuel Description
Fuel 011 Background
I
Ethyl ene Waste
II
III
IV
C-5,6 Fuel Oil
V
VI
VII
o2 (x)
6.3
7.2
9.4
4.9
8.1
5.6
5.5
co2 (%)
9.8
10.3
9.3
11.5
8.4
9.7
9.5
N2 (5)
73.0
74.0
73.4
74.2
69.6
68.8
67.9
H20 (5)
10.9
8.5
7.9
9.4
12.7
14.2
14.7
HC1 (%)
-
.
-
-
1.15
1.64
2.35
C12 (ppm)
-
-
-
-
<0.05
15
56
CO (ppm)
12
17
17
22
20
17
17
NOX (ppm)
170
200
150
520
145
140
120
HC (ppm)
10
5-10
5-15
5
10-25
35-65
30-35
S02 (ppm)
15-20
100-200
100-200
100-200
5
6
5
-------
mainly at identifying and quantifying any unburned waste or hazardous
partial combustion products. The production of potentially toxic levels
of trace metals from burning these wastes is also examined. Where
quantified species are calculated in mg/nP in the sample gas stream, the
gas volume data used to make these calculations are summarized in Appendix
B. The level of interest for this program is defined as 0.1 mg/m3 of sam-
ple gas, which represents the threshold level of the bulk of the most
toxic species as defined by OSHA and other health and safety organizations.
5.2.1 Organic Composition
The organic analyses were divided into: 1) quantitative searches by
gas chromatography for uncombusted constituents from the waste material or
other specific compounds that could be expected to be present, and 2)
qualitative surveys to identify unexpected compounds.
5.2.1.1 Results of Quantitative Analyses
The results of the gas chromatographic analyses on the extracts of the
combustion zone filter and sorbent trap samples are presented in Table 5-2.
The table shows that benzene, known to be present in the ethylene waste,
and mesityl oxide (identified subsequently by GC/MS) were found in the
combustion zone filter extracts. Mesityl oxide (4-methyl-3-pentene-2-one),
an industrial solvent and rust remover, was found in MA-IV. It is likely an
artifact of contamination during sampling or sample preparation.
Results of the analysis of the sorbent trap extracts by gas chroma-
tography indicated that a total of five compounds are present in the seven
sorbent trap extracts. Chromatograph retention times for these compounds
did not coincide with retention times for the waste constituents thus no
identification could be made on that basis. These compounds were arbi-
trarily identified as Compounds A through E based on chromatographic
retention time. Their concentration levels, based on equal detector
response, are presented in Table 5-2. Each sample had at least one of
these compounds and six samples had two or more. Combined GC/MS was used
to help resolve the composition of the extracts and identify the five un-
known compounds. Four samples were selected for GC/MS in such a way that
all five unknowns were contained in at least one of the selected samples.
Results of the GC/MS analysis indicated the presence of compounds pre-
dominantly silicones and oxygenated organics. None of the representative
waste constituents were found in these sorbent trap extracts. No further
analyses were performed on these samples since: 1) they are not part of the
original composition waste, 2) they represent low ppm levels in the sample
gases, and 3} there is no evidence of toxic characteristics in what is
known of the compounds.
5.2.1.2 Qualitative Survey Analysis
Samples were surveyed by gravimetric, infrared spectrometry (IR), and
low resolution mass spectrometry (LRMS) techniques. Since the qualitative
results correlate more to sample type than to waste burn, the data are dis-
cussed in the following order:
61
-------
Table 5-2. Results of Gas Chromatographlc Analysis
Run Number and
Fuel Description
Fuel Oil Background
MA-I
Ethyl ene Waste
MA-I I
MA-III
MA- IV
C-5,6/Fuel Oil
MA-V
MA- VI
MA-VI1
Sample Gas.
Volume (mj)
5.28
5.20
4.80
5.07
4.15
5.53
4.54
Compound
Benzene
{Note 1)
(Note 1)
Mesityl
Oxide
(Note 1)
Benzene
Benzene
Amount
Found (mg)
1
None Detected
Above Back-
ground
ii n
12
None Detected
Above Back-
ground
3
3
Concentration of
Detected Compounds
(mg/m3)
0.2
-------
a. Combined particulate filter and probe wash extracts
b. Sorbent trap extracts
c. Grab gas samples
a. Combined Probe Wash and Parti oil ate Filter Extracts
The combustion zone filters were extracted with methylene chloride
using the procedures described in Section 4. Table 5-3 presents the
amount of residues found In the combustion zone filter extracts and the
resulting concentrations in the sampled gas. The values have been cor-
rected for the amount of material extracted from the control, an unused
filter. The amount of material found after extraction of the control
filter was only somewhat less than that found on each of the sample
filters and its qualitative nature was also similar. This is Indicative
of low level contamination which often plagues trace organic analysis,
even though these filters were fired in a muffle furnace carefully pre-
conditioned, handled in clean glassware, and extracted with chromatographic
grade solvent.
The extractable organic material from the combustion zone filters
analyzed by IR and LRMS consisted of alkyl hydrocarbons and alkyl esters
of phthalic acid which are not considered toxic. No evidence of any of
the starting toxic materials was seen. No additional effort was made to
fully identify the esters and/or hydrocarbons present. A detailed dis-
cussion of observations made during preparation of the filters is found in
Appendix D.
Table 5-3. Organic Material Extracted From Combustion Zone Filters
Run No.
Fuel Oil
Background
I
Ethyl ene Waste
II
III
IV
C-5,6 Fuel Oil
V
VI
VII
Amount of Material
Extracted (mg)
0.80
0.55
0.65
4.40
1.50
1.55
1.35
Sample Gas Volume,
Met Basis (m3)
at 1 atm and 21°C
5.28
5.20
4.80
5.07
4.15
5.53
4.54
Concentration
Level in Gas
(mg/m3)
0.15
0.11
0.14
0.87
0.36
0.28
0.30
63
-------
b. Sorbent Traps
This section presents the results of the survey analyses performed on
the extracts of the sorbent traps from the combustion zone sampling train.
The amounts of organic materials found by extraction and gravimetry in the
samples are presented in Table 5-4.
Examination of the IR and LRMS data leads to the conclusion that cer-
tain classes of compounds commonly used as processing aids and found in
many industrial products and processes are present. There is evidence to
believe these compounds are indeed present in the combustion gases, but
were also found in the control and blank samples which were never exposed
to combustion gases. Details of control and blank analyses are provided
in Appendix D.
Table 5-4 shows that significant amounts of extractable materials were
found and at levels considerably higher in the sample traps than the un-
used sorbent trap which was used as a control. The IR and LRMS data indi-
cated hydrocarbon oils or greases, fatty acid compounds and phthalic acid
esters. The amounts of these materials found in the sample traps are so
much greater than the amounts found in the several control experiments
that either or both of the following conclusions must be reached: 1) these
species were actually present in the combustion gases, or 2) they were
originally in the sorbent trap resins despite prior clean up, and the
exposure to conditions inherent during sampling somehow aids in their
release from the resin matrix and subsequent solvent extraction.
Table 5-4. Organic Material Extracted From Sorbent Traps, Survey Analysis
Sampl e
MA-I-CG-ST
MA-II-CG-ST
MA-III-CG-ST
MA-IV-CG-ST
MA-V-CG-ST
MA-VI-CG-ST
MA-VII-CG-ST
Control (Unused
Sorbent Trap)
Amount of Material
Extracted
(Corrected for
Control) (mq)
60.67
49.49
70.32
100.86
92.60
133.58
119.76
7.36
b ample Gas
Volume, Wet
Basis
(m3 @ 1 atm
and 21 °C)
5.28
5.20
4.80
5.07
4.15
5.53
4.54
4.00
Assigned
Concentration
Level in Sample
Gas (mg/m3)
11.5
9.5
14.6
19.9
22.3
24.2
26.4
1.8
64
-------
c. Grab Gas Samples
Grab gas samples were taken from the combustion zone sampling train
downstream of the sorbent trap during train operation. Analysis of these
grab gas samples for possible organic species passing through the solid
sorbent traps was performed by LRMS. Inspection of the mass spectra indi-
cates that only propane and butane were present in the samples at rather
constant levels which never exceeded 10 pom. No other organic materials
were present in the sample. The calculated lower limit of detection for
other compounds of similar organic composition under the same analytical
conditions is 2 ppm. Table 5-5 summarizes the results of the analysis.
The levels are reported as ppm propane since a propane standard was used
to determine mass spectrometer response factors.
Table 5-5. Analysis of Organics in Grab Gas Samples
Sample
MA-I-CG-GG
MA-II-CG-GG
MA-III-CG-GG
MA-IV-CG-GG
MA-V-CG-GG
MA-V-HA-GG
MA-VI-CG-GG
MA-VI-HA-GG
MA-VII-CG-GG
MA-VII-HA-GG
Organic Content
(ppm as Propane)
9
9
9
9
9
8
9
9
9
8
Those samples with the designation "CG"
in their labels were taken from the
sampling valve downstream of the sorbent
trap in the combustion zone sampling
train. Those with a "HA" were taken
downstream of the gas chromatograph
which was being used as a total hydro-
carbon analyzer during the last three
runs with the chlorinated waste.
65
-------
5.2.2 Inorganic Characterization
Inorganic elemental concentrations were determined by analysis of the
particulate filters and aqueous impinger samples. The filters were first
weighed to determine total participate loading. The loadings obtained are
reported in Appendix D, Section D.7, along with other weights taken at
subsequent steps of filter preparation for analysis.
After total participate weights were measured, the combustion zone
filters were solvent extracted for organic species and the filters were
then photographed. Figure 5-1 shows a photograph of the particulate fil-
ters obtained from sampling the combustion gases from Tests I through VII,
in order going from left to right and top to bottom. The sooty nature of
the filters from the C-5,6 tests agrees with observations made during these
three tests that relatively larger amounts of burner residue (clinker) were
being formed. The clinker itself is described in Section 5.3.3. The
irregular dark and light markings on the filters shown in the photograph
are the result of prior folding of the filters for solvent extraction.
Trace metals on the particulate filters were put into solution by
acid digestion of the filters. The liquid collected in the impingers was
combined into one sample and an aliquot of that was acidified for trace
metal analyses. The filter digests and acidified impinger solutions from
Tests I (No. 2 background), IV (ethylene waste), and VI (C-5,6 waste) were
surveyed semi-quantitatively by inductively coupled argon plasma optical
emission spectrometry (ICPOES).
Quantitative analysis for selected elements were then performed by
atomic absorption spectrometry (AAS) on all samples. The elements to be
analyzed were selected by the following criteria:
1. Potentially toxic (e.g., Pb, Cd, Sb, and Hg)
2. Present at significant levels as determined by
the: survey analysis of the wastes (Section 4.1),
and/or survey analysis of the filter acid digests
and impinger solutions.
Using these criteria, four elements were selected for AAS analysis,
Cr, Pb, Co, and Sb. Although not among the most toxic of metals, Cu, Zn
and Mn were also analyzed for an additional comparison of the ICPOES and
AAS results. The ICPOES analysis indicated only very low levels of any
toxic elements (see Appendix D, Section D.8), which is confirmed by the
AAS results shown in Table 5-6, with the exception of the element Pb.
Results from analysis of the stack samples are also included in this table
for purposes of comparison.
In those cases where elements were either not detected or were mea-
sured at levels less than or equal to background levels, a less than
value is given. As expected, metal concentrations are fairly constant for
Runs II, III, and IV in which the ethyl ene waste was fed at approximately
the same rate for each test. On the other hand, for Runs V, VI, and VII,
in which the C-5, 6 was was fed at increasingly higher concentrations, a
66
-------
-
'
Figure 5-1. Filters From Combustion Zone Gas Sampling After Solvent Extraction
-------
Table 5-6. Toxic Trace Metals at Level of Interest {mg/m ) by MS
Run I
Stack
Combustion
zone
Run II
Stack
Combustion
zone
Run III
Stack
Combustion
zone
Run IV
Stack
Combustion
zone
Run V
Stack
Combustion
zone
Run VI
Stack
Combustion
| zone
Run VII
Stack
Combustion
zone
Copper
Impingers Filter
NGTB
<0.003
NGTB
0.022 <0.001
NGTB
<0.001
0.022 0.002
NGTB
<0.001
O.Ull 0.002
NGTB
<0.005
0.021 0.002
0.014
0.071 u.015
0.008
O.OB6 0.008
0.022
0.059 0.013
Chromium
mplngers Filter
ND
<0.001
ND ND
<0.001 <0.001
NO
<0.001
ND
<0.001 0.003
ND
<0.001
ND NGTB
<0.001 <0.001
ND
<0.001
ND
<0.001 0.001
0.008
0.002 0.030
0.014
0.001 0.045
0.037
0.001 0.14
Lead
mplngers Filter
0.001
0.002 0.002
0.001
0.003 0.003
0.001
ND
<0.001 0.002
0.003
0.002 0.002
0.01 0.003
0.55 0.010
0.01 0.005
0.95 0.006
0.05 0.006
1.40 0.008
Zinc
mplngers Filter
NGTB
<0.2
NGTB
0.005 <0.008
NGTB
<0.05
NGTB
0.009 <0.06
NGTB
<0.1
NGTB
0.010 <0.007
NGTB
<0.07
NGTB
o.on
-------
corresponding rise in the level of metals emitted can be seen. Of the
metals analyzed by AAS, only chromium is trapped predominently on the
filter, which substantiates the need for careful metal analysis of the
impinger solutions.
In addition to the trace metal analyses, the impinger solutions from
the C-5,6 tests were also analyzed by wet chemical methods for the HC1,
Cl2, and C02 scrubbed from the combustion gases by the caustic in the
impingers. The volumes of HC1 and C02 were needed in order to correct the
gas sample volumes as measured by the dry gas meter. The levels of free
chlorine, a potentially toxic specie, were determined to be <0.005, 15,
and 56 ppm for Tests V, VI, and VII, respectively. Complete results of
these wet chemical analyses are described in Appendix D, Section D-8.
5.3 Final Emissions
The stack gas, scrubber water and burner head residue (clinker) that
make up the final emissions from the SUW^burner incineration system were
sampled during each test and analyzed to determine the environmental safety
of the tests at TMC.
5.3.1 Stack Gases
Stack effluents were sampled during the tests with both a standard EPA
Method 5 train and Gastec® detection tubes. The Method 5 train was used
to obtain samples for the determination of parti oilate> loading and ele-
mental composition of the particulate, and the Gastec© tubes were used to
measure potentially hazardous gaseous species such as hydrocarbons, HC1,
C12> and COClp.
5.3.1.1 Samples Obtained From the Method 5 Train
The particulate filters obtained from sampling the stack gas are shown
in Figure 5-2. The filters from Tests I through VII are presented in order
going from left to right and top to bottom.
Particulate mass loading was determined by adding the weight of partic-
ulate found on the filter to the weight of the residue obtained by evap-
orating the probe washes to dryness. This gross weight is then corrected
for the appropriate blank or control values and divided by the total gas
sample volume. The resulting values, in both metric and English units,
for the particulate emitted from the stack are presented in Table 5-7.
After weighing, the filters were acid digested and selected samples
were surveyed for trace metals by inductively coupled argon plasma optical
emission spectroscopy (ICPOES). The ICPOES results (presented in Appendix
D, Section D.8) indicated very low levels of all toxic elements as did the
quantitative analyses by atomic absorption spectrometry (AAS) presented in
Table 5-6. Because of the relatively high level of lead and its pre-
dominante collection in the impingers from the combustion gas, lead was
also determined in the stack impingers and found to be quite low.
69
-------
•vj
o
Figure 5-2. Filters From Stack Gas Sampling
-------
Table 5-7. Participate Mass Loadings
Test No.
I
II
III
IV
V
VI
VII
parti GUI ate Loading
Mg/m3
21.2
20.9
25.0
20.4
36.2
43.8
113.5
Grains/Std. Ft3
0.0092
0.0091
0.0109
0.0089
0.0158
0.0192
0.0494
In addition to the seven elements discussed in Table 5-6, the element
sodium was also determined by MS in the digests of the stack filters from
the C-5,6 runs since it was suspected that the relatively high weights on
these samples were due to NaCl produced by the caustic scrubber. Indeed,
if all the Na measured (after correction for the background of the control
filters) is assumed to be present as NaCl, the weight of NaCl accounts for
85 to 90 percent of the stack particulate, as shown in Table 5-8. This
calculation is probably a good average since the values for the weight of
sodium salts would be higher for any sodium carbonate present and lower
for any sodium hydroxide present.
(ft
5.3.1.2 GastecwTube Measurements
Gastec^detection tubes were used to analyze for low concentration
gases that could be present. Sulfur oxides were measured in this way
and these results reported in Table 5-1. Other gases which were checked
using Gastec
-------
Table 5-8. Percent Sodium Salts as NaCl on Stack Filters
Run
No.
V
VI
VII
Total Wt. of
Material
on Filter (mg)
38.1
93.9
139.4
Wt. of Na
Recovery by
Acid Digestion
(mg)
13.0
33.5
46.50
Wt. Sodium
Salts
as NaCl (mg)
33.04
85.14
118.18
Percent NaCl
of Parti cul ate
on Filter
(mg)
87
91
85
Table 5-9. Gastec^Tube Results
Sampling Location
Stack
Stack
Gas Chromatograph
Outl et
Run No.
All
V. VI
and VII
V, VI
and VII
Detector Tube
Gasoline (aliphatic)
HC1
Cl2
COC12
CoCl2
Limits of
Detecti on
<0.005
percent
<0.2 ppm
<0.33 ppm
<0.1 ppm
<0.1 ppm
The only practical opportunity for testing the combustion zone gases
with Gas tec® tubes to check for any production of phosgene was at the
outlet of the GC sample bypass line. The GC, as previously mentioned,
was substituted for the Beckman HC monitor during the C-5,6 runs.
72
-------
5.3.2.1 Organic Characterization
Quantitative
The various scrubber water samples were extracted with Freon to remove
and analyze the organic materials in them. The concentrated extracts were
then analyzed by gas chromatography. The results are summarized in Table
5-10.
Table 5-10.
Results of Gas Chromatographic Analysis
of Scrubber Water Samples
Run Number
Fuel Oil
Background
I
Ethyl ene
Waste
II
III
IV
C-5,6/Fuel
on
V
VI
VII
City Water
Control
Fresh
Caustic
Water
Control
Scrubber Water
Volume of
Scrubber Water
Extracted
(Liters)
1.00
1.00
0.50
1.00
0.50
1.00
1.00
1.00
1.00
Low Boiling
Material in
Scrubber Water
Extract
(mg)a
26
73
30
28
60
20
26
53
12
Higher Boiling
Material in
Scrubber Water
Extract
(mg)a
0.7
0.8
0.4
0.5
<0.1
<0.1
<0.1
<0.1
<0.1
Concentration
of Extracted
Material in
Scrubber Water
(mg/1)
27
74
61
28
120
20
26
53
12
a See Section 4.4.2 for definition of low boiling and higher boiling
materials. Instrument response (calibration) for low and higher
boilers was determined with standard injections of benzene and
naphthalene respectively.
73
-------
No evidence of the ethylene waste or C-5,6 waste constituents were
found in the chromatographic data. The levels reported in Table 5-10
represent three artifacts that are present in the control samples and the
fuel oil background samples as well as the test samples. It can be rea-
sonably concluded that these materials are not generated as a result of
burning waste.
Qualitative Survey
An aliquot of the concentrated organic extract from the scrubber water
samples was also subjected to the survey analysis for unexpected combustion
by-products etc. The results of the gravimetric determination of the
residue is presented in Table 5-11. A trend of changing organic content in
the scrubber water than the No. 2 oil baseline burn or the ethylene waste
burns. The premature shutdown of Test VII test is not believed to have
caused the higher value for the scrubber water extract sample. Instead
the low values associated with Runs V and VI, are considered to be in-
consistently low.
Table 5-11.
Results of Analysis of Concentrated Organic
Extracts of Scrubber Water Samples
Sampl e
City Water9
Fuel Oil
Background
MA-I-SW
Ethylene Waste
MA-II-SW
MA-III-SW
MA-IV-SW
Control
MA-V-FSWb
C-5.6/Fuel Oil
MA-V-SW
MA-VI-SW
MA-VII-SW
Volume of
Scrubber
Water
Extracted
(liters)
1.00
1.00
1.00
0.50
1.00
1.00
0.50
1.00
1.00
Amount of Extracted
Material by
Gravimetry
(mi Hi grams)
0.53
84.42
70.0
62.0
77.1
16.3
0.9
0.7
33.2
Concentration
in Scrubber Water,
mg/ liter
0.5
84.4
70.0
124.0
77.1
16.3
1.8
0.7
33.2
a City water was used as the wet scrubber feed for Tests MA-I through MA-IV.
This sample is therefore the control sample for the first four tests.
b A control sample of fresh caustic scrubber solution was prepared by blend-
ing concentrated caustic with city water according to the blending propor-
tions used for Test MA-V. Therefore, this sample is the control for tests
MA-V through MA-VII.
-------
Survey analyses by infrared spectrophotometry (IR) and low resolution
mass spectrometry (LRMS) were performed on the residues obtained from the
evaporation of the Freon solvent in which the residues were dissolved. The
IR data for all the scrubber water extract samples were all extremely simi-
lar, and while no evidence of the waste constituents were seen, the spectro
indicated hydrocarbon oils, greases, fatty acid esters. The LRMS data
supported the IR findings but also detected minor amounts of phthalate
esters.
5.3.2.2 Inorganic Characterization
The scrubber water samples, aliquots of which had been acidified in the
field, were analyzed by AAS for several elements of interest. Identified
for analysis as a result of being detected in the representative waste
samples were Cu, Cr, Pb, Zn, Mn. Cobalt (Co) an antimony (Sb) were found
in the combustion test samples and were added to the test matrix. The re-
sults of those analyses, corrected for appropriate background are presented
in Table 5-12.
The Los Angeles County Sanitation District Phase I concentration limits
for point source discharge to a sewer were consulted. These heavy metal
concentrations are all at least a factor of ten below the most stringent
maximum allowable concentration for these listed metals which is 10 mg/liter
(ppm). It is concluded that these scrubber waster were sewered in accor-
dance with local regulations.
Table 5-12. Elemental Analysis of Scrubber Water by AAS (ppm)
Run No.
I
II
III
IV
V
VI
VII
Cu
0.10
0.11
0.07
0.13
0.15
0.14
0.23
Cr
0.21
0.29
0.01
0.38
0.18
0.18
0.35
Pb
0.08
0.02
0.02
0.02
0.21
0.31
0.54
-
Zn
0.37
0.82
0.12
0.74
0.23
0.21
0.17
Mn
0.48
0.42
0.06
0.59
0.19
0.12
0.23
Co
—
—
—
—
0.051
0.074
0.12
Sb
—
— — -
— — —
— — -
0.18
0.22
0.24
75
-------
5.3.3 Solid Residues
The solid residue (clinker) samples were obtained when the burner head
was opened after each test burn. The clinker was then scraped off the bur-
ner head as completely as possible and wrapped in aluminum foil. The
quantities obtained were as follows:
Fuel Oil
Run I
Ethylene Waste
Run II
Run III
Run IV
C-5.6/Fue1 Oil
Run V
Run VI
Run VII
None (probably <50g)
295g
245g
533g
662g
2327g
1480g
It must be noted, however, that these weights in no way represent the
total weight of clinker produced in each burn. There was evidence from both
feed pressure readings and acoustic observations that the clinker tended to
build up, break off, and travel down the tail pipe through the scrubber.
In fact, on the last run (VII), the tip of the mullite probe at the end of
the tailpipe for the on-line instruments was believed to be broken by
clinker hitting it.
The clinkers were analyzed in duplicate by a micro-analytical com-
bustion technique for their carbon and hydrogen content. The black, powdery,
coke-like appearance of the clinkers indicated that they would probably be
95 to 99 percent elemental carbon. The micro analysis results, shown in
Table 5-13, supported this conclusions.
Table 5-13. C and H Analysis of Clinkers
Element
Percent
Carbon
Percent
Hydrogen
Total
Run II
98.46
0.61
99.07
Run III
97.07
0.79
97.86
Run IV
99.30
0.52
99.82
Run V
88.90
1.19
90.09
Run VI
97.27
0.40
97.67
Run VII
98.02
0.40
98.42
76
-------
5.3.3.1 Organic Analysis
Portions of the clinker from each test were extracted for orgam'cs
with methylene chloride solvent in a Soxhlet extractor. The extract was
further prepared for analysis using the procedures in Section 4.4. The
concentrated extracts were analyzed by GC for waste constituents as
well as other unexpected by-products. The survey analysis, gravimetry,
IR, LRMS were also applied to the extracts.
Quantitative
The results of the GC analysis of the clinker extracts for the runs
with ethylene wastes are presented in Table 5-14. It shows that detectable
amounts of extractable material were present in the clinker from MA-II,
MA-III and MA-IV. Furthermore, the retention times on the gas chromatograms
for the sample extracts are the same as those of the ethylene waste repre-
sentative sample. It is concluded that uncombusted ethylene waste is pre-
sent in the clinker at the levels shown in Table 5-14. This is borne out
by the IR and LRMS survey analysis on the same extracts which also indicated
residuals of the original waste. Two compounds were found by GC to be
present in the clinker extracts from the C-5,6 tests (Table 5-14). The
retention times for the two peaks do not coincide with retention times for
any of the compounds in the representative sample of C-5,6 which was
chromatographed under the same conditions. Further work to identify these
two peaks present at these relatively low levels was not carried out since
LRMS and IR of, the clinker from these tests (discussed next) did not indi-
cate the presence of any chlorinated hydrocarbon species.
Survey
Aliquots of the concentrated clinker extracts were mildly evaporated
and the residues subjected to gravimetry, IR and LRMS. The amount of
residue obtained after mild evaporation of the clinker extracts are pre-
sented in Table 5-15. The ethylene test clinkers clearly have more
extractables than the C-5,6 clinker.
Inspection of the IR spectra of the clinker extracts from the tests
with the ethylene manufacturing wastes (Runs II, III and IV) revealed that
all three samples contain aliphatic and aromatic hydrocarbons. Comparison
of these spectra with that of the waste feed indicates that uncombusted or
incompletely combusted waste feed is present in the clinker. The clinker
has a strong odor characteristics of dicyclopentadiene, a constituent of
the waste.
The LRMS data for the clinker extracts from the ethylene waste tests
Runs II, III and IV also indicate the presence of alkyl substituted
naphthalenes and cyclohexadiene. There is no evidence of any polynuclear
aromatic hydrocarbons larger than naphthalene.
The IR and LRMS survey analyses of the clinker extracts from the C-5,6
tests (Runs V, VI and VII) showed that the three samples have essentially
the same composition. The IR spectra indicate that these extracts were all
aliphatic hydrocarbons, the source of which was likely the auxiliary fuel
that was mixed with the C-5,6 waste prior to burning. The clinker extract
77
-------
Table 5-14. Results of Analysis of Clinker Extracts by Gas Chromatography
Run Number
and Fuel
Description
Fuel Oil
Background
I
Ethyl ene
Waste
II
III
IV
C-5.6/
Fuel Oil
V
VI
VII
Weight
of Clinker
Sample
Extracted
(g)
Low Boiling
Material in
Clinker
Extract
(mg)a
Higher Boiling
Material in
Clinker
Extract
(mg)a
Concentration
of Extracted
Material in
Clinker (mg/g
Clinker)
NOT APPLICABLE NO CLINKER RECOVERED
10.360
9.102
15.654
13.166
15.160
14.110
25
16
27
11
8
14
0.5
0.8
0.8
<0.1
<0.1
<0.1
2.4
1.9
1.8
0.8
0.5
1.0
Comments
Extracted
materi al s
were iden-
tified as
ethyl ene
waste con-
stituents
Extracted
materi al s
not iden-
tTTied as
C-5,6 waste
constituents
a See Section 4.4.2 for definition of low boiling and higher boiling materials.
Instrument response (calibration) for low and high boilers was determined
with standard injections of benzene and maphthalene respectively.
Table 5-15. Residue Extracted from Clinker
mg extracted
g clinker
I
Ethyl ene
Run II
4.73
Run III
21.12
Run IV
3.71
C-5,6
Run V
0.52
Run VI
0.28
Run VII
0.15
78
-------
from MA-VII also contained eitheraldehydes, ketones, esters, or possibly
a mixture of them as shown by IR. Neither the IR nor the LRMS data show
any trace of C-5,6 similar chlorinated species, or POM. The level of
detection of the IR for these materials was experimentally determined to
be about 5 percent in the extractables.
Samples of this solid clinkers themselves were also directly examined
by LRMS (no extraction) before and after they were solvent extracted. The
following observations were made:
• The ethylene waste test clinkers contained the waste con-
stituents confirming the GC analysis. Examination of the
solvent extracted clinker by the same LRMS technique showed
only a trace organic content. The extraction of the clinker
was thus estimated to be more than 95 percent complete
based upon mass spectrometer instrument response.
• The unextracted clinker from the C-5,6 tests contained
benzene and/or toluene, aliphatic hydrocarbons, and
monatomic and diatomic chlorine combustion products.
Hexachlorocyclopentadiene (C-5,6) and other chlorinated
orgam'cs were specifically searched for, but no trace
of any chlorinated compounds were detected in the un-
extracted clinker. Levels of detection are estimated
at 0.1 and 1 percent of the clinker. In addition, the
typical 149 AMU peak for phthalate esters was found in
all of these C-5,6 clinker samples. This is significant
because these clinker samples were not extracted or
otherwise processed except for grinding, and the prob-
ability of contamination during grinding is very small
indeed. Thus, the presence of phthalates in the
clinker is strong evidence for the presence of phthalates
in C-5,6 waste/fuel oil feed used in the tests. This
can exaplain the presence of at least part, if not all,
of these phthalates in the samples from C-5,6 testing.
Examination of the extracted clinker showed only traces
of hydrocarbon pickets and methylene chloride, indi-
cating an extraction efficiency similar to that found
for the ethylene test clinkers.
5.3.3.2 Inorganic Analysis
Samples of the clinkers were also analyzed by optical emission
spectroscopy (OES) for elemental inorganic composition. The results of
this analysts are presented in Table 5-16.
Certain other elements whose presence was indicated in the represen-
tative samples by analysis (see Section 4) were, in turn not detected by
OES. Table 5-17 presents these elements of interest and their limits of
detection below which they would not be seen by OES when present in a
carbon matrix similar to the clinker.
79
-------
Table 5-16. Inorganic Composition of Clinkers (ppm)
Element
Al
B
Ca
Cr
Cu
Fe
Pb
Mg
Mn
Ni
Si
Na
Run II
30
10
10
300
5
1000
20
50
10
200
50
ND***
Run III
100
100
20
10
100
200
30
800
10
NO**
500
ND***
Run IV
100
50
50
100
50
3000
30
800
20
1000
500
ND***
Run V
30
10
20
10
5
70
50
30
ND*
ND**
10
2000
Run VI
30
30
10
30
10
30
10
30
10
ND**
20
ND***
Run VII
30
30
10
30
5
100
10
' 30
10
ND**
20
ND***
<5 ppm
**<10 ppm
**
<100 ppm
Table 5-17.
Element
As
Ba
Co*
Hg
Sb*
V*
Zn
OES Detection Limits
Estimated Detection
Limit (ppm)
100
500
5
1000
50
30
200
* Would only be likely to be present in C-5,6 test samples (Runs V, VI
and VII).
80
-------
6. WASTE INCINERATION COST
Individual economic analyses were performed to determine the costs of
incinerating, on an industrial basis, the annual source plant productions
of the two waste materials tested at The Marquardt Company (TMC). The
economic analyses were divided into capital investment and annual operating
costs. The economic analysis for each disposal facility, for the SUE(Jy
incinerator and scrubbing system portion, was based on equipment prices,
estimates of manpower requirements, and fuel and power consumptions obtained
from TMC. The costs of other portions of the disposal facilities and
associated labor were estimated using the Happel Method, data from Guthrie
("Capital Cost Estimating", Chemical Engineering, March 24, 1969) and
standard engineering reference methods. Equipment costs were adjusted to
January 1976 prices using the Marshall Steven Cost Indexes. Land prices
and transportation costs are not included in the two disposal plant cost
estimates. Both incineration plants are assumed to be at the manufacturing
plant generating the waste to be disposed.
6.1 CAPITAL INVESTMENT
The capital investment for the facility to incinerate 15 million
kilograms per year of ethylene manufacturing wastes shown in Table 6-1 is
based upon a design concept which employs in parallel one 1500 liters/hour
(400 GPH) and one 760 liters/hour (200 GPH) SUE® burner-incinerator-scrubber
system, fed by a common automatic feed system. The facility costs include,
as common use items, an ethylene manufacturing waste storage tank (16 hour
storage capacity), a scrubbing water supply system (24-hour storage capacity),
and a settling pond with a 24-hour retention capacity. It was assumed that
the only fuel needed for the incinerators was the ethylene manufacturing
wastes.
The size of the facility was based on a ninety percent plant operating
factor and 15 million kilograms per year of ethylene wastes with a density
of 910 g/liter. The 2270 liters/hr, ethylene wastes incineration plant has
a nominal capacity of 10 percent over the required capacity of 2070 liters/
hr.
The total capital investment for the ethylene manufacturing waste
incineration facility is estimated at $1,818,000. It should be noted that
over one-third of this capital investment is directly due to scrubbing
equipment and scrubbing related facilities which may not be required for
incinerating a fuel as "clean" as the ethylene manufacturing wastes tested.
The capital investment for the plant required to incinerate 4.5 million
kilograms per year of the hexachlorocyclopentadiene wastes mixture shown in
Table 6-2 is based on sizing the incinerator conservatively on the heat of
combustion resulting from feeding a 1:2 weight ratio of Cede waste and
No. £-/uel oil. This is equivalent in size to a 1500 liters/hr (400 GPH)
SUE U5 burner incinerator. The plant design concept includes the associated
scrubber system, an automated feed system, a CcClg waste storage tank (with
24-hour storage), a No. 2 fuel oil storage tank (z4-hour storage), and the
caustic soda solution storage tank (24-hour capacity). Auxiliary equipment
81
-------
Table 6-1. Capital Investment
15 Million kg./Yr. Ethylene Manufacturing Waste Incineration Plant
Equipment
fo\ ^
1-SUE w Basic incinerator, with scrubbing system
1-Ethylene Manufacturing Waste Storage Tank system
1 -Scrubbing Water Supply System - Tank
- Pumps
1 -Automated feed system, waste
1 -Scrubber Wastes Collection Settling Pond
Piping (at 45%)
Foundations (4%)
Buildings (4%)
Structures (4%)
Fire Protection (0.75%)
Electrical (4.5%)
Painting & Cleanup (0.75%)
Material & Labor
Overhead (30%)
Total Erected Cost (130%)
Engineering Fee (10% erected)
Contingency Fee (10% Erected)
Total Capital Investment
Size
2,270 LPH
33,300 liters
2,650,000 liters
8,950-17,900 watts
2,650,000 liters
$1,165,175
350,000
1,515,000
151,500
151,500
$1,818,000
Estimated
Eouioment
272,750
11,500
100,000
50,000
20,000
454,250
225,000
20,000
20,000
20,000
3,750
22,500
3,750
769,250
Costs
Labor
45,425
225,000
30,000
14,000
4,000
22,500
32,500
22,500
395,925
* Includes burner, incinerator, reaction tailpipe, Venturi scrubber, scrubber tank,
pump, air supply system, waste pump.
-------
00
CO
Table 6-2. Capital Investment
4.5 Million kg./Yr. Hexachlorocyclopentadiene Waste Incineration Plant
Eau lament
1-SUE ^ Basic incinerator, with scrubbing system*
1-CgClg Waste Storage Tank, Heated, SS
1-No. 2 Fuel Oil Storage Tank (& Pump)
1 -Scrubbing Water Supply System-Tank
-Pump
1-50% Caustic Soda Storage Tank & Pump
1- Automated. feed system, waste & oil
1-20% HC1** Storage tank & pump, rubberlined
1-pH Control System, scrubber pond
1-Scrubber Wastes Collection & Settling Pond
Piping (45%)
Foundations (4%)
Buildings (4%)
Structures (4%)
Fire Protection (0.75%)
Electrical (4.5%)
Painting & Cleanup (0.75%)
Material & Labor
Overheat (30%)
Total Erected Cost (130%)
Engineering Fee (10%)
Contingency (10%)
Total Capital Investment
Size
1 ,500 LPH
7,950 liters
31 ,400 gal
1,890,000 liters)
17,900 watts }
22,710 liters
16,870 liters
1,890,000 liters
$1,044,400
313,300
1,358,000
136,000
136,000
$1,630,000
Estimated
Eauioment
171,250
22,500
10,000
80 000
UW ) WWW
8,300
100,000
12,200
10,000
15,000
429,500
193,300
17,200
17,200
17,200
3,200
19,300
3,200
700,100
Costs
Labor
43,000
193,300
25,800
12,000
3,500
19,500
28,000
19,200
344,300
* Includes burner, incinerator, reaction tailpipe, Venturi scrubber, scrubber tank,
pump, air supply system, waste pump.
From plant waste HC1.
**
-------
also includes the scrubbing water supply system tank (24-hour capacity)
and pump, the scrubber wastes collection/settling pond (24-hour storage),
the pH control system, and the storage tank (24-hour capacity) for the
plant waste HC1 used to adjust pH.
The estimated capital investment for the CsClg waste mixture is
$1,630,000. This estimate, as noted above, is conservative since the tests
at TMC showed that only a 1:1 feed ratio between waste and No. 2 fuel oil
was required to achieve satisfactory incineration.
6.2 ANNUAL OPERATING COSTS
The annual operating costs consist of annual labor, chemical and
utility costs plus cost of capital, equipment depreciation, maintenance,
taxes and insurance. The labor costs have been calculated on the number of
personnel assigned to operate the incinerator and the automatic waste
loading equipment at rates prevalent in the chemical industry. The costs
for supervision, supplies and payroll-related expense are also included.
The utility costs include electricity and scrubber water; solid waste
disposal costs are shown separately. In the plant for hexachlorocyclopenta-
diene waste incineration, the costs shown also include the annual require-
ments for No. 2 fuel oil and sodium hydroxide. The amount of No. 2 fuel
oil consumed was based on a 1:1 weight ratio with Cede wastes (shown to
have proper incineration characteristics by tests at Tftc). Annual electric-
ity usage was based on estimates from TMC. Scrubber water and caustic
soda requirements were calculated from plant test data.
The annual operating costs for the plant to incinerate 15 million
kilograms per year of ethylene manufacturing wastes are summarized in
Table 6-3. The estimated annual operating expense for this plant is
$1,037,500 or $69.31/metric ton. As noted earlier for "Capital Investment",
approximately one-third of this annual cost for the disposal of ethylene
wastes is due to scrubber-associated systems which may not be mandatory in
this case.
The annual operating costs for the plant to incinerate 4.5 million
kilograms per year of hexachlorocyclopentadiene wastes are summarized in
Table 6-4. The estimated annual expense based on hexachlorocyclopentadiene
wastes is $2,211,700 or $487.59/metric ton.
Both disposal plant annual cost estimates were premised on 24-hour per
day operation, 330 days per year. The "cost of capital" shown is based on
the assumption that private debt financing is used for each disposal plant.
84
-------
Table 6-3. Annual Operating Cost
15 Million kg./Yr. Ethylene Manufacturing Waste Incineration Plant
Item Cost - $/Yr.
Depreciation (15% of plant investment) 272,700
Cost of Capital (10% of plant investment) 181,800
Maintenance (8% of plant investment) 145,400
Utilities 86,400
Electric power 290KW(7920) ($0.015) = 34,500
Scrubber water 787 x 107 liters/yr @ $0.066/1000 liters = 51,900
Solid Waste disposal @ 6.50/metric ton for
n wt. processed = 15 x 10° (.02) ($6.50) = 2,000 2,000
1000
» Labor x 312,800
Incinerator Operator 1 x 24 x 365 x
Incinerator Operator Helper 1 x 24 x 365 x $6.50 /= 184,000
Automatic Feed System Operator 1 x 24 x 365 x
$7.50^
$6.50 > =
$7.00j
Supervision (15% Operating Labor) = 27,600
Supplies (20% Operating Labor) = 36,800
Payroll Related Expense (35% Labor) = 64,400
Taxes & Insurance @ 2% Plant Investment .36.400
Total $1,037,500
Cost per metric ton of ethylene manufacturing waste = $69.31
-------
Table 6-4. Annual Operating Cost
4.5 Million kg./Yr. Hexachlorocyclopentadiene Waste Incineration Plant
Item Cost - $/Yr.
Depreciation (15% of plant investment) 244,500
Cost of Capital (10% of plant investment) 163,000
Maintenance (8% of plant investment) 130,400
Utilities 60,400
Electric power 190KW{7920) ($0.015) = 22,600
Scrubber water 572 x 107liters/yr @ $0.066/1000liters= 37,800
Supplies, Chemical g 1,266,000
Fuel Oil, No. 2, 4.5 x 10° kg = 32,530 bbl<3 $13.00/bbl= 423,000
NaOH (5035 solution) 5.35 x 106 kg 6 $158 metric ton = 843,000
c» HC1 - Plant waste at zero value, 10.2 x 106 kg
Solid Waste Disposal & 5.90/ton for 2% of waste processed 2,000
Labor 312,800
Incinerator Operator 1 x 24 x 365 x
Incinerator Operator Helper 1 x 24 x 365 x $6.50 } =184,000
Automatic Feed System Operator 1 x 24
$7.50 1
365 x $6.50 >
x 365 x $7.00J
Supervision (152 labor)
Supplies (20% labor)
Payroll Related Expense (35% labor)
Taxes & Insurance @ 2% of Plant Investment
= 27,600
= 36,800
= 64,400
Total
32,600
$2,211,700
Cost per metric ton of C,-C1, waste =$487.59
o 6
-------
APPENDIX A
ASSESSMENT OF ENVIRONMENTAL
IMPACT OF DESTRUCTING CHEMICAL
WASTES AT THE MARQUARDT COMPANY
87
-------
ASSESSMENT OF ENVIRONMENTAL IMPACT OF DESTRUCTING CHEMICAL WASTES
AT
THE MARQUARDT COMPANY
16555 SATICOY STREET
VAN NUYS, CALIFORNIA
The Marquardt liquid injection incinerator, located in Van Nuys,
California, is adjacent to the Van Nuys Municipal Airport. This incinerator
will be used for the destruction and evaluation of the following wastes:
1) Ethylene manufacturing waste
2) Hexachlorocylcopentadiene (C-5,6)
The Los Angeles Air Pollution Control District has been notified of
the intent to thermally destruct these wastes at the Marquardt facility.
The incinerator is operated under L.A. APCO Permit No. P55506.
The SUE® (SJJdden .Expension) burner, derived from aerospace engine
technology, has demonstrated high effectiveness in incinerating waste
propellents and solvents as well as chlorinated hydrocarbon herbicides
and pesticides. The burner was designed for a continuous feed rate of 50
to 60 gallons per hour of liquid wastes, and a maximum wall temperature of
3000°F. TMC's incinerator facility is a well-instrumented research type
installation equipped with a high energy venturi scrubber for control of
atmospheric emissions. Scrubber water is analyzed by TMC personnel before
discharge to the municipal water treatment system. Solid residue is dis-
posed of in an approved landfill after analysis by TMC.
The area surrounding the incinerator facility includes other Marquardt
test areas, the Van Nuys airport, and residential areas. The nearest
residential area is a trailer court approximately 400 yards from the incin-
erator. Prevailing winds are from the west (toward the airport runways) at
usually moderate velocity. The parking lot and nearby residential streets
are lined with trees. No wildlife was observed in the immediate area.
Local vehicular traffic is not affected by this program, since drums
are delivered only at the start of the test series. Operation of the
incinerator is not noisy compared to other equipment, such as ramjets,
tested at TMC. Only a white steam plume from the wet scrubber can be
observed when the incinerator is operating. The incinerator facility oper-
ates 8 hours per pay only when tests are scheduled. A test crew consists
of four to five persons.
The potential detrimental environmental effects are expected to result
from: 1) storage and handling of wastes prior to destruction, 2) emissions
occurring during tests, and 3; disposal of liquid and solid residue remain-
ing after combustion. The most significant hazard would result from contact
with waste liquid and/or fumes during a spill. The polymeric oil waste from
ethylene manufacture has physical characteristics (viscosity, density, color)
and elemental composition (88.7% C, 8.61% H, 1.32% S) similar to the No. 2
88
-------
oil. The principal hazard of this waste is its flammability. The CgClg
waste contains approximately 56 percent CrCle, 33 percent octachlorocycTo-
pentene (C5C18), and a number of other chlorinated cyclic compounds. CsClg
is known to be moderately toxic by ingestion, inhalation or through skin
contact. The 1973 Toxic Substances List provides the following data on
C5C16:
Oral Toxicity (rat): LD5Q - 505 mg/kg
Dermal Toxicity (rabbit): lowest lethal dose - 430 mg/kg
Inhalation Toxicity:
LC5Q (rat) - 3.1 ppm (3hr)
Lowest lethal dose (mouse) - 1.4 ppm (3 hr)
Lowest lethal dose (rabbit) - 1.4 ppm (3 hr)
LCgo (guinea pig) - 3.2 ppm (7 hr)
Toxicity data on CgClg are not available. The toxicity of CgClg is probably
in the same range as CgClg.
Storage and Handling
TMC has established safety procedures for handling hazardous materials,
including propellents and explosive materials. A detailed safety plan
has been prepared for this program, including waste handling procedures.
Wastes will be received by truck in 55-gallon drums, which will be inspected
for leaks and stored in a bonded area until transferred into the run tank.
All transfer operations will be conducted within a diked, electrically
grounded area by trained personnel using appropriate equipment and protec-
tive clothing. Any leaks or spills will be flushed and incinerated.
Incineration Tests
Operating temperature and residence time of the liquid injection incin-
erator should provide essentially complete combustion of the wastes, result-
ing in harmless exhaust emissions. On-line monitoring of gases from the
combustion zone will be utilized as an indication of combustion efficiency.
In addition, stack emissions (downstream of scrubber) will be checked for
hazardous gaseous species using Gastec® analyzer tubes for specific gases
and vapors. Caustic scrubbing solution will be used for chlorine removal.
Disposal of Residue
Residue material from the incineration process will consist of scrubber
water and ash. Liquid residue from the scrubber will be analyzed before
discharge to the municipal water treatment plant by TMC personnel. Solid
residues (ash) will also be tested by the TMC personnel prior to approved
landfill. Wastes remaining in the storage/run tanks and wash down liquids
from any spills will be incinerated at the conclusion of the test program.
89
-------
APPENDIX B
SAMPLE TRAIN OPERATION AND
SAMPLE VOLUME DATA
Preceding page blank
91
-------
APPENDIX B
SAMPLE TRAIN OPERATION AND
SAMPLE VOLUME DATA
For each test, data was collected on the operation of the two sampling
trains. This information is presented in Table B-l. The percent of
isokinetic at which the gas samples were drawn from both the stack and
combustion zone sampling sites was calculated from the following equation
given in EPA Method 5:
• •
/i C77 "i1n\ f/n no267 in. Hg-cu ft\ v + _m /p + H \] T
V-677IeTJ ^°-OOZ67 ml - OR ) vw * Tm Tbar 13.6 j) 's
1 = ^TP /'7854 DnZ\
5 s V 144 /
The terms of this equation are defined, and values measured for each
test burn are summarized in Table B-l. It was assumed for the purpose of
these calculations that Ps at the stack was nearly ambient or slightly
positive, and that the 0.5-inch diameter combustion zone probe would behave
roughly like a 0.5-inch nozzle.
Tables B-2 and B-3 summarize the sample gas and collected water volume
data.
92
-------
TABLE B-l. GAS VOLUME DATA
Weights/Conditions
Volumes at Standard
Conditions (ft3)
V .
meter
V
water
u ,
C12
w ,
VHC1
u
CO 2 scrubbed
Mole Fractions
B ,.
water
Cl
U2
BHC1
Gas Molecular
Weight (Ib/lb-mole)
M..
dry
•w
Run I
SG
62.71
46.93
.452
30.05
24.63
CG
70.81
15.6'
.109
30.0
28.7
Run II
ST
64.68
30.62
.347
30.11
25.91
CG
72.57
11.00
.085
30.11
29.08
Run III
SG
47.72
24. 6'i
.366
30.02
25.62
CG
160.22
9.K
.079
30.02
29.0
Run IV
SG
46.84
33.75
.444
30.25
24.81
CG
66.65
12.23
.093
30.25
29.11
R'in V
SG
46.76
30.15
.417
29.94
24.96
CG
129.72
14.98
o.oot
1.68(
*
.127
.000
.012
Run \I
SG
82.99
39.72
.349
30.02 30.09
28.57
25.87
CG
169.04
22.75
0.00
3.20
0.27
.142
.000
.016
30.19
28.56
tur VII
ST
45.75
31.19
.430
30.10
24.90
CG
136.29
19.55
0.00
3.75
0.57
.147
.000
.024
30.25
28.60
\o
to
insufficient sample to analyze.
-------
TABLE B-2. FINAL CORRECTED GAS VOLUMES
Run
Run I
Stack
Combustion Zone
Run II
Stack
Combustion Zone
Run III
Stack
Combustion Zone
Run VI
Stack
Combustion Zone
Run V
Stack
Combustion Zone
Run VI
Stack
Combustion Zone
Run VII
Stack
Combustion Zone
Dry
T~
fr
62.71
170.87
64.68
172.57
47.72
160.22
46.84
166.65
46.76
131.41
82.99
172.52
45.75
140.63
3
1.78
4.84
1.83
4.89
1.35
4.54
1.33
4.72
1.32
3.72
2.35
4.89
1.30
3.98
Met
. 3
ft
109.64
186.51
95.30
183.57
72.37
169.32
80.59
178.88
76.91
146.39
122.71
195.27
76.94
160.18
3
m
3.11
5.28
2.70
5.20
2.05
4.80
2.28
5.07
2.18
4.15
3.48
5.53
2.18
4.54
-------
TABLE B-3. LIQUID IMPINGER VOLUME DATA
c . Starting Volume Volume Final
bample Before Test (ml) After Test (ml) Sample
Volume (ml)
MA-I-SR-CI
MA-I-CG-CI
MA-I-CG-C1-A
Mft-II-SO-Cl
MA-II-C6-C1
•W-II-CG-CI-A
.'1A-III-SG-C1
r'A-III-CG-CI
'1A-III-CG-CI-A
MA-IV-SG-C1
-A-iv-cr.-ci
'1A-IV-CG-C1-A
flA-V-SG-Cl
'4A-V-CG-LI
MA-V-CG-L1-A
MA-V-CG-CI
W-V-CG-CI-A
MA-VI-SG-CI
I1A-VI-CG-LI
MA-VI-C6-I.I-A
MA-VI-CG-CI
f1A-VI-CG-CI-A
F1A-VII-SG-CI
MA-VII-CG-LI
'1A-VII-CG-LI-A
MA-VII-CG-CI
MA-VII-CG-CI-A
ZOO
200
ZOO
200
200
200
200
200
200
500
200
200
1200
200
200
1000
200
1171
471
826
387
702
348
883
422
819
730
250
1019
1543
300
R34
1293
290
1171
221
250
826
117
250
702
98
250
883
172
250
819
500
500
150
100
1019
1293
250
200
100
834
1000
son
190
100
Volume of Water
Collected (ml)
971
271
626
187
502
148
683
222
619
230
50
819
343
100
634
293
90
95
-------
APPENDIX C
CALCULATION OF INCINERATOR PERFORMANCE
Preceding page blank
97
-------
APPENDIX C
CALCULATION OF INCINERATOR PERFORMANCE
The destruction efficiencies presented in Table 1-1 were calculated for
each of the tests where the appropriate on-line and hot zone samples were
taken. The data used in these calculations can be found in Tables 4-6,
5-2, 5-3, and 5-4.
The waste destruction efficiency (DEwaste) is based upon comparing the
waste input rate to the waste emitted rate.
DEwaste = waste Input^ waste emitted x , QQ% ((M }
Equation C-l restated in another form, is
DEwastfi = ^aste - j^V Ewaste] x 1QO% (C-2)
waste
where:
I «. - input rate of the waste = fuel feed rate x weight
waste fraction of waste in fuel. Units are milligrams
per second. Table 4-6.
VFR = volumetric flow rate of combustion gases from the
burner. Approximately equal to input rate Table
4-6. Units are cubic meters per second.
Ewaste = concentration of organic waste constituents in
the combustion gas as determined by G. C. It is
the sum of the waste constituents found in the
probe wash, filter and sorbent trap samples from
the combustion zone (Table 5-2). Units are milli-
grams per cubic meter.
Similarly the destruction efficiency for total organics
. ) compares the input rate of combined waste and auxiliary fuel to
organics
the emitted rate of all organic material found in the combustion zone samples,
' E
DEtotal organic = fue1 " tOta1 Or
-------
where:
*fue1 = Input rate of combined waste and fuel oil (when used).
Units are milligrams per second. Table 4-6.
total organics = sum of the concentrations of all organics
found in the combustion zone samples. (Tables 5-2,
5-3,5-4) Units are milligrams per cubic meter.
The waste destruction efficiency calculation for Test MA-V is presented
below as an example. Initially Iwaste is calculated, then VFR , and
finally Ewastg is taken from 5-2, added to the equation to calculate
^aste = 0.0341 kg/sec x 1 x 10 mg/kg
3
= 11,400 mg/sec
where:
0.0341 kg/sec = total fuel feed rate Table 4-6
1/3 = C-5,6 weight fraction in fuel
Volumetric Flow Rate of Gas
VFRr,ac = °-568 JSSL * mole x lOOOg/kg x 0.024 m3 x 294° K
gas sec 28.8g mole 273° K
=0.51 m /sec
where: 0.568 kg = air feed rate Table 4-6
sec
where: 0.568 kg = air feed rate Table 4-6
sec
mole air = estimated molecular weight of combustion gas
28.8g
0.024 itr = approximate molar volume of gas at 21°C
mole
99
-------
294° K = volume correction based on temperature.
273° K
Waste Destruction Efficiency
DEwaste = 11.400 - [0.51 x <0.2] 1(m (c_2)
11,400 x u
= >99.9991%
The destruction efficiencies for total organics. DEtQtal Organ1c
are calculated in a similar manner using equation C-3 and inputing the com-
bined waste and auxiliary fuel feed and the concentration total organic found
(Tables 5-2, 5-3 and 5-4).
100
-------
APPENDIX D
ANALYTICAL CHEMISTRY DETAILS
101
-------
APPENDIX D
ANALYTICAL CHEMISTRY DETAILS
The elements of this appendix consist of the interpretation of the
instrumental data that led to the results discussed in Section 5. As
part of the on-going effort to thoroughly characterize hardware and sample
behavior in this relatively new analysis technology, extra effort was made
to characterize certain control samples and monitor filter weights during
their preparation. Details of this effort are also documented in this
appendix.
D-l. Survey Analysis on Combustion Zone Filter Extracts
The material found in the solvent (methylene chloride) extracts of
the combustion zone filters was surveyed for its qualitative nature by
infrared spectrophotometry (IR) and low resolution mass spectrometry
(LRMS). The IR data indicates that two main classes of compounds are
present. The first is alkyl esters of phthalic acid. This commonly used
placticiser and lubricant may have been in the waste feeds or may be a
contaminant from the system. Whether one compound or several members of
the family of these compounds is present cannot be determined from the IR
scans alone. The second group of compounds is esters of fatty acids such
as adipic and sebacic acids. The evidence for this is that the 2800 to
2900cm~' IR region is more intense than that usually found on commonly
occurring phthalate esters. Further evidence is shown by the LRMS data.
The LRMS data support these conclusions drawn from the IR data. The
m/e 149 peak is quite evident, but is not the strongest peak in any of the
mass spectra. For all phthalate esters in the available literature, this
149 peak,which is caused by the stable "anhydride plus 1 hydrogen" ion, is
the strongest peak. Other m/e peaks in the sample such as 27, 29, 41, 43,
55, 57, 69, 71, etc., are stronger and this indicates that alkyl hydro-
carbon groups, classically represented by this peak pattern, predominate.
The presence of a peak at 167 atomic mass units (AMU) is indicative of
di-n-octyl phthalate and dimthyl phthalate being present. A peak at 185
AMU is strong evidence for adipate and sebacate (fatty acid) esters.
It is important to note that the samples show no evidence of any of
the hazardous constituents of the original waste feeds. That is to say,
none of the constituents of the ethyl ene waste nor C-5,6 and its chlori-
nated homologues can be seen in the filter extract samples from their
respective tests.
D-2. Survey Analysis of Control Sorbent Trap Extracts
This section presents the results of the analysis of solvent and
sorbent traps that were examined as part of the analysis of the traps
actually used in the sample trains during testing. This discussion is
important from an analytical view point because significant amounts of
background material were found in these control traps. As discussed later,
the materials found in both the samples and controls are quite similar and
102
-------
the discussion here is a prelude to the discussion of the materials ex-
tracted from the test burn samples. Furthermore, results gathered here
impact on the preparation and use of these traps in the future in order to
lower their background contribution.
Two solid sorbent trap control samples were analyzed. The require-
ment for two samples resulted from the fact that two lots of pre-cleaned
resin were used in the tests. The first control sample, MA-ST-C1, was
extracted using a standard soxhlet extractor with pre-extracted thimbles
into which the resin was placed after removal from the trap. The second
control, MA-ST-C2, was extracted using the special apparatus which performs
a Soxhlet-type extraction of the resin without its removal from the trap.
The survey analysis results of these control samples show some weight
differences depending on the extraction method used. The soxhlet extracted
resin yields more weighable residue (Table D-l). The table also shows that
the combined thimble and solvent blank also yielded a residue in spite of
the prior extraction of the thimble, and the use of the highly pure sol-
vents.
The first control trap, extract residue MA-ST-C1, contained esters of
phthalic acid, perhaps an amide, and substituted benzene compounds. This
material is present in spite of the extensive sorbent trap resin cleanup
and pre-extraction of the Soxhlet thimbles and apparatus.
Table D-l. Extracted Material from Control Samples
Sample
MA-ST-C1
Control Sorbent Trap
MA-ST-C2
Control Sorbent Trap
MA-ST-BLANK
Soxhlet Thimble
Solvent Blank
Material Extracted
by Pentane (mg)
1.370
0.430
0.810
Material Extracted
by Methanol (mg)
12.3501
4.490
1.635
1
The soxhlet thimble used for the C-l control was not pre-extracted
with methanol prior to extracting the sorbent resin with methanol.
The reported value is believed to be higher than would have been
achieved with a methanol pre-extracted thimble.
103
-------
The Infrared spectra of the, MA-ST-C1 residues were quite similar for
both pentane and methanol extracts. The qualitative information is as
follows:
Spectral
Region
3025
3000-2800
1740
1450
1680
1260
Assignment
vC-H
C-H bending
vOO
vC-N, 6-N-H or vC-O-C
Possible Compounds
aromatic, possibly phthalates
alkyl hydrocarbon
ester
alkyl hydrocarbon
amide
ester or amide
The LRMS data from the MA-ST-C1 sample displayed the following pattern.
AMU Assignment
27, 29, 41, 43, 55, 57, etc.
77, 91, 105
149
Alkyl hydrocarbon
Alkyl substituted benzene
compounds such as toluene,
ethyl benzene and xylene
Phthalate esters
The second control sorbent trap, sample MA-ST-C2, yielded consider-
ably less residue than the first control resin. The reason is believed to
be the elimination of the thimble as contributor of some of the contami-
nating materials, however the weight of residue obtained from resin itself
also indicates a source of potential contaminants. The IR spectrum of
the pentane extract residue was very weak and of little use. At the
most favorable sensitivity, only the 3000 to 2800cm-' region (alkyl hydro-
carbons) was discernable. The methanol extract residue showed only traces
of the methanol solvent and one peak at 1610 to 1520cm'1 (-NHg scissoring)
which suggests, but which does not confirm an amine.
The material extracted from the second resin control sample by
pentane had a LRMS peak pattern that was rather typical of the types of
compounds seen in the Cl control sorbent trap and in many other sorbent
trap extracts. Later discussion of other samples will refer to this
pattern. The significant aspects of the low resolution mass spectra are
as follows:
Mass Spectra Peak Pattern
A dense peak pattern commonly called
"pickets" containing stronger peak
pairs at 27, 29, 41, 43, 55, 57, 69,
71, etc., increasing in 14 AMU Incre-
ments (CHo). These peaks usually
extend up'through the 111, 113 AMU
peak pair befor they become no more
intense than surrounding peaks.
104
Assignment
Alkyl, normal and
branched hydrocarbons
either as a class of
compounds or as sub-
stituents on other
classes of compounds.
-------
Mass Spectra Peak Pattern Assignment
91, 105 Alkyl substituted
benzene compounds, e.g.,
toluene, ethyl benzene
styrene (benzoates are
also possible).
Ill, 129, 185 Adi pate, sebecate esters
(fatty acid esters)
149 Phthalate esters
241 Di-butyl sebecate
The methanol extract of the MA-ST-C2 resin trap contained not only
the same "picket" pattern, the substituted benzene pattern, and phthalate
peaks described immediately above; but also contained a 279 AMU peak
which, while not found in the TRW base of mass spectrometric data, has
been identified as characteristic of dodecanoic (lauryl) sulfonates com-
monly found in detergents and surfactants.
D-3. Survey Analysis of Thimble and Solvent Blanks
The background sample characterization for sorbent traps was completed
with an analysis of a third control sample obtained by extracting an
empty thimble with pentane and methanol after it had already been pre-
extracted with these solvents as per the standard resin extraction procedure.
The amounts of recovered material are small in relation to the other
samples but the LRMS data indicate the same basic types of compounds:
AMU Assignment
Dense "pickets" Alkyl hydrocarbons
149 Phthalates
111, 129, 185 Adi pate, sebecate esters
167 Di-methyl phthalate
256, 73 Palmitic acid esters
There was insufficient sample in the extract aliquot to obtain a useful
IR spectra.
D-4. Survey Analysis of Test Sorbent Trap Extracts.
This section presents the results of the survey analysis on the
extracted material from the sorbent traps used in the sample trains. The
procedures for the extracting these traps is discussed in Section 4.4, and
105
-------
the weight of extracted material is summarized in Table 5-4. in general,
the composition of these materials from tests with the same waste are quite
similar, and for ease of presentation the results are given in two groups:
(1) MA-I, V, VI, VII, relating to the tests with C-5,6/fuel oil, and
(2) MA-II, III, IV, relating to the tests with ethylene manufacturing
waste.
MA-I, V, VI, VII
The IR spectra for pentane and methanol (and benzene 1n the three
samples where it was used) extracts are very similar. However, there are
some differences from which inferences can be made with respect to the
overall composition. These differences are discussed below. The peaks
found in the pentane extract residues and which are useful in determining
the structure of these materials follow.
The pentane extracts of the solid sorbent traps yielded the IR
spectra and assignments below. The methanol extract residues display the
same pattern except that there is a trace of aromatic C-H absorption above
3000cm-l and the carbonyl (C=0) peak in the 1750 cm-1 is somewhat
stronger indicating that the methanol may be more effective in extracting
the ester material. There is no evidence in the IR spectra that indicates
the presence of any uncombusted C-5,6. The levels of sensitivity at which
C-5,6 would be seen in this sample matrix were discussed in Section 4.4.2
and are on the order of less than 5 percent of the residue from each
extract as shown in Table 5-4.
Spectral Region cm"1 Assignment
2950 vas CH3
2920 vas CH2
2850 vs CH3, CH2
1750 - 1730 VC=0 ester
1700 VC=0
1650 - 1500, weak phenyl nucleus or weak amide
1460 - 1450 6asCH3, CH2 scissoring
1380 - 1350, weak 6s CH3 bending of methyl group
from alkyls, esters
1300 - 1250 asym C-O-C stretch; esters, ethers
1100 - 1050 sym C-O-C stretch; esters, ethers
The LRMS data of the sorbent trap pentane extracts confirm the general
features of the IR spectra. The typical hydrocarbon fragment "picket"
pattern is evident in all samples. Alkyl substituted benzene compounds
106
-------
are indicated (77, 78, 79, 91, 105, 133 AMU). Phthalate esters are also
present but not as the predominant class of compounds. The presence of
esters of fatty acids (adipic, sebacic) are shown by the 111, 129, 185
peaks in several of these samples.
The methanol extracts of the sorbent traps contained the same com-
pounds. The 149 AMU phthalate peak was stronger in these extracts. A
45, 60, 74 AMU pattern was indicative of alkyl, monocarboxylic acids.
Some of the fatty acid compounds produce these peaks. As a matter of
routine examination of all LRMS data from all samples, mass spectral
evidence of hexachlorocyclopentadiene (C-5,6) and any other cholorinated
hydrocarbons were specifically and carefully searched for. None were
detected. The sensitivity of the mass spectrometer solids probe analysis
technique for searching for particular compounds in a sample is discussed
in Section 4.4.2. The detectable level for a compound under the worst
circumstance was 10 percent (w/w) of the sample residue being examined.
MA-II-. Ill, IV
The MA-II, III and IV sorbent trap extracts display differences in
the compositional nature of the extracted material especially between that
extracted by pentane and that extracted by methanol. The IR spectra for
the pentane extract residues of MA-II, III, and IV are essentially the
same as the pattern discussed earlier for MA-I-C6-ST. The MA-III
spectrum is weaker than the others, but the major peaks are apparent.
The aromatic composition of the MA-II, III, IV pentane extracts is
greater than that for the MA-I baseline run. This is consistent with the
high aromatic content of the ethylene waste. The spectrum of the MA-II
methanol extract residues show the strong alkyl and carbonyl patterns
discussed above with respect to the C-5,6 tests, but the MA-III and IV
spectra show extremely small amounts of absorption in these areas. The
esters and hydrocarbons previously mentioned are present at very low
levels in these two samples. The only peaks in these spectra valuable
for identification purposes are caused by small amounts of residual
methanol in the sample after evaporation.
The LRMS data confirms the observation that the esters, etc., are
present in these samples at lower levels relative to the others sorbent
trap extracts. The LRMS data from the sorbent trap pentane extracts for
Runs II, III, and IV again show distinct similarity to each other and
with the patterns earlier described for the sorbent trap control samples:
AMU Assignment
hydrocarbon pickets alkyl hydrocarbons or hydrocarbon
41, 43, 55, 57, etc. substituents on other compounds
73 (strong peak) C-O-R esters, ethers, etc., sub-
77, 78, 79, 91, 105 stituted benzene compounds
133 phenyl compounds, substituted
benzenes, benzoates, phthalates
107
-------
AMU Assignment
149 phthalates
167 dimethyl phthalate
185 ad1pate and sebacate esters
p+14 (-CH9) pattern hydrocarbon oils and greases
from ^2077 221, etc.,
out to AMU 600 and
higher
The LRMS data from the methanol extract residues of the sorbent traps
for Runs II, III, and IV show the same pattern discussed above, but the
patterns are less Intense and indicate that the esters and automatic
compounds are likely present at lower levels. This trend agrees with the
IR data discussed above. Different patterns are evident in the methanol
extracts but examination of the literature provided a multitude of com-
pounds which ionize into some of the fragments listed above but there is
not a sufficiently good fit of the literature data with the sample
spectra in other areas of the spectrum to arrive at even a tentative con-
clusion as to what types of compounds are presents. It is stressed how-
ever that this spectra bears little or no resemblance with that of the
original waste.
AMU Assignment
31 primary alcohols, esters, or ethers
32
48
64
65
80
81
82
95
110
113 possible furyl or similar ring ethers
Mass Spectrometer Analysis of Sorbent Trap Resins By Direct Probe
Additional LRMS direct probe analysis of the sorbent trap resins from
selected tests was carried out to 1) determine the nature of the material,
if any, which might still be absorbed on the resin after solvent extrac-
tion, and (2) gather additional information as to the source of the fatty
acid esters and phthalic acid esters which are found in many of the
samples.
108
-------
Four samples were selected for analysis:
• MA-ST-C1 pre-cleaned, unused resin
• MA-I-CG-ST base-line test with No. 2 oil
o MA-III-CG-ST second test with ethylene waste
o MA-V-CG-ST first test with C-5,6 waste/fuel oil
Resin samples from MA-III and MA-V were selected because the amounts of
materials extracted from the sorbent traps for these tests were signifi-
cantly lower than the recoveries obtained from the other tests with the
same waste, leading one to suspect that there might be unextracted mate-
rial still on the resin. A portion of the resin was placed in the solids
probe of the mass spectrometer. The probe block was heated from 50° to
350°C at 100°C increments and the thermally desorbed constituents were
then analyzed. This analysis provides no quantitative estimate of what
might be present on the resin, it only provides qualitative information.
The results of the mass spectra of desorbed materials are summarized
as follows:
Sample
MA-ST-C1
MA-I-CG-ST
MA-III-CG-ST
MA-V-CG-ST
Observations
Only residual benzene (the third and
final extraction solvent) and the
149 AMU phthalate fragment were noted
at all temperatures.
Benzene is the only compound seen at lower
temperatures. A pattern at 60 to 64 AMU is
apparent at 150°C and 250°C probe tempera-
tures. At higher temperatures after
benzene is driven off, a 91,105 pattern of
substitued benzene can be seen. Phthalate
ester peak at 149 is seen.
Alky! substituted benzene compounds
were detected along with benzene, fatty
acid esters and phthalate esters.
Benzene and alkyl substituted benzenes,
fatty acid and phthalic acid esters,
dimethyl phthalate. Chlorinated species
were specifically looked for and none were
found.
109
-------
The results of the analysis on the extracted resins from selected
runs show that no new classes of compounds were thermally desorbed from
the resins which had not already been determined as part of the solvent
extracted material. The fact that traces of phthalate esters were found
on the MA-ST-C1 control resin is evidence which supports the belief that
the resins retained these compounds in low levels in spite of the resin
cleanup procedures which were performed prior to the test program. The
presence of these materials does not reduce the effectiveness of searching
for toxic species; their presence only makes the task more difficult.
D-5. Survey Analysis of Scrubber Water Sample Extracts
The LRMS data from the carefully evaporated residues of the scrubber
water extracts for tests MA-I through MA-IV were quite similar. They
displayed a pattern shown by a mixture of alkyl hydrocarbons. The m/e 27
and 29 peaks are strongly present. These peaks are caused by the C2Ha
and C2H5 fragments, respectively. This two-peak pattern is repeated every
14 atomic mass units (AMUs), i.e., 41 and 43, 55 and 57 (propyl, butyl,
etc.). The 14 AMUs correspond to the -CH2 fragment which is added to the
molecular weight with increasing carbon number in a homologous series.
There is evidence of constituents with molecular weights greater than 350.
This is not uncommon with hydrocarbon oils. A compound with a molecular
weight around 350 would contain about 25 carbon atoms. Evidence of the
minor presence of phthalates is provided by the characteristic m/e 149
peak. The presence of substituted benzene compounds such as toluene,
xylene, ethyl benzene, etc., at trace levels is also indicated.
Three additional peaks are present and are not in the 14 AMU peak
pattern discussed above. These peaks occur at 247, 303, and 340. The
compound(s) causing these peaks could not be identified using the
available literature. Substituted benzenes and naphthalenes at low
levels are also indicated, but xylene and toluene are not among those present.
The LRMS data had weak spectral patterns since only very small amounts
of extract residue were found in the city water control sample, the fresh
caustic scrubber control sample and the spent scrubber waters from MA-V -
MA-VI as very little sample residue available for transfer to the mass
spectrometer sampling system. The only signal shown by these extracts is
the hydrocarbon cracking pattern discussed above. The highest detectable
molecular weights are 139/141 although it is believed that higher
molecular weight materials are likely to be present at undetected levels.
The LRMS spectrum of the MA-VII scrubber water extract residue is quite
similar to the MA-I through MA-IV spectra discussed above, except that the
m/e 247, 303, 340 pattern is absent. There is no evidence of any C-5,6 in
the residue. The estimated levels at which these materials are seen in the
mass spectrometer have been discussed in Section 4.4 and apply here.
D-6. Survey of the Ground Clinker by LRMS Direct Probe.
The LRMS data for the ground clinker form the tests with C-5,6/fuel
oil display the hydrocarbon pickets at m/e 27, 29, 41, 43, etc., up to 69,
110
-------
71. Traces of benzene and/or toluene were noted. Peaks at 35 and 70
were observed that can be described to monotomic and diatomic chlorine.
These material are expected as combustion products. Hexachlorocyclopen-
tadiene (C-5,6) and other chlorinated compounds organics were specifically
searched for, but no trace of any chlorinated compounds were detected in
the unextracted clinker. In addition, the typical 149 AMU peak for
phthalate esters was found in all of these C-5,6 clinker samples. This
is significant because these clinker samples were not extracted or other-
wise processed except for grinding, and the probability of contamination
during grinding is very small indeed. Thus, the presence of phthalates in
the clinker is strong evidence for the presence of these materials in C-5,6
waste/fuel oil feed used in the tests. This can explain the presence of
at least part, if not all, of these materials in the samples related to
C-5,6 testing. Examination of the extracted clinker showed only traces of
hydrocarbon pickets and methylene chloride, indication an extraction
efficiency similar to that found for the ethylene test clinkers.
D-7. Weight Behavior of Combustion Filters During Analytical Preparation.
Weights were obtained on both the stack and combustion zone filters
with their trapped particulate at several stages of their preparation as
part of the on-going effort to understand the behavior of these samples
during preparation and final analysis. They were first tared before
their use and then weighed after the tests to determine the amount of
particulate trapped. The weight of material extractable with solvent
was then determined as well as the weight loss upon a low temperature
plasma ashing. These weights are summarized in Table D-2. The data show
that the amount removed by solvent extraction of the combustion zone
filters ranges from 1 to 37 percent (13 percent average) of the total
lost by solvent extraction and plasma ashing combined. Since the
majority of the material removed by ashing is carbon in the form of soot,
this result is not surprising. However, there are certain data which
indicate that the solvent extraction process needs further study.
• Sample MA-II lost 66.3 percent of the original weight from
plasma ashing and yet the visible nature of the filter does not
indicate carbonaceous soot (Figure 5-1).
• All of the hot zone filters from tests I, II, III, and IV have
consistently larger losses (relative to tests V, VI, and VIII)
from ashing although the filters showed little or no visible
soot. These losses may represent organic materials that are
not being recovered by the solvent extraction. The amount of
material lost by plasma ashing, averaged for all seven tests,
represents 4 mg/m3 of sample gas. This value would be the maxi-
mum amount of unextracted organics if more of the particulate were
carbonaceous.
Ill
-------
Table D-2, Filter Heights Through Analyses (mg)
Sample
Run I
Stack
Combustion Zone
Run II
Stack
Combustion Zone
Run III
Stack
Combustion Zone
Run IV
Stack
Combustion Zone
Run V
Stack
Combustion Zone
Run VI
Stack
Combustion Zone
Run VII
Stack
Combustion Zone
Tare Weight
Before Test
605.6
617.1
609.9
606.6
647.2
656.8
652.8
653.6
643.8
655.7
658.8
658.7
658.2
655.6
Weight
After Test
627.9
653.1
641.6
709.7
668.0
697.0
679.4
690.1
681.9
684.0
752.7
700.4
797.6
711.9
Gain Due to
Parti cul ate
21.4
35.1
30.8
102.2
19.9
39.3
25.7
35.6
37.2
27.4
93.0
40.8
138.5
55.4
Loss on
Solvent
Extraction
0.8
0.6
0.6
4.1
1.5
1.6
1.4
Loss on
Plasma
Ashing
2.1
19.3
13.7
67.2
5.7
18.4
5.8
6.9
0.6
9.9
-2.8a
6.4
1.2
11.2
Remaining
Weight
on Filters
19.3
15.0
17.1
34.4
14.2
20.3
19.9
24.6
36.6
16.0
95. 8a
32.8
137.3
42.8
% Loss
9.8
57.3
44.5
66.3
28.6
48.4
22.6
30.9
1.6
30.7
-3.0a
19.6
0.9
22.7
a - Weight Gain
-------
D-8. Measurement of Particulate Loadings In the Combustion Zone vs Stack
Comparison of the measured participate loadings, shown in Table D-3,
between the combustion zone and stack values, also yields valuable in-
formation on the performance of the process and the ability of the
sampling and analytical techniques to accurately measure that performance.
Table D-3. Particulate Mass Loadings
Run No.
I
II
III
IV
V
VI
VII
Combustion Zone
mg/m
7.3
20.9
8.7
7.5
7.4
8.4
13.9
grains/std. ft3
0.0032
0.0091
0. 0038
0.0033
0.0032
0.0037
0.0060
Stack
mg/m
21.2
20.9
25.0
20.4
36.2
43.8
113.5
grains/std. ft3
0.0092
0.0091
0.0109
0.0089
0.0158
0.0192
0.0494
The fact that the concentration of particulate measured in the stack
(downstream of a participate scrubber) is higher than the level measured
in the combustion zone is considered to be the result of a combination of
four factors.
1. The stack was sampled superisokinetically. Rates at the
stack ranged from 125 to 200 percent of isokinetic. At
the hot zone, the rates were approximately 30 to 50
percent of isokinetic even though the hot zone train was
operated at its maximum flow rate since it was not pos-
sible to put a nozzle on the hot zone probe. Superiso-
kinetic sampling results in a bias toward a higher than
actual small particulate/large particulate ratio.
Sampling at lower than the isokinetic rate yields a
bias in the opposite direction.
2. There was no nozzle on the hot zone probe. The use of
a nozzle at Marquardt was ruled out as being impossible
to protect and/or cool in the hot, corrosive combustion
zone. It does appear possible, however, that the absence
of a nozzle could also have resulted in substantially
reduced collection of the large particulates at the
combustion zone.
113
-------
3. The combustion zone probe rinsings were not evaporated
to dryness and weighed as were the stack probe rinses.
The hot zone probe rinses were not evaporated because
of the likelihood of losing or altering any organics
which might be present. The significance of this
factor can be estimated from the fact that the probe
rinse weights from the stack were on the average equal
to 30 percent of the filter particulate weight.
4. Salts generated by the HC1 1n the C-5,6 runs from the
caustic in the scrubbing process were entrained in the
scrubber mists and were carried to the stack filter,
where they evaporated leaving a solid residue which was
weighed as part of the particulate. A more efficient
demister would likely reduce the salt emissions.
D-9. Inorganic Survey Analyses by ICPOES
To determine the inorganic composition of the particulate from the
combustion and stack gases, selected filter acid digests and impinger
solutions were first surveyed for 32 elements by ICPOES. These results
are shown in Table D-4.1 Normally the ICPOES method is, in itself, highly
accurate. However, solutions with high solids contents such as the
samples from caustic impingers and acid digested filters, create Inter-
ferences in the method which reduce its accuracy. Thus the ICPOES data
does not agree with the MS analyses as well as would generally be
expected.
In addition to the elements shown in Table D-4, the following ele-
ments are determined by ICPOES but were not detected in the submitted
samples: Au, As, Be, Eu, Pb, P, Se, Te, W, and U.
D-10. Wet Chemical Analyses for HC1. CU. and C0g
The caustic impinger solutions were analyzed for chloride, free
chlorine, and carbonate to determine the amounts of each which were
scrubbed from the combustion gases of the C-5, 6 tests. The results are
listed in Table D-5. The large, 2-liter impinger (described in Section
4.3.2) was analyzed separately from the standard Greenburg-Smith Impingers
to determine the relative scrubbing efficiencies for each specie. No
free chlorine was found in the Impingers from test V, thus a less than
value representing the detection limit of the technique is reported.
114
-------
Table D-4. Elemental Survey of Selected Samples (mg/m3)
ELEMENT
Ag
Al
Ba
Ca
Cd
Co
Cr
Cu
Fe
K
Kg
Mn
Mo
Na
N1
Si
Sr
T1
V
Zn
RUN I
COMBUSTION ZONE
IMPINGERS
ND
0.006
ND
0.025
ND
ND
ND
0.018
0.007
1.6
0.006
ND
ND
ND
ND
1.4
ND
ND
ND
0.004
FILTER
NGTB
<0.001
NGTB
<0.004
NGTB
<0.004
NGTB
<0.032
<0.001
ND
<0.001
<0.001
0.001
ND
NGTB
< 0.009
<0.001
ND
NGTB
<0.22
NGTB
<0.002
...
NGTB
<0.001
<0.001
ND
NGTB
<0.002
STACK
FILTER
NGTB
<0.001
NGTB
<0.11
NGTB
<0.14
NGTB
<0.53
<0.002
ND
0.001
< 0.001
0.002
ND
NGTB
<0.14
< 0.001
ND
NGTB
<1.7
NGTB
<0.002
...
NGTB
<0.007
<0.001
ND
NGTB
< 0.079
RUN IV
COMBUSTION ZONE
IMPINGERS
ND
0.052
ND
0.024
ND
ND
ND
0.019
0.039
1.5
0.008
ND
ND
ND
ND
1.6
0.0003
0.004
0.0004
0.009
FILTER
NGTB
<0.001
NGTB
<0.005
NGTB
<0.001
NtfB
<0.028
<0.001
ND
0.001
0.001
0.021
ND
NGTB
<0.007
0.001
ND
NGTB
<0.34
O.,002
—
NGTB
<0.001
ND
ND
NGTB
< 0.003
STACK
FILTER
NGTB
<0.001
NGTB
<0.030
NGTB
<0.035
NGfB
<0.33
0.002
ND
0.001
0.001
0.011
ND
NGTB
< 0.085
0.001
ND
NGTB
<1.5
NGTB
< 0.007
...
NGTB
<0.002
NO
ND
NGTB
< 0.020
RUN VI
COMBUSTION ZONE
IMPINGERS
0.010
ND
ND
0.056
0.0004
0.028
0.009
0.099
0.016
a
0.009
ND
0.076
a
0.002
1.2
0.017
ND
0.002
0.020
FILTER
NGTB
<0.001
NGTB
<0.009
NGTB
<0.001
NGTB
<0.048
<0.001
0.001
0.019
0.003
0.38
ND
NGTB
<0.013
0.022
ND
NGTB
<0.42
0.055
...
NGTB
<0.001
<0.001
NO
NGTB
0.005
STACK
FILTER
NGTB
<0.001
NGTB
<0.094
NGTB
<0.11
NGtB
<0.23
0.001
ND
0.008
0.003 ;
0.24
ND
NGTB
« 0.071
0.022
ND
6.2
1
0.056 •
...
NGTB
'0.004
-0.001
ND
NGTB
0.15
N6TB - Not greater than backoround
a - The caustic implnger solution had very
high backgrounds of Na and K
ND - Not detected
-------
Table D-5. Caustic Impinger Analyses
Sample
Run V
Large Impinger
G.-S? Impingers
Total
Run VI
Large Impinger
G.-S. Impingers
Total
Run VIII
Large Impinger
G.-S. Impingers
Total
Chloride
g Cl
66.900
3.175
70.075
130.075
3.120
133.195
151.800
4.350
156.150
liters HC1
47.74
90.76
106.41
Chlorine
mg Cl
<0.5
<0.5
<1
18.5
203.4
221.9
471.0
322.5
793.5
HterSCl2
<0.003
<0.076
0.269
Carbonate
g C02
6.635
7.680
14.315
22.950
6.989
29.939
liters C02
7.82
16.31
•e
Q
I
* Greenburg - Smith
------- |