EPA/600/2-91/032a
July 1991
THE FATE OF TRACE METALS IN A ROTARY KILN
INCINERATOR WITH A SINGLE-STAGE IONIZING
WET SCRUBBER
Volume I — Technical Results
By
D. J. Fournier, Jr. and L. R. Waterland
Acurex Corporation
Environmental Systems Division
Incineration Research Facility
Jefferson, Arkansas 72079
EPA Contract 68-C9-0038
Work Assignment 0-3
EPA Project Officer: R. C. Thurnau
Technical Task Manager: Gregory J. Carroll
Waste Minimization, Destruction, and Disposal Research Division
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
-------
TECHNICAL REPORT DATA
(Ptcast reed Instructions on Ihr rrsfric ttc/ort compltlr
REPORT MO.
EPA/600/2-9l/03Za
4. TITLE AND SU8TITI.E
THE FATE OF TRACE METALS IN A ROTARY KILN INCINERATOR
WITH A SINGLE-STAGE IONIZING WET SCRUBBER- Volume I
Technical Results
5. REPORT DATE
July 1991
6. PERFORMING ORGANIZATION cooe
7. AUTHORtSI
D. J. Fournier, Jr.
L. R. Waterland
b. performing organization report no.
s. performing organization name and aooress
Acurex Corporation
Jefferson, AS. 72079
10. pbqobam element no.
11. contract/grant no.
68-C 9-0038
12. SPONSORING agency name ANO AOORE5S
Risk Reduction Engineering Laboratory—Cin., OH
Office of Research and Development
U. S. Environmental Protection Agency
Cincinnati, Ohio 45268
10. TVPE OF REPORT ANO PERIOD COVERED
Project Report
1«. SPONSORING AGENCY COOE
EPA/600/14
IS. SUPPLEMENTARY NOTES
Gregory J. Carroll
FTS: 684-7948
Commercial: 513/569-7948
16. abstract
A series of pilot-scale incineration tests was performed at EPA's Incineration Research
Facility (IRF) in Jefferson, Arkansas, to evaluate the fate of trace metals fed to a rotary kiln
incinerator equipped with an ionizing wet scrubber (IWS) for particulate and acid gas control.
Test variables were kiln temperature, ranging from 816" to 92T°C (1500* to 1700T); afterburner
temperature, ranging from 982" to 1204'C (1800° to 2200°F); and feed chlorine content, ranging
from 0 to 8 percent. The test program evaluated the fate of five hazardous constituent trace
metals (arsenic, barium, cadmium, chromium, and lead) and four nonhazardous constituent
trace metals (bismuth, copper, magnesium, and strontium).
The test results indicate that cadmium and bismuth were relatively volatile, with an
average of less than 40 percent discharged with the kiln ash. Arsenic, barium, chromium,
copper, lead, magnesium, and strontium were relatively nonvolatile, with an average of greater
than 80 percent discharged with the kiln ash. Observed relative metal volatilities generally
agreed wiih the volatilities predicted based on vapor pressure/temperature relationships, with
the exception of arsenic which was much less volatile than predicted. The volatility of cadmium,
bismuth, and lead increased as kiln temperature was increased; the discharge distributions of the
remaining metals were not significantly affected by changes in kiln temperature. Apparent
scrubber collection efficiencies for the metals averaged 22 to 71 percent, and were generally
higher for the less volatile metals. The overall average metal collection efficiency was 43 percent.
1?. .
KEY WORDS ANO DOCUMENT ANALY5IS
DESCRIPTORS
b,IDENTIFIERS/OPEN ENDED TERMS
C. COSATI Ficld/Gioup
Incineration
Hazardous Waste
Trace Metals
Partitioning
Particle Size Distribution
Air Pollution Control Equipment
Emissions
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
t9. SECURITY Class I fins Report/
UNCLASSIFIED
21. NO. Of PAGES
145
20. SECURITY CLASS ,r/iU ptitei
UNCLASSIFIED
22. PRICE
EPA Form 2230-1 (H«*. 4-77) pxiviouj edition is 00jolete
-------
NOTICE
The information in this document has been funded by the United States Environmental
Protection Agency under Contract 68-C9-0038 to Acurex Corporation. It has been subjected to
the Agency's peer and administrative review, and it has been approved for publication as an EPA
document. Mention of trade names or commercial products does not constitute endorsement
or recommendation for use.
ii
-------
Today's rapidly developing and changing technologies and industrial products and
practices frequently carry with them the increased generation of materials that, if improperly
dealt with, can threaten both public health and the environment. The U.S. Environmental
Protection Agency is charged by Congress with protecting the Nation's land, air, and water
resources. Under a mandate of national environmental laws, the agency strives to formulate and
implement actions leading to a compatible balance between human activities and the ability of
natural systems to support and nurture life. These laws direct the EPA to perform research to
define our environmental problems, measure the impacts, and search for solutions.
The Risk Reduction Engineering Laboratory is responsible for planning, implementing,
and managing research, development, and demonstration programs to provide an authoritative,
defensible, engineering basis in support of the policies, programs, and regulations of the EPA
with respect to drinking water, wastewater, pesticides, toxic substances, solid and hazardous
wastes, and Superfund-related activities. This publication is one of the products of that research
and provides a vital communication link between the researcher and the user community.
This report describes a series of tests conducted at the EPA's Incineration Research
Facility to evaluate the fate of trace metals fed to a rotary kiln incinerator. The testing is an
extension of the work described in the EPA report, "The Fate of Trace Metals in a Rotary Kiln
Incinerator with a Venturi/Packed Column Scrubber," and is similar to that work in all respects
with the exception of the air pollution control system utilked. For further information, please
contact the Waste Minimization, Destruction, and Disposal Research Division of the Risk
Reduction Engineering Laboratory.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
-------
ABSTRACT
A 3-week series of nine pilot-scale incineration tests was performed at the U.S.
Environmental Protection Agency's (EPA's) Incineration Research Facility (IRF) in Jefferson,
Arkansas, to evaluate the fate of trace metals fed to a rotary kiln incinerator equipped with a
single-stage ionizing wet scrubber for control of particulates and acid gas. Test variables were
kiln temperature, ranging from 816° to 927° C (1500° to 1700° F); afterburner temperature,
ranging from 982° to 1204°C (1800° to 2200°F); and feed chlorine content, ranging from 0 to
8 percent.
The test results indicated that cadmium and bismuth were relatively volatile, with an
average of less than 40 percent discharged with the kiln ash. Arsenic, barium, chromium, copper,
lead, magnesium and strontium were relatively nonvolatile, with an average of greater than
80 percent discharged with the kiln ash. Observed relative metal volatilities generally agreed
with the volatilities predicted based on vapor pressure/temperature relationships, with the
exception of arsenic, which was much less volatile than predicted. Cadmium, bismuth, and lead
were more volatile at higher kiln temperature; the discharge distributions of the remaining
metals were not significantly affected by kiln temperature.
Enrichment of metals in the fine-particulate fraction was observed at the afterburner
exit, with an average of roughly 50 percent of flue-gas particulate metal in the less-than-10-pm
size range. The distributions of the more-volatile metals were shifted to fine particulate more
so than those of the less-volatile metals. Both increased kiln temperature and the addition of
chlorine to the synthetic waste feed caused a shift of metals to fine particulate.
Apparent scrubber collection efficiencies for the metals averaged 22 to 71 percent, and
were generally higher for the less-volatile metals. The overall average metal collection efficiency
was 43 percent. It should be noted that industrial applications of ionizing wet scrubbers are
typically in multiple stages and, as such, would be expected to collect metals more efficiently than
the single-stage scrubber at the IRF.
This report was submitted in fulfillment of Contract 68-C9-0038 by Acurex Corporation
under the sponsorship of the U.S. Environmental Protection Agency. This report covers work
conducted during July and August 1989.
iv
-------
TABLE OF CONTENTS
FOREWORD iii
ABSTRACT iv
FIGURES vii
TABLES x
1 INTRODUCTION 1
2 FACILITY DESCRIPTION AND TEST CONDITIONS 3
2.1 Rotary Kiln Incineration System Description 3
2.1.1 Incinerator Characteristics 3
2.1.2 Air Pollution Control System 3
2.2 Synthetic Test Mixture 6
2.3 Test Conditions 7
3 SAMPLING AND ANALYSIS PROCEDURES 15
3.1 Sampling Procedures 15
3.2 Analysis Procedures 23
4 TEST RESULTS 27
4.1 Synthetic Waste Feed Composition 27
4.2 Continuous Emission Monitor Data 30
4.3 Flue Gas Flowrates and Particulate and HC1 Emissions 34
4.4 Trace Metals Discharge Data 34
4.4.1 Average Trace Metal Discharge Distributions 39
4.4.2 Effects of Incinerator Operating Conditions on Metal
Distributions 46
4.4.3 Flue Gas Metal Phase Distributions 74
4.4.4 Metal Distributions Among Flue Gas Particulate by
Particle Size in the Afterburner Exit 74
4.4.5 Apparent Scrubber Collection Efficiencies 92
v
-------
TABLE OF CONTENTS (concluded)
4.5 Volatile organic constituent concentrations and POHC destruction and
removal efficiencies 103
5 CONCLUSIONS 110
6 QUALITY ASSURANCE 113
6.1 Volatile Organic Analyses 114
6.1.1 Volatile Organic Analysis of Clay/Organic Liquid Feed, Kiln
Ash, and Scrubber Blowdown Samples 119
6.1.2 Volatile Organic Compounds in Flue Gas 119
6.2 Metals Analysis 122
REFERENCES 132
vi
-------
FIGURES
Number Eag£
1 Schematic of the IRF rotary kiln incineration system 4
2 Actual versus target operating conditions 14
3 Sampling protocol 16
4 Generalized CEM sample gas flow schematic 18
5 Distribution of metals in the discharge streams expressed as a
percent of the metal fed 43
6 Normalized distribution of metals in the discharge streams 44
7 Summary of mass balance closures around the kiln ash, scrubber
exit flue gas and scrubber liquor 46
8 Arsenic discharge distributions expressed as a percent of the metal fed ... . 54
9 Normalized arsenic discharge distributions . 55
10 Barium discharge distributions expressed as a percent of the metal fed .... 56
11 Normalized barium discharge distributions 57
12 Bismuth discharge distributions expressed as a percent of the metal fed ... 58
13 Normalized bismuth discharge distributions 59
14 Cadmium discharge distributions expressed as a percent of the
metal fed 60
15 Normalized cadmium discharge distributions 61
16 Chromium discharge distributions expressed as a percent of the
metal fed 62
vii
-------
FIGURES (continued)
Number Page
17 Normalized chromium discharge distributions 63
18 Copper discharge distributions expressed as a percent of the metal fed .... 64
19 Normalized copper discharge distributions 65
20 Lead discharge distributions expressed as a percent of the metal fed 66
21 Normalized lead discharge distributions 67
22 Magnesium discharge distributions expressed as a percent of the
metal fed 68
23 Normalized magnesium discharge distributions 69
24 Strontium discharge distributions expressed as a percent of the
metal fed 70
25 Normalized strontium discharge distributions 71
26 Average of metal distributions in the afterburner exit flue gas
particle size fractions 84
27 Effect of kiln temperature on the distribution of metals in the
afterburner exit flue gas particle size fractions 86
28 Effect of afterburner temperature on the distribution of metals in
afterburner exit flue gas particle size fractions 87
29 Effect of feed chlorine content on the distribution of metals in the
afterburner exit flue gas particle size fractions 88
30 Size distribution of the flue gas particulate in the afterburner exit 90
31 Arsenic distributions in the afterburner exit flue gas particulate 93
32 Barium distributions in the afterburner exit flue gas particulate 94
33 Bismuth distributions in the afterburner exit flue gas particulate 95
34 Cadmium distributions in the afterburner exit flue gas particulate 96
viii
-------
FIGURES (concluded)
Number Page
35 Chromium distributions in the afterburner exit flue gas particulate 97
36 Copper distributions in the afterburner exit flue gas particulate 98
37 Lead distributions in the afterburner exit flue gas particulate 99
38 Magnesium distributions in the afterburner exit flue gas particulate 100
39 Strontium distributions in the afterburner exit flue gas particulate 101
40 Apparent ionizing wet scrubber efficiencies for metals 102
41 Apparent scrubber collection efficiencies for metals showing
associated variations with changes in kiln exit temperature and
waste feed chlorine content 105
ix
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
TABLES
Page
Design characteristics of the IRF rotary kiln incineration system 5
Target trace metal integrated feed concentrations 8
Target test conditions 8
Kiln operating conditions 9
Afterburner operating conditions 10
Air pollution control system operating conditions 11
Actual versus target operating conditions 13
Continuous emission monitors 17
Sampling and analysis matrix 20
Multiple metals sampling train impinger reagents 22
Stack gas Method sampling train impinger reagents 23
Summary of test samples 24
Volatile organic compounds routinely analyzed by GC/FTD at the IRF .... 26
POHC concentrations in clay/organic Liquid feed samples 28
Aqueous spike solution metals concentrations 29
Clay/organic liquid mixture metal concentrations 31
Integrated feed metal concentrations 32
CEM data 33
x
-------
TABLES (continued)
Number Page
19 Flue gas flowrates 35
20 Flue gas particulate data 36
21 HC1 emission data 36
22 Synthetic waste feedrates and kiln ash discharge rates 38
23 Summary of metal discharge distributions in the kiln ash and
afterburner exit flue gas 40
24 Summary of metal discharge distributions in the kiln ash, scrubber
exit flue gas, and scrubber liquor 41
25 Summary of metal mass balance closure 45
26 Metal discharge distributions and mass balance closure 47
27 Normalized metal discharge distributions 50
28 Summary of Due gas metal particulate and vapor/dissolved phase
distributions 75
29 Phase distributions of flue gas metals in the afterburner and
scrubber exit flue gas 76
30 Metal distributions in the afterburner exit flue gas particulate by
particle size 80
31 Summary of apparent scrubber efficiency ranges and averages 102
32 Apparent scrubber collection efficiencies 104
33 Flue gas POHC concentrations 107
34 POHC DREs 108
35 Flue gas volatile PIC concentrations 109
36 Precision, accuracy, and completeness objectives 115
37 Volatile organic sample hold times 116
xi
-------
TABLES (concluded)
Number Page
38 Volatile organic analyses practicable quantitation limits achieved 118
39 Clay/organic liquid feed sample POHC analysis results 120
40 Volatile organic constituent recovery from clay/organic liquid feed
matrix spike samples 121
41 Volatile organic constituent recovery from kiln ash and scrubber
liquor matrix spike samples 122
42 Octane surrogate recovery from VOST traps 123
43 4-Bromofluorobenzene surrogate recover)' from VOST traps 124
44 Volatile organic constituent recovery from VOST matrix spike samples . . . 125
45 Metals analyses practicable quantitation limits achieved 127
46 Recovery of metals from matrix spike samples 128
47 Aqueous metals spike solution analysis accuracy and precision 129
48 Clay absorbent metals analysis precision 131
xii
-------
SECTION 1
INTRODUCTION
The hazardous waste incinerator performance standards, promulgated by the
Environmental Protection Agency (EPA) in January 1981 under the Resource Conservation and
Recovery Act (RCRA), established limits on incinerator emissions of particulate, HCl, and
hazardous organic constituents. The standards established the latter limits by requiring that at
least 99.99 percent destruction and removal efficiency (DRE) be achieved for the principal
organic hazardous constituents (POHCs) in wastes fed to an incinerator. Emissions of hazardous
constituent trace metals were only regulated indirectly, via the particulate standard. However,
risk assessments to date have suggested that these metal emissions may represent the largest
component of the total risk to human health and the environment from properly operated
incinerators.
Despite its importance, the data base on trace metal emissions from incinerators is
currently sparse. Data on the effects of waste composition and incinerator operation on trace
metal emissions are particularly lacking. In response to these data needs, an extensive series of
tests was conducted at EPA's Incineration Research Facility (IRF) in Jefferson, Arkansas, with
support from the Office of Solid Waste (OSW), to investigate the fate of trace metals fed to a
rotary kiln incinerator equipped with a single-stare ionizing wet scrubber (hereafter referred to
as the scrubber) for particulate and acid gas control.
The purpose of this test program was to extend the data base on the trace metal
composition of stack emissions and residual discharges to support continuing regulatory
development by OSW. This program was a continuation of previous IRF a test program that
employed a venturi scrubber/packed column scrubber as the primary air pollution control system
(APCS).1
A primary objective of this test program was to investigate the fate of five hazardous
constituent trace metals fed to a rotary kiln incinerator in a synthetic solid waste matrix. The
hazardous trace metals investigated were arsenic, barium, cadmium, chromium, and lead. Of
interest were the partitioning, particle size distribution, flue gas phase distribution, and scrubber
efficiency for each of the metals, as a function of incinerator operating temperatures and feed
chlorine content.
In another OSW-sponsored effort within EPA's Risk Reduction Engineering Laboratory
(RREL), a mathematical model has been developed under an independent EPA contract to aid
in predicting the relative distributions of trace metals in emissions and discharges from
incinerators. Thus, a second objective of this test program was to supply data to evaluate the
predictive capabilities of this model, and to perhaps guide further model refinement. To support
1
-------
this objective, an additional four nonhazardous constituent trace metals were included in the test
feed material—bismuth, copper, magnesium, and strontium.
The test program to address the above objectives was comprised of a series of nine
parametric tests. This report summarizes the results of the test program. Section 2 describes
the IRF rotary kiln incineration system in which the tests were performed, outlines the test waste
feed characteristics, and summarizes the incinerator operating conditions for each of the tests.
Section 3 discusses the emissions and discharge stream sampling and analysis protocols.
Section 4 presents test program results. Section 5 summarizes test conclusions. Section 6
discusses the quality assurance and quality control (QA/QC) aspects of the test program. All
of the Appendices referenced in this volume are contained in Volume II of the report.
2
-------
SECTION 2
FACILITY DESCRIPTION AND TEST CONDITIONS
The IRF's rotary kiln incinerator system (RKS) was used in this test program. A
description of the system is presented in Section 2.1, followed by a description of the synthetic
waste mixture in Section 2.2. The test matrix and operating conditions are described in
Section 2.3.
2.1 ROTARY KILN INCINERATION SYSTEM DESCRIPTION
As indicated in Figure 1 and Table 1, the RKS consists of a primary combustion
chamber, a transition section, and a fired afterburner chamber. The primary APCS for these
tests consisted of a quench section followed by a single-stage ionizing wet scrubber manufactured
by Air Plastics, Inc. Downstream of the primary APCS a backup APCS, consisting of a demister,
carbon-bed adsorber and a high-efficiency particulate (HEPA) filter, is in place. The backup
system is designed to ensure that organic compound and particulate emissions to the atmosphere
are negligible during testing at less-than-optimal conditions. The main components of the RKS
and its APCS are discussed in more detail in the following subsections.
2.1.1 Incinerator Characteristics
The rotary kiln combustion chamber has an inside diameter of 1.0 m (39 in) and is 2.5 m
(8.2 ft) long. The chamber is lined with 19 cm (7.4 in) of refractory encased in a 9.5-mm
(0.375-in) thick steel shell. The chamber volume, including the transition sections, is 1.9 m3
(67.3 ft3). Four steel rollers support the kiln barrel. A variable-speed DC motor coupled with
a reducing gear transmission turns the rotary kiln. Rotation speeds can be varied from 0.2 to
1.5 rpm.
The afterburner chamber has a 0.91-m (3-ft) inside diameter and is 2.74 m (9 ft) long.
The afterburner chamber wall is constructed of a 15-cm (6-in) layer of refractory encased in a
6.3-cm (0.25-in) thick carbon steel shell. The volume of the afterburner chamber is 1.80 m3
(63.6 ft3).
2.1.2 Air Pollution Control System
As previously mentioned, combustion gas exiting the afterburner flows through a primary
APCS consisting of a quench section and, a scrubber system. The quench section reduces the
temperature of the combustion gas to approximately 81°C (178°F). The cooled flue gas then
enters the scrubber. An ionizer located at the entrance can be operated at up to 10,000 volts.
3
-------
SiNQLE<6TAQE ION2WQ
wetscrubber
SCRUBBER
LIQ'JOR
RECIRCULATION
ATMOSPHERE
AFTERBURNER
ID PAN
NATURAL
QAS.
LIQUID ~
PEED
CARBON BID HERA
ADSORBER FILTER
AFTERBURNER
SCRUBBER
LOUOR
RECIRCULATION
MODULAR PRIMARY AIR
POLLUTION CONTROL
DEVICES
REDUNDANT AIR
POLLUTION CONTROL
SYSTEM
ROTARY KILN
INCINERATOR
Figure 1. Schematic
or the
IRF rotary kiln incineration system.
4
-------
TABLE 1. DESIGN CHARACTERISTICS OF THE IRF ROTARY KILN INCINERATION
SYSTEM
Characteristics of the Kiln Main Chamber
Length
2.49 m (8 ft-2 in)
Diameter, outside
1.37 m (4 fl-6 in)
Diameter, inside
Nominal 1.00 m (3 ft-3_5 in)
Chamber volume
1.90 m' (673 ft3)
Construction
0.95 cm (0.375 in) thick cold-rolled steel
Refractory
18.7 cm (7.375 in) thick high alumina castable refractory, variable depth to produce
a frustroconical effect for moving solids
Rotation
Clockwise or counterclockwise, 0.2 to 15 rpm
Solids retention time
1 hr (at 0.2 rpm)
Burner
North American burner rated at 800 kW (2.7 MMBtu/hr) with liquid feed
capability
Primary fuel
Natural gas
Feed system:
Liquids
Positive displacement pump via water-cooled lance
Sludges
Moyno pump via front face, water-cooled lance
Solids
Metered twin-auger screw feeder or fiberpack ram feeder
Temperature (max)
1010"C (1850T)
Characteristics of the Afterburner Chamber
Length
3.05 m (10 ft)
Diameter, outside
1.22 m (4 ft)
Diameter, inside
0.91 m (3 ft)
Chamber volume
1.80 m' (63.6 ft5)
Construction
0.63 on (0.25 in) thick cold-rolled steel
Refractory
15.2 cm (6 in) thick high alumina castable refractory
Gas residence time
0.8 to 2.5 s depending on temperature and excess air
Burner
North American Burner rated at 800 kW (2.7 MMBtu/hr) with liquid feed
capability
Primary fuel
Natural gas
Temperature (max)
1200°C (2200T)
Characteristics of the Single-Stage Ionizing Wet Scrubber APCS
System capacity,
85 m'/min (3000 acfm) at 78"C (172°F) and 101 kPa (14.7 psia)
inlet gas flow
Pressure drop
15 kPa (6 in W.C.)
Liquid flow
230 L/min (60 gpm) at 345 kPa (50 psig)
pH control
Feedback control by NaOH solution addition
Characteristics of the Venturi/Packed-Column Scrubber APCS (Not in service during this test program)
System capacity,
107 m'/min (3773 acfm) at 1200"C (2200°F) and 101 kPa (14.7 psia)
inlet gas flow
Pressure Drop
Venturi scrubber
7.5 kPa (30 in W.C.)
Packed column
1.0 kPa (4 in W.C.)
Liquid flow
Venturi scrubber
77.2 L/mln (20.4 gpm) at 60 kPa (10 psig)
Packed column .
116 L/min (30.6 gpm) at 69 kPa (10 psig)
pH control
Feedback control by NaOH solution addition
5
-------
The scrubber volume is 3.4 m3 (120 ft3). A 15-cm (6-in) deep liquor reservoir is maintained at
the bottom of the scrubber. The rest of the scrubber is packed with polypropylene ballast
saddles. Scrubber liquor is continuously sprayed along the top of the saddles with a maximum
liquid flowrate of 230 L/min (60 gpm). Acid gases and particulate are removed as the flue gas
passes through the ionizer, then through the scrubber liquor and packing. The liquor is a dilute
NaOH solution and is monitored continuously by a pH sensor. An integral pH controller
automatically meters the amount of NaOH needed to maintain the required pH for proper HC1
removal.
A demister located downstream of the scrubber removes most of the suspended liquid
droplets. In a typical commercial incinerator system, the flue gas would be vented to the
atmosphere at this point. However, a backup APCS is in place at the IRF. The flue gas passes
through a bed of activated carbon designed to adsorb remaining vapor phase organic compounds.
Typically, the carbon bed operates at 77°C (170°F). Because the flue gas is saturated with
moisture and is cooled as it flows through the ducts, condensate is continuously formed. The
condensate accumulates in the carbon bed and drains through a bottom tap, then is pumped to
the blowdown storage tanks.
A set of HEPA filters designed to remove remaining suspended particulate from the flue
gas is located downstream of the carbon adsorption bed. An induced-draft fan draws and vents
the effluent gas to the atmosphere.
22 SYNTHETIC TEST MIXTURE
The synthetic waste fired throughout the test program included a mixture of organic liquids
added to a clay absorbent material. Trace metals were incorporated by spiking an aqueous
mixture of the metals of interest onto the organic liquid-containing solid material. The synthetic
waste was fed to the rotary kiln via a twin-auger screw feeder at a nominal rate of 63 kg/hr
(140 Ib/hr).
The organic liquid base that supplied the heat content and POHC concentrations to the
synthetic waste consisted of toluene, with varying amounts of tetrachloroethylene and
chlorobenzene added to provide a range of synthetic waste chlorine contents. Waste chlorine
content was varied from zero (no chlorinated POHCs added) to nominally 8 percent. The waste
was prepared by adding the premixed organic liquid to the clay absorbent material in a portable
cement mixer. The mixer opening was covered and the mixer operated for 15 min, yielding a
homogeneous mixture containing about 30 percent (by weight) organic liquid. Approximately
65 lb of mix was prepared per batch. After mixing, the test mixture was poured into 55-gal
drums, each holding approximately 140 kg (300 lb).
All trace metals of interest, except magnesium and chromium, were introduced into the
kiln by metering an aqueous spike solution of the metals into the clay/organic liquid mixture at
the screw feeder, just prior to feed introduction. The aqueous spike solution was prepared in
glass containers, and contained trace amounts of four of the five test program hazardous
constituent trace metals (arsenic, barium, cadmium and lead) and greater amounts of three of
the four test program nonhazardous constituent trace metals (bismuth, copper, and strontium).
Previous analysis of the test clay absorbent material showed it to contain an average of 53 ppm
6
-------
of chromium and 2.2 percent magnesium. Thus, chromium and magnesium were not included
in the aqueous spike solution, but were instead inherent in the feed from the clay absorbent
matrix. The spike solution was metered at a rate that produced the final nominal synthetic waste
feed concentrations noted in Table 2. A gear pump was used to inject the trace metal aqueous
spike solution at a nominal flowrate of 2 L/hr. The clay/organic mixture and spike solution
feedrates were controlled independently to provide the desired waste concentrations.
23 TEST CONDITIONS
The test variabiles were the chlorine content of synthetic waste feed, kiln temperature, and
afterburner temperature. Seven specific combinations of these variables were selected as test
points based on a factorial experimental design for three variables varied over three levels, as
shown in Table 3. One condition was tested in triplicate to give an indication of measurement
precision. All tests were designed to have the same nominal kiln exit flue gas 02 (11.5 percent),
afterburner exit flue gas 02 (8.0 percent), and synthetic waste feedrate (63 kg/hr (140 lb/hr) of
which 18 kg/hr (40 lb/hr) was the organic liquid matrix). The test conditions reflect typical
industrial hazardous waste incinerator operation.
The kiln and afterburner operating conditions achieved for each test are summarized in
Tables 4 and 5, respectively. Table 6 provides a similar summary of the APCS operating
conditions for each test. Control room records of the operating parameters recorded at 15-min
intervals are given in Appendix A. Appendix A also contains plots of the exit flue gas
temperature and 02 concentration data for the kiln and afterburner, recorded at 40-s intervals
on a PC data acquisition system. Flue gas sampling times, major events, and the cumulative
amounts of the clay/organic liquid mixture and the trace metal spike solution fed are also
indicated on these plots in Appendix A.
The ranges and averages of the temperature, continuous emission monitor (CEM) and
scrubber pH data presented in Tables 4, 5 and 6 were developed using the computer- recorded
data during periods of flue gas sampling. The values given for the remaining parameters were
taken from the control room data log book during periods of flue gas sampling.
Feed interruptions of the clay/organic liquid and metal spike feed matrix occurred during
Tests 2, 4, 7 and 9 while flue gas sampling was in progress. These events were generally caused
by the auger screw feeder jamming, although a brief total system shutdown did occur during
Test 7. To minimize the impact of feed interruptions on the flue gas samples, flue gas sampling
was temporarily stopped until feeding resumed and steady-state operating conditions were
reestablished. Two exceptions to this occurred at the end of Tests 4 and 9, during which the flue
gas sampling trains continued to operate through feed interruptions. As shown in the operation
plots in Appendix A, these feed interruptions caused brief excursions of the primary parametric
variables. However, since these excursions were brief in comparison to the overall flue gas
sampling time, their impact on the data was minima].
Since variations in the feedrate of the clay/organic mixture occurred during several tests,
the logbook data for the clay /organic liquid and metal spike solution feeds have been tabulated
in Appendix A to facilitate the analysis of data potentially affected by feed variations (e.g.,
metals mass balance and POHC DRE calculations). The information in Table 4 associated with
-------
TABLE 2. TARGET TRACE METAL INTEGRATED FEED CONCENTRATIONS
1
Synthetic solid hazardous
waste concentration
Metal
(ppm) |
Hazardous constituent trace metals
Arsenic
Barium
Cadmium
Chromium
Lead
Other trace metals
Bismuth
Copper
Magnesium
Strontium
50
400
10
40
50
400
400
16,000
400
TABLE 3. TARGET TEST CONDITIONS
Feed mixture CI
Kiln exit
Afterburner exit
content
temperature
temperature
Test
<%)
°C (°F)
°C (°F)
1
0
871 (1600)
1093 (2000)
2
4
816 (1500)
1093 (2000)
3
4
927 (1700)
1093 (2000)
4
4
871 (1600)
1093 (2000)
5
4
871 (1600)
1204 (2200)
6
4
871 (1600)
982 (1800)
7*
4
871 (1600)
1093 (2000)
8*
4
871 (1600)
1093 (2000)
9
8
871 (1600)
1093 (2000)
O
r-
r-
o
o
If)
"Test points 7 and 8 are replicates of test point 4
-------
TABLE 4. KILN OPERATING CONDITIONS
Parameter
Terf I
(8/17/89)
Test 2
(8/2/89)
Test 3
(8/4/89)
Ttsl 4
(8/1/89)
TestS
(8/16/89)
Ifcrt 6
(8/15/89)
Test 7
(8/9/89)
Test 8
(8/11/89)
Test 9
(7/28/89)
Natural gas feed rule
average, scm/hr 10.7 10.7
(scfh) (379) (379)
kW III III
(kBlu/hr) (379) (379)
Combustion air flowraie
average, scmflir 173 93
(scfh) (6,120) (3.280)
Total air flowraie*
average, sctnAir 604 506
(«cfli) (21340) (17,870)
Pressure
average. Pa -10.0 -12.4
(in WC) (-0.04) (-0.05)
Rxil lemperatire
range, °C
(°P)
average, °C
CP) (1.652) (I.5CT7)
Exit O,
range, % II I to 14 9 7.9 to 13.4
average, % 13.6 (12-0)* 12.3
Clay/organic mixture
average feedrate, kg/hr 64.6 54.0
(Ib/hr) (142) (119)
Clay/organic mixture
healing value, kJ/kg 9,790 8,260
(Btu/lb) (4^20) (3.560)
Clay/organic mixture
heal input, kW 176 124
(kBtu/hr) (601) (424)
Total heM input, kW 287 235
(kBtu/hr) (980) (803)
Calculated avenge flue gas
residence lime, • 2.5 3.6
883 lo 917 791 to 840
(1,622 to 1.682) (1,455 to 1,544)
900 819
21.3
(751)
220
(751)
240
(8.470)
749
(26.470)
-14.9
(-0.06)
907 to 951
(1,665 to 1,744)
929
(1.704)
10.1 to 13.3
11.9
62.0
(137)
8,520
(3.670)
147
(502)
367
(1.253)
2.2
17.2
(606)
178
(606)
160
(5.660)
614
(21.690)
-12.4
(-0-05)
12.5
(441)
129
(441)
268
(9.480)
585
(20.640)
-8.7
(-0.035)
13 6
(4*2)
141
(482)
152
(5.370)
603
(21,280)
-124
(-005)
11.9
(420)
123
(420)
156
(5.520)
635
(22.420)
-12.4
(-0.05)
802 to 961 865 to 895 868 to 906 862 to 904
(1.476 to 1.762) (1.589 to 1,643) (1,595 to 1.663) (1,584 to 1,659)
877 883 887 881
(1,610)
9.0 to 16 6
11.9
52.4
(116)
8,450
(3,640)
123
(421)
301
(1.025)
2.8
(1.625)
9.0 to 15 5
12.1
64.1
(Ml)
8,330
(3,590)
149
(507)
278
(948)
29
(1.629)
8 5 to 13 A
11.9
64.0
(141)
8.520
(3.670)
152
(518)
293
(1,000)
2.8
(1.618)
110 to 13.2
12.6
65.4
(144)
8,870
(3.820)
161
(551)
285
(971)
2.7
13.0
(460)
135
(460)
267
(9,440)
640
(22,620)
-14.9
(0.06)
858 to 893
(1,557 to 1.640)
879
(1.614)
118 to 13 6
12.6
62.7
(138)
8,750
(3.770)
153
(521)
288
(981)
2.7
O
19 5 ul
(689)
202
(689)
83
(2.930)
699
(24,690)
-11.2
(-0.W5)
R00 to 946
(1,471 lo 1.734)
881
(1.617)
14.0 lo 17.2
14.8 (12.9)*
56.7
(125)
6,750
(2.910)
107
(364)
309
(1.053)
2.4
'Including inleakage.
bKiln exit CEM sampling system leak developed during the lest. O, concentrations determined using a Bacharach Fryrite 02 analyzer are also given.
-------
TABLE 5. AFTERBURNER OPERATING CONDITIONS
Parameter
Test 1
(8/17/89)
Test 2
(8/2/89)
Tert 3
(8/4/89)
Test 4
(8/1/89)
Test S
(R/16/89)
Test 6
(8/15/89)
Test 7
(8/9/89)
Test 8
(8/11/89)
Tent 9
(7/28/89)
Natural gas feedrate
average, scm/hr
(acft)
kW
(kBtu/hr)
Combustion air Howrate
avenge, scmAir
(sefli)
E*it lempcmure
range, "C
(°F)
average, "C
(°F)
En it Oj
range, %
average, %
Exit CO,
range, *
avenge, *
Calculated avenge flue gas
residence lime, I
66.2
(2.138)
6*5
(2.338)
441
(15,560)
1,076 to 1,097
(1,969 to 2,006)
1,088
(1,990)
7.6 to 9.0
8.3
6.9 to 8.1
7.5
I.I
55.9
(1.975)
579
(1.975)
465
(16,430)
1,083 to 1,109
(1,981 to 2,029)
1,095
(2.002)
64 to 8.7
7.7
7.1 to 9.0
7.7
1.3
586
(2.071)
607
(2,071)
455
(16,050)
1,073 to 1,108
(1,964 to 2,026)
1,092
(1.998)
5.8 10 8 3
7.3
72 to 8.8
7.9
1.2
55.4
(1.956)
573
(1.956)
457
(16,140)
78.6
(2,776)
814
(2,776)
511
(18,050)
43.5
(1.537)
450
(1.537)
408
(14,390)
53.8
(1.900)
557
(1,900)
486
(17,170)
55.4
(1.956)
573
(1.956)
486
(17.170)
l.083lo 1,124 1,145 to 1,176 1,011 to 1,027 1,074 to 1,114 l,078lol,106
(1,982 to 2,055) (2,093 to 2,148) (1.851 to 1,880) (1,965 to 2.037) (1.973 to 2.023)
1.096
(2.006)
4 5 to 11.8
7.3
4.7 to 10.0
7.9
12
1.163
(2.125)
52 to 8.8
7.9
6.9 to 8.7
7 5
1.1
1.017
(1.863)
5 A to 9A
8.6
6.8 to 9J
73
13
1.103
a,oi 8)
60 to 82
7.6
7.5 to 8.8
7.9
1.2
1.097
(2.007)
7 J to 8 6
8.1
7 .1 to 8 I
7.6
1.5
O
45.7 u
(1.613)
473
(1.613)
334
(11.780)
1,073 to 1,104
(1,964 to 2,019)
1,087
(1.988)
6.7 to 10 J
8 1
4 4 to 81
7.3
1.4
-------
TABLE 6. AIR POLLUTION CONTROL SYSTEM OPERATING CONDITIONS
Test I Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9
Parameter (8/17/89) (8/2/89) (8/4/89) (8/1/89) (8/16/89) (8/15/89) (8/9/89) (8/11/89) (7/28/89)
Quench chamber scrubber liquor
flowratc
.average, L/min 68 68 76 68 68 68 68 68 68
(gpm) (18) (18) (20) (18) (18) (18) (18) (18) (18)
Ionizing wet scrubber liquor
flowrate
average, L/min 170 159 170 121 170 170 170 170 193
(gpm) (45) (42) (45) (32) (45) (45) (45) (45) (51)
Scrubber liquor pH
range 7.4 to 7.5 5.8 to 9.4 8.6 to 9.2 8.4 to 9.7 6.8 to 7.7 5.7 to 8.2 7.2 to 8.7 5.1 to 7.5 7.8 to 8.6
average (7.45) (7.9) (8.8) (9.0) (7.3) (7.0) (8.1) (6.6) (8.1)
Ionizer voltage, volts 9000 3000 0 3000 9400 9000 9000 9000 3000
Scrubber blowdown flowrate
average, L/min 1.5 1.9 2.6 1.9 1.5 1.9 1.9 2.3 1.9
(gpm) (0.4) (0.5) (0.7) (0.5) (0.4) (0.5) (0.5) (0.6) (0.5)
Scrubber makeup flowrate
average, L/min 25 23 23 21 22 23 38 24 22
(gpm) (6.6) (6.2) (6.2) (5.5) (5.7) (6.1) (10.0) (6.4) (5.7)
Scrubber liquor temperature
average, °C 73 74 75 74 72 72 70 70 77
(°F) (163) (165) (167) (165) (162) (162) (158) (158) (170)
Scrubber inlet gas temperature
average, °C 79 81 80 81 81 78 80 80 77
(°F) (174) (178) (176) (178) (178) (172) (176) (176) (170)
Scrubber exit gas temperature
average, °C 78 79 78 79 79 77 78 78 78
(°F) (172) (174) (172) (174) (174) (170) (172) (172) (172)
-------
the synthetic feed mixture are test averages, useful for determining gross average feedrates.
However, the analysis of data potentially affected by variations in feedrate incorporate the
Appendix A feedrate information for the corresponding time frames of interest.
Table 7 summarizes the actual incinerator operating condition (temperatures and Due gas
Oj levels) ranges and averages for each test during flue gas sampling, and compares these
with the respective target conditions. Figure 2 presents a graphical summary of the incinerator
temperature data from Table 7.
Table 7 and Figure 2 illustrate that the average kiln exit temperatures were within 16° C
(29s F) of target temperatures for all tests except Test 1. For Test 1, the average kiln exit
temperature was 29°C (52°F) high. Table 7 and Figure 2 also indicate that the average
afterburner exit temperatures were within 10'C (18°F) of target temperatures for all tests
except Tests 5 and 6. For Test 5, the average afterburner temperature was 41°C (75°F) below
the target of 1204° C (2200°F), while for Test 6 the average afterburner temperature was 35° C
(63°F) above the target of 982°C (1800°F). The relatively wide variations in the range of kiln
temperatures for Tests 4 and 9 are associated with short temperature fluctuations following feed
interruptions near the end of the tests.
The data in Table 7 show that the average kiln exit flue gas 02 was generally within about
1 percent of the target of 11.5 percent for all tests except Tests 1 and 9. Post-test calibrations
of the continuous emission monitor (CEM) sampling system at the kiln exit indicated that a
sampling system leak had developed during these tests. However, 02 readings obtained during
these tests with an Orsat analyzer were within 1.5 percent of the target. Finally, the data in
Table 7 show that the average afterburner exit flue gas 02 concentrations were within 1 percent
of the target of 8.0 percent for all of the tests.
12
-------
>
TABLE 7. ACTUAL VERSUS TARGET OPERATING CONDITIONS
Kiln exit Afterburner exit
Temperature °C (°F) Temperature "C (°F) _
_ Flue gas 02 (%) Flue gas O, (%) °
Actual (Target: 11.5%) Actual (Target: 8.0%) w
Test
date
Target
Minimum
Maximum
Average
Range
Average
Target
Minimum
Maximum
Average
Range
Average
1 '
8/17/89
871 (1600)
883 (1622)
971 (1682)
900(1652)
11.1 to 14.9
13.6"
1093 (2000)
1076 (1969)
1097 (2006)
1088 (1990)
7.6 to 9.0
8.3
2
8/2/89
816 (1500)
791 (1455)
840(1544)
819 (1507)
7.9 to 13.4
12.3
1093 (2000)
1083 (1981)
1109(2029)
1095 (2002)
6.4 to 8.7
7.7
3
8/4/89
927 (1700)
907 (1665)
951 (1744)
929(1704)
10.1 to 13.3
11.9
1093 (2000)
1073 (1964)
1108 (2026)
1092 (1998)
5.8 to 8.3
7.3
4
8/1/89
871 (1600)
802 (1476)
961 (1762)
877 (1610)
9.0 to 16.6
11.9
1093 (2000)
1083 (1982)
1124 (2055)
1096 (2006)
4.5 to 11.8
7.3
5
8/16/89
871 (1600)
865 (1589)
895 (1643)
885 (1625)
9.0 to 15.5
12.1
1204 (2200)
1145 (2093)
1176 (2148)
1163 (2125)
5.2 to 8.8
7.9
6
8/15/89
871 (1600)
868 (1595)
906(1663)
887 (1629)
8.5 to 13.4
11.9
982 (1800)
1011 (1851)
1027 (1880)
1017 (1863)
5.4 to 9.4
8.6
7
8/9/89
871 (1600)
862 (1584)
904 (1659)
881 (1618)
11.0 to 13.2
12.6
1093 (2000)
1074 (1965)
1114 (2037)
1103 (2018)
6.0 to 8.2
7.6
8
8/11/89
871 (1600)
858 (1577)
906(1663)
879 (1615)
11.8 lo 13.6
12.6
1093 (2000)
1078 (1973)
1118(2044)
1098 (2008)
7 J to 8.7
8.1
9
7/28/89
871 (1600)
800 (1471)
946(1734)
881 (1617)
14 to 17.2
14.8*
1093 (2000)
1074 (1984)
1104(2019)
1087 (1988)
6.7 to 10.5
8.1
u>
'Posttest calibrations indicate a sampling line leak developed. Orsat analyzer measurement suggest actual kiln exit flue gas 02 levels were 12.0 percent for
Test 1 and 12.9 percent for Test 9.
-------
© TARGET CONDITION
i-Ih ACHIEVED CONDITION.
1 INTERSECTION IS
MEAN CONDITION;
BARS DENOTE RANGE
OVER TEST.
©
©
1 '""S"
f
—1
=, gr
1
1 VII
1
\
<
" ' 1
¦A
k> l
h?
1
©
©
800
850 900
KILN EXIT TEMPERATURE fC)
950
Figure 2. Actual versus target operating conditions.
14
-------
SECTION 3
SAMPLING AND ANALYSIS PROCEDURES
Sampling and analysis protocols for the test series were designed to meet research
objectives and IRF permit-compliance requirements. Figure 3 identifies the sampling point
locations. Stack sampling (sampling point 7) was for permit compliance. In general, sampling
for each test consisted of the following:
• Obtaining a composite sample of the feed materials (clay/organic liquid mixture,
and aqueous metal spike solution) and of the kiln ash
• Obtaining several samples of the scrubber blowdown water over time
• Obtaining samples of the flue gas at the afterburner and scrubber exits for
particulate and vapor phase metal analyses, and for volatile organic hazardous
constituent analyses
• Various combinations of CEM sampling of flue gas 02, CO, C02, NO,, and total
unburned hydrocarbon (TUHC) at the kiln, afterburner, and scrubber exits, and
in the stack
• Obtaining samples of the flue gas at the afterburner and scrubber exits, and at the
stack for total particulate and HC1
The CEMs available at the IRF and the locations they monitored for these tests are
summarized in Table 8. This monitoring arrangement was employed in all tests. Figure 4
illustrates the generalized CEM flue gas conditioning and flow distribution system at the IRF.
Four independent systems, as illustrated in Figure 4, are in place so that appropriately
conditioned sample gas from four separate sampling locations can be routed to any of the
available monitors listed in Table 8. The CEM setup described in Table 8, with appropriate gas
conditioning per Figure 4, was employed throughout this test program.
Details of the extractive sampling and analysis procedures are discussed in the following
subsections.
3.1 SAMPLING PROCEDURES
Table 9 summarizes the sampling and analysis matrix for the test series. As indicated,
the incinerator flue gas was characterized at three locations: the afterburner exit, the scrubber
15
-------
FAN
HE PA
FILTER
AFTER-
BURNER
OUENCH
SECTION
KILN
CARBON
BED
IONIZING WET
SCRUBBER
1 2 3 4 5 6
SAMPLE LOCATIONS
Feeds and residuals Continuous monitors Flue gas
Method 17 Method 5
with multiple with multiple
metals metals
impingers Impingers
Ss Clay/ Metal (particulate, - (particulate, VOST Method 5
Sampling organic spike Kiln Scrubber CO, Unhealed metals, and metals, Bnd (volatile (particulate
point liquid solution ash blowdown 02 CO] NO, TUHC HCI) HCI) organics) and HCI)
1 X X
2 X
3 X
4 X X X X X
5 X
6 XXX XX
7 XXX X* X
Tests 4, 7, and 8 only.
Figure 3. Sampling protocol.
-------
TABLE 8. CONTINUOUS EMISSION MONITORS
Monitor
Location Constituent Manufacturer Model
Principle
Range
Kiln exit
Afterburner
exit
0,
Stack
CO
CO,
Scrubber exit 0,
NO,
Unheated
TUHC
CO
CO,
Unheated
TUHC
Beckman
Beckman
Horiba
Horiba
Unheated Shimadzu
TUHC
Teledync
Thermo
Electron
Shimadzu
Honba
Horiba
Shimadzu
Teledyne
755
755
VIA 500
PIR 2000
GC Mini
326A
10AR
VIA 500
PIR 2000
GC Mini
326A
Paramagnetic
Paramagnetic
NDIR
NDIR
FID
Fuel cell
Chemiluminc scent
GC Mini FID
NDIR
NDIR
FID
Fuel cell
0-10%
0-25%
0-50%
0-100%
0-10%
0-25%
0-50%
0-100%
0-50 ppm
0-500 ppm
0-20%
0-80%
0-10 ppm to
0-2,000 ppm in
multiples of 2
0-5%
0-10%
0-25%
0-75 ppm to
0-10,000 ppm in
multiples of 2
0-10 ppm to
0-2,000 ppm in
multiples of 2
0-50 ppm
0-500 ppm
0-20%
0-80%
0-10 ppm to
0-2,000 ppm in
multiples of 2
0-5%
0-10%
0-25%
17
-------
r
FILTER
PUMP
FILTER
HIGH
BAY
SAMPLE
PORT
VENT
PERMA PURE
DRYER
ICE BATH/
WRINGER
CONTROL
ROOM
CALIBRATION
GAS
HEATED FLOW
CONTROL UNIT
DISPOSABLE
ABSORPTION
UNfTS-
FLOW
CONTROL
UNIT
SAMPLE GAS MANIFOLD
HEATED
TUHC
MONITOR
Oz
MONITOR
Mbcad Na and Ca hydroxides for
add gas removal
CO CO 2 NO, UNHEATED
MONrTOR MONtTOR MONrTOR TUHC
MONITOR
^ HEATED TO
|_ J 150 TO 175*C
Figure 4. Generalized CEM sample gas now schematic.
18
-------
exit, and the stack. Characterization at the afterburner and scrubber exit locations supported
test objectives. Stack gas sampling was performed to demonstrate compliance with the IRF's
operating permit.
The flue gas sampling performed at the afterburner exit was designed to measure flue
gas particulate load and size distribution, HC1, trace metal vapor phase and particulate emissions
by particle size range, and volatile organic hazardous constituent emissions.
Volatile organic hazardous constituent emissions were sampled using the Method 00302
volatile organic sampling train (VOST) protocol. All other parameters listed above were
measured in the afterburner exit flue gas using a variation of a Method 17 sampling train*. The
impingers used in the afterburner exit Method 17 train are noted in Table 10. After sampling,
the contents of impingers 1 and 2 were combined and 100 mL taken for HQ analysis. The
remaining contents of impingers 1 and 2, along with the contents of impingers 3 and 4, were
preserved to a pH of less than 2 with HN03 for later metals analysis. The sampling probe was
washed with acetone. The probe wash was desiccated and weighed, then resuspended in acetone
for later metals analysis.
The modified Method 17 train at the afterburner exit also collected particulate in an
oversized alundum thimble placed within the sampling probe. After obtaining the total
particulate weight, the particulate in these samples was divided according to the terminal velocity
in air using a centrifugal classifier. This device is described in the ASME Power Test Code 28\
Weights of the resulting nine size cuts were recorded. These samples were later recombined into
five size cuts and subjected to metals analysis to provide data on metals distribution by
particulate size.
The flue gas sampling performed at the scrubber exit was designed to measure the same
parameters as measured in the afterburner exit flue gas, with the exception that particle sizing
and metals analysis by particle size were not performed. Accordingly, VOST (Method 0030)2
sampling at the scrubber exit was performed. The particulate levels in the scrubber exit were
expected to be significantly lower than those in the afterburner exit, making the collection of
sufficient particulate for particle sizing difficult. In addition, the high water content of the
scrubber exit flue gas would have caused a moist and agglomerated particulate catch.
Classification of this catch would likely have not been representative of the actual particulate size
distribution in the scrubber exit flue gas. Thus, instead of a modified Method 17 train, a
Method 5 train3 was run at the scrubber exit.
The impingers used in the scrubber exit Method 5 train were the same as those used in
the afterburner exit Method 17 train (Table 10). After sampling, the filter and probe wash
(acetone) from the Method 5 train were recovered and desiccated to constant weight per the
Method 5 procedure. The probe wash was then resuspended in acetone for later metals analysis.
Impinger collection, aliquoting, combining, and preservation for the scrubber exit Method 5 train
were exactly the same as performed for the afterburner exit Method 17 train discussed above.
A Method 5 train was also used to sample the stack gas for each test to determine
compliance with the IRF operating permit. Particulate load and HC1 levels were measured. A
19
-------
TABLE 9. SAMPLING AND ANALYSIS MATRIX
Sampling
Analysis
Sample
Location
procedure
Parameter Method
Frequency
Clay/organic
Preparation S005"
C,Hg, CjCl^,
Purge and trap by
1 composite/
liquid mixture
storage drum
C6HSC1
Method 50306;
formulation
GC/FID analysis
(Appendix F)
Ultimate
ASTM D 3176-74 (C,
1 composite/
analysis (C, H,
H, O), ASTM D 2361-
formulation
O.CI)
66 (CI)
As, Ba, Bi, Cd,
Digestion by Method
1 composite/
Cr, Cu, Pb,
3050, ICAP analysis by
formulation
Mg, and Sr
Method 6010"
Aqueous
Kiln inlet Grab/tap
As, Ba. Bi, Cd,
Digestion by Method
1 composite/
metal spike
Cr, Cu, Pb,
3010, ICAP analysis by
test
solution
Mg, and Sr
Method 6010b
Kiln ash
Kiln ash pit S005*
As, Ba, Bi, Cd.
Digestion by Method
1/tesl
Cr, Cu, Pb,
3050, ICAP analysis by
Mg, and Sr
Method 6010b
Volatile organic
Purge and trap by
1/lest
hazardous
Method 5030",
constituents
GC/FTD analysis
(Appendix F)
TCLP leachate
TCLP extraction1,
1 /test
for As, Ba, Bi,
digestion by Method
Cd, Cr, Cu, Pb,
3010, ICAP analysis
Mg, and Sr
by Method 6010b
Scrubber
Blowdown Grab/tap
As, Ba, Bi, Cd,
Digestion by Method
At least
blowdown
discharage
Cr, Cu, Pb,
3010, ICAP analysis by
5/test
water
Mg, and Sr
Method 60101*
Volatile organic
Purge and trap by
1/test
hazardous
Method 5030*",
constituents
GC/FID analysis
(Appendix F)
Flue gas
Afterburner Method 17d
exit
Particulate load
Method 17d
1/test
Particulate size
ASME PTC 28e
1/test
distribution
"(5), Harris, et al.
b(2), SW-846.
c(6), CFR TCLP citation.
d(3), CFR reference methods.
e(4), AS ME PTC 28.
(continued)
20
-------
TABLE 9. (continued)
Sample
Location
Sampling
procedure
Analysis
Parameter
Method
Frequency
Flue gas
(continued)
Afterburner
exit
(continued)
Method 17d
(continued)
Method 0030b
Scrubber exit Method 5d
Suck gas
Method 0030*
Suck Method 5*
downstream
of carbon
bed/HEPA
filter
Method 0030h
HCl
As, Ba, Bi, Cd,
Cr, Cu, Pb,
Mg, and Sr
Volatile organic
hazardous
constituents
Paniculate load
HCl
As, Ba, Bi, Cd,
Cr, Cu, Pb,
Mg, and Sr
\blaule organic
hazardous
constituents
Particulate load
HCl
Volatile organic
hazardous
constituents
Analysis of impinger
solution for CI by
specific ion electrode
(Appendix F)
Digestion by Method
3010 or 3050, ICAP
analysis by Method
60l0b (impingers and
particulate by size)
Thermal desorption,
purge and trap by
Method 504(f, GC/FID
analysis (Appendix F)
Method 5i
Analysis of impinger
solution for CI" by
specific ion electrode
(Appendix F)
Digestion by Method
3010 or 3050, ICAP
analysis by Method
6010* (impingers and
particulate)
Thermal desorption,
purge and trap by
Method 5040\ GC/FID
analysis (Appendix F)
Method 5"1
Analysis of impinger
solution for CI* by
specific ion electrode
(Appendix F)
Thermal desorption,
purge and trap by
Method 5040b, GC/FID
analysis (Appendix F)
1/tcst
1 /test
3 trap pairs/
test
1/test
1/Lest
1/test
3 trap pairs/
test
1/test
1/test
3 trap pairs/
test. Tests 4,
7, and 8 only
b(2), SW-846.
-------
TABLE 10. MULTIPLE METALS SAMPLING TRAIN IMPINGER REAGENTS
Impinger g
number Reagent Quantity £
1 Empty m
2 O.lNNaOH 100 mL
3 5% HNOj and 10% H2Oj 100 mL
4 5% HNOj and 10% H202 100 mL
5 Silica gel 750g
Method 5 train, with impinger contents as noted in Table 11, was used at this location. In
addition, VOST (Method 0030)2 sampling was performed at the stack location during Tests 4,
7, and 8.
All scrubber exit Method 5 sampling trains collected at least 2.8 m3 (100 ft3) of flue gas
over a nominal 3-hr sampling period. With the exception of Test 9, all afterburner exit
Method 17 sampling trains collected at least 19.8 m3 (700 ft1) of flue gas over the same nominal
3-hr sampling period. This period began no less than 1/2 hour after the start of test mixture
feed. Test mixture feed continued until all sampling was completed. For Test 9, the afterburner
exit Method 17 train collected 15.2 m3 (537 ft3). Sampling durations and schedules for each test
are noted on the incinerator operating condition plots in Appendix A.
In addition to the flue gas sampling described above, samples of the clay/organic liquid
feed mixture, the aqueous trace metal spike solution, the kiln ash, and the scrubber blowdown
were collected for analysis. Each test was performed using waste feed from one drum of feed
prepared. Composite feed samples from each drum were collected by trier sampling at three
locations in the drum cross-section5. Each of these composites was ultimately analyzed for
toluene, tetrachloroethylene, chlorobenzene, arsenic, barium, bismuth, cadmium, chromium,
copper, lead, magnesium, strontium, and ultimate composition. Samples for volatile organic
analysis were collected from each drum's composite sample when the contents of that drum were
first fed into the kiln. These samples were sealed in 40-mL VOA vials at that time.
A grab sample of each test's aqueous metal spike solution was collected from a sampling
port in the aqueous metal solution feed system. These samples were preserved with HN03 at
a pH of less than 2 for later trace metals analysis.
For scrubber blowdown sampling, a 100-mL grab sample was taken at the start of flue
gas sampling, and hourly thereafter until sampling was completed. A final blowdown sample was
taken at the end of the flue gas sampling period. Each individual blowdown sample was
preserved with HN03 at a pH of less than 2.
22
-------
TABLE 11. STACK GAS METHOD 5 SAMPLING TRAIN IMPINGER REAGENTS
g
Impinger g
number Reagent Quantity £
1 0.1 N NaOH 100 mL ^
2 0.1 N NaOH 100 mL
3 0.1 N NaOH 100 mL
4 Silica gel 750g
A composite kiln ash sample was collected from the ash pit after the completion of each
test. The ash pit contents were weighed and a clean, empty ash collection drum was placed in
the collection pit for the subsequent test.
All sampling activities completed in this test program were performed in accordance
with the test plan for this program and the methods cited therein and above, with the few
exceptions noted above.
32 ANALYSIS PROCEDURES
Table 12 summarizes the number of samples collected during the nine tests. Particulate
load determinations from the Methods 17 and 5 samples were performed at the IRF, in
accordance with the respective method procedures, prior to size classification and combination
for trace metals analysis as discussed in Section 3.1. Chloride analyses for determining HC1
emissions were performed on aliquots of appropriate impinger solution combinations via specific
ion electrode analysis at the IRF according to the procedure given in Appendix F.
Analysis of VOST (Method 0030)2 traps was performed at the IRF by thermal
desorption, purge and trap gas chromatography/flame ionization detector (GC/FID) analysis.
Thermal desorption, purge and trap was performed in accordance with Method 50402. Analysis
was by capillary column GC/FID using the procedure given in Appendix F. The 22 volatile
organic compounds routinely determined via this method at the IRF were analyzed. These
compounds are listed in Table 13.
All samples for trace metals analysis were preserved as noted in Section 3.1 and
submitted to the National Center for Toxicological Research (NCTR) inorganic laboratory for
analysis. Metals analysis was performed by inductively coupled argon plasma (JCAP)
spectroscopy per Method 60102. Samples were digested appropriately prior to analysis.
Method 3050 was used for solid samples; Method 3010 was used for aqueous liquid samples2.
Kiln ash samples were also extracted by the toxicity characteristic leaching procedure (TCLP)6.
The leachates were digested by Method 3010 and analyzed for the nine test metals by ICAP
spectroscopy per Method 6010.
23
-------
TABLE 12. SUMMARY OF TEST SAMPLES
""* 1 1 rv.
CD
Number of samples o
Sample type
Analyte
Each test
Total
Gay/organic liquid feed mixture
C7Hg, C2C14, C6HsC1
2 to 3
21
Ultimate analysis (C, H, O, Q)
2 to 3
22
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
2 to 3
22
Aqueous metal spike solution
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1 to 2
10
Kiln ash
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1 to 2
11
Volatile organic hazardous constituents
1 to 2
11
Kiln ash TCLP leachate
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1 to 2
11
Scrubber blowdown
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
At least 5
60
Volatile organic hazardous constituents
1
9
Afterburner exit flue gas:
Method 17 train8:
<2 )im particulate
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
2 to 4 |im particulate
As, Ba, Bi, Cd, Cr, Cu, Pb. Mg, and Sr
1
9
4 to 10 |im particulate
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
10 to 30 |im particulate
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
>30 |im particulate
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg. and Sr
1
9
Probe wash
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
1st and 2nd impingers
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
CI-
1
9
3rd impinger
As, Ba, Bi, Cd, Cr. Cu, Pb, Mg, and Sr
1
9
4th impinger
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
VOST (Method 0030)
Sample trap pair
Volatile organic hazardous constituents
3
27
Scrubber exit flue gas:
Method 5 train8:
Probe wash
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
Filter
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
9
1st and 2nd impingers
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
CI-
1
9
3rd impinger
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
4th impinger
As, Ba, Bi, Cd, Cr, Cu, Pb, Mg, and Sr
1
9
(continued)
24
-------
TABLE 12. (continued)
Number of samples
Sample type
Analyte
Each test
Total
Scrubber exit flue gas
(continued):
VOST (Method 0030)
Sample trap pair
Volatile organic hazardous constituents
3
27
Stack gas:
VOST (Method 0030)
Sample trap pair
Volatile organic hazardous constituents
3 (Tests 4,
9
7, 8 only)
Field spike trap pair
Volatile organic hazardous constituents
1
9
Field blank trap pair
Volatile organic hazardous constituents
1
9
Lab blank trap pair
Volatile organic hazardous constituents
1
9
Method 5 train4:
1st and 2nd impingcrs
cr
1
9
3rd impinger
cr
1
9
•Particulate load measured by weighing the dessicated probe wash and thimble or filter prior
to digesting for metals analysis.
25
-------
TABLE 13. VOLATILE ORGANIC COMPOUNDS ROUTINELY
ANALYZED BY GC/FID AT THE IRF
Methylene chloride
Benzene
1,1 -Dichloroethane
1,1,2-Trichloroethane
t-1,2-Dichloroethylene
Hexane
Chloroform
Bromoform
1,2-Dichloroethane
Tetrachloroethylene + Tetrachloroe thane
1,1,1-Trichloroethane
Toluene
Carbon tetrachloride
Chlorobenzene
Bromodichloromethane
Ethyl benzene
1,2-Dichloropropane
1,3-Dichlorobenzene
t-1,3-Dichloropropylene
1,2-Dichlorobenzene
Trichloroethylene
1,4-Dichlorobenzene
The clay/organic liquid feed composite samples were analyzed at the IRF for toluene,
tetrachloroethylene and chlorobenzene via purge and trap by Method 50302, with GC/FID
analysis according to the procedure given in Appendix F. Composite samples of the kiln ash and
the scrubber blowdown were similarly analyzed for the 22 volatile organic compounds listed in
Table 13. Composite clay/organic liquid feed samples were sent to Galbraith Laboratories in
Knoxviile, Tennessee, for ultimate analysis by the ASTM methods noted in Table 3-2.
All analyses completed during this test program followed the procedures documented in
the methods cited above.
26
-------
SECTION 4
TEST RESULTS
Results from the test program are discussed in this section. The discussion is subdivided
as follows: measured synthetic waste feed composition; CEM data; flue gas flowrates; particulate
and HC1 emissions; trace metals test results; volatile organic emissions; and POHC DREs.
These areas are discussed in the following subsections.
4.1 SYNTHETIC WASTE FEED COMPOSITION
Section 2 discussed the targeted synthetic waste feed composition planned for each test.
Section 3 noted the various waste feeds sampled and analyzed during each test to verify the
actual composition of the feeds prepared. This section discusses the results of the feed analyses.
Table 14 summarizes the clay/organic liquid mixture POHC concentrations measured
for each test's composite sample and compares these to target values. The data in Table 14
confirm close agreement between measured and target composition for all tests, although
evidence of some evaporative loss of all POHC components between preparation and sampling
is apparent.
Table 15 summarizes the analyzed aqueous spike solution trace metal concentrations.
Analyzed concentrations were comparable to prepared concentrations for all metals in Tests 2,
3, 4, 5, 6, and 9. For Test 1, the analyzed concentrations were about 20 percent lower that the
prepared concentrations for all metals, while for Test 8 the analyzed concentrations were lower
by about 30 percent. For Test 7, the analyzed concentration for bismuth was 30 percent lower
than the prepared concentration; the analyzed concentration for arsenic was only 41 percent of
the prepared concentration. Concentrations of the other Test 7 metals were comparable to the
prepared concentrations.
The low analyzed concentrations reported for Tests 1 and 8 were likely due to sample
dilution caused by residual purge water remaining in the sampling line used to obtain the spike
solution samples. Low analyzed concentrations for Test 7 arsenic and bismuth could not be
explained. Therefore, during the initial data analysis both the analyzed and prepared
concentration values were used. The reduced data for Tests 1 and 8, and for bismuth in Test 7,
did not differ sufficiently in either case to alter the overall interpretation of the test results. The
analyzed spike solution concentration values were thus used in the data reduction calculations.
The analyzed arsenic value reported for Test 7, however, was treated as an outlier, and the
prepared concentration of 1590 mg/L was used in the calculations. Additional information
27
-------
TABLE 14. POHC CONCENTRATIONS IN CLAY/ORGANIC LIQUID FEED SAMPLES
Weight % in mixture
Chlorine
Test
Test date
Toluene
Tetrachloroethylene
Chlorobenzene
content"
Mixture 1 target
28.6
0
0
0
composition
Measured composition
1
8/17/89
23.1
0
0
0
Mixture 2 target
21.7
3.4
3.4
4
composition
Measured composition
2
8/2/89
17.2
3.0
2.9
3.5
3
8/4/89
17.8
3.0
2.9
3.5
4
8/1/89
17.7
3.0
2.9
3.5
5
8/16/89
17.3
3.2
3.0
3.7
6
8/15/89
17.8
3.1
3.0
3.6
7
8/9/89
18.6
3.1
2.9
3.5
8
8/11/89
18.3
3.3
3.0
3.8
Mixture 3 target
14.9
6.9
6.9
8
composition
Measured composition
9
7/28/89
11.6
6.0
5.6
6.9
'Calculated based on measured tetrachloroethylene and chlorobenzene concentrations.
28
-------
TABLE 15. AQUEOUS SPIKE SOLUTION METALS CONCENTRATIONS
Prepared spike Analyzed spike solution concentration (ppm as metal)
solution
Spike
concentration
Test I
Test 2
Test 3
Test 4
Test 5
Test 6
Test 7
Test 8
Test 9
Metal
compound
(ppm as metal)
(8/17/89)
(8/2/89)
(8/4/89)
(8/1/89)
(8/16/89)
(8/15/89)
(8/9/89)
(8/11/89)
(7/28/89)
Arsenic
AsjOj
1,590
1,290
1,620
1,620
1,480
1,580
1,610
648*
1,070
1,600
Barium
Ba(N03)2
12,700
10,400
12,800
13,000
11,800
12,700
12,950
12,700
8,800
12,500
Bismuth
Bi(N03)3.5H20
12,700
9,390
11,800
11,900
11,000
11,600
12,000
8,920
8,450
11,900
Cadmium
Cd(N03)2
318
255
315
324
292
313
321
308
217
318
Copper
Cu(N03)2.3H20
12,700
9,990
12,200
12,300
11,400
12,200
12,500
12,200
8,670
12,200
Lead
Pb(NOj)2
1,590
1,180
1,460
1,470
1,350
1,450
1,470
1,440
1310
1,450
Strontium
Sr(NOj)2
12,700
10,700
13,100
13,300
12,200
13,100
13,350
13,100
9,170
12,900
'Rejected as an outlier. Prepared concentration (1,590 mg/L) used in the calculation of integrated feed concentration.
-------
supporting the approach used to handle the analyzed spike solution data is discussed in
Section 6.
Previous data on the clay matrix composition indicated that the clay materia] contained
53 ppm of chromium and 2.2 percent magnesium. These background concentrations were
sufficiently high that chromium and magnesium were not included in the aqueous metals spike
solution. To confirm these values, and to determine the background clay contribution to the
integrated feed concentrations, samples of the clay/organic liquid mixture from each test were
analyzed for each of the metals. Results of these analyses are summarized in Table 16.
Table 17 combines the clay/organic liquid mixture analysis results from Table 16 with
the aqueous spike solution data from Table 15 to give the concentration of each metal in the
integrated feed for each test. The clay/organic liquid feedrate data and the aqueous spike
solution feedrate data obtained during the period of scrubber exit flue gas sampling noted in
Appendix A were used to determine the individual component concentrations.
42 CONTINUOUS EMISSION MONITOR DATA
Table 18 summarizes the CEM data obtained for each of the tests during periods of flue
gas sampling. The values in the table were developed from the data recorded by the PC data
acquisition system. Graphs of the CEM data for each of the tests are included in Appendix A.
As noted in Section 2, interruptions in the clay/organic liquid feed occurred during several tests.
The CEM data plotted in Appendix A illustrate the effects of these interruptions on flue gas
constituent concentrations. As noted in Section 2, flue gas sampling was suspended during most
feed interruptions until feeding resumed and steady-state conditions were reestablished.
Therefore, deviations caused by feed interruptions during which flue gas sampling was stopped
are not reflected in the data in Table 18. Since flue gas sampling continued through the
interruptions at the end of Tests 4 and 9, associated variations in the CEM data were included
in generating the values in Table 18.
A leak in the CEM sampling train at the kiln exit was identified during post-test
calibration following Tests 1 and 9. A sampling train leak was suspected during the course of
these tests based on unexpected increases in kiln exit 02 concentrations. To verify this suspicion,
kiln exit flue gas samples were also analyzed at about 30-min intervals with an Orsat analyzer
to determine 02 concentrations. The averages of these values (individual Orsat measurements
are included in Appendix A) are also included in Table 18, and confirm that flue gas 02
concentrations were lower than indicated by the CEM, and comparable to 02 concentrations
during the other tests.
The data in Table 18 show that, for most tests, CO levels at the afterburner exit and at
the stack remained below the 5-ppm detection limit of the monitors. Afterburner CO spikes of
about 250 and 600 ppm accompanied flameouts during Tests 7 and 9, respectively. However, as
noted above, sampling was suspended during these periods. Average NO, concentrations at the
scrubber exit ranged from 37 to 49 ppm. A spike of about 400 ppm accompanied a flameout
during Test 7; otherwise, NO, concentrations varied little during each test. Hydrocarbon
emissions, as measured by both heated and unheated monitors, remained below the instrument
detection limit of 1 ppm throughout the test program.
30
-------
TABLE 16. CLAY/ORGANIC LIQUID MIXTURE METAL CONCENTRATIONS
Concentration (mg/kg clay organic liquid mixture)
Test 1
Test 2
Test 3
Test 4
Test 5
Test 6
Test 7
Test 8
Ttst 9 |
Metal
(8/17/89)
(8/2/89)
(8/4/89)
(8/1/89)
(8/16/89)
(8/15/89)
(8/9/89)
(8/11/89)
(7/28/89) '
Arsenic
<4.4
<4.2
<4.2
<4.2
4.3
<3.6
<3.6
4.5
<4.2
Barium
21
25
24
23
23
21
22
23
24
Bismuth
<6.0
<6.5
<6.5
<6.5
<3.9
<3.9
<3.9
<3.9
<6.5
Cadmium
1.4
1.4
1.2
13
1.4
1.3
1.3
1.5
1.7
Chromium
45
38
38
38
42
39
42
42
37
Copper
25
26
26
27
24
24
26
25
27
Lead
3.9
<1.7
<1.7
<1.7
3.3
<2.8
3.9
3.3
<1.7
Magnesium
18,800
18,800
19,050
19,900
18,100
18,400
19,100
18,900
18,500
Strontium
30
32
32
33
31
31
31
32
31
-------
TABLE 17. INTEGRATED FEED METAL CONCENTRATIONS
Metal concentration (ppm)
Test I
Test 2
Test 3
Test 4
Test 5
Test 6
Test 7
Test 8
Test 9
Metal
(8/17/89)
(8/2/89)
(8/4/89)
(8/1/89)
(8/16/89)
(8/15/89)
(8/9/89)
(8/11/89)
(7/28/89)
Arsenic
41
56
53
46
51
44
49'
36
52
Barium
335
460
450
390
395
375
395
295
430
Bismuth
280
400
390
340
340
330
265
260
390
Cadmium
9
12
12
10
11
10
10
8
12
Chromium
45
38
38
38
42
39
42
42
37
Copper
325
445
430
380
380
365
385
290
425
Lead
39
50
48
42
45
40
46
44
47
Magnesium
18,800
18,800
19,100
19,900
18,100
18,400
19,100
18,900
18,500
Strontium
350
480
470
410
415
395
415
315
455
'Assuming 1,590 mg/L in the aqueous spike solution.
-------
TABLE 18. CEM DATA
Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9
Parameter (8/17/89) (8/2/89) (8/4/89) (8/1/89) (8/16/89) (8/15/89) (8/9/89) (8/11/89) (7/28/89)
Kiln exit:
0.2, range, %
average, %
Afterburner exit:
02, range, %
average, %
C02, range, %
average, %
CO, range, ppm
average, ppm
Scrubber exit:
02, range, %
average, %
NO„, range, ppm
average, ppm
Stack:
02, range, %
average, %
C02, range, %
average, %
CO, range, ppm
average, ppm
11.1 - 14.9
13.6 (12.0*)
7.6 - 9.0
8.3
6.9-8.1
7.5
12.1 - 12.8
12.4
35-40
37
12.4 - 13.0
12.7
4.4 - 5.0
4.7
<5
7.9 - 13.4 10.1 - 13.3 9.0- 16.6 9.0- 15.5 8.5 - 13.4 11.0 - 13.2 11.8 - 13.6 14.0- 17.2
12.3 11.9 11.9 12.1 11.9 12.6 12.6 14.8(12.9')
6.4 - 8.7 5.8 - 8.3 4.5 - 11.8 5.2 - 8.8 5.4 - 9.4 6.0 - 8.2 7.5 - 8.6 6.7 - 10.5
7.7 7.3 7.3 7.9 8.6 7.6 8.1 7.6
7.1 -9.0 7.2 - 8.8 4.7 - 10.0 6.9- 8.7 6.8 -9.5 7.5 - 8.8 7.1 - 8.1 4.4 -8.1
7.7 7.9 7.9 7.5 7.3 7.9 7.6 7.3
<5 <5 <5 <5 <5 <5 <5 <5-11
8
11.3-12.5 10.8-12.0 10.5-14.3 11.3-12.2 12.0-12.8 10.4-11.8 11.4-12.1 10.9-13.3
11.7 11.6 11.6 11.7 12.4 11.4 11.8 11.5
29-46 38-43 20 -49 36-47 35 -41 37 -48 38 -44 38 -53
43 40 41 44 38 45 40 49
11.5-12.8 11.2-12.3 10.9-14.5 11.9-12.9 12.3-12.9 10.8-12.2 11.8-12.5 11.2-13.6
11.9
4.5 - 5.4
5.0
7-11
10
11.9
4.7 - 5.4
4.9
<5
11.9
3.4 - 5.6
4.9
7-18
9
12.2
4.6 - 5.2
5.0
<5
12.5
4.2-5.1
4.8
not in
service
11.7
5.1 -5.9
5.3
<5
12.1
4.8 - 5.4
5.1
not in
service
11.9
3.6 - 5.5
5.1
8-11
9
'Average of data obtained using a Bacharach Fryrite® Orsat Oj analyzer. A leak in the kiln exit CEM sampling system was identified during
post-test calibration.
-------
43
FLUE GAS FLOWRATES AND PARTICULATE AND HC1 EMISSIONS
Flue gas flowrates were determined using data from the Que gas sampling trains at the
afterburner and scrubber exits and at the stack. To confirm these measured values, the flowrates
were also calculated by carbon balance assuming complete combustion of the natural gas and
synthetic organic waste feed. CEM data for both 02 and C02 were used in the calculations.
Table 19 lists the flowrates determined by the flue gas sampling trains, and the flowrates
estimated based on combustion analysis using both measured flue gas 02 and C02. Good
agreement was observed between the flowrates determined by the three methods. The increased
flowrate between the afterburner exit and scrubber exit was caused by the addition of air at the
scrubber inlet as part of its normal operation, as well as duct inleakage. Inleakage between the
scrubber exit and stack averaged less than 3 percent. All data analysis requiring information on
flue gas flowrates employed the values determined using the sampling train data.
Flue gas particulate emissions were measured at the afterburner exit, scrubber exit, and
stack. As shown in Table 20, stack particulate concentrations ranged from 28 to 151 mg/dscm
at 7 percent 02. All levels were below the hazardous waste incinerator performance standard
of 180 mg/dscm at 7 percent 02. For all tests, the scrubber exit particulate concentrations were
less than the afterburner exit particulate concentrations by an average of about 50 percent. With
the exception of Tests 1 and 6, stack particulate concentrations were equal to, or less than,
particulate concentrations in the scrubber exit. The stack particulate concentration measured
for Test 6 was the highest value of any sample obtained during the test program, and may have
been due to a sampling error.
As shown in Table 21, stack HC1 levels were nondetectable at detection limits up to
7.3 mg/dscm for all tests. Corresponding HC1 emission rates were less than 14 g/hr for these
tests, substantially below the hazardous waste incinerator performance limit of 1.8 kg/hr.
Table 21 also includes information on chlorine feedrates and afterburner and scrubber
exit HC1 emission rates. Good correlation was observed between chlorine feedrates and HC1
emission rates at the afterburner exit. Scrubber exit HC1 levels were at or below detection limits
of up to 5 mg/dscm in Tests 1 through 8. All scrubber HCI collection efficiencies in Tests 1
through 8 were greater than 99 percent. Scrubber exit HCI levels were measurable at
154 mg/dscm (235 g/hr) for Test 9, the test with the highest feed chlorine content.
4.4 TRACE METALS DISCHARGE DATA
This section discusses the distributions of the trace metals fed to the RKS among the
discharge streams sampled and analyzed. For the metal fraction discharged with the kiln ash,
distributions were determined based on the ash composite concentration and the total weight of
the kiln ash. For the metal fraction discharged through the Que gas and scrubber liquor,
distributions were based on the mass flowrate of the given element. The mass flowrate was
calculated from the analyzed concentration of each metal in a given stream, and the total mass
flowrate of that stream. Appendix C includes all the laboratory analysis reports that serve as the
basis for the given stream concentrations. Appendix D contains the flue gas sampling train data,
including flue gas stream flowrate information.
34
-------
TABLE 19. FLUE GAS FLOWRATES
Kiln
Afterburner
Scrubber exit
Stack
Test
Test
date
Combustion
analysis by
% O,
Measured
Combustion
analysis by
% CO,
Combustion
analysis by
% O,
Measured
Combustion
analysis by
% O,
Measured
Combustion
analysis by
% CO,
Combustion
analysis by
% O,
1 '
8/17/89
dscm/hr
713 [586*]
1,090
1,410
1,350
2,310
1,990
2300
2.240
2,050
(dscf/hr)
(25,200) 1(20.700")]
(38,500)
(49,700)
(47,500)
(81,500)
(70,100)
(81300)
(79300)
(72,500)
2
8/2/89
dscm/hr
491
1,000
1,130
1,080
1,580
1,540
1.710
1,740
1,550
(dscf/hr)
(17300)
(35,400)
(40.000)
(38.000)
(55,900)
(54,400)
(60,400)
(61,600)
(54,900)
3
8/4/89
dscm/hr
723
1,030
1,320
1,250
1,890
1,820
1,940
2,130
1,870
(dscf/hr)
(25,500)
(36,200)
(46,600)
(44,100)
(66,800)
(64,200)
(68,500)
(75,100)
(66,000)
4
8/1/89
dscm/hr
593
1,030
1,180
1,130
1,790
1,640
1,800
1.900
1,680
(dscf/hr)
(20,900)
(36,200)
(41,600)
(39,800)
(63,100)
(57,700)
(63,600)
(67,100)
(59,200)
5
8/16/89
dscm/hr
567
1,090
1,550
1,450
2,230
2,060
2310
2320
2,150
(dscf/hr)
(20,000)
(38,400)
(54,600)
(51,100)
(78,800)
(72,800)
(81,700)
(81,900)
(76.100)
6
8/15/89
dscm/hr
583
1,060
1,120
1.050
1,790
1.520
1,880
1,710
1,530
(dscf/hr)
(20,600)
(37,300).
(39,600)
(37,100)
(63,300)
(53,700)
(66,400)
(60300)
(54.000)
7
8/9/89
dscm/hr
617
1,040
1,170
1,100
1,780
1,540
1,800
1,740
1,590
(dscf/hr)
(21,800)
(36,600)
(41,200)
(38,800)
(62,800)
(54,200)
(63,700)
(61,400)
(56,100)
8
8/11/89
dscm/hr
622
817
1.230
1,160
1,830
1,630
1,910
1,840
1,700
(dscf/hr)
(22,000)
(28,900)
(43,500)
(40,800)
(64,600)
(57,600)
(67300)
(64,800)
(60,100)
9
7/28/89
dscm/hr
884 |677']
940
1,140
1,070
1,520
1,440
1,660
1,630
1,500
(dscf/hr)
(31,200) 1(23.900*))
(33,200)
(40300)
(37,700)
(53,700)
(50,900)
(58,500)
(57.700)
(52,800)
"Calculated using data obtained by a Bacharach Fryritc® Orsat 02 analyzer A leak in the kiln exit CEM sampling train was identified during
post-test calibration.
-------
TABLE 20. FLUE GAS PARTICULATE DATA
Particulate concentration
(mg/dscm at 7% Oj)
Test Test date Afterburner exit Scrubber exit Stack
1
8/17/89
117
17
43
2
8/2/89
70
28
28
3
8/4/89
143
35
35
4
8/1/89
84
75
37
5
8/16/89
77
59
54
6
8/15/89
144
42
151
7
8/9/89
101
36
28
8
8/11/89
116
81
45
9
7/28/89
126
62
55
TABLE 21. HCI EMISSION DATA
cr
Afterburner exit
HCI
Scrubber exit
HCI
Scrubber
HCI
collection
Stack HCI
Test
Test date
feedrate
(kg/hr)
(g/dscm) (kg/hr)
(mg/dscm)
(gflir)
efficiency
(%)
(mg/dscm)
(g/hr)
1
(8/17/89)
0
0.21
023
<0.4
<1.0
>99.6
<5.7
<13.2
2
(8/2/89)
1.89
2.32
2.32
<4.8
<7.7
>99.7
<5.1
<8.8
3
(8/4/89)
2.17
1.37
1.41
<4.9
<9.3
>99.3
<5.6
<10.9
4
(8/1/89)
1.83
1.96
2.01
<4.7
<8.4
>99.6
<5.7
<10.3
5
(8/16/89)
2.37
2.32
2.52
4.6
10.3
99.6
<5.4
<12.5
6
(8/15/89)
2.30
1.77
1.87
<1.1
<2.0
>99.9
<5.2
<9.8
7
(8/9/89)
2.35
2.21
2.29
<3.1
<5.5
>99.8
<6.1
<11.1
8
(8/11/89)
2.38
2.86
2.34
<0.5
<0.9
>99.9
<7.3
<14.0
9
(7/28/89)
3.91
3.81
3.58
154
235
93.4
<6.4
<10.7
36
-------
Table 22 summarizes the quantities of synthetic waste fed and kiln ash discharged in
each test. It is noteworthy that the fraction of the clay matrix feed which was accounted for by
the kiln ash discharge was relatively constant, averaging about 89 percent (range of 86 to
92 percent) in all tests. The moisture content of the clay was about 7 percent. Therefore, about
4 percent of the clay fed should theoretically be accounted for by ash entrained in the kiln Que
gas and carried into the afterburner.
As discussed in Section 3, a sample of ash from each test was subjected to TCLP
extraction and the leachates analyzed for the nine trace metals. Leachate concentrations for all
metals were low, with the exception of magnesium which averaged about 100 mg/L. Maximum
leachate concentrations of the hazardous constituent metals were as follows: arsenic, 0.26 mg/L;
barium, 1.12 mg/L; cadmium, 0.02 mg/L; chromium, 0.16 mg/L; and lead, 0.09 mg/L. These
levels are well below the regulatory limits defined in the recently promulgated TCLP regulations7.
The amount of a given metal removed by the scrubber deserves some discussion. Prior
to each test the RKS was operated for a minimum of 18 hours, fired on natural gas alone.
Following this period, synthetic waste feed was started. Flue gas sampling was begun only after
at least half an hour of feeding had elapsed. At the conclusion of flue gas sampling, feeding was
stopped and the RKS was returned to natural gas firing to maintain kiln temperature. Kiln
rotation continued until all ash was discharged from the kiln into the ash collection pit.
During natural gas firing, scrubber blowdown continued with fresh water scrubber
makeup introduced to keep the scrubber liquor loop full. Consequently, the scrubber loop was
purged of a previous test's metal buildup while firing with natural gas such that at the beginning
of a subsequent test, the scrubber liquor contained very little to no test trace metal. Thus, over
the time span of a given test, the recirculating scrubber liquor metal concentrations increased
from near zero to a higher value. Furthermore, at the end of a test a significant portion of the
metal fed to the RKS during the test and removed by the scrubber remained in the 1420-L
(375-gal) scrubber liquor loop.
To account for the nonsteady-state scrubber operation during a test, resulting from the
above mode of operation, the scrubber blowdown sampling protocol specifically called for
obtaining a scrubber blowdown sample at the beginning of a test, hourly thereafter, and at the
end of a test. This allowed the buildup of metal concentrations in the scrubber liquor to be
determined.
The mass flowrate of metal removed by the scrubber is calculated as the sum of two
terms. The first is that removed in the scrubber blowdown. This term was calculated from the
blowdown rate noted in Table 6, and the average blowdown concentration measured from the
blowdown samples taken during the period of flue gas sampling. The second term accounts for
the buildup of metals in the scrubber liquor loop caused by the nonsteady-state scrubber
operation. This term is calculated from the rate of metal concentration increase (slope of the
concentration versus time plots) during the period of flue gas sampling, and the liquor loop
volume of 1420 L (375 gal). The product of the concentration increase rate and the scrubber
liquor loop volume represents the average "flowrate" of metal into the scrubber liquor loop.
Appendix B contains graphs of the concentrations of the test metals in each blowdown sample
37
-------
TABLE 22. SYNTHETIC WASTE FEEDRATES AND KILN ASH DISCHARGE RATES
Synthetic waste Ted
(kg/lb) Kiln ash
Feed
Aqueous
Fraction of
Test
duration
Clay/organic
Clay
metal
Discharged
clav fed
Test
date
(hr)
liquid
fraction
solution
(kg/lb)
(%)
1
8/17/89
4.9
313 (690)
241 (530)
10 (22)
211(466)
88
2
8/2/89
8.4
450 (992)
346 (762)
15 (35)
311 (686)
90
3
8/4/89
3.4
211 (466)
161 (355)
7(16)
147 (323)
91
4
8/1/89
5.1
240 (530)
183 (404)
8(18)
167 (368)
91
5
8/16/89
4.7
303 (667)
231 (510)
10 (22)
204 (449)
88
6
8/15/89
5.1
328 (723)
250 (551)
10 (22)
215 (474)
86
7
8/9/89
4.4
257 (567)
194 (427)
8(18)
170 (375)
88
8
8/11/89
4.7
299 (659)
225 (497)
10 (22)
208 (458)
92
9
7/28/89
6.7
305 (673)
235 (517)
10 (22)
211 (466)
90
'Test 9 with the high chlorine content feed was performed first; Test 1 with no feed
chlorine was performed last. Test 2 through 8 were performed in random order.
in each test. Through linear regression analysis, a linear estimate of the rate of concentration
increase was determined and is indicated on each graph.
Experimental and laboratory data were reduced and analyzed to address the
experimental objectives discussed in Section 1. Data obtained during this program provided
information on the following:
• Distribution of metals among the kiln ash, Que gas, and scrubber liquor streams
• Effects of the primary test variables of kiln exit temperature, afterburner exit
temperature, and waste feed chlorine content on metal distributions
• Afterburner exit and scrubber exit flue gas metal distributions between the solid
(particulate) and the vapor/dissolved (impinger catch) phase
• Distribution of metals in the particulate in the afterburner exit Que gas by particle
size range
• Apparent scrubber collection efficiency for each metal
38
-------
The test results related to the above are summarized and discussed in the following
subsections. Relative metal discharge distributions from all nine tests are first summarized and
discussed, independent of the RKS operating conditions, in Section 4.4.1. The relationships
between the discharge distributions of the metals and the RKS operating conditions are then
discussed in Section 4.4.2. The tables in Section 4.4.1 therefore contain only ranges and averages
of test results; the individual values obtained for each test are contained in the tables and figures
in Section 4.4.2.
Using the format established in Sections 4.4.1 and 4.4.2, the flue gas phase distribution
data are first summarized, then discussed with respect to variations in incineration operating
conditions, in Section 4.4.3. Section 4.4.4 discusses the afterburner particulate data in similar
order. Section 4.4.5 discusses the apparent scrubber removal efficiencies for each metal.
For many samples, laboratory analysis results showed that several metals were not
detected at sample-specific and metal-specific detection limits. Ill the data evaluation, two
calculations were performed: one with the analysis result assumed to be zero, and one with the
analysis result assumed to be the detection limit. In many cases, distribution conclusions were
not affected. For these cases the values given were derived by assuming the metals were present
in the sample at the detection limit. However, in those cases where distribution conclusions
could be affected, distributions are reported as ranges.
4.4.1 Average Trace Metal Discharge Distributions
As previously noted, a major objective of this test series was to identify the relative
distribution of each metal among the several RKS discharge streams. Table 23 summarizes the
relative distributions between the kiln ash and afterburner exit flue gas. Table 24 is a similar
summary of the relative distributions among the kiln ash, scrubber exit flue gas, and scrubber
liquor.
In both tables, the first set of columns represents the fraction of the metal fed accounted
for by the noted discharge (e.g., kiln ash, afterburner exit flue gas). The range (low, high)
exhibited over the nine tests performed and the average for all nine of the tests are also noted.
The second set of columns in the tables represents fractions normalized to the total
amount measured in the discharge streams, closed around the afterburner exit or the scrubber
discharges. Two discharge streams were considered for closure around the afterburner exit: the
kiln ash discharge and the afterburner exit flue gas. Three discharge streams were considered
for closure around the scrubber discharges: the kiln ash discharge, the scrubber liquor, and the
scrubber exit flue gas. The normalized fractions represent discharge distributions as they would
have been had mass balance closure for the metal been 100 percent; the sum of the normalized
values for each element in the discharge streams in each table is 100.
Since the first set of columns are percent of metal fed, the kiln ash fractions are the
same in Tables 23 and 24. However, normalized kiln ash values change between the two tables
because mass balance closures experienced around the kiln ash/afterburner exit flue gas
discharges differed from those experienced around the kiln ash/scrubber system discharges.
39
-------
TABLE 23. SUMMARY OF METAL DISCHARGE DISTRIBUTIONS IN THE KILN ASH AND
AFTERBURNER EXIT FLUE GAS
Distribution Normalized distribution
(% of metal fed) (% of total measured) olatility
Metal
Low
High
Average
Low
High
Average
iciupciaiuic
°C (CF)'
Kiln ash
Arsenic
38
90
59
89
96
93
32 (90)
Barium
17
60
26
92
99.6
97
849 (1560)
Bismuth
5
41
21
25
87
72
621 (1150)
Cadmium
3
32
13
7
75
46
216(420)
Copper
23
75
44
83
%
91
1116 (2040)
Lead
43
148
90
77
98
94
627 (1160)
Magnesium
62
123
99
99.8
99.97
99.93
1549 (2820)
Strontium
15
60
27
98.4
99.9
99.4
1454 (2650)
Afterburner
exit flue gas
Arsenic
3
10
5
4
11
7
Barium
0.1
1.5
0.6
0.4
8
3
Bismuth
3
48
9
13
75
28
Cadmium
7
43
15
25
93
54
Copper
2
11
5
4
17
9
Lead
2
13
5
2
23
6
Magnesium
0.03
0.13
0.06
0.03
0.2
0.07
Strontium
0.05
0.29
0.13
0.1
1.6
0.6
Temperature at which the effective vapor pressure of the metal is 10"6 atm8.
40
-------
TABLE 24. SUMMARY OF METAL DISCHARGE DISTRIBUTIONS IN THE KILN ASH,
SCRUBBER EXIT FLUE GAS, AND SCRUBBER LIQUOR
Distribution
(% of metal fed)
Normalized distribution
(% of total measured)
Volatility
Metal
Low
High
Average
Low
High
Average
IV U1UI w
°C (°F)'
Kiln ash
Arsenic
38
90
59
80
95
89
32 (90)
Barium
17
60
26
87
99
95
849 (1560)
Bismuth
5
41
21 .
9
76
41
621 (1150)
Cadmium
3
32
13
7
60
25
216(420)
Chromium
70
199
146
90
98
94
1613 (2935)
Copper
23
75
44
77
95
86
1116 (2040)
Lead
43
148
90
71
92
82
627 (1160)
Magnesium
62
123
99
99.4
99.8
99.7
1549 (2820)
Strontium
15
60
27
94
99
96
1454 (2650)
Scrubber exit flue gas
Arsenic
2
6
4
3
10
6
Barium
0.1
0.5
0.3
0.2
2
1
Bismuth
7
31
18
13
60
35
Cadmium
13
52
30
32
80
59
Chromium
3
9
5
2
9
4
Copper
3
8
5
3
18
10
Lead
4
23
13
7
24
12
Magnesium
0.02
0.1
0.07
0.01
0.2
0.1
Strontium
0.1
0.7
0.4
0.1
3
2
Scrubber liquor
Arsenic
2
7
3
2
11
5
Barium
0.3
2.4
0.9
0.6
11
4
Bismuth
3
23
12
5
44
24
Cadmium
2
14
8
4
35
16
Chromium
0.8
5.7
2.8
1
4
2
Copper
0.3
3.5
1.7
I
9
4
Lead
0.4
17
7
1
14
6
Magnesium
0.1
0.4
0.2
0.1
0.4
0.2
Strontium
0.1
0.7
0.4
0.6
4
2
"S
Temperature"at which the effective vapor pressure of the metal is 10"6 atm8.
41
-------
Chromium is not included in Table 23. The afterburner exit flue gas sampling train
contained a stainless steel condenser between the alundum thimble filter (employed to collect
the large quantities of particulate needed for size fractionation) and the impinger train. As a
consequence, concentrations of chromium in all afterburner exit sampling train impinger
solutions far exceeded what could have realistically been expected in the flue gas.
The last columns in Tables 23 and 24 contain a value termed volatility temperature8.
This is the temperature at which the effective vapor pressure of the metal is 10"* atm. The
effective vapor pressure is the sum of the equilibrium vapor pressure of all species containing
the metal. It reflects the quantity of metal that would vaporize under a given set of conditions.
A vapor pressure of 10"6 aim is selected because it represents a measurable amount of
vaporization. The lower the volatility temperature of the metal, the more volatile it is expected
to be.
Figure 5 summarizes the data in Table 23 for the discharge distributions, expressed as
a percent of the metal fed. Figure 6 similarly summarizes the data in Table 24 for the
normalized discharge distributions of each metal closed around the kiln ash, scrubber exit flue
gas, and scrubber liquor. In both figures, the bar for each metal represents the range in the
fraction accounted for by each discharge stream over the nine tests, with the average fraction
over the nine tests noted by the midrange tick mark. The metal discharge distribution data are
plotted versus the metal's volatility temperature.
Figure 6 indicates a correspondence between observed volatility and volatility
temperature for all the metals tested, except arsenic. With the exception of arsenic, average
normalized kiln ash fraction s generally increased with increasing volatility temperature. Based
on the normalized data, cadmium and bismuth tended to be relatively volatile and were less
prevalent in the kiln ash than were the more refractory metals. Kiln ash accounted for the major
fraction of arsenic, lead, barium, copper, strontium, magnesium, and chromium. Note that the
shift between volatile behavior and refractory behavior occurs in the general range of kiln
temperatures used in these tests (816° to 927°C (1500° to 1700°F)).
Based on volatility temperature, arsenic is expected to be the most volatile element, and
on that basis should be found at the lowest percentage in the kiln ash. However, the data show
arsenic to be apparently refractory, remaining largely with the kiln ash. The volatility
temperature for arsenic is based on the vapor pressure of As^j, the most volatile arsenic
compound under oxidizing combustion conditions. The fact that arsenic is significantly less
volatile than expected (were ASjO, the predominant arsenic species) suggests that either some
other, less volatile arsenic compound (perhaps an arsenate) was preferred, or that some other
chemical interaction, such as strong adsorption to the clay, occurred.
Table 25 summarizes the mass balance closures for each metal. Mass balance closure
represents the fraction of the feed metal that is accounted for by the discharges. Closures
around the kiln ash discharge and the afterburner exit Que gas, and around the kiln ash and
scrubber discharges (flue gas and scrubber liquor), are tabulated. The data in Table 25 show that
average mass balance closure ranged from 27 percent for barium to 99 percent for magnesium,
around the kiln ash/afterburner exit flue gas, with an overall average of 52 percent for all metals.
Individual metal closures ranged from 8 to 153 percent. Average mass balance closures around
42
-------
KILN ASH
I
co
< a
Z LU
zf
Q LL
§S
iz —'
200 400 ' 600 800 1000 1200
VOLATILITY TEMPERATURE fC)
SCRUBBER EXIT FLUE GAS
1400
1600
CD
g
? s
C LL
9 -i
£ K
O LU
W 2
Z LL
Z 9
Q si
f-
a
<
iz
60
50
40
30
20
10
-
Cd
.4.
¦Bi
-Pb
T As
i
Ba
1 £
Cu
Cr
I
X .
I
1 * ,
Sr
1 *
200 400 600 800 1000 1200
VOLATILITY TEMPERATURE ("C)
1400
1600
Figure 5. Distribution of metals in the discharge streams expressed as a percent of the
metal fed. Bar indicates range observed over all nine tests. Average is noted by
the midrange tick mark.
43
-------
KJLN ASH
100
200 400 600 800 1000 1200
VOLATILITY TEMPERATURE ('C)
SCRUBBER EXIT FLUE GAS
1400
1600
O 80
fxO 40
100
600 000 1000 1200
VOLATILITY TEMPERATURE ('C)
SCRUBBER UQUOR
tc Q
m UJ
cd cc
m =>
=> <£
cr <
w 2
Z
2
o
s
200 400 600 800 1000 1200
VOLATILITY TEMPERATURE ("C)
1400
1600
I
1400
1600
Figure 6. Normalized distribution or metals in the discharge streams. Bar Indicates range
observed over all nine tests. Average is noted by the midrange tick mark.
44
-------
TABLE 25. SUMMARY OF METAL MASS BALANCE CLOSURE
Mass balance closure
(% of metal fed)
Closed around kiln ash discharge Closed around kiln ash discharge
and afterburner exit flue gas and scrubber discharges
Metal
Low
High
Average
Low
High
Average
Arsenic
43
100
63
47
95
66
Barium
17
60
27
17
60
27
Bismuth
8
63
31
35
63
50
Cadmium
13
46
28
36
68
50
Chromium
a
—
—
77
204
154
Copper
28
85
48
30
81
50
Lead
56
153
95
47
177
110
Magnesium
63
123
99
63
123
99
Strontium
15
60
27
15
60
28
= Not applicable.
the kiln ash and scrubber discharges ranged from 27 percent for barium to 154 percent for
chromium. The overall average closure for all metals was 65 percent. Individual metal closures
ranged from 15 to 204 percent.
Figure 7 summarizes the data in Table 25 for the mass balance closures achieved around
the kiln ash and scrubber discharges. The bar for each metal represents the range of closures
over the nine tests, with the average noted by the midrange tick mark.
There are several explanations for the imperfect mass balance closures experienced
during these tests. Slag buildup in the afterburner chamber and particulate accumulation in the
connecting ductwork were observed during the test series. An accumulation of sludge was also
observed in the bottom of the scrubber and in the recirculation tank. These deposits were very
likely sources of unaccounted-for metal loss.
In addition, it is possible that in some cases the sample digestion method
(Method 30502) was not sufficiently aggressive in completely liberating the metals from the solid
samples. This is suspected to have occurred to some extent since in some cases an estimated
90 percent of the solid material remained following the digestion process. In accordance with
the method, the digestate was subjected to analysis for metals; the remaining solid material was
not analyzed further.
45
-------
220
200
—180
S £'160
<£<140
< tn lH 120
S 2 J100
3°0 80
2 sl 60
40
20
0
Figure 7. Summary of mass balance closures around the kiln ash, scrubber exit flue gas and
scrubber liquor. Bar Indicates range observed over all nine tests. Average is noted
by the midrange tick mark.
This observation could explain the high closures observed for chromium. As previously
discussed, chromium was not included in the trace metal aqueous spike solution, but fed only by
virtue of its presence in the clay matrix. It is possible that the feed chromium was not fully
quantified in the feed due to incomplete recovery of chromium from the feed clay samples
during the sample digestion process. Having been exposed to the high temperatures of the kiln
in the presence of chlorine, the chromium in the ash may have become more amenable to
digestion.
Nevertheless, the mass balance closures achieved in these tests are considered quite
good. Typical good trace metal mass balance closure results, based upon past experience with
combustion sources, are in the 30 to 200 percent range.
4.42 EfTects of Incinerator Operating Conditions on Metal Distributions
A major objective of this test series was to identify the effects of the RKS operating
conditions and the waste feed chlorine content on the distribution of the metals among the
incinerator discharges. As discussed in Section 2, the test variables were kiln exit temperature,
afterburner exit temperature, and waste feed chJorine content. The discussion in Section 4.4.1
considered all nine parametric tests performed as a whole, without mention of variations in test
conditions. In this subsection, the data from each test are presented in a format that facilitates
interpretation with respect to the primary test variables.
Table 26 summarizes the metal discharge distributions and mass balances, with the
discharge distributions expressed in terms of the fraction of the metal fed. Table 27 summarizes
the normalized distributions in both the kiln ash and afterburner exit flue gas discharges, and in
the kiln ash and scrubber discharges. In cases where some samples were reported as containing
less than the detection limit of a given metal, the distributions are reported as ranges. Endpoints
of the ranges correspond to an assumed sample concentration of zero and a concentration
corresponding to the measurement detection limit. As discussed in Section 4.4.1, afterburner exit
flue gas chromium data are not reported.
i
I
46
-------
TABLE 26. METAL DISCHARGE DISTRIBUTIONS AND MASS BALANCE CLOSURE
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wt %)
i
Target:
R16
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit = 1093°C; chlorine =
4%
Kiln exit = 871°C; chlorine =
4%
Kiln exil = 871°C; AB exit = 1093°C
Test, average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
Discharge distribution (% of metal fed)
Arsenic
Kiln ash
59.0
37.8
55.4
68.0
53.0
52.8
37.8
55.4
68.0
53.2
58.8
37.8
55.4
68.0
90.0
Afterburner exit flue gas
2.5
4.7
3.1
4.8
3.7
5.4
4.7
3.1
4.8
2.6
3.4
4.7
3.1
4.8
10.3
Scrubber exit flue gas
0.6-3.1
1.7-4.9
1.1-3.8
0.9-4.9
1.8-4.6
0.7-1.8
1.7-4.9
1.1-3.8
0.9-4.9
2.4-5.6
0.7-4.7
1.7-4.9
1.1-3.8
0.9-4.9
0.4-2.7
Scrubber liquor
1.7-2.0
4.3
2.5
3.9
7.0
2.7
4.3
2.5
3.9
2.6
2.2
4.3
2.5
3.9
2.4
Mass balance closure
61.5
42.6
58.5
72.8
56.7
58.2
42.6
58.5
72.8
55.9
62.2
42.6
58.5
72.8
100.3
around kiln/afterburner
Mass balance closure
61-64
44-47
59-62
73-77
62-65
56-57
44-47
59-62
73-77
58-62
62-66
44-47
59 62
73-77
93 95
around kiln/scrubber
.u Rarium
Kiln ash
37.2
16.5
19.8
27.1
18.2
17.8
16.5
19.8
27.1
19.1
20
16.5
19.8
27.1
59.9
Afterburner exit flue gas
0.2
0.4
0.1
0.8
0.7
1.5
0.4
0.1
0.8
0.7
0.4
0.4
0.1
0.8
0.3
Scrubber exit flue gas
0.3
0.1
0.3
0.4
0.3
0.1
0.1
0.3
0.4
0.4
0.5
0.1
0.3
0.4
o
©
LA
©
Scrubber liquor
0.3
0.5
0.4
0.9
0.7
0.8
0.5
0.4
0.9
2.4
1.4
0.5
0.4
0.9
0.4
Mass balance closure
37.4
16.9
19.9
27.9
18.9
19.3
16.9
19.9
27.9
19.8
20.4
16.9
19.9
27.9
60.2
around kiln/afterbumer
Mass balance closure
37.8
17.1
20.5
28.4
19.2
18.7
17.1
20.5
28.4
21.9
21.9
17.1
20.5
28.4
60.4
around kiln/scrubber
nismuth
Kiln ash
41.3
7.3
28.6
30.5
4.8
22.2
7.3
28.6
30.5
10.6
15.5
7.3
28.6
30.5
29.2
Afterburner exit flue gas
6.2
4.5
4.2
5.0
3.0
4.6
4.5
4.2
5.0
3.6
47.8
4.5
4.2
5.0
5.7
Scrubber exit flue gas
10.4
18.8
21.8
16.4
24.4
15.7
18.8
21.8
16.4
31.4
12.5
18.8
21.8
16.4
6.5
Scrubber liquor
2.8
16.0
12.9
11.6
22.9
8.2
16.0
12.9
11.6
11.2
7.3
16.0
12.9
11.6
12.9
Mass balance closure
47.4
11.8
32.8
35.5
7.7
26.8
11.8
32.8
35.5
14.2
63.3
11.8
32.8
35.5
34.9
around kiln/afterburner
Mass balance closure
54.5
42.1
63.2
58.5
52.1
46.1
42.1
63.2
58.5
53.1
35.3
42.1
63.2
58.5
48.6
around kiln/scrubber
(continued)
-------
TABLE 26. (continued)
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wt %)
Target:
816
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit = 1093°C; chlorine =
4%
Kiln exit = 871°C; chlorine =
4%
Kiln exit = 871
°C; AR exit = 1093°C
Test average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
Discharge distribution (% of metal fed)
Cadmium
Kiln ash
31.8
5.0
11.7
18.1
<3.0
16.0
5.0
11.7
18.1
7.2
<3.0
5.0
11.7
18.1
17.4
Afterburner exit flue gas
10.6
12.7
6.8
8.8
10.0
13.9
12.7
6.8
8.8
8.6
42.9
12.7
6.8
8.8
19.9
Scrubber exit flue gas
19.0
37.3
31.6
30.9
23.7
30.9
37.3
31.6
30.9
51.8
28.7
37.3
31.6
30.9
12.7
Scrubber liquor
2.3
7.7
6.6
10.4
14.1
8.8
7.7
6.6
10.4
8.6
4.4
7.7
6.6
10.4
9.1
Mass balance closure
42.4
17.7
18.4
26.9
10-13
29.9
17.7
18.4
26.9
15.8
43-46
17.7
18.4
26.9
37.3
around kiln/afterburner
Mass balance closure
53.0
50.0
49.9
59.4
38-41
55.7
50.0
49.9
59.4
67.7
33-36
50.0
49.9
59.4
39.2
around kiln/scrubber
"
Chromium
Kiln ash
166.4
182.5
145.8
152.0
125.0
135.7
182.5
145.8
152.0
139.6
69.9
182.5
145.8
152.0
199.0
Afterburner exit flue gas
i
i
—
—
—
—
—
—
—
—
—
—
—
—
—
Scrubber exit flue gas
1.2-4.2
2.6-5.6
2.5-4.9
1.1-3.9
1.7-5.0
3.5
2.6-5.6
2.5-4.9
1.1-3.9
5.7-8.9
6.7
2.6-5.6
2.M.9
1.1-3.9
0.4-3.1
Scrubber liquor
2.6
2.4
1.2
4.9
1.9
5.7
2.4
1.2
4.9
4.0
0.8
2.4
1.2
4.9
1.8
Mass balance closure
around kiln/afterburner
Mass balance closure
173.2
190.6
151.9
160.7
131.9
144.9
190.6
151.9
160.7
152.6
77.4
190.6
151.9
160.7
203.9
around kiln/scrubber
Copper
Kiln ash
44.2
23.0
39.9
46.8
40.1
36.6
23.0
39.9
46.8
36.6
53.9
23.0
39.9
46.8
74.6
Afterburner exit flue gas
2.5
4.8
5.6
3.8
4.6
3.2
4.8
5.6
3.8
1.7
3.5
4.8
5.6
3.8
10.6
Scrubber exit flue gas
3.6
4.5
5.5
6.1
4.3
3.9
4.5
5.5
6.1
8.2
2.6
4.5
5.5
6.1
2.5
Scrubber liquor
0.5
2.5
1.4
1.2
3.5
1.0
2.5
1.4
1.2
1.2
0.3
2.5
1.4
1.2
3.5
Mass balance closure
46.7
27.8
45.5
50.7
44.7
39.8
27.8
45.5
50.7
38.2
57.4
27.8
45.5
50.7
85.2
around kiln/afterburner
Mass balance closure
48.3
30.0
46.9
54.2
47.9
41.5
30.0
46.9
54.2
45.9
56.8
30.0
46.9
54.2
80.6
around kiln/scrubber
¦— = Not applicable. (continued)
-------
TABLE 26. (continued)
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wt %)
Target:
816
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit = 1093°C; chlorine =
4%
Kiln exit = 871
°C; chlorine = 4%
Kiln exil
r-
00
II
°C; AB exil = 1093°C
TesL average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
Discharge distribution (% of metal fed)
Lead
Kiln ash
119.3
110.1
76.9
86.4
67.3
88.4
110.1
76.9
86.4
75.0
42.5
110.1
76.9
86.4
147.6
Afterburner exit flue gas
4.4
2.8
9.5
1.8
2.0
. 4.3
2.8
9.5
1.8
2.1
13.0
2.8
9.5
1.8
4.9
Scrubber exit flue gas
10.0
16.5
11.9
11.1
14.1
11.2
16.5
11.9
11.1
23.0
1.5-4.4
16.5
11.9
11.1
12.5
Scrubber liquor
1.1
9.6
5.2
4.6
13.1
4.4
9.6
5.2
4.6
5.3
0.4
9.6
5.2
4.6
17.3
Mass balance closure
123.6
112.8
86.4
88.2
69.4
92.7
112.8
86.4
88.2
77.1
55.5
112.8
86.4
88.2
152.5
around kiln/afterburner
Mass balance closure
130.3
136.2
94.0
102.1
94.5
104.1
136.2
94.0
102.1
103.3
47.4
136.2
94.0
102.1
177.3
around kiln/scrubber
Magnesium
Kiln ash
111.3
104.9
101.2
106.1
83 J
96.4
104.9
101.2
106.1
101.0
62.4
104.9
101.2
106.1
122.9
Afterburner exit flue gas
0.06
0.05
0.03
0.M
0.05
0.1
0.05
0.03
0.04
0.03
0.1
0.05
0.03
0.04
0.05
Scrubber exit flue gas
0.07
0.02
0.1
0.09
0.1
0.05
0.02
0.1
0.09
0.1
0.09
0.02
0.1
0.09
0.02
Scrubber liquor
0.09
0.2
0.1
0.09
0.4
0.2
0.2
0.1
0.09
0.1
0.1
0.2
0.1
0.09
0.2
Mass balance closure
111.3
105.0
101.2
106.1
83.4
96.5
105.0
101.2
106.1
101.1
62.6
105.0
101.2
106.1
123.0
around kiln/afterburner
Mass balance closure
111.4
105.1
101.4
106.2
83.8
96.6
105.1
101.4
106.2
101.2
62.6
105.1
101.4
106.2
123.1
around kiln/scrubber
Strontium
Kiln ash
36.9
14.8
19.9
26 J
18.4
18.0
14.8
19.9
26.3
20.3
26.3
14.8
19.9
26.3
59.7
Afterburner exit flue gas
0.1
0.05
0.1
0.1
0.1
0.3
0.05
0.1
0.1
0.1
0.2
0.05
0.10
0.1
0.1
Scrubber exit flue gas
0.4
0.1
0.6
0.7
0.5
0.2
0.1
0.6
0.7
0.6
0.7
0.1
0.6
0.7
0.1
Scrubber liquor
0.3
0.2
0.4
0.4
0.7
0.5
0.2
0.4
0.4
0.4
0.1
0.2
0.4
0.4
0.3
Mass balance closure
37.0
14.8
20.0
26.4
18.5
18.3
14.8
20.0
26.4
20.4
26.5
14.8
20.0
26.4
59.8
around kiln/afterburner
Mass balance closure
37.6
15.1
20.9
27.4
19.6
18.7
15.1
20.9
27.4
21.3
27.1
15.1
20.9
27.4
60.1
around kiln/scrubber
-------
TABLE 27. NORMALIZED METAL DISCHARGE DISTRIBUTIONS
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wt %)
Target:
816
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit = 1093°C; chlorine =
4%
Kiln exit = 871°C; chlorine =
4%
Kiln exit
= 871°C; AB exit = 1093°C
Test average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
Normalized discharge distribution (%)
Arsenic
Kiln ash
95.9
88.9
94.7
93.4
93.5
90.7
88.9
94.7
93.4
95.2
94.5
88.9
94.7
93.4
89.7
Afterburner exit flue gas
4.1
11.1
5.3
6.6
6.5
9.3
11.1
5.3
6.6
4.8
5.5
11.1
5.3
6.6
10.3
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Kiln ash
92-96
80-87
90-94
89-94
82-86
92-94
80-87
90-94
89-94
87-92
89-95
80-87
90-94
89-94
95-97
Scrubber exit flue gas
1-5
4-11
2-6
1-6
3-7
1-3
4-11
2-6
1-6
4-9
1-7
4-11
2-6
1-6
0.5-2
Scrubber liquor
3
9
4
5
11
5
9
4
5
4
4
9
4
5
3
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Barium
Kiln ash
99.5
97.6
99.4
97.0
96.3
92.3
97.6
99.4
97.0
96.2
97.9
97.6
99.4
97.0
99.6
Afterburner exit flue gas
0.5
2.4
0.6
3.0
3.7
7.7
2.4
0.6
3.0
3.8
2.1
2.4
0.6
3.0
0.4
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Kiln ash
98.6
96.3
96.3
95.4
94.8
94.9
96.3
96.3
95.4
87.3
91.2
96.3
96.3
95.4
99.3
Scrubber exit flue gas
0.7
0.9
1.7
1.5
1.6
0.8
0.9
1.7
1.5
2.0
2.2
0.9
1.7
1.5 0.1-0.2
Scrubber liquor
0.8
2.8
2.0
3.1
3.6
4.2
2.8
2.0
3.1
10.8
6.6
2.8
2.0
3.1
0.6
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Bismuth
Kiln ash
87.0
61.8
87.1
86.0
61.4
82.8
61.8
87.1
86.1
72.7
24.5
61.8
87.1
86.1
83.7
Afterburner exit flue gas
13.0
38.2
12.9
14.0
38.6
17.2
38.2
12.9
14.0
27.3
75.5
38.2
12.9
14.0
16.3
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Kiln ash
75.8
17.3
45.2
52.2
9.1
48.1
17.3
45.2
52.2
18.5
44.1
17.3
45.2
52.2
60.1
Scrubber exit flue gas
19.0
44.6
34.5
28.1
46.9
34.1
44.6
34.5
28.1
60.1
35.3
44.6
34.5
28.1
13.4
Scrubber liquor
5.2
38.1
20.4
19.8
44.0
17.8
38.1
20.4
19.8
21.4
20.6
38.1
20.4
19.8
26.5
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
(continued)
-------
TABLE 27. (continued)
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wl %)
Target:
816
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit = 1093°C; chlorine =
4%
Kiln exit = 871°C; chlorine =
4%
Kiln exit
= 871°C; AB exit = 1093°C
Test average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
Normalized discharge distribution (%)
Cadmium
Kiln ash
75.0
28.2
63.3
67.4
<23
53.6
28.2
63.3
67.4
45.4
<7
28.2
63.3
67.4
46.7
Afterburner exit flue gas
25.0
71.8
36.7
32.6
>77
46.4
71.8
36.7
32.6
54.6
>93
71.8
36.7
32.6
53.3
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Kiln ash
59.9
10.0
23.4
30.5
<7
28.7
10.0
23.4
30.5
10.6
<8
10.0
23.4
30.5
44.5
Scrubber exit flue gas
35.7
74.6
63.4
52.0
58-61
55.5
74.6
63.4
52.0
76.6
80-86
74.6
63.4
52.0
32.3
Scrubber liquor
4.3
15.4
13.2
17.4
35-39
15.8
15.4
13.2
17.4
12.8
12-14
15.4
13.2
17.4
23.2
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Chromium
Kiln ash
a
Afterburner exit flue gas
Total
97.6
Kiln ash
96.1
95.8
96.0
94.6
94.8
93.7
95.8
96.0
94.6
91.5
90.3
95.8
96.0
94.6
Scrubber exit flue gas
0.7-2.4
1.4-3.0 1.7-3.2 0.7-2.4 1.3-3.8
2.5
1.4-3.0
1.7-3.2 0.7-2.4 3.8-5.8
8.7 1.4-3.0 1.7-3.2 0.7-2.4 0.2-1.5
Scrubber liquor
1.5
1.3
0.8
3.0
1.5
3.8
1.3
0.8
3.0
2.6
1.1
1.3
0.8
3.0
0.9
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Copper
Kiln ash
94.7
82.7
87.7
92.4
89.7
92.1
82.7
87.7
92.4
95.5
93.8
82.7
87.7
92.4
87.6
Afterburner exit flue gas
5.3
17.3
12.3
7.6
10.3
7.9
17.3
12.3
7.6
4.5
6.2
17.3
12.3
7.6
12.5
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Kiln ash
91.5
76.6
85.2
86.5
83.7
88.2
76.6
85.2
86.5
79.0
94.9
76.6
85.2
86.5
92.5
Scrubber exit flue gas
7.5
14.9
11.8
11.3
8.9
9.4
14.9
11.8
11.3
18.3
4.6
14.9
11.8
11.3
3.1
Scrubber liquor
0.9
8.5
3.0
2.2
7.4
2.4
8.5
3.0
2.2
2.7
0.6
8.5
3.0
2.2
4.4
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
= Not applicable.
(continued)
-------
TABLE 27. (continued)
Test:
2
4
7
8
3
6
4
7
8
5
I
4
7
8
9
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wt %)
Target:
816
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit = 1093°C; chlorine =
4%
Kiln exit = 871°C; chlorine =
4%
Kiln exit
= 871°C; AB exit = 1093°C
Test average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
Normalized discharge distribution (%)
Lead
Kiln ash
96.5
97.6
89.0
98.0
97.1
95.4
97.6
89.0
98.0
97.0
76.7
97.6
89.0
98.0
96.8
Afterburner exit flue gas
3.5
2.4
11.0
2.0
2.9
4.6
2.4
11.0
2.0
3.0
23.3
2.4
11.0
2.0
3.2
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Kiln ash
91.5
80.8
81.9
84.6
71.3
R5.0
80.8
81.9
84.6
70.9
90-96
80.8
81.9
84.6
83.2
Scrubber exit flue gas
7.7
12.1
12.6
10.9
15.0
10.8
12.1
12.6
10.9
23.7
3-9
12.1
12.6
10.9
7.0
Scrubber liquor
0.9
7.1
5.5
4.5
13.8
4.3
7.1
5.5
4.5
5.4
1
7.1
5.5
4.5
9.7
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Magnesium
Kiln ash
99.95
99.95
99.97
99.96
99.94
99.87
99.95
99.97
99.96
99.97
99.80
99.95
99.97
99.96
99.96
Afterburner exit flue gas
0.05
0.05
0.03
0.04
0.06
0.13
0.05
0.03
0.04
0.03
0.20
0.05
0.03
0.04
0.04
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Kiln ash
99.85
99.81
99.77
99.83
99.44
99.78
99.81
99.77
99.83
99.74
99.66
99.81
99.77
99.83
99.84
Scrubber exit flue gas
0.07
0.02
0.10
0.09
0.12
0.05
0.02
0.10
0.09
0.12
0.15
0.02
0.10
0.09
0.01
Scrubber liquor
0.08
0.17
0.13
0.08
0.44
0.17
0.17
0.13
0.08
0.13
0.19
0.17
0.13
0.08
0.15
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Strontium
Kiln ash
99.6
99.7
99.5
99.5
99.1
98.4
99.7
99.5
99.5
99.6
99.3
99.7
99.5
99.5
99.9
Afterburner exit flue gas
0.4
0.3
0.5
0.5
0.9
1.6
0.3
0.5
0.5
0.4
0.7
0.3
0.5
0.5
0.1
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Kiln ash
98.2
97.7
95.5
95.9
93.7
96.2
97.7
95.5
95.9
95.1
96.4
97.7
95.5
95.9
99.3
Scrubber exit flue gas
0.9
0.7
2.6
2.5
2.5
1.4
0.7
2.6
2.5
2.9
2.4
0.7
2.6
2.5
0.1
Scrubber liquor
0.9
1.6
1.8
1.6
3.8
2.4
1.6
1.8
1.6
2.0
1.2
1.6
1.8
1.6
0.6
Total
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
-------
Three groupings of tests appear in Tables 26 and 27. Each grouping corresponds to a
series of tests in which one test variable (kiln exit temperature, afterburner exit temperature, or
waste feed chlorine content) was varied, with the other two variables held nominally constant at
their midpoints. The center point, which was tested in triplicate (Tests 4, 7 and 8), is included
in all groupings. The data in Tables 26 and 27 represent the individual test data that served as
the basis for the ranges and averages noted in Tables 23 and 24.
Figures 8 through 25 show the variations in the discharge distributions for each of the
test metals. There are two figures for each metal. The first shows the discharge distributions
through the kiln ash, afterburner exit flue gas, scrubber exit flue gas and the scrubber liquor
expressed as a fraction of the metal fed. The second shows the discharge distributions
normalized to the total measured in the kiln ash, scrubber exit flue gas and the scrubber liquor.
Both methods of presenting metal discharge distributions are useful in the process of
data interpretation. Expressing distributions as a percent of the metal fed provides a more
accurate representation of individual discharge stream fractions if there are metal losses in the
incineration system that are not quantified. As discussed, there is some evidence of such losses,
possibly affecting all discharge streams. However, this method introduces a source of test-to-test
data variability caused by less than perfect mass balance closure.
The use of normalized distributions corrects for this source of data variability. Mass
balance closure varied from test to test in this program, as it has in all past experience in
measuring trace metal discharges from combustion sources. Thus, variations in metal discharge
distributions expressed as a percent of the metal fed are affected by both actual variations due
to changes in test variables, and by individual test mass balance closure. Variations in
normalized distributions more closely reflect those due to changes in test variables by "forcing"
100 percent mass balance closure for each test.
The following discussion incorporates distributions presented as both a fraction of the
metal fed and normalized to the total amount measured. TTie greatest credence is placed on
conclusions supported by both methods of presentation.
Three bar charts are shown in each figure corresponding to the three groupings of tests
in Tables 26 and 27. Thus, each bar chart in Figures 8 through 25 corresponds to a series of
tests in which one test variable was varied, with the other two variables held nominally constant.
In these figures, values for the replicate tests (Tests 4, 7 and 8) have been averaged and
plotted as a single bar. The observed range for these three tests is also indicated. Since Tests
4, 7 and 8 are replicate tests conducted under the same nominal test conditions, they are
indications of degree of data variability.
Before proceeding, the general effects of chlorine on the distribution of metals to the
kiln ash fraction should be noted. With the exception of bismuth and cadmium, metal
distributions to the kiln ash represented as a fraction of the metal fed increased significantly
when waste feed chlorine content increased from 4 to 8 percent. (The data for bismuth and
cadmium are inconclusive based on the observed range of ash distributions in the replicate tests.)
53
-------
Variable: Klin Exit Temperature
FraoHon ol Matal F«d (St
100 - —
BO
eo
40
20
0
Kiln itfi *8 tluf SI* SE flue gaa llquot
100
80
60
40
20
0
Kiln ttn AB flu* gat 8E Hue gas Uouor
Variable: Waste Feed Chlorine Content
Fraction of Malsi fad (*!
100
BO
ao
*o
20
O
Figure 8. Arsenic discharge distributions expressed as a percent of the metal fed,
54
Bl»c Dine CDuic
Variable: Afterburner Exit Temperature
Ftsoifon el M«t«i F«d («)
E308? c CD 1088 c Oa«M c
*
10
T.t-
m
Bo* ~< * CDs *
0~
if
zir.
a
it
6-
f^T]
Ssig
TOftS 1
0
s
B
1
Klin «ah AS Hub qib SE (lu> oat Liquor
-------
Variable: Kiln Exit Temperature
Fraollan of lots! Measured (*)
awe CDenc CJb27 0
00
BO
Variable: Afterburner Exit Temperature
Fioollon ol Total Measired (2)
ioa
0B2 C CD 1063 C CD 1204 c
00
BO
se Hue e»s
Variable: Waste Feed Chlorine Content
Fiaollon ot Total Measured (1)
16
E3o» ~ < l ~) »
tt
10
_
"1
!
i
1
0
-H
11
d
Kim Mh 8E hue gu Liquor
Figure 9. Normalized arsenic discharge distributions.
55
-------
Variable: Kiln Exit Temperature
60
60
40
30
20
10
Fraction of Motal Fed
-------
Variable.- Kiln Exit Temperature
Fraolior o1 Tolat Mmii/sK (S)
lOCr
06
eo
86
10
—0
Batec CDsnc Oasrc
ISM
JB2
Kiln ash
SE Hue oea
Liquoi
Variable: Afterburner Exit Temperature
Fraction ol Total Msaairsd {*)
ioq
oar C CD *003 C CD *204 c
06
eo
BG
Variable: Waste Feed Chlorine Content
Fraotlon of Total Measured (ft)
KMn uh SE Hue Qtl LlQUOr
Figure 11. Normalized barium discharge distributions.
57
-------
Variable: Klin Exit Temperature
Fraction or Maul Fad (V)
60
<0
30
20
10
Kiln ash AB (Iub gas SE Hue gas Liquor
Variable: Afterburner Exit Temperature
Fieolion oi Melal Fad (l)
1 982 C O 1099 C C3 1204 C
1
Kiln ash AS flua gas SE llu« gas Uouor
Variable: Waste Feed Chlorine Content
Fraollon ol Malel Fad 4*)
Kiln uh
A6 llua gu
8E llua gaa
Liquor
Figure 12. Bismuth discharge distributions expressed as a percent of the metal Fed.
58
-------
Variable: Kiln Exit Temperature
Fr Kit Ion Of Tola! Measured
ESewC Obtic C3s2r c
wxvvsyt
KKn Mh
8E fius qm
LfQuor
Variable: Afterburner Exit Temperature
Frsotion of Total Maasurad |*J
mz c ~ toes c dl t?cw c
Kllrt aati 8E Hue gas - LICUO*
Variable: Waste Feed Chlorine Content
Fraction of Tola! Maaiurad <«)
ft D« •
SSM.'f
Kiln iah
SE tiua dm
Liquor
Figure 13. Normalized bismuth discharge distributions.
59
-------
Variable: Klin Exit Temperature
eo
60
40
30
20
10
0
Variable: Afterburner Exit Temperature
Fr aclion ot Metal Fea <%)
ec
40
30
20
10
0
eo
60
40
30
20
10
0
Kiln aih ab Hue gas 5E due gaa Liquor
Figure 14. Cadmium discharge distributions expressed as a percent of the metal fed.
Frao'lon ol Mats' Fed (B)
Haiec CDaric c
:
1
If
1
i
W-r
1
-
fp
a
rSr
Kim aah AB Hue gas SE Hue gaa Liquor
ESeeic CD toss c ~ t2o< c
Jt
P
Ml
m
KB
liil
Kiln asfi
AB Hue gas
SE flu* gaa
Liquor
Variable: Waste Feed Chlorine Content
Fraotlon of Metal Fad (4)
-------
Variable*. Kiln Exit Temperature
60
eo
40
20
Kiln sBh SE Hue gaa Liquor
Variable: Afterburner Exit Temperature
Ftaolion of Tola Mesemad (*)
00
00
*0
20
Klin ash SE Hue oas Liquor
Variable: Waste Feed Chlorine Content
Freotion oi Tola) M»a»urad (4)
80
60
40
20
0
Figure 15. Normalized cadmium discharge distributions.
Fraotlon ot Total MMaurBd ft)
018 C
~ BTl C
~ WJC
•\r-L;
EE3 082 C
~ 1003 C
CD 1204 c
8
id
3
1
M
H
8
r^i
liquor
-------
Variable: Kiln Exit Temperature
200
160
100
60
Kiln »»h SE gae Lk*JO(
Variable: Afterburner Exit Temperature
Fraction ot Metal Fed {%)
200
160
100
to
0
Variable: Waste Feed Chlorine Content
Frsollon ot Metal fed 1%)
200
160
100
60
0
Figure 16. Chromium discharge distributions expressed as a percent of the metal Ted.
Fraction of M«tgl Fed (ft)
10
ESewc CZDs'i c G3«27c
Liquor
Kiln BBh
LHncx
-------
Variable: Kiln Exit Temperature
Fraolion of "lotai Measured (ft)
EEeiec lDbtic CUe?7c
Kim sen
SE flue gaa
LIQuOt
iocr
Variable: Afterburner Exit Temperature
Frootlon of Total Measured {%)
ESeasc C_J low c lZj i2cn c
>UV,Vvf
>v.wv
IlViWaWi
86-
Kiin aeh
SE flue QB6
Liquor
Variable: Waste Feed Chlorine Content
Fraction o! Tola' Measured {%)
Vr*i'
Klin aah
SE hue gas
Liquor
Figure 17. Normalized chromium discharge distributions.
63
-------
Variable: Kiln Exit Temperature
Frsollon of Metal Fed (*)
SO
10
at# c CDenc dtnc
60
40
20
BE tlue
Kiln
AB Hue OAS
Variable: Afterburner Exit Temperature
Fraotlcn at Melsl Fed (»)
60
10
b«2 c CD iocs c C3 ecu c
60
40
20
s.
Liquor
Variable: Waste Feed Chlorine Content
Fraotlon ot Mttel F«d (S)
eo
40
ZD
Afi due ga»
6E llus gu
Figure 18. Copper discharge distributions expressed as a percent of the metal Ted.
64
-------
Variable: Klin Exit Temperature
Fraollon of Total Mmii*ad (»)
10Q
ewe Dene CDOJ' c
20
60
m
60
a
70
Kiln ash
Variable: Afterburner Exit Temperature
Fraction ol Total Maaairad {%)
20
E3ae? c Giotic CDucmc
fix
11
PI
10
T
::rr
0
Sps
1.
Variable: Waste Feed Chlorine Content
Fraotion of Total Measured (S)
ioq
00
10
80
70
8E llua gaa
Figure 19. Normalized copper discharge distributions.
65
-------
160
126
100
76
60
26
Variable: Kiln Exit Temperature
FreoUor of Uslol Fed (•)
26
20
»C
Istec Dene E3e2'c
Kiln aeh AB Hue gas SE flue gss
LMjuoi
Variable: Afterburner Exit Temperature
160
126
100
Freotlon of Mela) Ped {%}
982 C
I I
a 1^04 c
160
126
100
Kiln aeh AS Hi* oae SE Hut qib Liquor
Variable: Waste Feed Chlorine Content
Freolton cl Mslal Fed |l)
Kiln uh
AB liue gti SE flu* gtt
Llquot
Figure 20. Lead discharge distributions expressed as a percent of the metal fed.
66
-------
Variable: Kiln Exit Temperature
Fraction of Totii Meaaised (®)
KHn ash 6E rtue Q&a LlQuor
Variable: Afterburner Exit Temperature
Fracdon of Total Measured (2)
10a
Klin ash SE flue Qaa Liquor
Variable: Wa9te Feed Chlorine Content
Fraollon ol Total Measured l%)
10CT
80
60
TO
20
lorn CZU« ~(!
,\ -.1
I
10
'
!"¦"
1Z,
' 3 J
0
'< ^Ut.
i
,t!SB
B itii _
WM
y||jj
KHn aaft
8E fiua oae Lfguof
Figure 21. Normalized lead discharge distributions.
67
-------
Variable: Kiln Exit Temperature
Fraotlon of Mal&l Fed {%)
126
eiec CDbtic OasTC
0-4
100
76
02
"I
60
26
0
Kiln ash
Variable: Afterburner Exit Temperature
Fraction oi Mate' Fad (%)
_5-
UffBiC O10C3C CDucm c
0.4
1
Ip
s
1
Wl\i
0.2
¦ "
I
m
II .
n
to
as
\\\\\w
m
iJ-L
i
0
KMn aah AB Hje pas SE flue Qaa Liquor
Variable: Waste Feed Chlorine Content
Fraction of Metal Fed (95)
126
0.4
100
76
0.2
60
0
Figure 22. Magnesium discharge distributions expressed as a percent of the metal fed.
68
-------
Variable: Kiln Exit Temperature
Fraction ot Total MMtKtd I#)
99(
1.0
0.6
Seiec mi 87i c CD m c
Klin ash 6E Hue gsa Uqw
Variable: Afterburner Exit Temperate
Fraalion ol Total Mtumd (%l
ioq
09. E
ee
1.0
PI
m
is
"is
IS'
i,
w
wm
0.5
)b8:o Dmiic C3 ooj c
¦i
Kiln ash 6E Mue gsa Liquor
Variable: Waste Feed Chlorine Content
Freotlon ol Total Maaaurad i%)
Kiln Mh 6E Hue gas Liquor
Figure 23. Normaliied magnesium discharge distributions.
69
-------
Variable: Kiln Exit Temperature
Fraotlon of M«tal Foe (%)
l8»C CD 871 c Qiztc
0.5
is
in
m
W,
»m\vs
i
B
K\V\V§
|i
p
Kiln aeh AS Hub gst 6E flue o»« Llquo'
Variable: Afterburner Exit Temperature
Fiaollon d1 Metal Fafl (X)
062 C LJ1000 C L_J 12CM C
eo
60
40
30
20
10
Klin aBh AB (Il» e«s SE 11 Lie gas Liquor
Variable: Waste Feed Chlorine Content
Fraollon of Malal Fad («.)
lot ~'< CDat
06
Kiln aah AB llua c«e 6E tlua gaa Liquor
Figure 24. Strontium discharge distributions expressed as a percent of the metal Ted.
70
-------
Variable: Kiln Exit Temperature
ioq
Fraction of Total MMiurid 1%)
Oe< c Cj07ic O027 c
mvM
Kiln aoh SE Hue Q06 LlQuOf
Variable: Afterburner Exit Temperature
Ffaotioi ol Tolei MseBmec {%)
ioq
B6
eo
10
1 ea: c CD toes c C3eo
-------
The same was true for chromium, lead and magnesium when waste feed chlorine content
increased from 0 to 4 percent.
The reason for this increase is unclear. For many metals, increased volatility evidenced
by lower ash fractions is expected during incineration in the presence of chlorine, since metal
chlorides are often more volatile than metal oxides or elemental metals. One explanation is that
exposure to kiln temperatures in the presence of chlorine leads to increased metal solubility,
resulting in increased recovery of metals from the ash during the process of sample digestion.
As discussed above, the digestion process often did not fully digest all the solid sample. Thus,
some portion of a given metal may have remained in the residual undigested solid that was not
subjected to further analysis.
The data suggest that, when incinerated in the presence of chlorine, metals remaining
in the kiln ash may become more susceptible to the digestion procedure used. This is expected
if metal chlorides are more soluble than metal oxides or elemental metal. The net effect is a
higher apparent recovery of metals in the ash with increased waste chlorine content.
The following discussion addresses the individual metal distribution data in turn.
Arsenic
Figures 8 and 9 show that, in these tests, neither changes in kiln nor in afterburner
temperature influenced the discharge distributions of arsenic within the limits of data variability
established by the triplicate test conditions. As discussed above, arsenic is expected to be much
more volatile than observed in these tests, based on its volatility temperature. Given this
expected greater volatility, discharge distributions might have been expected to vary with kiln
temperature. The apparent lack of kiln temperature effect, together with the relatively high
fraction discharged with kiln ash, indicates that arsenic behaved as a relatively refractory metal
in the kiln during these tests.
Barium
Figures 10 and 11 indicate that with increased kiln temperature there was a decrease
in the ash fraction of barium. Based on Figure 10, there were no clear effects of changes in
afterburner temperature or chlorine content on barium distributions.
Bismuth
Figures 12 and 13 show that increased kiln temperature caused a significant increase in
the volatility of bismuth, with significant decreases in the kiln ash fraction and corresponding
increases in the scrubber exit flue gas and scrubber liquor fractions. This relationship with
temperature is expected since the kiln temperatures used in these tests are in the general range
of the volatility temperature of bismuth. Neither afterburner temperature nor waste feed
chlorine content significantly altered the discharge distribution of bismuth, within the limits of
data variability established by the triplicate test condition.
72
-------
Cadmium
Figures 14 and 15 show that the volatility of cadmium, as with bismuth, also significantly
increased with increased kiln temperature. This is as expected, since the volatility temperature
of cadmium is somewhat below the kiln temperatures tested. Within the data variability, no clear
trends in the discharge distributions were observed with varying afterburner temperature.
Chromium
Figures 16 and 17 indicate that neither kiln temperature nor afterburner temperature
affected the discharge distributions of chromium within the limits of data variability,
Copper
Figures 18 and 19 show that neither kiln temperature nor afterburner temperature
affected copper discharge distributions, within the limits of data variability.
Lead
Figures 20 and 21 show that, as with bismuth and cadmium, increased kiln temperature
caused a noticeable increase in the volatility of lead, with significant decreases in the kiln ash
fraction and corresponding increases in the scrubber exit flue gas and scrubber liquor fractions.
This is as expected, since the kiln temperatures used in these tests are in the general range of
the volatility temperature of lead. Although the volatility of lead increased with higher kiln
temperature, lead still remained relatively refractory and was found primarily in the ash.
The scrubber exit flue gas fraction of lead nearly doubled as the afterburner temperature
increased from 1093° to 1204° C (2000° to 2200° F). It is unclear whether this variation is
related to changes in test conditions or possible measurement errors.
Magnesium
Figures 22 and 23 show a slight decrease in the kiln ash fraction of magnesium with
increasing kiln temperature. Neither afterburner temperature nor waste feed chlorine content
noticeably affected magnesium discharge distributions. The increase in magnesium recovery
from the kiln ash with increased feed chlorine content is consistent with the observations for the
other metals.
Strontium
Figures 24 and 25 indicate that there was a small decrease in the ash fraction of
strontium with increased kiln temperature. There were no clear effects of afterburner
temperature (or chlorine content) on strontium distributions. Interestingly, strontium
distributions were nearly the same as barium distributions in each test. Both metals also had
much lower overall mass balance closures than did the other metals.
73
-------
4.43 FJue Gas Metal Phase Distributions
Table 28 summarizes the ranges and averages of the relative distribution of each of the
metals between the particulate and vapor/dissolved phases in the afterburner and scrubber exit
flue gas. The particulate phase fraction is defined as that found in the probe wash and filter
catches of the sampling train. The vapor/dissolved phase fraction is defined as that found in the
sampling train impingers. Presumably, any metal present in the vapor phase at the sampling
train probe exit would be captured and accounted for in the impingers. In addition, much of any
metal present as water-soluble salts would also be collected in the impingers as it would weep
through the filter in solution. Arsenic, bismuth, cadmium, chromium and lead were reported at
below detection limits in the impingers of the scrubber exit sampling train. The values given in
Table 28 were calculated assuming these metals were present at the detection limit.
The data in Table 28 show that barium was found largely in the vapor/dissolved phase
in both the afterburner exit flue gas and the scrubber exit flue gas. This suggests that the barium
species predominant in the flue gas is quite water-soluble. Arsenic, bismuth, copper, and lead
were found mainly in the vapor/dissolved phase in the afterburner exit. In contrast, bismuth,
copper, and lead were found mainly in the particulate phase in the scrubber exit. Conclusions
regarding arsenic are limited by the wide ranges established by calculations based on detection
limits. The observations suggest that arsenic, bismuth, copper, and lead may be present mainly
as water-soluble species in the afterburner exit flue gas that are effectively captured by the
scrubber so that, at least for bismuth, copper, and lead the scrubber exit distributions favor the
particulate phase.
Phase distributions for magnesium and strontium varied widely from test to test in both
the afterburner exit flue gas and scrubber exit flue gas. Cadmium distributions in the afterburner
exit flue gas also varied widely from test to test, although cadmium was found predominantly in
the particulate phase fraction in the scrubber exit flue gas.
Table 29 summarizes the distribution of each metal between the particulate and the
vapor/dissolved phase in the afterburner and scrubber exit flue gas for each test. The data in
Table 29 represent the individual test data comprising the ranges and averages noted in Table 28.
Again, distributions affected by samples with nondetectable levels of a metal are represented by
ranges. The range boundaries reflect the assumption that the samples in question contained zero
and the detection limit of the metal, respectively. The data in Table 29 show no consistent
relationship between particulate and vapor/dissolved phase ratios and test variables, within limits
of data variability established by the replicate test measurements.
4.4.4 Metal Distributions Among Flue Gas Particulate by Particle Size in the
Afterburner Exit
Recall from the sampling and analysis protocol discussion in Section 3 that the flue gas
at the afterburner exit was sampled with a sampling train designed to collect a large (greater
than 1 g) sample of particulate so that this particulate could be size-fractionated, and trace metal
distributions among the particulate as a function of particle size could be determined. Results
of the particulate particle size sizing measurements are given in Appendix E. Particle size
fractions were combined to give five size fraction samples for trace metals analyses. These size
74
-------
TABLE 28. SUMMARY OF FLUE GAS METAL PARTICULATE AND VAPOR/DISSOLVED PHASE DISTRIBUTIONS
Afterburner exit flue gas Scrubber exit flue gas*
Vapor/dissolved phase Particulate phase Vapor/dissolved phase Particulate phase
fraction (%) fraction (%) fraction (%) fraction (%)
Metal Low High Average Low High Average Low High Average Low High Average
Arsenic
61
93
81
7
39
19
<53
<82
<68
>18
>47
>32
Barium
68
98
89
2
32
11
66
90
81
10
34
19
Bismuth
74
97
87
3
26
13
<1
<6
<4
>94
>99
>96
Cadmium
18
73
51
27
82
49
<2
<9
<5
>91
>98
>95
Chromium
b
—
—
—
—
<25
<81
<54
>19
>75
>46
Copper
59
90
78
10
41
22
4
70
24
30
96
76
Lead
42
95
73
5
58
27
<7
<66
<21
>34
>93
>79
Magnesium
16
76
51
24
84
49
12
85
65
15
88
35
Strontium
26
83
53
17
74
47
36
87
71
13
64
29
'Values are derived using analytical detection limits for samples reported as below detection limits.
b— = Not applicable.
-------
TABLE 29. PHASE DISTRIBUTIONS OF FLUE GAS METALS IN THE AFTERBURNER AND SCRUBBER EXIT FLUE GAS
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9:
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wt %) '
Target:
816
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8 |
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit = 1093°C; chlorine
= 4%
Kiln exit = 871
°C; chlorine =
4%
Kiln exit = 871
°C; AB exit =
1093°C
Test average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
%
Arsenic
Afterburner exit:
Particulate
20
11
15
11
24
27
11
15
11
14
39
11
15
11
7
Vapor/dissolved phase
80
89
85
89
76
73
89
85
89
86
61
89
85
89
93
Scrubber exit:
Particulate
>24
>37
>34
>22
>41
>47
>37
>34
>22
>43
>19
>37
>34
>22
>18
Vapor/dissolved phase
<76
<63
<66
<78
<59
<53
<63
<66
<78
<57
<81
<63
<66
<78
<82
Barium
Afterburner exit:
Particulate
32
4
19
5
11
10
4
19
5
2
7
4
19
5
11
Vapor/dissolved phase
68
96
81
95
89
90
96
81
95
98
93
96
81
95
89
Scnibbcr exit:
Particulate
16
34-85
11
12
13
29
34-85
11
12
10
12
34-85
11
12
31-59
Vapor/dissolved phase
84
15-66
89
88
87
71
15-66
89
88
90
88
15-66
89
88
41-69
Bismuth
Afterburner exit:
Particulate
3
9
4
16
22
26
9
4
16
19
16
9
4
16
6
Vapor/dissolved phase
97
91
96
84
78
74
91
96
84
81
84
91
96
84
94
Scrubber exit:
Particulate
>96
>97
>97
95
>98
>99
>97
>97
95
98
94-99
>97
>97
95
>94
Vapor/dissolved phase
<4
<3
<3
5
<2
<1
<3
<3
5
2
1-6
<3
<3
5
<6
(continued)
-------
TABLE 29. (continued)
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9i
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wt %)
i
Target:
816
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8 '
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit
= 1093°C; chlorine =
4%
Kiln exit
= 871°C; chlorine =
4%
Kiln exit
= 871°C; AB exit = 1093°C
Test average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
%
Cadmium
Afterburner exit:
Particulate
27
31
59
78
48
82
31
59
78
49
28
31
59
78
37
Vapor/dissolved phase
73
69
41
22
52
18
69
41
22
51
72
69
41
22
63
Scrubber exit:
Particulate
>93
>96
>96
>94
>94
>98
>%
>96
>94
>97
>93
>96
>96
>94
>91
Vapor/dissolved phase
<7
<4
<4
<6
<6
<2
<4
<4
<6
<3
<7
<4
<4
<6
<9
Chromium
Afterburner exit:
Particulate
a
Vapor/dissolved phase
Scrubber exit:
Particulate
>32
>46
>51
>31
>37
>75
>46
>51
>31
>64
>59
>46
>51
>31
>19
Vapor/dissolved phase
<68
<54
<49
<69
<63
<25
<54
<49
<69
<36
<41
<54
<49
<69
<81
Copper
Afterburner exit:
Particulate
23
13
10
22
15
41
13
10
22
33
24
13
10
22
15
Vapor/dissolved phase
77
87
90
78
85
59
87
90
78
67
76
87
90
78
85
Scrubber exit:
Particulate
91
96
61
66
89
81
96
61
66
77
30
96
61
66
91
Vapor/dissolved phase
9
4
39
34
11
19
4
39
34
23
70
4
39
34
9
*— - Not applicable.
(continued)
-------
TABLE 29. (continued)
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9
Primary variable:
Target:
Kiln exit temperature (°C)
816 871 871 871 927
Afterburner exit temperature (°C)
982 1093 1093 1093 1204
Feed chlorine content (wt %) ,
0 4 4 4 8 i
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held constant:
AB exit
= 1093°C; chlorine
= 4%
Kiln exit
= 871
°C; chlorine =
4%
Kiln exit
= 871°C; AB exit =
1093°C
Test average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
%
Lead
Afterburner exit:
Particulate
10
12
5
47
34
58
12
5
47
26
11
12
5
47
39
Vapor/dissolved phase
Scrubber exit:
90
88
95
53
66
42
88
95
53
74
89
88
95
53
61
Particulate
>80
>84
>83
>79
>84
>93
>84
>83
>79
>88
>34
>84
>83
>79
85-95
Vapor/dissolved phase
<20
<16
<17
<21
<16
<7
<16
<17
<21
<12
<66
<16
<17
<21
5-15
Magnesium
Afterburner exit:
Particulate
24
33
61
56
39
79
33
61
56
35
26
33
61
56
84
Vapor/dissolved phase
Scrubber exit:
76
67
39
44
61
21
67
39
44
65
74
67
39
44
16
Particulate
19
88
15
17
15
41
88
15
17
22
19
88
15
17
77
Vapor/dissolved phase
81
12
85
83
85
59
12
85
83
78
81
12
85
83
23
Strontium
Afterburner exit:
Particulate
63
43
30
44
63
67
43
30
44
17
25
43
30
44
74
Vapor/dissolved phase
Scrubber exit:
37
57
70
56
37
33
57
70
56
83
75
57
70
56
26
Particulate
18
>64
13
18
13
36
>64
13
18
18
20
>64
13
18
62
Vapor/dissolved phase
82
<36
87
82
87
64
<36
87
82
82
80
<36
87
82
38
-------
fractions were nominally less than 2 jim, 2 to 4 |im, 4 to 10 jim, 10 to 30 jim, and greater than
30 jim. From the analyses of these size fractions, distributions of each of the test trace metals
among the particle size ranges in the afterburner exit flue gas were obtained.
Table 30 summarizes the metal distributions by particle size range in the afterburner exit
flue gas particulate. As before, three groupings of tests appear in the table, where each grouping
corresponds to a series of tests in which one test variable (kiln exit temperature, afterburner exit
temperature, or waste feed chlorine content) was varied, with the other two variables held
nominally constant. The center point, which was tested in triplicate (Tests 4, 7 and 8), is
included in all groupings. The tabulated values represent the cumulative percent of the total of
each metal found in the particulate sample within the indicated range. The size ranges are less
than 2 jim, less than 4 jim, less than. 10 |im, less than 30 |im, and greater than 30 |im, and
correspond to the nominal size cuts resulting from the size fractionation of the particulate
sample.
Metals find their way into flue gas particulate via two pathways. In one pathway, the
metal remains in a condensed phase through the entire incinerator system and is carried out of
the system with entrained ash in the combustion gas. In the second pathway, the metal vaporizes
at some point in the incinerator, then recondenses when the flue gas cools. Both vaporization
and condensation can occur locally under proper conditions. Vaporized metal can condense
homogeneously into condensation nuclei that grow into a very fine fume, or they can condense
heterogeneously onto existing flue gas particulate. In both mechanisms the tendency is to enrich
(be found at high per mass concentration) in fine particulate. In the former mechanism, fume
particles are very fine. In the latter mechanism, the surface-to-mass ratio is higher for fine
particles than for coarse particles. Since condensation onto available surface is a per surface
area event, this also leads to enrichment in fine particulate.
Via the above mechanisms, the distribution of a given metal among flue gas particle size
ranges is strongly influenced by the extent to which the metal vaporizes in the incineration
system. Refractory metals that do not vaporize significantly should be relatively evenly
distributed in the flue gas particulate size ranges on a per mass (mg/kg particulate) basis.
Volatile metals should enrich in the fine particulate fractions, with enrichment tendency
increasing with increasing volatility. Thus, incinerator operating variables that affect metal
volatility are expected, in turn, to affect the distribution of the volatile metal among the particle
size ranges analyzed. These operating variables include the kiln temperature, afterburner
temperature, and the chlorine content of the waste feed. These are noted in Table 30, along
with the secondary variables of flue gas velocity in the kiln and afterburner and the particulate
loading (mg/actual wet m3) in the afterburner exit. Flue gas velocities are noted because they
influence the degree of entrapment of particulate into the flue gas. Flue gas particulate loading
is noted because it is a measure of the degree of entrainment of particulate into the flue gas.
Figure 26 shows the metal distributions among the particle size ranges, averaged over
all nine tests. The figure contains three plots representing the cumulative percent less than 4,
less than 10, and greater than 30 jim, respectively. (Plots for the less than 2 jim fraction are not
shown since this fraction contained small amounts of most metals.) The average cumulative
percent values for the total particulate in each size range are also indicated. The data show a
relationship between the relative volatility of each metal (as indicated by its volatility
79
-------
TABLE 30. METAL DISTRIBUTIONS IN THE AFTERBURNER EXIT FLUE GAS PARTICULATE BY PARTICLE SIZE
—8
Tcsl: 24783 64785 1 4 7 8 9 g
Primary variable: Kiln cxillcmpcraturc (°C) Afterburner exit temperature (°C) Feed chlorine content (wl %) q
Target: 816 871 871 871 927 982 1093 1093 1093 1204 0 4 4 4 8 2
Tcsl average: 819 877 881 879 929 1017 1096 1103 1098 1163 0 3.5 3.5 3.8 6.9
Held constant: AB exit = 1093°C; chlorine = 4% Kiln exit = 871°C; chlorine = 4% Kiln exit = 871°C; AB exit = 1093°C
Test average: 1095 10% 1103 1098 1092 887 877 881 879 885 900 877 881 879 881
Tcsl average: 3.5 3.5 3.5 3.8 3.5 3.6 3.5 3.5 3.8 3.7 1088 1096 1103 1098 1087
Flue gas velocity, m/s
Kiln 0.68 0.88 0.91 0.91 1.13 0.87 0.88 0.91 0.91 0.84 0.96 0.88 0.91 0.91 1.00
Afterburner 2.07 2.14 2.15 1.75 2.16 2.01 2.14 2.15 1.75 2.44 2.45 2.14 2.15 1.75 1.93
Afterburner flue gas particulate loading, mg/m3
13 17 20 21 28 28 17 20 21 3 20 17 20 21 24
Particulate size range,
lim Cumulative % in the particle size range
oo Total Sample
° <2 4.4 3.4 5.1 4.2 4.4 5.0 3.4 5.1 4.2 4.6 2.7 3.4 5.1 4.2 4.9
<4 10.7 8.8 17.4 15.5 15.9 18.0 8.8 17.4 15.5 8.6 9.0 8.8 17.4 15.5 12.9
<10 19.4 23.5 42.1 36.7 26.8 35.4 23.5 42.1 36.7 16.2 19.6 23.5 42.1 36.7 32.8
<30 33.1 37.9 52.8 47.2 36.9 48.2 37.9 52.8 47.2 24.7 33.6 37.9 52.8 47.2 48.4
>30 66.9 62.1 47.2 52.8 63.1 51.8 62.1 47.2 52.8 75.3 66.4 62.1 47.2 52.8 51.6
Arsenic
<2 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0
<4 8.4 1.9 18.7 20.1 37.5 22.0 1.9 18.7 20.1 9.0 8.2 1.9 18.7 20.1 13.8
<10 25.6 51.5 74.0 82.5 82.9 81.2 51.5 74.0 82.5 16.5 41.3 51.5 74.0 82.5 55.7
<30 72.1 78.3 80.4 92.5 89.0 90.5 78.3 80.4 92.5 56.4 69.0 78.3 80.4 92.5 77.7
>30 27.9 21.7 19.6 7.5 11.0 9.5 21.7 19.6 7.5 43.6 31.0 21.7 19.6 7.5 22.3
Barium
<2 5.6 2.3 1.9 1.4 3.6 1.3 2.3 1.9 1.4 5.2 1.5 2.3 1.9 1.4 3.4
<4 13.4 8.3 18.0 16.3 25.8 15.2 8.3 18.0 16.3 12.4 12.6 8.3 18.0 16.3 10.7
<10 24.0 41.3 53.9 46.0 52.3 44.6 41.3 53.9 46.0 27.1 28.7 41.3 53.9 46.0 42.6
<30 58.1 78.7 70.5 67.4 75.0 65.5 78.7 70.5 67.4 66.7 54.3 78.7 70.5 67.4 66.5
>30 41.9 21.3 29.5 32.6 25.0 34.5 21.3 29.5 32.6 33.3 45.7 21.3 29.5 32.6 33.5
(continued)
-------
TABLE 30. (continued)
Test:
2
4
7
8
3
6
4
7
8
5
1
4
7
8
9 f
Primary variable:
Kiln exit temperature (°C)
Afterburner exit temperature (°C)
Feed chlorine content (wt %)
h
c
Target:
816
871
871
871
927
982
1093
1093
1093
1204
0
4
4
4
8 t
Test average:
819
877
881
879
929
1017
1096
1103
1098
1163
0
3.5
3.5
3.8
6.9
Held, constant:
AB exit
= 1093°C; chlorine
= 4%
Kiln exit
= 871°C; chlorine =
4%
Kiln exit
= 871°'
C; AB exit = 1093°C
Test average:
1095
1096
1103
1098
1092
887
877
881
879
885
900
877
881
879
881
Test average:
3.5
3.5
3.5
3.8
3.5
3.6
3.5
3.5
3.8
3.7
1088
1096
1103
1098
1087
Flue gas velocity, m/s
Kiln
0.68
0.88
0.91
0.91
1.13
0.87
0.88
0.91
0.91
0.84
0.96
0.88
0.91
0.91
1.00
Afterburner
2.07
2.14
2.15
1.75
2.16
2.01
2.14
2.15
1.75
2.44
2.45
2.14
2.15
1.75
1.93
Afterburner flue gas particulate loading, mg/m3
13
17
20
21
28
28
17
20
21
13
20
17
20
21
24
Particulate size range,
(im
Cumulative % in the particle size range
Bismuth
<2
0.0
0.0
0.0
0.0
1.2
0.3
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.0
1.2
<4
10.4
6.1
8.7
1.5
30.5
13.2
6.1
8.7
1.5
3.1
7.6
6.1
8.7
1.5
17.7
<10
22.5
54.2
31.5
5.7
63.7
41.4
54.2
31.5
5.7
26.9
33.9
54.2
31.5
5.7
53.3
<30
74.6
97.6
42.7
10.5
75.5
47.9
97.6
42.7
10.5
26.9
52.8
97.6
42.7
10.5
68.3
>30
25.4
2.4
57.3
89.5
24.5
52.1
2.4
57.3
89.5
73.1
47.2
2.4
57.3
89.5
31.7
Cadmium
<2
0.4
0.8
0.2
0.1
1.0
0.4
0.8
0.2
0.1
0.7
0.1
0.8
0.2
0.1
1.1
<4
10.3
9.3
17.1
16.6
37.5
21.6
9.3
17.1
16.6
4.8
4.8
9.3
17.1
16.6
16.0
<10
30.7
54.1
72.5
77.1
80.5
74.7;
54.1
72.5
77.1
34.1
24.7
54.1
72.5
77.1
55.9
<30
84.1
85.1
79.8
84.0
86.6
82.5
85.1
79.8
84.0
54.1
49.5
85.1
79.8
84.0
76.5
>30
15.9
14.9
20.2
16.0
13.4
17.5
14.9
20.2
16.0
45.9
50.5
14.9
20.2
16.0
23.5
Chromium
<2
0.3
0.8
0.4
0.1
0.8
0.5
0.8
0.4
0.1
0.4
0.0
0.8
0.4
0.1
1.5
<4
4.3
4.1
10.3
0.9
13.2
6.6
4.1
10.3
0.9
1.6
0.4
4.1
10.3
0.9
17.8
<10
10.0
30.3
28.1
3.5
29.0
18.0
30.3
28.1
3.5
10.0
1.8
30.3
28.1
3.5
50.6
<30
30.4
49.1
37.9
7.2
37.3
25.6
49.1
37.9
7.2
20.5
6.0
49.1
37.9
7.2
70.3
>30
69.6
50.9
62.1
92.8
62.7
74.4
50.9
62.1
92.8
79.5
94.0
50.9
62.1
92.8
29.7
(continued)
-------
TABLE 30. (continued)
Test:
Primary variable:
Target:
Test average:
Meld constant:
Test average:
Test average:
2 4 7 8 3
Kiln exit temperature (°C)
816 871 871 871 927
819 877 881 879 929
AB exit = I093°C; chlorine = 4%
1095 1096 1103 1098 1092
3.5 3.5 3.5 3.8 3.5
6 4 7 8 5
Afterburner exit temperature (°C)
982 1093 1093 1093 1204
1017 10% 1103 1098 1163
Kiln exit = 871°C; chlorine = 4%
887 877 881 879 885
3.6 3.5 3.5 3.8 3.7
1 4 7 8 9 S
Feed chlorine content (wt %) o
0 4 4 4 8 &
0 3.5 3.5 3.8 6.9
Kiln exit = 871°C; AB exit = 1093°C
900 877 881 879 881
1088 1096 1103 1098 1087
Kiln
Afterburner
Flue gas velocity, m/s
0.68 0.88 0.91 0.91 1.13 0.87 0.88 0.91 0.91 0.84
2.07 2.14 2.15 1.75 2.16 2.01 2.14 2.15 1.75 2.44
0.% 0.88 0.91 0.91 1.00
2.45 2.14 2.15 1.75 1.93
13
17
Afterburner flue gas particulate loading, mg/mJ
20 21 28 28 17 20 21 13
20
17
20
21
24
Particulate size range,
pm Cumulative % in the particle size range
Copper
<2
1.2
1.4
0.6
0.5
2.2
0.7
1.4
0.6
0.5
1.6
0.4
1.4
0.6
0.5
2.0
<4
8.9
6.1
17.5
17.1
30.1
21.0
6.1
17.5
17.1
5.8
5.7
6.1
17.5
17.1
15.4
<10
21.8
37.5
59.1
61.5
63.8
64.9
37.5
59.1
61.5
26.5
19.3
37.5
59.1
61.5
50.0
<30
56.4
63.9
69.5
73.4
74.5
75.8
63.9
69.5
73.4
50.6
41.2
63.9
69.5
73.4
69.9
>30
43.6
36.1
30.5
26.6
25.5
24.2
36.1
30.5
26.6
49.4
58.8
36.1
30.5
26.6
30.1
Lead
<2
0.0
1.4
0.3
0.2
1.6
0.5
1.4
0.3
0.2
0.0
0.3
1.4
0.3
0.2
1.5
<4
7.8
9.7
16.5
19.3
34.4
22.7
9.7
16.5
19.3
4.0
4.5
9.7
16.5
19.3
16.3
<10
21.4
43.9
57.1
60.3
68.1
69.3
43.9
57.1
60.3
26.6
19.1
43.9
57.1
60.3
53.1
<30
58.2
66.9
64.2
68.0
76.6
75.7
66.9
64.2
68.0
26.6
41.6
66.9
64.2
68.0
73.5
>30
41.8
33.1
35.8
32.0
23.4
24.3
33.1
35.8
32.0
73.4
58.4
33.1
35.8
32.0
26.5
Magnesium
<2
1.6
1.0
0.7
0.4
1.6
0.6
1.0
0.7
0.4
1.6
0.3
1.0
0.7
0.4
2.6
<4
8.0
4.2
13.9
11.6
26.8
16.7
4.2
13.9
11.6
7.2
6.1
4.2
13.9
11.6
10.1
<10
17.9
45.1
62.8
51.3
62.5
65.0
45.1
62.8
51.3
41.4
25.9
45.1
62.8
51.3
43.8
<30
52.5
85.5
75.7
68.1
74.0
80.5
85.5
75.7
68.1
75.6
54.2
85.5
75.7
68.1
65.0
>30
47.5
14.5
24.3
31.9
26.0
19.5
14.5
24.3
31.9
24.4
45.8
14.5
24.3
31.9
35.0
(continued)
-------
TABLE 30. (continued)
Test:
Primary variable:
Target:
Test average:
Held constant:
Test average:
Test average:
24783 64785
Kiln exit temperature (°C) Afterburner exit temperature (°C)
816 871 871 871 927 982 1093 1093 1093 1204
819 877 881 879 929 1017 1096 1103 1098 1163
14 7 8
Feed chlorine content (wt %)
0 4 4 4
0 3.5 3.5 3.8
U>
9 §
o
CO
Ul
6.9
AB exit = 1093°C; chlorine = 4%
1095 1096 1103 1098 1092
3.5 3.5 3.5 3.8 3.5
Kiln exit = 871°C; chlorine = 4%
887 877 881 879 885
3.6 3.5 3.5 3.8 3.7
Kiln exit = 871°C; AB exit = 1093°C
900 877 881 879 881
1088 1096 1103 1098 1087
Kiln
Afterburner
Flue gas velocity, m/s
0.68 0.88 0.91 0.91 1.13 0.87 0.88 0.91 0.91 0.84
2.07 2.14 2.15 1.75 2.16 2.01 2.14 2.15 1.75 2.44
0.% 0.88 0.91 0.91 1.00
2.45 2.14 2.15 1.75 1.93
13
17
Afterburner flue gas particulate loading, mg/m3
20 21 2K 28 17 20 21
13
20
17
20
21
24
Particulate size range,
|im Cumulative % in the particle size range
Strontium
<2
1.3
2.9
2.0
1.2
2.9
1.3
2.9
2.0
1.2
4.3
1.2
2.9
2.0
1.2
3.4
<4
3.4
7.9
18.7
16.7
26.8
16.0
7.9
18.7
16.7
10.4
13.7
7.9
18.7
16.7
9.2
<10
6.1
40.0
51.6
45.2
54.0
43.1
40.0
51.6
45.2
23.1
30.9
40.0
51.6
45.2
41.0
<30
86.0
75.9
69.8
67.5
75.3
63.2
75.9
69.8
67.5
61.7
56.4
75.9
69.8
67.5
66.0
>30
14.0
24.1
30.2
32.5
24.7
36.8
24.1
30.2
32.5
38.3
43.6
24.1
30.2
32.5
34.0
-------
16
12
Ourmatlw pwoOTt < 4 mlorone
¦Pb »Ba ,cu
Sr* -sample
Mg
Bh
• a
600 1000. (600
Volatility temperature (*C)
2000
CurrtjIatlvB peroant < 10 mlcrone
« AS * Cd
sample
600 1000 1600
Volatility temperature ("C)
2000
60
eo
40
20
Cumulative peroent > 30 irtorona
a
sample
Bl
W
Ba Sr. 9Mq
,As »Cd
600 KKW 1600
Volatility temperature (*C)
2000
Figure 26. Average or metal distributions in the afterburner exit flue gas particle size fractions.
84
-------
temperature noted on the horizontal axis) and its propensity for redistribution to finer
particulate. This is indicated by the higher fractions of the metals with lower volatility
temperatures in the less than 4 and less than 10 (im particle size fractions and their correspond-
ing lower fractions in the greater than 30 (im particle size fraction.
The observed behavior is as expected, since the extent of metal vaporization is
temperature-dependent. Metals with lower volatility temperatures should experience a greater
degree of vaporization than metals with higher volatility temperatures. Interestingly, with respect
to enrichment in fine particulate, arsenic behaves as a volatile metal, in contrast to its behavior
as a more refractory metal as measured by its overall partitioning to kiln ash as discussed in
Sections 4.4.1 and 4.4.2. Evidently, the form of the arsenic determining its behavior is relatively
nonvolatile at the kiln temperatures tested. However, any arsenic in entrained flyash carried out
of the kiln becomes exposed to the higher afterburner temperatures. If the form of arsenic that
determines its behavior is volatile at the higher afterburner temperatures, the entrained arsenic
can vaporize in the afterburner, then recondense and be enriched in fine particulate as observed
in these tests.
Figures 27, 28, and 29 show the variations in the distributions of the metals among the
particulate size fractions as a function of the three parametric test variables: kiln temperature,
afterburner temperature, and waste feed chlorine content. Cumulative distributions of less than
4 )im and less than 10 |im for each metal and for the total particulate sample are given. The
three figures correspond to the three groupings of tests in Table 30. Thus, each figure
corresponds to a series of tests in which one test variable was varied, with the other two variables
held nominally constant. Values for the three center replicate test points are averaged and
plotted as a single point.
The effect of variations in the kiln temperature is shown in Figure 27. For the test
conducted with a nominal kiln temperature of 816°C (1500°F), Figure 27 indicates that the
distributions of metals in the particle fractions of less than 4 jim and less than 10 pm were
essentially the same as those for the overall particulate sample. About 10 percent of the total
particulate sample was less than 4 pm and about 10 percent of most metals were in the less than
4 |im size fraction; about 20 percent of the total particulate was less than 10 |im and about
20 percent of most metals were in the less than 10 jim particulate. A slight shift to finer
particulate (less than 10 |im) is seen for arsenic and cadmium. Metals present in Que gas
particulate solely via entrapment are expected to exhibit a size distribution similar to that of the
overall flue gas particulate, provided no significant vaporization of metals from the entrained
particulate occurs. Thus, at low kiln temperature, the data suggest that the metals were present
in the flue gas particulate primarily through entrainment, and did not experience significant
vaporization.
With increased kiln temperature from 816° to 871°C (1500° to l600eF), the total
afterburner exit flue gas particulate load shown in Table 30 increased from 13 to about 20 mg/m3
of flue gas, suggesting increased entrainment of particulate as the calculated average kiln exit flue
gas velocity increased from 0.68 to 0.90 m/s. Little redistribution of the metals with the
increased temperature is observed in the less than 4 pm size fraction. The small increase in the
total particulate fraction (to about 13 percent) in this range was accompanied by a similar
increase in this fraction for most metals.
85
-------
40
30
20
10
Variable: Kiln Exit Temperature
Cumulative peroent < 4 mlorone
>
if)
«-¦
aCd
Bl
C
Pb n
Cu
c
Ba
Sri
3
P mq
a 1
s
i <
i
sample
~
ft
i
I
'
'
<
>
¦
Cr
100
80
60
40
20
0
600 1000 1600 2000
Volatility temperature ("C)
o 816*C (1500'F) * 871*C (1600-F) ° 927'C (1700'F)
Cumulative percent < 10 microns
„Cd
Bi
Pb C
i;
i n B8 <
i;
Sr i
3 <
E
<
< >
»
~
Jsample
i
<
i i
0 v
Cr
•
0 600 1000 1600 2000
Volatility temperature (*C)
° 816'C (1600'F) * 871'C (1600*F) o Q27'C (1700T)
Figure 27. Effect of kiln temperature on the distribution or metals in the afterburner exit flue
gas particle size fractions.
86
-------
30
26
20
15
10
6
0
Variable: Afterburner Exit Temperature
Cumulative percent < 4 microns
As
fCd
Pb
Cu
-
Ba
A
Sr 1
AQ ,
i
^sample
41 j
Bi
f i
T
;
O
•
I
i
t
a
A
.i
Jo
600 1000 1500
Volatility temperature (°C)
2000
100
eo
60
40
20
$ 982'C (1800*F) * 1093*C (2000'F) ° 1204*C (2200T)
Cumulative percent < 10 microns
As ^
Cd
a
I f
Pb
Cu
Mg
Ba
Ti
Sr
_Q_
isample
at
0 500 1000 1600 2000
Volatility temperature (#C)
o 062*C (1800T) * rooa'C (2000'F) D 1204*C (2200*F)
Figure 28. Effect of afterburner temperature on the distribution of metals in afterburner exit
flue gas particle size fractions.
87
-------
20
15
10
Variable: Waste Feed Chlorine Content
Cumulative percent < 4 microns
c
Bi
Cd
Pb
! ~Cu
:
~ As *
(
]Ba <
~
Sri
Mg
*
~
¦
i
<
a
y
sample
V 0
1 1 1
0
0
600 1000 1600
Volatility temperature (°C)
2000
80
60
40
20
^0* *4% D 8 *
Cumulative percent < 10 microns
As Cd
I T
o t
o
1 Bi rn Pb ,
1 1Ba
*Cu Mg*
' Srt 1
a
(
o|
*1 ®
> 1
i
0
*sample
, <
~
600 1000 1600
Volatility temperature (*C)
« 01 * 4 SB D 8 %
2000
Figure 29. Effect of feed chlorine content on the distribution of metals in the afterburner exit
flue gas particle size fractions.
86
-------
There was a similar small increase in the fraction of total particulate in the less than
10 pm range with increased kiln temperature from 816° to 871°C (1500° to 1600°F). However,
with the exception of chromium and bismuth, a much greater increase in the fraction of metals
in this range is observed. Furthermore, Figure 27 shows that the redistribution of metals to this
range generally varies with the relative volatilities of the metals, in that metals with lower
volatility temperatures were enriched to a greater degree in the less than 10 pm particulate.
It is interesting that the redistribution of metals to finer particle size fractions with kiln
temperature increased from 816° to 871°C (1500° to 1600°F) was seen in the less than 10 pm
particle size fraction but not in the less than 4 |im size fraction. An explanation is found in
Figure 30, which shows the fraction of total particulate accounted for by each size fraction as a
function of the test variables. Once again, values for the three center point replicate test
conditions (Tests 4, 7, and 8) were averaged and plotted as a single bar. The observed ranges
for the three tests are indicated in the figure. Since Tests 4, 7, and 8 are replicate tests
conducted under the same nominal test conditions, variations in test data among them are an
indication of data variability.
The upper part of Figure 30 shows that, while the fraction of total particulate in the less
than 4 pm size range perhaps slightly increased with increased kiln temperature from 816" to
871°C (1500° to 1600°F), the fraction in the 4 to 10 pm size range was significantly increased
with this temperature change. A corresponding increase in the surface area available to
accommodate recondensing volatile metal in this size range would have occurred, with the result
that general metal redistribution to this size range is possible.
With further increased kiln temperature from 871° to 927°C (1600° to 1700°F), the
afterburner exit flue gas particulate load shown in Table 30 increased from about 20 to 28 mg/m3
of flue gas, suggesting additional entrainment of particulate as the calculated average kiln exit
flue gas velocity further increased from about 0.90 to 1.13 m/s. The particle size distribution of
the total sample was not significantly altered, as shown in Figure 27. However, with this further
increase in kiln temperature, a significant redistribution of all metals to particulate of less than
4 (im is seen.
The data shown in Figure 27 suggest that, as the kiln temperature is increased, there
may be a selective redistribution of metals to sequentially finer particulate in the flue gas. In
addition, the effect of kiln temperature on metal redistribution to finer particulate correlates with
the relative metal volatility of the metals. More volatile metals were redistributed to a greater
extent.
Variations in afterburner temperature also appear to affect both the distribution of flue
gas particulate, and the distribution of metals among the particulate. The data in Table 30 show
that, with increased afterburner temperature, there was a reduction in the particulate loading in
the flue gas on a mg/m3 basis. The particulate load at the afterburner exit decreased from 28
to about 20 mg/m3 as the afterburner temperature increased from 982° to 1093°C (1800° to
2000°F). The particulate load further decreased to 13 mg/m3 of flue gas as the afterburner
temperature increased to 1204°C (2200°F).
89
-------
Variable: Kiln Exit Temperature
eo
oo
40
SO
0
« 4 u 4 - 10 u 10 - 30 14 ~ 30 u
Variable: Afterburner Exit Temperature
FraoHon of tola! among perlloulata (ft)
SO
eo
40
20
0
Variable: Waste Feed Chlorine Content
Freollon of lata' Among parlloulale (%)
BO
eo
40
20
0
<*u *-Ku 10-30u > SO u
Fr«o!kan ot total anoog parlloiriale <*)
Figure 30. Size distribution of the flue gas particulate in the afterburner exit.
90
-------
Figure 30 shows that the overall particle size distribution shifted to coarse particulate
as afterburner temperature increased. This shift was most pronounced as the afterburner
temperature increased from 1093° to 1204°C (2000° to 2200° F). Specifically, the fraction of
afterburner exit particulate in the greater than 30 pm size range increased from about 50 to
75 percent. This shift to coarse particulate is most likely due to elimination of finer particulate,
through melting or softening and coalescing to larger particulate.
Figure 28 shows that the changes in the distributions of the test metals with afterburner
temperature correspond to the changes in the overall particulate distributions. That is, as
afterburner temperature is increased, the fraction of each metal in the less than 4 |im and the
less than 10 pm paniculate decreased, as did the fraction of total particulate in these size ranges.
This reflects the shift in overall particulate particle size distribution to coarser particulate with
increased afterburner temperature, as'noted above. Interestingly, at the low and intermediate
afterburner temperatures, the fraction of the metals with lower volatility temperatures in the less
than 10 ^m particulate was greater than corresponding fractions for the metals with higher
volatility temperatures (bismuth excepted). This suggests that volatiliza-
tion/condensation/enrichment in fine particulate of the metals with lower volatility temperatures
occurred. This enrichment likely became masked by the overall particle size distribution shift
to coarse particulate at the higher afterburner temperature.
The effects of the waste feed chlorine content on total particle and metal-specific size
distributions are shown in Figure 29. When feed chlorine was increased from 0 to 4 percent, the
fraction of total particulate in the less than 10 pm fraction increased from 20 lo about
35 percent. This is expected if the presence of chlorine in the feed serves to increase the
volatility of some feed inorganic constituents. When reviewing the data, the effects of chlorine
were taken to be most significant when the metal distributions were shifted more than the
distributions of the total particulate sample. Thus, flue gas particle size distributions for barium
and strontium and, to a lesser extent, arsenic, bismuth, and magnesium, were not significantly
affected by waste feed chlorine concentrations. For these metals, the magnitude of the shift to
the finer particulate fractions was about the same as the shift for the total particulate sample,
primarily reflecting the shift in the particulate sample size distribution.
Chlorine had a more pronounced effect on the particle size distributions of cadmium,
chromium, copper, and lead. For cadmium, copper, and lead, the shift to finer particulate
occurred with the initial feed chlorine content increase from 0 to 4 percent. The distribution of
these metals in particulate of less than 10 jim increased from about 20 to about 55 percent. No
additional redistribution occurred with the further feed chlorine content increase to 8 percent.
Chromium distributions in particulate of less than 10 pm increased with both feed chlorine
content increases, from 2 to 20 to 50 percent with chlorine increased from 0 to 4 to 8 percent.
The results for chromium are quite interesting. Chromium is the least volatile of the
metals investigated in the test program. However, chromium showed the greatest relative
redistribution with increased chlorine in the waste feed. It is suspected that chromium in the
entrained particulate may have formed chromyl chloride (Cr02Cl2) in tests with feed chlorine
present, with increased formation occurring with increased feed chlorine content. Chromyl
chloride is quite volatile. If its formation is increasingly favored as feed chlorine content
91
-------
increases, chromium redistribution to finer particulate would be expected to increase
correspondingly with increasing feed chlorine content.
Figures 31 through 39 show the distributions of each test metal in the four afterburner
exit flue gas particle size fractions analyzed as a function of the test parameters varied. The
format of Figures 31 through 39 is the same as that of Figure 30 for the total particulate
distributions. The figures offer further support for the observations stated above.
In summary, all three test variables affected the distribution of the test trace metals in
the afterburner exit flue gas particle size fractions. Increasing kiln temperature caused a
redistribution of the metals to finer particle size ranges. The degree of redistribution increased
with increasing metal volatility (decreasing volatility temperature). Increasing feed chlorine
content also caused a redistribution of some of the metals to fine particulate. Both are expected
observations and are consistent with the volatilization/condensation mechanism of volatile metal
enrichment in fine particulate. Increasing afterburner temperature caused a shift of overall
sample particle size distribution to coarser particulate, most likely because of fine particles
melting or softening and coalescing into larger particles. A corresponding shift in metal
distributions to coarse particulate occurred.
4.4.5 Apparent Scrubber Collection Efficiencies
The apparent scrubber efficiency for collecting flue gas metals was determined for each
test, The apparent scrubber efficiency represents the ratio of the normalized metal fraction
measured in the scrubber liquor to the sum of the normalized metal fractions measured in the
scrubber liquor and scrubber exit flue gas. As discussed in Section 4.4.1, an accumulation of
sludge in the bottom of the scrubber and in the recirculation tank suggests that there may have
been a loss of metals in the scrubber loop. If the blowdown sample analyzed did not fully reflect
the overall scrubber loop metal concentrations, the apparent scrubber collection efficiencies
measured during the tests could possibly be biased low. Thus, caution should be exercised in
interpreting the apparent scrubber collection efficiency data discussed below.
Table 31 summarizes the observed range in, and average of, the apparent scrubber
efficiencies for removing each metal from the flue gas stream over all nine tests. The data in
Table 31 show that average metal collection efficiencies ranged from 22 to 71 percent; the overall
average for all metals was 43 percent. Figure 40 summarizes the data in Table 31. The bar for
each metal represents the range of apparent scrubber efficiencies over the nine tests, with the
overall average for the nine tests noted by the midrange tick mark. The apparent scrubber
efficiency for each metal is plotted against the metal's volatility temperature. Also included in
the figure are values for arsenic and chromium calculated by setting scrubber exit flue gas sample
concentrations reported as less than the analytical detection limit to zero, instead of the
detection limit. Arsenic and chromium were the only metals with scrubber efficiencies
significantly affected by sample concentrations reported to be below detection limits.
Figure 40 shows that there is a wide range in the apparent scrubber collection
efficiencies observed for each metal over the nine tests. However, average collection efficiencies
were generally higher for the less volatile metals. This would be expected for particulate control
devices that collect particulate via inertial impact mechanisms such as a venturi scrubber. As
92
-------
Variable: Kiln Exit Temperature
Fraolion ol total among oamoulala 1*)
i
=?£¦
w
1
e:
BSP8
bmc dene G3B27C
B
wsxj
iiJ
r'.-T
1
M
Ph
8
r~
4- .
flL
wr
-S
Variable: Afterburner Exit Temperature
Fiaollon ol tola) among cartloulBtt (*!
SO
Buic CD 1083 C Dbwc
40
> 30 u
10 - 30 l*
4 - 10 u
Variable-. Waste Feed Chlorine Content
Fraollon of total among parlloulals 1%)
40
20
10 - 30 U
> 30 u
4 - K>u
« « U
Figure 31. Arsenic distributions in the afterburner exit flue gas particulate.
93
-------
Variable: Klin Exit Temperature
00
40
20
0
< 4 u 4 - 10 m 10 - 30 U » 30 U
Variable: Afterburner Exit Temperature
Freotlon ol total among particulate (95)
Ouesc CDijoj c
B
1
I
I
$r:
....
<
%
.
I
1
n
-L
77?
$
f
111
-V-
w
1
< 4 u * - 10 u 10 -30u > GO u
Variable: Waste Feed Chlorine Content
Frsotlon or total among parlloulets («)
eo
40
20
0
« i, u 4 - 10 u 10 - 30 u > 00 V
Fraotioi of total among pa'tloutete ft)
Dene d)R7iC CZ3e2r C
Figure 32. Barium distributions in the afterburner exit flue gas particulate.
94
-------
Variable: Kiln Exit Temperature
60
60
40
20
0
Variable: Afterburner Exit Temperature
Fraction of tola1 among partloufsle (*)
eo
eo
40
20
0
. l w i-lOu 10 - 30 W » 30 u
Variable: Waste Feed Chlorine Content
Fraction or total among Dertloulete (X)
BO
eo
40
20
0
(III 4-Wu 10 - 30 U >30u
Figure 33. Bismuth distributions in the afterburner exit llue gas particulate.
95
Fraotlon of lolat among parOoutate <%)
> inn ft mas* c
M
m
—
1
1
s
m
ii
m
-vv-
-
¦tx
IfT
1
r-;.
s
< 4 u 4 - 10 U >0 - 30 u * 30 u
-------
Variable: Kiln Exit Temperature
F'aotlon of total among partlouiata I*)
Iffi eio c
L_J 827 C
' 4 u
4 - 10 U
10 - 30 u
> 30 V
Variable: Afterburner Exit Temperature
Fraotlon of total among particulate (%)
« 4 U
4 - 10 u
10 - 30 v
> 30 v
Variable: Waste Feed Chlorine Content
Fraotion ol total among partloutale («)
<* v
4 - »0 u K) - 30 ii
> 30 u
Figure 34. Cadmium distributions in the afterburner exit flue gas particulate.
96
-------
Variable: Klin Exit Temperature
Fraction o< total among pa'tloulala (S)
100
Bto c CDenc dn'C
76
CO
0
10 - 30 U
« - io u
« * u
Variable: Afterburner Exit Temperature
Fr#cHor> of totat among paHloulatt (%)
^ 1762 C ED 1009 C O UO< C
¦k'
_
I
J-
¦m
< 4 U 4 - 10 U 10 - 30 u . > 30 u
Variable: Waste Feed Chlorine Content
Fraction ol total among oemouifite (S)
ESSoa Dm Qii
it
ll
J-
&w\w|
1
H
Figure 35. Chromium distributions in the afterburner exit flue gas particulate.
-------
Variable: Klin Exit Temperature
eo
40
20
0
Variable: Afterburner Exit Temperature
Fraoiion of total araoQ pertiouiale (%}
60
40
20
0
eo
40
20
0
Figure 36. Copper distributions in the afterburner exit flue gas particulate.
Fraction ol total among particulate («)
BSfflsiec CDeric
113 027 c
-I'
1
P
|
J-
Is
flj
1
i
pv
11$
I
'ID 4 - 10 u 10 - 30 u > 30 u
082 C CD 1003 C CD 1201 c
» 30 u
4 - to U
to - 30 M
< 4 u
Variable: Waste Feed Chlorine Content
Fraction ol total among particulate (%)
* - K>U
tO -30u
> 30 u
-------
Variable: Kiln Exit Temperature
60
eo
40
20
a
BO
eo
40
20
0
<«U 4 - 10 u 10 - 30 u ' 30 u
Variable: Waste Feed Chlorine Content
Fraotlon ot total among parlloulit* (%)
60
eo
40
20
0
37. Lead distributions in the afterburner exit flue gas particulate.
99
Fraotlon of total among particulate (V)
* 4 u
4 - » U
10 • 30 u
> 30 u
Variable: Afterburner Exit Temperature
Fraction of total among partiouiett (%)
E3m2 c CD looa c CD t2tM c
—7
-
_
1
WAW
1
1
X
-T
IF
ii
¦
E
-------
Variable: Klin Exit Temperature
Fiaolloo or total among partlouisl* (#]
<1 11
4 - K) u to - 30 u
> 30 u
Variable: Afterburner Exit Temperature
Frtoilon of Iota! «monf pariloultl# (%}
« 4 u 4 • 10 u 10 ' 30 u > 30 m
Variable: Waste Feed Chlorine Content
Fraction ol total among particulate (1)
< 4 u
* - 10 u
(0 - 30 u
> 90 u
Figure 38. Magnesium distributions in the afterburner exit flue gas particulate.
100
-------
Variable: Kiln Exit Temperature
Fraotlon ol lota! among psrlloulal* (II
ewe dime CDaj7 c
eo
40
20
0
10 - 30 u
« 4 U
Variable: Afterburner Exit Temperature
FrMllori ol total among particulate (1)
IS 062 c CD toca c CD i2cm c
r
"P
iP
y
ii
¦
1
1
< 4 u 4-IOU to - 30 u > 30 U
Variable: Waste Feed Chlorine Content
Fraction ol total among parttoulsts (S)
80
IUC CD 10S3 C CD 1204 c
eo
40
20
0
4 - K>u
<4 u
• 30 u
Figure 39. Strontium distributions In the afterburner exit flue gas particulate.
101
-------
TABLE 31. SUMMARY OF APPARENT SCRUBBER EFFICIENCY
RANGES AND AVERAGES
Apparent scrubber removal efficiency g
(%)
MeLal
Low
High
Average
Arsenic
32
60
44
Barium
52
85
71
Bismuth
22
66
40
Cadmium
11
42
22
Chromium
13
61
35
Copper
11
58
26
Lead
9
58
30
Magnesium
47
94
67
Strontium
34
80
53
a
w
LU
100
60
60
40
20
Apparent eorubber eMkrienoy (%)
Mg
• AS (DL-O)
-
•Ba
-Ct (DL-O)
As
Bi
Sr-
(DL)
.Cd
-Pb
-Cu
m
Ct
-------
noted in Section 4.4.4, redistribution of the more volatile metals to fine particulate was clearly
observed in these tests. Inertial collection devices are significantly more efficient in collecting
coarse particulate than fine particulate. However, an ionizing wet scrubber uses the ionizing
section specifically to enhance collection efficiency for fine particulate. The data in Figure 40,
however, suggest that the fine particulate collection efficiency is still lower than that for coarser
particulate.
Table 32 summarizes the apparent scrubber efficiencies measured for each metal in each
test. Once again, three groupings of tests appear, where each grouping corresponds to a series
of tests in which one test variable (kiln exit temperature, afterburner exit temperature, or waste
feed chlorine content) varied, with the other two variables held nominally constant. The center
point, which was tested in triplicate (Tests 4, 7 and 8), is included in all groupings. The values
in Table 32 were used to determine the ranges and averages in Table 31.
Figure 41 presents two graphs corresponding to the groupings of tests in Table 32 for
the variables of kiln exit temperature and waste feed chlorine content. Collection efficiencies
for arsenic, barium, strontium, magnesium and chromium do not vary directly with any of the
test variables, within the limits of data variability established by the replicate test conditions. For
these metals, bars are used to indicate the range of efficiencies observed over the five tests
corresponding to the associated grouping in Table 32.
Collection efficiencies for cadmium, bismuth, lead and copper apparently varied with
changes in both kiln temperature and waste feed chlorine content. In Figure 41, values for the
three center replicate test points are averaged and plotted as a single point. Apparent collection
efficiencies for these metals increased with increased kiln temperature and increased waste feed
chlorine content.
Increased scrubber collection efficiency might be expected with increased feed chlorine
content if the presence of chlorine leads to the formation of more soluble metal chlorides.
However, it is not clear why increased kiln temperature would lead directly to increased
collection efficiency. Recall from Section 4.4.2, however, that kiln ash discharge fractions for
cadmium, bismuth and lead all decreased markedly with increased kiln temperature, with
associated increases in the flue gas and scrubber liquor fractions, This observation suggests that
the apparent collection efficiency determined for each metal may increase with higher flue gas
loading of the metal.
Apparent scrubber collection efficiencies for metals did not vary with afterburner exit
temperature, within the limits of data variability established by the replicate test conditions.
4.5 VOLATILE ORGANIC CONSTITUENT CONCENTRATIONS AND POHC
DESTRUCTION AND REMOVAL EFFICIENCIES
As noted in Section 3, one ash and one blowdown sample from each test were analyzed
by purge and trap, GC/FID for the 22 volatile organic constituents listed in Table 13. Results
of these analyses are included in Appendix C, and indicate that none of the compounds sought
were present above detection limits in the ash. Chloroform, bromodichloromethane and
bromoform were found in the scrubber blowdown samples. These compounds are typically found
103
-------
TABLE 32. APPARENT SCRUBBER COLLECTION EFFICIENCIES
Test:
Primary variable:
Target:
Test average:
Held constant:
Test average:
Test average:
2 4 7 8 3
Kiln exit temperature (°C)
816 871 871 871 927
819 877 881 879 929
AB exit = 1093°C; chlorine = 4%
1095 10% 1103 1098 1092
3.5 3.5 3.5 3.8 3.5
6 4 7 8 5
Afterburner exit temperature (°C)
982 1093 1093 1093 1204
1017 1096 1103 1098 1163
Kiln exit = 871°C; chlorine = 4%
887 877 881 879 885
3.6 3.5 3.5 3.8 3.7
1 4 7 8 9
Feed chlorine content (wt %)
0 4 4 4 8
0 3.5 3.5 3.8 6.9
Kiln exit = 871°C; AB exit = 1093°C
900 877 881 879 881
1088 1096 1103 1098 1087
Collection efficiency (%)
Arsenic
38-73
46-71
40-68
45-81
60-80
60-79
46-71
40-68
45-81
32-52
32-77
46-71
40-68
45-81
47-86
Barium
52
77
55
67
69
83
77
55
67
85
75
77
55
67
77-87
Bismuth
22
46
37
41
48
34
46
37
41
26
37
46
37
41
66
Cadmium
11
17
17
25
37
22
17
17
25
14
13
17
17
25
42
Chromium
38-68
30-45
19-28
56-81
28-54
61
30-45
19-28
56-81
31-41
11
30-45
19-28
56-81
37-77
Copper
11
36
20
16
45
20
36
20
16
13
12
36
20
16
58
Lead
10
37
30
29
48
28
37
30
29
19
9
37
30
29
58
Magnesium
53
89
57
47
76
77
89
57
47
52
56
89
57
47
94
Strontium
49
69
41
39
60
64
69
41
39
41
34
69
41
39
80
-------
Variable: KILN EXIT TEMPERATURE
si 100
>
o
w so
o
LU
cr
UJ
CD
cc
z>
cc
o
en
UJ
cc
<
CL
CL
<
60
40
20
J As (DL=0) ? Mg
^ ^fcr(DL.O)
T 1Ba
1 AS(DL) rrj
"HA
p
~ [
» Bi
Cd
i
fpb
Cu £ Cr (DL)
k
A
<
. | ¦
' 6 *
l l
•
en
cv
r--
O
Q
(/>
0 500 1000 1500
VOLATILITY TEMPERATURE (*C)
~ 816 'C (1500 *F) A 871 "C (1600 "F) ~ 927 "C (1700-F)
2000
Variable: WASTE FEED CHLORINE CONTENT
— 100
>
o
UJ
o
UJ
oc
UJ
CD
CD
z>
DC
O
CO
LU
CO
<
a.
0.
<
n 80
60
40
20
o 1
Mg'
p
t As (DL-0)
~
Ba -
i
~
Sr
\
i
k
I
Cr (DLsO)
As (DL) Bi
? a j l
I
Pb CU
I
<
k
¦
k
I
Cd 1
k
i
•
Cr
f
(DL)
£
A
•
4
1
-1
• ~<
_ l_
•
500 1000 1500
VOLATILITY TEMPERATURE (*C)
~ 0% A 4% a 8%
2000
Figure 41. Apparent scrubber collection efficiencies for metals showing associated variations
with changes in kiln exit temperature and waste feed chlorine content.
105
-------
in chlorine-treated potable water supplies such as that used for the scrubber system makeup
water.
The three compounds introduced in the synthetic waste feed as principal organic
hazardous constituents (POHCs) for these tests were toluene, chlorobenzene, and tetrachloroeth-
ylene. The flue gas concentrations of these compounds measured at the three locations sampled
(the afterburner exit, the scrubber system exit, and the stack (three tests)) during the tests are
summarized in Table 33. The data in the table show that POHC concentrations in the stack
were substantially higher than in the scrubber exit or afterburner exit for Tests 7 and 8.
Tetrachloroethylene concentrations were similarly higher for Test 4. The higher concentrations
in the stack gas are due to the removal of previously absorbed compounds from the carbon bed
by the steam stripping action of the saturated flue gas. This phenomenon has been observed
during past testing at the IRF.
Table 34 combines the flue gas POHC concentration data from Table 33 with the flue
gas flowrate data from Table 19, the synthetic liquid waste feed composition data from Table 14,
and the liquid waste feedrate data from Appendix A corresponding to the period of VOST
sampling, to give the POHC DREs at the three locations sampled. All POHC DREs were
greater than 99.99 percent at the afterburner exit and the scrubber exit for all tests. The lower
DREs at the stack are likely due to stripping of previously absorbed POHC as the flue gas passes
through the carbon bed.
As noted in Section 3, all VOST traps were analyzed by thermal desorption, purge and
trap, GC/FID for the volatile organic constituents listed in Table 13. This list includes several
potential volatile products of incomplete combustion (PICs), in addition to the POHCs. Table 35
summarizes the flue gas concentrations of the non-POHC volatile organic constituents detected
in the afterburner exit flue gas and scrubber exit flue gas during the tests. Tabulated values
represent the average of the three VOST trap pairs obtained during each test. For determining
the average emission rate, practical quantitation limits (PQLs) were used for trains containing
concentrations less than the PQL.
Table 35 shows that the major PICs detected were hexane, chloroform, carbon
tetrachloride, and benzene. Concentrations in the afterburner exit ranged from about 1 to
15 ng/dscm. Concentrations in the scrubber exit were about the same as in the afterburner exit.
The chloroform increase during Tests 2, 5, 8 and 9 was probably due to the release of chloroform
from the scrubber makeup water. As mentioned, the scrubber water source is a chlorine-treated
potable supply in which trihalomethane (THM) compounds are typically found. These
compounds volatilize readily; the scrubber essentially acts as a purger, releasing the THM into
the flue gas stream with the entrained water vapor.
106
-------
TABLE 33. FLUE GAS POHC CONCENTRATIONS
POHC concentration (jift/dscm)
Afterburner exit
Scrubber exit
Stack"
Test
Test
date
Toluene
Tetrachloro-
ethylene
Chloro-
benzene
Toluene
Tetrachloro-
ethylene
Chloro-
benzene
Toluene
Tetrachloro-
ethylene
Chloro-
benzene
lb
8/17/89
20
C
—
26
—
—
—
—
—
2
8/2/89
60
12
9.2
93
16
9.0
—
—
-
3
8/4/89
28
6.0
3.4
19
6.2
2.6
—
—
4
8/1/89
230
60
42
290
60
42
290
160
26
5
8/16/89
1.4
1.6
1.4
9.0
2.5
3.6
—
—
—
6
8/15/89
8.8
2.3
1.8
14
5.4
5.3
—
—
—
7
8/9/89
26
13
5.4
12
2.9
1.9
530
140
47
8
8/11/89
13
2.9
2.3
16
4.3
2.0
1400
360
150
9
7/28/89
280
180
130
180
96
53
—
—
—
"Stack gas sampling conducted during Tests 4, 7, and 8 only.
bNo tetrachloroethylene or chlorobenzene in waste feed for Test 1.
c— = Not applicable.
-------
TABLE 34. POHC DREs
DRE (%)
Afterburner exit
Scrubber exit
Stack
Test
Test
date
Toluene
Tetrachloro-
ethylene
Chloro-
benzene
Toluene
Tetrachloro-
ethylene
Chloro-
benzene
Toluene
Tetrachloro-
ethylene
Chloro-
benzene
1
8/17/89
99.9999
a
—
99.9996
—
—
—
—
—
2
8/2/89
99.9994
99.9993
99.9994
99.9984
99.9985
99.9991
—
—
—
3
8/4/89
99.9997
99.9997
99.9998
99.9997
99.9994
99.9997
—
—
—
4
8/1/89
99.9974
99.9961
99.9971
99.9944
99.9932
99.9949
99.9942
99.981
99.9968
5
8/16/89
99.9999
99.9999
99.9999
99.9998
99.9997
99.9996
—
—
—
6
8/15/89
99.9999
99.9999
99.9999
99.9998
99.9995
99.9995
—
—
—
7
8/9/89
99.9998
99.9993
99.9997
99.9997
99.9996
99.9997
99.9922
99.987
99.9957
8
8/11/89
99.9999
99.9999
99.9999
99.9997
99.9996
99.9998
99.977
99.967
99.984
9
7/28/89
99.9946
99.9935
99.9949
99.9944
99.9943
99.9966
—
—
—
= Not applicable.
-------
TABLE 35. FLUE GAS VOLATILE PIC CONCENTRATIONS
Test:
1
2
3
4
5
6
7
8
9
Test date:
8/17/89
8/2/89
8/4/89
8/1/89
8/16/89
8/15/89
8/9/89
8/11/89
7/28/89
Compound
Afterburner exit flue gas concentration (|ig/dscm)
Hexane
1.5
2.0
1.4
1.3
<1.2
1.6
1.3
1.3
3.1
Chloroform
3.8
4.0
2.6
2.5
4.4
5.8
11
8.0
15
Carbon tetrachloride
5.9
1.7
1.8
2.5
2.1
<1.5
7.6
5.9
<1.5
Benzene
<1.4
2.4
<1.4
2.3
1.5
3.1
17
3.2
3.7
Compound
Scrubber exit flue gas concentration (^ig/dscm)
Hexane
1.8
2.8
2.2
2.9
1.5
5.3
1.4
7.4
3.2
Chloroform
4.5
52
6.0
5.6
90
6.2
11.6
58
35
Carbon tetrachloride
<1.5
1.6
1.7
<1.5
6.3
2.2
1.8
<1.5
<1.5
Benzene
9.0
3.9
2.1
4.2
3.5
29
1.9
4.2
4.1
-------
SECTION 5
CONCLUSIONS
A series of pilot-scale incineration tests was performed to evaluate the fate of trace
metals fed to a rotary kiln incinerator equipped with a single-stage ionizing wet scrubber for
particulate/acid gas control. The test series consisted of nine tests designed to parametrically
evaluate the fate of five hazardous constituent trace metals (arsenic, barium, cadmium,
chromium, and lead), and four nonhazardous constituent trace metals (bismuth, copper,
magnesium, and strontium) fed to the incinerator. Test variables were kiln exit temperature,
afterburner exit temperature, and waste feed chlorine content. A factorial experimental matrix
was used in which kiln exit temperature was varied from nominally 816° to 927°C (1500° to
1700DF), afterburner exit temperature was varied from nominally 982° to 1204°C (1800° to
2200° F), and waste feed chlorine content was varied from nominally 0 to 8 percent.
For all tests, the waste feed consisted of a synthetic mixture prepared by combining an
organic liquid mixture with a clay absorbent material. The organic liquid mixture consisted of
toluene with varying amounts of tetrachloroethylene and chlorobenzene added to give the desired
feed chlorine content. This synthetic solid waste material contained nominally 0.4 kg of organic
liquid to 1 kg of clay absorbent. This mixture was introduced to the kiln via a screw feeder.
Test trace metals were prepared in an aqueous solution. This solution was metered into the
clay/organic liquid mixture in the screw feeder.
Conclusions from the parametric trace metal tests include the following:
• Based on the normalized discharge distribution data, cadmium and bismuth were
relatively volatile metals with an average of less than about 40 percent of the
discharged metal being present in kiln ash. Arsenic, barium, chromium, copper,
lead, magnesium and strontium were relatively nonvolatile with an average of
greater than 80 percent of the discharged metal present in the kiln ash.
• Observed metal volatilities generally agreed with the order predicted by metal
volatility temperatures (temperature at which the effective vapor pressure of the
metal is 10"4 atm), with the notable exception of arsenic. Arsenic has the lowest
volatility temperature of the metals tested, but was observed to be one of the least
volatile of the metals. This suggests that As^Oj was not a predominant arsenic
species in the incinerator, or that the arsenic is adsorbed to the clay/ash matrix.
• Kiln temperature affected the relative volatility of the observed most volatile
metals (cadmium and bismuth), and of lead. The fractions of these metals
discharged in the kiln ash decreased with increasing kiln temperature.
110
-------
• Waste feed chlorine content appears to have affected the fraction of metals
recovered in the kiln ash. With increased feed chlorine content from 4 to
8 percent, a significant increase in the fraction of each metal found in the ash was
observed. The implication is that the fraction of the metals recovered in the
digestion of the kiln ash samples may have been higher in the cases in which the
feed contained chlorine. If true, these observations suggest that the digestion
procedure (Method 3050) was not sufficiently aggressive in fully liberating the
metals from the solid sample.
• Afterburner exit temperature did not clearly affect metal partitioning between the
scrubber exit flue gas and scrubber liquor discharge streams
• Metal flue gas phase distribution (particulate/vapor, dissolved phase) data showed
no consistent relationship with test variables
• Enrichment of metals in the fine-particulate fraction of the afterburner exit
particulate was observed, with an average of roughly 60 percent of the flue gas
particulate metal in the less-than-10-nm size range, compared to an average of
about 30 percent of the total particulate sample. The distributions of the more
volatile metals were shifted to fine particulate more so than the less volatile
metals. Arsenic behaved as a volatile metal with respect to its distributions among
the afterburner exit flue gas particle size ranges.
• Each test variable affected the distributions of at least some of the metals among
the flue gas particulate particle size ranges. Size distributions of the metals most
nearly reflected the overall sample particle size distribution for Test 2 (the lowest
kiln temperature), Test 5 (highest afterburner temperature), and Test 1 (no
chlorine in the waste feed) very little redistribution among the particulate was
observed. For these three tests, about 20 to 25 percent of each metal and the
total particulate sample were found in the less than 10 jim particulate.
• Increased kiln temperature from 816° to 927 °C (1500° to 1700 °F) caused the
average distributions to shift from roughly 20 percent less than 10 nm to an
average of 60 percent less than 10 |im for all test metals except chromium. For
cadmium, copper, and lead, an increase in waste feed chlorine content from 0 to
4 percent caused their distributions to shift from roughly 20 percent less than
10 jim to 55 percent less than 10 nm. No further effects with feed chlorine
increased to 8 percent were observed for these metals. For chromium, increasing
chlorine content from 0 to 4 to 8 percent caused a corresponding shift of 2 to 20
to 50 percent in particulate less than 10 pm.
• The average apparent scrubber collection efficiencies for metals ranged from 22
to 71 percent, and generally increased with decreasing metal volatility. The
overall average collection efficiency for all metals was 43 percent. It should be
noted that industrial applications of ionizing wet scrubbers are typically in multiple
stages and, as such, would be expected to collect metals more efficiently than the
single-stage scrubber at the IRF.
Ill
-------
• Within the limits of data variability, none of the test variables affected scrubber
collection efficiencies for arsenic, barium, strontium, magnesium and chromium.
Apparent scrubber collection efficiencies for cadmium, bismuth, lead, and copper
increased with increased kiln temperature and with increased waste feed chlorine
content. Afterburner temperature had no discernable effect on apparent scrubber
collection efficiencies for any of the metals.
As discussed in Section 6, the data quality objectives established in the quality assurance
project plan (QAPP) for this test program were met with few exceptions. The discussion in
Section 6 suggests that these few exceptions had no impact on test program conclusions.
112
-------
SECTION 6
QUALITY ASSURANCE
All samples analyzed to obtain the data reported in this report were taken at the IRF
by members of the IRF operating staff. All samples were collected and/or recovered in
accordance with methods appropriate to their eventual analysis. After appropriate preservation,
the samples were relinquished to the custody of the onsite Sample Custodian. The Sample
Custodian subsequently directed the splitting of samples and the transport of these to the
appropriate laboratories for analysis. The following were performed in IRF onsite laboratories:
• Volatile organic analyses
• TCLP extractions
• Sample digestions for metals analyses
• Matrix spike sample preparation
Metals analyses were performed in the inorganic laboratories of NCTR, the host site for the
IRF. Soil feed ultimate analyses (C, H, O, N, S, and CI) analyses were performed by Galbraith
Laboratories, Knoxville, Tennessee.
The sample chain-of-custody procedures described in the QAPP for these tests were
followed without deviation. No compromise in sample integrity occurred. Sample tracking
records are given in Appendix C.
A number of quality assurance (QA) procedures were followed to assess the data quality
of the laboratory analytical measurements performed in these tests. In addition, QA efforts
performed to ensure that data quality is known for the particulate and CEM measurements
involved adherence to Reference Method procedures and CEM manufacturer's specifications.
No deviations from the QAPP occurred for these measurements. The major laboratory QA
efforts focused on the following measurements:
• Volatile organic compounds in the feed, kiln ash, and scrubber blowdown samples
• Volatile organic compounds in the VOST traps
• Metals in the feed and incinerator discharges
113
-------
The data quality objectives (DQOs) for these measurements are listed in Table 36. The QA
procedures performed included:
• Analyzing replicate feed mixture samples for toluene, chlorobenzene, and
tetrachJoroethylene
• Preparing a clay/organic liquid feed mixture matrix spike and analyzing replicate
samples for spike recovery
• Spiking all VOST train resin traps with two method surrogates and measuring
surrogate recovery
• Preparing nine matrix spike VOST traps and analyzing these for spike recovery
• Preparing matrix spike and matrix spike duplicate samples for the nine test metals
in the following matrices: kiln ash, scrubber blowdown, and multiple metals train
impinger solutions
Results of these QA procedures are discussed in the following subsections.
6.1 VOLATILE ORGANIC ANALYSES
Strict adherence to the requirements and procedures of the respective methods used to
sample and analyze for the volatile organic compounds of interest in this program served as the
basis for assuring that the data generated in these tests were of known quality. This involved:
• Performing sampling train and analytical instrument calibrations as required by
the respective methods
• Following all method procedures without deviation
• Analyzing all samples within method hold time limitations
During this test program, all instruments were calibrated as required by the respective
methods, at a frequency in accordance with the methods. The instruments met calibration
specifications required by the methods before sample analyses were performed. In addition, all
method procedures were followed without deviation.
Table 37 summarizes the sample collection and analysis dates for all samples analyzed
for volatile organics. Six clay/organic liquid feed samples exceeded the specified analysis hold
time of 2 weeks by 1 to 2 days. All kiln ash and scrubber blowdown samples were analyzed
within the 2-week analysis hold time.
Table 38 summarizes the volatile organic practicable quantitation limits (PQLs) achieved
for the kiln ash, scrubber blowdown and VOST samples. The POL was defined to be five times
the blank analysis signal (noise level) of the instrument.
114
-------
TABLE 36. PRECISION, ACCURACY, AND COMPLETENESS OBJECTIVES
Measurement parameter
Measurement/
analytical method
Reference
Conditions
Precision
(% RSD*
or RPDb)
Accuracy
(%)
Completeness
<%)
Volatile organic compounds
in clay/organic liquid feed,
kiln ash, and scrubber
blowdown
Purge and trap by Method
5030, GC/FID analysis by
method in Appendix F
SW-846C,
Appendix F
Direct injection
50
50 to 140
80
Volatile organic compounds
in flue gas
Thermal desorption purge
and trap by Method 5040,
GC/FID analysis by
method in Appendix F
SW-846C,
Appendix F
Thermal
desorption
100
50 to 150
80
Metals
ICAP by Method 6010
SW-846C
Acid digestion
and ICAP
analysis
25
75 to 125
80
'Percent relative standard deviation.
bRelative percent difference.
c(2).
-------
TABLE 37. VOLATILE ORGANIC SAMPLE HOLD TIMES
Analysis hold time
Sample
Collection date
Analysis date
(days)
Clay/organic liquid feed
Test 1
Drum 1
8/16/89
8/18/89
2
Drum 2
8/16/89
8/18/89
2
Test 2
Drum 1
8/1/89
8/16/89
15
Drum 2
8/2/89
8/16/89
14
Drum 3
8/2/89
8/18/89
16
Test 3
Drum 1
8/2/89
8/18/89
16
Drum 2
8/2/89
8/18/89
16
Test 4
Drum 1
7/31/89
8/8/89
9
Drum 2
7/31/89
8/8/89
9
Drum 3
8/1/89
8/16/89
15
Test 5
Drum 1
8/15/89
8/18/89
3
Drum 2
8/16/89
8/18/89
2
Drum 3
8/16/89
8/18/89
2
Test 6
Drum 1
8/10/89
8/18/89
8
Drum 2
8/10/89
8/18/89
8
Test 7
Drum 1
8/9/89
8/18/89
9
Drum 2
8/9/89
8/18/89
9
Test 8
Drum 1
8/9/89
8/18/89
9
Drum 2
8/9/89
8/18/89
9
Test 9
Drum 1
7/25/89
8/7/89
13
Drum 2
7/25/89
8/7/89
13
Matrix spike (10)
9/11/89
9/12/89
1
Kiln ash
Test 1
8/17/89
8/18/89
1
Test 2
8/3/89
8/16/89
13
Test 3
8/4/89
8/8/89
4
Test 4
8/1/89
8/8/89
7
Test 5
8/16/89
8/18/89
2
Test 6
8/15/89
8/18/89
3
Test 7
8/9/89
8/16/89
7
Test 8
8/11/89
8/18/89
7
Test 9
7/28/89
8/7/89
10
(continued)
116
-------
TABLE 37. (continued)
Analysis hold time
Sample
Collection dale
Analysis date
(days)
Scrubber blowdown
Tesi 1
8/17/89
8/18/89
1
Test 2
8/2/89
8/8/89
6
Tesi 3
8/4/89
8/8/89
4
Test 4
8/1/89
8/7/89
6
Test 5
8/16/89
8/18/89
2
Test 6
8/15/89
8/18/89
3
Test 7
8/9/89
8/16/89
7
Test 8
8/11/89
8/18/89
7
Test 9
7/28/89
8/7/89
10
Method 0030 trains (VOST)
Test 1
8/17/89
8/17/89
0
Test 2
8/2/89
8/2/89
0
Test 3
8/4/89
8/4/89
0
Test 4
8/1/89
8/1/89
0
Test 5
8/16/89
8/16/89
0
Test 6
8/15/89
8/15/89
0
Test 7
8/9/89
8/9/89
0
Test 8
8/11/89
8/11/89
0
Test 9
7/28/89
7/28/89
0
7/31/89
3
Matrix spikes (1 per test)
Same as above
Same as above
0
Blanks (1 per test)
Same as above
Same as above
0
117
-------
TABLE 38. VOLATILE ORGANIC ANALYSES PRACTICABLE
QUANTITATION LIMITS ACHIEVED
PQL
Compound
Kiln ash
(mg/kg)
Scrubber
blowdown
(jig/L)
Method 0030
trains (VOST)
(pg/dscm)
Hexane
80
4.8
1.2
1,1-DichIoroe thane
100
6.3
1.6
Chloroform
110
6.4
1.6
1,1,1 -Trichloroethane
85
5.1
1.3
Carbon tetrachloride
100
6.0
1.5
Benzene
93
5.6
1.4
Trichloroelhylene
99
5.9
1.5
1,2-Dichloropropane
96
5.8
1.4
Bromodichl oromethane
120
7.0
1.7
t-1,3-Dichloropropylene
63
3.8
0.95
Toluene
81
4.9
1.2
1,1,2-Trichloroeihane
110
6.5
1.6
Tetrachloroethylene
110
6.5
1.6
Chlorobenzene
95
5.7
1.4
Ethyl benzene
90
5.4
1.4
Bromoform
140
8.5
2.1
1,3-Dichlorobenzene
96
5.8
1.4
1,4-Dichlorobenzcne
91
5.5
1.4
1,2-Dichlorobenzene
120
7.0
1.7
118
-------
6.1.1 Volatile Organic Analysis or Clay/Organic Liquid Feed, Kiln Ash, and Scrubber
Blowdown Samples
As discussed in Section 3, a composite sample of the feed mixture from each drum
prepared was collected and analyzed for the volatile organic hazardous constituents toluene,
chlorobenzene, and tetrachloroethylene. Each drum of a given test mixture was prepared by
mixing together weighed amounts of the individual liquid organic constituents, then mixing
together weighed amounts of the liquid mixture and the clay sorbent. Synthetic waste feed for
a given test was taken from two or three drums depending on test conditions and durations.
Thus, comparing analysis results for each drum of a target test mixture preparation gives a
measure of the precision and accuracy of the entire feed preparation, sampling, and analysis
procedure.
Table 39 provides such a comparison. For each target test mixture preparation, analysis
results for a sample of each drum prepared are compared to the target composition and to each
other. The accuracy of the feed preparation, sampling, and analysis operation is measured by
the ratio of the analyzed concentration of a mixture component to the target value (termed
"Recovery" in Table 39). The precision of the procedure is measured by the percent relative
standard deviation (% RSD) or relative percent difference (RPD) of all samples analyzed.
The data in Table 39 show that analyzed concentrations ranged from 74 to 101 percent
of target concentrations over 21 samples analyzed. Since all measurements were within the 50
to 140 percent recovery range, the accuracy DQO for the measurement was met. The precision
of the procedure, as measured by the % RSD or RPD of the analysis of different drum samples
of a given target formulation, ranged from 4 to 16 percent. Since all were within the precision
DQO of 50 percent for this measurement, this DQO was met.
Ten samples were taken of a clay/organic liquid feed mixture matrix spike prepared
under controlled laboratory conditions (as opposed to bulk feed sample preparation) and
analyzed to determine volatile organic constituent matrix spike recovery. The data in Table 40
show that the recoveries ranged from 75 to 115 percent. All were within the accuracy DQO
range for this measurement of 50 to 140 percent recovery. Thus, this DQO was also achieved.
Furthermore, the % RSD of the replicate spike analyses ranged from 7.1 to 9.1 percent, all
within the DQO for precision of 50 percent.
Matrix spike and matrix spike duplicate samples were also prepared under laboratory
conditions using a kiln ash sample and a scrubber liquor blowdown sample. These samples were
analyzed to determine matrix spike recovery. For these matrices, the data in Table 41 show that
the recoveries ranged from 93 to 101 percent. The RPD of the duplicate spikes ranged from 0
to 4 percent. Thus, the DQOs for both accuracy and precision were met for these matrices.
6.12 Volatile Organic Compounds in Flue Gas
To assess measurement accuracy of the Method 0030 (VOST) sample analyses, each
resin trap was spiked with the surrogates octane and 4-bromofluorobenzene prior to use in
sampling. Surrogate recovery was then measured. The VOST tubes of the third trap pair at the
scrubber exit for Test 7 were broken during sample analysis. Thus, no values are reported for
this trap pair.
119
-------
TABLE 39. CLAY/ORGANIC LIQUID FEED SAMPLE POHC ANALYSIS RESULTS
Toluene
Tetrachloroethylene
Chlorobenzene
Measured
Measured
Measured
concentration
Recovery*
concentration
Recovery*
concentration
Recovery*
Test Test date
Drum
(wt %)
(%>
(wt %)
(%)
(wt %)
(%)
Mixture 1 Target concentration
28.6
0
0
1 8/17/89
1
23.6
83
b
—
—
—
2
22.7
79
—
—
—
—
RPD (%)
4
Mixture 2 Target concentration
21.7
3.4
3.4
2 8/2/89
1
16.9
78
2.9
86
2.8
84
2
18.4
85
3.2
95
3.1
92
3
16.4
76
2.8
83
2.8
82
3 8/4/89
1
19.4
89
3.1
92
3.1
90
2
16.3
75
2.9
86
2.8
82
4 8/1/89
1
18.2
84
3.1
92
2.9
86
2
17.0
78
3.0
88
2.8
81
3
17.9
83
3.0
89
2.9
84
5 8/16/89
1
18.6
86
3.4
100
3.1
91
2
16.1
74
3.1
91
2.8
83
3
17.3
80
3.2
93
3.0
89
6 8/15/89
1
19.2
88
3.4
99
3.3
96
2
16.4
76
2.8
81
2.7
80
7 8/9/89
1
20.0
92
3.2
95
3.1
91
2
17.3
80
2.9
85
2.8
82
8 8/11/89
1
18.8
87
3.4
101
3.1
91
2
17.8
82
3.2
93
2.9
85
% RSD
6.4
6.4
5.4
Mixture 3 Target concentration
14.9
6.9
6.9
9 7/28/89
1
11.0
74
5.7
83
5.1
74
2
12.2
82
6.4
92
6.0
87
RPD (%)
10
12
16
"Ratio of measured concentration to prepared concentration.
b— = Not included in mixture.
-------
TABLE 40. VOLATILE ORGANIC CONSTITUENT RECOVERY FROM CLAY/ORGANIC LIQUID FEED MATRIX
SPIKE SAMPLES
Matrix spike sample
1
2
3
4
5 6
7
8
9
10
Compound
Spike recovery
(%)
% RSD
Toluene
76
94
89
84
88 83
86
75
90
77
7.3
Tctrachlorocthylene
97
111
112
99
110 96
112
88
115
91
9.1
Chloro benzene
88
105
103
95
101 94
99
85
104
89
7.1
-------
TABLE 41. VOLATILE ORGANIC CONSTITUENT RECOVERY FROM KILN ASH AND
SCRUBBER LIQUOR MATRIX SPIKE SAMPLES
Toluene
Tetrachloroethylene
Chlorobenzene
Spike
recovery
Matrix (%)
Duplicate
spike RPD
(*>
Spike
recovery
(%)
Duplicate
spike RPD
(%)
Spike Duplicate
recovery spike RPD
(%) (%)
Q
W
Kiln ash
Scrubber
96
100
0
4
93
99
91
101
Table 42 summarizes octane recovery from the VOST trap pairs. The data in the table
show that octane recovery ranged from 0 to 164 percent, with 5 of 82 recoveries not within the
DQO range of 50 to 150 percent. Measurement completeness was 94 percent, which met the
DQO for recovery completeness of 80 percent.
Table 43 similarly summarizes 4-bromofluorobenzene recovery from the VOST trap
pairs. The data show that 4-bromofluorobenzene recovery ranged from 0 to 185 percent, with
11 to 82 recoveries not within the DQO range of 50 to 150 percent. Measurement completeness
(percent of analyses performed meeting the DQO) was 87 percent, which met the DQO for
recovery completeness of 80 percent.
As a further measure of VOST analysis accuracy, and as a measure of analysis precision,
nine matrix spike VOST trap pairs were prepared and analyzed. Results of these analyses are
summarized in Table 44. The low recoveries for Test 3 suggest that a leak developed during the
sample desorption process. Recoveries for the remaining tests were very good, ranging from 92
to 122 percent; 5 of 45 recoveries were not within the DQO range of 50 to 150 percent.
Measurement completeness was 89 percent, which met the DQO for recovery completeness of
80 percent. The % RSD of the matrix spike analyses for all samples was 32 percent, all within
the DQO for precision of 100 percent.
As part of the routine analytical procedure, laboratory blanks were prepared and
analyzed before proceeding with the test samples. Sample analysis began only after the blanks
indicated that there was no system contamination by volatile organics above detection limits, and
that internal standard and surrogate recoveries confirmed the absence of system leaks. Field
blanks were also generated and analyzed for each test to identify cases of potential sample
contamination. Analysis results for these samples are included in Appendix C along with the
VOST results. These data indicate that no significant contamination occurred during handling
of the VOST tubes during testing.
62 METALS ANALYSIS
As with the volatile organic analyses, the metals analyses were performed in strict
adherence to, without deviation from, method procedures. This included instrument calibration
procedures, frequency of calibration, and specifications for acceptable calibration.
122
-------
TABLE 42. OCTANE SURROGATE RECOVERY FROM VOST TRAPS
Afterburner exit traps
Scrubber system exit traps
Other traps
Octane
Octane
Octane
recovery
recovery
recovery
Sample
(%)
Sample
m
Sample
(%)
Stack Samples
Test 1 (8/17/89)
Test 1 (8/17/89)
Test 4 (8/1/89)
1st trap pair
100
1st trap pair
107
1st trap pair
110
2nd trap pair
101
2nd trap pair
107
2nd trap pair
95
3rd trap pair
85
3rd trap pair
105
3rd trap pair
104
Test 2 (8/2/89)
Test 2 (8/2/89)
Test 7 (8/9/89)
1st trap pair
99
1st trap pair
106
1st trap pair
114
2nd trap pair
101
2nd trap pair
100
2nd trap pair
113
3rd trap pair
97
3rd trap pair
104
3rd trap pair
106
Test 3 (8/4/89)
Test 3 (8/4/89)
Test 8 (8/11/89)
1st trap pair
103
1st trap pair
99
1st trap pair
0
2nd trap pair
97
2nd trap pair
93
2nd trap pair
0
3rd trap pair
2
3rd trap pair
108
3rd trap pair
164
Test 4 (8/1/89)
Test 4 (8/1/89)
Field blank traps
1st trap pair
109
1st trap pair
110
7/28/89
101
2nd trap pair
102
2nd trap pair
107
B/l/89
92
3rd trap pair
96
3rd trap pair
106
8/2/89
104
Test 5 (8/16/89)
8/4/89
102
Test 5 (8/16/89)
8/9/89
98
1 st trap pair
99
1st trap pair
104
8/11/89
98
2nd trap pair
101
2nd trap pair
103
8/15/89
98
3rd trap pair
102
3rd trap pair
106
8/16/89
104
Test 6 (8/15/89)
Test 6 (8/15/89)
8/17/89
103
1st trap pair
102
1st trap pair
125
2nd trap pair
103
2nd trap pair
113
3rd trap pair
101
3rd trap pair
122
Test 7 (8/9/89)
Test 7 (8/9/89)
Matrix spike traps
1st trap pair
101
1st trap pair
96
7/28/89
97
2nd trap pair
101
2nd trap pair
105
8/1/89
100
3rd trap pair
102
8/2/89
96
Test 8 (8/11/89)
Test 8 (8/11/89)
8/4/89
8/9/89
17
103
1st trap pair
102
1st trap pair
116
8/11/89
95
2nd trap pair
102
2nd trap pair
105
8/15/89
101
3rd trap pair
99
3rd trap pair
116
8/16/89
96
Test 9 (7/28/89)
Test 9 (7/28/89)
8/17/89
101
1st trap pair
116
1st trap pair
114
2nd trap pair
112
2nd trap pair
111
3rd trap pair
115
3rd trap pair
112
123
-------
TABLE 43. 4-BROMOFLUOROBENZENE SURROGATE RECOVERY FROM VOST TRAPS
Afterburner exit traps
4-Bromo-
fluorobenzene
recovery
Scrubber system exit traps
4-Bromo-
fluorobenzene
recovery
Other traps
4-Bromo- ^
fluorobenzene m
recovery
Sample
(%)
Sample
(%)
Sample
(%)
Stack Samples
Test 1 (8/17/89)
Test 1 (8/17/89)
Test 4 (8/1789)
1st trap pair
76
1st trap pair
93
1st trap pair
110
2nd trap pair
94
2nd trap pair
78
2nd trap pair
12
3rd trap pair
87
3rd trap pair
48
3rd trap pair
62
Test 2 (8/2/89)
Test 2 (8/2/89)
Test 7 (8/9/89)
1st trap pan
95
1st trap pair
102
1st trap pair
98
2nd trap pair
99
2nd trap pair
88
2nd trap pair
88
3rd trap pair
81
3rd trap pair
88
3rd trap pair
61
Test 3 (8/4/89)
Test 3 (8/4/89)
Test 8 (8/11789)
1st trap pair
61
1st trap pair
89
1st trap pair
0
2nd trap pair
62
2nd trap pair
75
2nd trap pair
185
3rd trap pair
0
3rd trap pair
81
3rd trap pair
113
Test 4 (8/1789)
Test 4 (8/1/89)
Field blank traps
1st trap pair
94
1st trap pair
95
7/28/89
103
2nd trap pair
98
2nd trap pair
98
8/1/89
93
3rd trap pair
96
3rd trap pair
97
8/2/89
104
8/4/89
105
Test 5 (8/16/89)
Test 5 (8/16/89)
8/9/89
89
1st trap pair
89
1st trap pair
42
8/11/89
104
2nd trap pair
99
2nd trap pair
101
8/15/89
94
3rd trap pair
55
3rd trap pair
77
8/16/89
106
Test 6 (8/15/89)
Test 6 (8/15/89)
8/17/89
106
1st trap pair
53
1st trap pair
106
2nd trap pair
79
2nd trap pair
23
3rd trap pair
93
3rd trap pair
105
Test 7 (8/9/89)
Test 7 (8/9/89)
Matrix spike traps
1st trap pair
94
1st trap pair
24
7/28/89
101
2nd trap pair
82
2nd trap pair
96
8/1/89
104
3rd trap pair
91
8/2/89
96
Test 8 (8/11/89)
Test 8 (8/11789)
8/4/89
8/9/89
4
109
1st trap pair
94
1st trap pair
91
8/11/89
102
2nd trap pair
93
2nd trap pair
61
8/15/89
103
3rd trap pair
77
3rd trap pair
85
8/16/89
95
Test 9 (7/28/89)
Test 9 (7/28/89)
8/17/89
101
1st trap pair
94
1st trap pair
100
2nd trap pair
100
2nd trap pair
83
3rd trap pair
108
3rd trap pair
0
124
-------
TABLE 44. VOLATILE ORGANIC CONSTITUENT RECOVERY FROM VOST MATRIX SPIKE SAMPLES
% recovery
Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9
Compound (8/17/89) (8/2/89) (8/4/89) (8/1/89) (8/16/89) (8/15/89) (8/9/89) (8/11/89) (7/28/89) % RSD
Benzene 100 101 6 107 102 108 106 108 101 33
Trichlorocthylene 119 117 11 121 118 122 120 119 117 32
Toluene 98 97 7 102 96 96 96 103 97 33
Tetrachloroethylenc 108 104 9 108 107 109 107 107 106 32
Chlorobenzenc 96 92 5 95 93 94 93 93 92 33
-------
The first test samples were obtained on July 25, 1989. All test samples were
appropriately preserved as required by the analysis method. All metals analyses were completed
before December 14, 1989, within the analytical hold time limit of 6 months.
The achieved PQLs for the metals in the various sample matrices are summarized in
Table 45. Again, the PQL was defined to be five times the blank instrument analysis signal
(noise Jevel). The only method blank sample requiring analysis was the filter blank from the flue
gas metals sampling train filter at the scrubber exit. Metals in the filter blank were reported to
be below the PQLs given in Table 45.
For determining the accuracy and precision of the metals analyses, matrix spike and
matrix spike duplicate samples were prepared in the laboratory. The matrices employed were
kiln ash, scrubber blowdown, and multiple metals train impinger solutions. For each matrix, the
added spike for each metal was two times the typical level found on original analysis or five
times the analytical method detection limit for that metal.
Kiln ash matrix spike samples were prepared by adding the metals in an aqueous
solution to an ash sample from Test 1. Specific matrix spike solutions were prepared in the
laboratory, corresponding to the scrubber blowdown, the solution from combined impingers 1 and
2 of the multiple metals sampling train, and the solution from impingers 3 and 4 of the multiple
metals train. Separate matrix spikes were prepared to represent the afterburner exit and
scrubber exit sampling trains because of the different concentrations of metals observed in the
impinger solutions of the trains during the tests.
Table 46 summarizes the metals matrix spike and matrix spike duplicate sample analysis
results. Recoveries ranged from 56 to 446 percent, with 48 of 54 recoveries (89 percent) falling
within the accuracy DQO of 75 to 125 percent, thus satisfying the accuracy completeness DQO
of 80 percent. The RPD values ranged from 0 to 14 percent, all within the precision DQO of
25 percent, thus meeting the precision completeness DQO of 80 percent.
As a further measure of trace metal analytical accuracy and precision, recall from
Section 4.1 that a composite sample of the aqueous metal spike solution for each test was
analyzed for the 7 spiked test trace metals. These can be considered field aqueous solution
matrix spike samples. Comparing analysis results from test to test will give a measure of the
accuracy and precision of the entire aqueous spike solution preparation, sample collection and
preservation, and laboratory analysis process.
Table 15 summarized the results of all aqueous metal spike solution analyses. Table 47
presents these results in different form. Specifically, Table 47 tabulates aqueous spike solution
matrix spike recoveries where recovery is defined as the analyzed concentration divided by the
target preparation concentration. The data in Table 47 show that analysis recoveries ranged
from 41 to 105 percent.
For 6 of 9 solutions, recoveries were comparable to prepared concentrations for all
metals. Analyzed concentrations for Tests 1 and 8 were uniformly lower than the spiked
concentrations for all metals. This is consistent with sample dilution prior to analysis. As
discussed in Section 4.1, both the analyzed concentrations and the prepared concentrations were
used during the initial data analysis. Conclusions resulting from the reduced data were not
126
-------
TABLE 45. METALS ANALYSES PRACTICABLE QUANTITATION LIMITS ACHIEVED
PQL
Matrix
Arsenic
Barium
Bismuth
Cadmium
Chromium
Copper
Lead
Magnesium
Strontium
Clay/organic liquid Teed
(mg/kg)
4.2
0.5
6.5
0.5
3.6
0.6
2.8
3.7
0.3
Kiln ash (mg/kg)
3.4
0.6
6.5
0.5
3.3
0.6
2.6
0.3
0.4
Kiln ash TCLP leachatcs
(Ug/L)
90
7.0
130
10
85
14
65
23
7.0
Scrubber blowdown
(Hg/L)
100
10
130
10
85
10
40
10
10
Multiple metals trains:
Probc wash
(|ig/samplc)
9.0
11
7.0
1.0
7.0
1.0
6.0
7.0
0.5
Filter (|Xg/Tillcr)
7.0
4.0
7.0
1.0
4.0
1.0
4.0
7.0
0.5
Impingcr solution
(Hg/L)
90
20
100
10
65
10
60
30
10
-------
TABLE 46. RECOVERY OF METALS FROM MATRIX SPIKES
Matrix Arsenic Rarium Bismuth Cadmium Chromium Copper Lead Magnesium Strontium
Kiln ash:
Spike recovery (%) 75 92 108 446 101 94 138 56 101
Duplicate spike RPD (%) 0.2 14 5.7 1.1 0.7 3.8 2.0 6.4 11
Scnibbcr liquor:
Spike recovery (%) 84 97 95 102 91 93 97 92 97
Duplicate spike RPD (%) 3.1 1.4 1.0 1-7 4.6 1.3 1.8 0.6 1.8
Afterburner exit impinger
solutions:
Impingcrs 1 and 2:
Spike recovery (%) % 101 99 71 97 101 103 99 103
Duplicate spike RPD (%) 8.3 5.2 6.0 6.4 5.5 5.8 6.5 6.7 6.9
Impingcrs 3 and 4:
Spike recovery (%) 74 106 102 88 97 99 110 92 107
Duplicate spike RPD (%) 11 5.7 3.9 2.9 13 2.9 2.6 7.9 1.6
Scnibber exit impinger
solutions:
Impingers 1 and 2:
Spike recovery (%) 99 98 98 89 91 % 203 103 102
Duplicate spike RPD (%) 0.3 5.4 7.3 12 8.3 1.7 1.5 4.2 2.8
npingers 3 and 4:
Spike recovery (%) 88 95 96 91 86 92 97 89 %
Duplicate spike RPD (%) 2.3 1.0 3.3 0 7.4 0.4 0.5 2.8 2.1
-------
TABLE 47. AQUEOUS METALS SPIKE SOLUTION ANALYSIS ACCURACY AND PRECISION
Prepared
spike solution
concentration
(ppm as
metal)
Analysis recovery (%)"
. 0)
3
o
t-
Metal
Tfcst I
(8/17/89)
n-st
(8/2/89)
Tfcst 3
(8/4/89)
Ttest 4 Tfcst 5
(8/1/89) (8/16/89)
Tfcst 6
(8/15/89)
Tfe st 7
(8/9/89)
Tfest 8
(8/11/89)
Test 9
(7/28/89)
Q
CO
UJ
%RSD
Arsenic
1,590
81
102
102
93 99
101
41
67
101
23
Barium
12,700
82
101
102
93 100
102
100
69
98
11
Bismuth
12,700
74
93
94
87 91
94
70
67
94
11
Cadmium
318
80
99
103
92 98
101
97
68
100
11
Copper
12,700
79
96
97
90 96
98
%
68
%
10
Lead
1,590
74
92
92
85 91
92
91
82
91
6
Strontium
12,700
84
103
105
96 103
105
103
72
102
11
to
v©
'Analyzed amount (sec Table 4-2)/prepared amount.
-------
significantly different and the analyzed spike concentrations were used in the data reduction
calculations. As reported in Section 4.1, the analyzed arsenic value for Test 7 was treated as an
outlier, and the prepared concentration of 1590 mg/L was used in the calculations.
The above discussion aside, the data in Table 47 show that 54 of 63 determinations
(86 percent) were within the accuracy DQO of 75 to 125 percent recovery, with the completeness
objective for this measurement being 80 percent.
The data in Table 47 also show that the precision of individual metal analyses for the
aqueous spike solutions ranged from 6 to 20 % RSD. All were within the DQO for this
measurement of 25 percent. Thus, the measurement precision DQO was met.
Finally, recall from Section 4.1 that clay matrix samples for all parametric trace metal
tests were analyzed for the nine test trace metals. Comparing the results of these analyses gives
a measure of the analytical precision of the metal analyses in the clay matrix. Table 6-13
provides this comparison for the 6 metals detected above PQL in all clay samples. As indicated
in Table 48, the demonstrated analytical precision for the 6 metals reported above detection
limits ranged from 2.6 to 9.9 % RSD. Again, all these results were within the precision DQO
of 25 % RSD.
130
-------
TABLE 48. CLAY/ORGANIC LIQUID FEED ANALYSIS PRECISION
Measured clay/organic liquid mixture sample metal concentration (mg/kg)
Test 1
Test 2
Test 3
Test 4
Test 5
Test 6
Test 7
Test 8
Test 9
Metal
(8/17/89)
(8/2/89)
(8/4/89)
(8/1/89)
(8/16/89)
(8/15/89)
(8/9/89)
(8/11/89)
(7/28/89)
% RS
Barium
21
25
24
23
23
21
22
23
24
5.8
Cadmium
1.4
1.4
1.2
1.3
1.4
1.3
1.3
1.5
1.7
9.9
Chromium
45
38
38
38
42
39
42
42
37
6.4
Copper
25
26
26
27
24
24
26
25
27
4.2
Magnesium
18,800
18,800
19,050
19,900
18,100
18,400
19,100
18,900
18,500
2.6
Strontium
30
32
32
33
31
31
31
32
31
2.6
-------
REFERENCES
1. Fournier, Jr., D. J., W. E. Whitworth, and L. R. Waterland. "Pilot-Scale Evaluation of the
Fate of Trace Metals in a Rotary Kiln Incinerator with a Venturi Scrubber/Packed Column
Scrubber." Acurex Draft Report prepared under EPA Contract 68-03-3267, October 1989.
2. "Test Methods for Evaluating Solid Waste: Physical/Chemical Methods." EPA SW-846,
3rd ed., November 1986.
3. 40 CFR Part 60, Appendix A.
4. "Determining the Properties of Fine Particulate Matter." ASME Power Test Code 28.
5. Harris, J. C., et al. "Sampling and Analysis Methods for Hazardous Waste Incineration."
EPA-600/8-84-002, February 1984.
6. 40 CFR Part 261, Appendix II.
7. 40 CFR Part 261.24.
8. Barton, R. G., et al. "Development and Validation of a Surrogate Metals Mixture."
Proceedings of at the Fifteenth Annual Research Symposium: Remedial Action. Treatment
and Disposal of Hazardous Waste. EPA/60019-90/006, February 1990.
132
------- |