EPA/600/R-92/003a
March 1992
EMISSIONS OF METALS, CHROMIUM AND NICKEL SPECIES, AND ORGANICS
FROM MUNICIPAL WASTEWATER SLUDGE INCINERATORS
VOLUME I: SUMMARY REPORT
Prepared by:
Robin R. Segall
Entropy Environmentalists, Inc.
Research Triangle Park, North Carolina 27709
William G. DeWees
DEECO, Inc.
Cary, North Carolina 27519
F. Michael Lewis
Mountain View, California 94040
EPA Contract No. 68-CO-0027
Work Assignment No. 0-5
Technical Managers:
Harry E. Bostian, Ph.D.
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
Eugene P. Crumpler
Office of Water
U.S. Environmental Protection Agency
Washington, D.C. 20460
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This material has been funded wholly or in part by the United States
Environmental Protection Agency's Risk Reduction Engineering Laboratory and Office
of Water (OW) under Contract No. 68-02-4442, Work Assignment No. 81; Contract No.
68-02-4462, Work Assignment No. 90-108; and Contract No. 68-C0-0027, Work
Assignment No. 0-5. It has been subject to the Agency's 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
-------
FOREWORD
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 research and the user community.
The problem of disposing of primary and secondary sludge generated at municipal
wastewater treatment facilities is one of growing concern. Sludge of this type may
contain toxics such as heavy metals and various organic species. Viable sludge disposal
options include methods of land disposal or incineration. In determining the
environmental hazards associated with incineration, the Risk Reduction Engineering
Laboratory and the Office of Water have sponsored a program to monitor the emissions
of metals and organics from a series of municipal wastewater sludge incinerators. The
following document presents a summary of the results and testing procedures from all
five test sites (Sites 5, 6, 7, 8, and 9)-1'2^4-5-6-7-8
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
iii
-------
ABSTRACT
The U.S. Environmental Protection Agency (EPA) Office of Water (OW) has
been developing new regulations for sewage sludge incinerators and EPA's Risk
Reduction Engineering Laboratory (RREL) has been assisting OW in the collection of
supporting data. There is particular concern regarding chromium and nickel species in
the emissions from incineration of municipal wastewater sludge because of the associated
cancer risk. OW has drafted risk-based sludge regulations under Section 405d of the
Clean Water Act which have been published for comment in the Federal Register.
Volume 54, No. 23, February 6, 1989. Final regulations are scheduled for publication in
the Federal Register in January 1992.
EPA sponsored testing at five sewage sludge incinerators under this study (Sites 5,
6, 7, 8, and 9). Four incinerators tested under a previous project conducted by Radian
Corporation are included in the Site numbering convention used. At three of the five
incinerators in the present project, a wide range of data oh emissions of metals,
hexavalent chromium, nickel subsulfide, polychlorinated dibenzodioxins and furans
(PCDD/PCDFs), semivolatile and volatile organic compounds, carbon monoxide (CO),
and total hydrocarbon (THC) were determined. Two multiple hearth incinerators and
one fluidized bed incinerator were tested. All three of these incinerators employed
venturi/scrubbers for controlling air emissions. On the fluidized bed unit a pilot-scale
wet electrostatic precipitator (ESP) was installed. A full-scale wet ESP was installed on
one of the multiple hearth units. Feed sludge was tested for metals, moisture, and
carbon and hydrogen content. The other two test sites included an evaluation of
hexavalent chromium methods (Site 5) and an evaluation of continuous emission
monitoring systems for carbon monoxide and total hydrocarbon (Site 7).
Of the metals measured, chromium, lead, and nickel consistently had the highest
feed rate to the incinerators. Cadmium and lead had the highest emission factors of the
metals fed to the incinerators. The emission control devices at the multiple hearth
incinerators had similar removal efficiencies for particulate matter, chromium, and
nickel, with lead and cadmium removal efficiencies being less than particulate matter.
At the fluidized bed incinerator, the venturi/scrubber had the highest removal efficiency
by a scrubber system without discriminating between metals and particulate matter. The
wet ESPs were effective in further removal of the metals and particulate matter emitted
from the venturi/scrubbers.
The hexavalent chromium test method developed for this program provided
acceptable results for the measurement of hexavalent chromium without artifact
formation at the outlet locations. The ratio of hexavalent chromium to total chromium
was highest (8.3 - 42%) when lime was used for sludge conditioning, during good
combustion conditions, and with the long residence times required for combustion of
iv
-------
sludge in a multiple hearth incinerator. The ratio of hexavalent chromium to total
chromium in the emissions from a fluidized bed incinerator (despite relatively high total
chromium levels) was very low (<2%), probably due to the short sludge retention time
in the incinerator and the absence of alkaline material in the sludge.
The ratio of nickel subsulfide to total nickel was extremely low (< 10%) under
both normal combustion and improved combustion conditions.
PCDD/PCDFs, semivolatile organic compounds, and volatile organic compounds
were measured in the controlled emissions from both types of incinerators. The wet ESP
was effective in controlling the emissions of these compounds.
The combustion efficiency at both multiple hearth incinerators was improved
during the test programs. The process operating conditions established for the second
series of test runs at Site 6 and Site 9 greatly reduced the concentrations of carbon
monoxide (CO) and total hydrocarbon (THC) emissions. A good correlation was seen
between CO emissions and the THC emissions. The fluidized bed incinerator displayed
better combustion efficiency than could be achieved with the multiple hearth
incinerators.
v
-------
TABLE OF CONTENTS
Section Page
Disclaimer ii
Foreword iii
Abstract iv
List of Figures vii
List of Tables viii
Acknowledgement ix
1.0 Introduction 1-1
2.0 Sampling and Analytical Procedure 2-1
3.0 Results and Discussion 3-1
3.1 Metals and particulate 3-1
3.2 Hexavalent chromium 3-5
3.3 Nickel speciation 3-8
3.4 PCDD/PCDFS and semivolatile and volatile compounds 3-8
3.5 Carbon monoxide and total hydrocarbon monitoring 3-14
4.0 Conclusions 4-1
4.1 Metals and particulates 4-1
4.2 Hexavalent chromium 4-1
4.3 Nickel subsulfide 4-2
4.4 Continuous emission monitoring of CO and THC 4-3
4.5 Semivolatile organics 4-3
4.6 Volatile organics 4-3
4.7 Conclusions on Study 4-4
References 5-1
VI
-------
LIST OF FIGURES
Number Page
1 Schematic of multiple metals/particulate sampling train 2-4
2 Sample recovery procedures for multiple metals train 2-5
3 Sample preparation and analysis scheme for multiple metals
trains 2-6
4 Schematic of recirculating reagent impinger train for
hexavalent chromium 2-7
5 Sample recovery scheme for hexavalent chromium
impinger train 2-9
6 Schematic of inlet location recirculating reagent
impinger train for hexavalent chromium 2-10
7 Analytical protocol for quadruplicate recirculatory train
hexavalent chromium sampling at midpoint and outlet
locations 2-11
8 Schematic of nickel/nickel subsulfide sampling train 2-12
9 Schematic of sample recovery procedures for nickel trains 2-14
10 Schematic of the MM5 train for semivolatile organics
and PCDD/PCDF 2-16
11 Semivolatile organic train sample recover scheme 2-17
12 Extraction for semivolatile organic samples 2-18
13 Schematic of volatile organic sampling trains 2-19
14 Correlation of combustion efficiency and hexavalent to
total chromium ratio 3-7
15 Total hydrocarbon emissions versus carbon monoxide
emissions, Site 6 3-15
16 Total hydrocarbon emissions versus carbon monoxide
emissions, Site 9 3-16
vii
-------
LIST OF TABLES
Number Page
1 Characteristics of incinerators and sludges at the
five sites 1-5
2 Specific elements and compounds of interest 2-2
3 Typical process monitoring data 2-21
4 Feed rates for metals in the sludge (g/hr) 3-2
5 Particulate and metals stack emission factors for steady
state (low CO) and normal operations 3-2
6 Metals and particulate removal efficiency across the
various control devices 3-4
7 Ratio of metals to particulate matter emissions under
steady state (low CO) and normal operation 3-5
8 Hexavalent chromium sampling results 3-6
9 PCDD/PCDF emissions summary 3-10
10 Semivolatile emissions summary for outlet and midpoint
at Site 9 3-11,12
11 Volatile organics emissions summary 3-13
viii
-------
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the following invaluable contributions to the
efforts described in this report: Dr. Joseph E. Knoll of the Quality Assurance Division of
EPA for advice and assistance on hexavalent chromium sampling and analysis, Dr.
Vladimar Zatka of Zatka Chemical Consulting Company for advice and analytical work
on nickel speciation, Dr. Nolan F. Mangelson of Brigham Young University for
instrumental analysis of chromium and nickel species, Dr. Kate K. Luk of Research
Triangle Institute for metals analysis, Ms. Elizabeth J. Arar and Dr. Stephen Long of
Technology Applications, Inc. for IC/PCR and ICP/MS analysis of hexavalent and total
chromium, Dr. Yves Tondeur of Triangle Laboratories for trace organic analysis, and Dr.
Scott C. Steinsberger, formerly of Entropy Environmentalists, Inc. for his tireless effort
and ingenuity in developing new methodologies.
ix
-------
1.0 INTRODUCTION
The U.S. Environmental Protection Agency (EPA) Office of Water (OW) has
been developing new regulations for sewage sludge incinerators and EPA's Risk
Reduction Engineering Laboratory (RREL) has been assisting OW in the collection of
supporting data. There is particular concern regarding chromium and nickel species in
the emissions from incineration of municipal wastewater sludge because of the associated
cancer risk. OW has drafted risk-based sludge regulations under Section 405d of the
Clean Water Act which have been published for comment in the Federal Register.
Volume 54, No. 23, February 6, 1989. Final regulations are scheduled for publication in
the Federal Resister in January 1992.
¦ - ¦¦ i i *
The draft regulations are based on the risk incurred by the "most exposed
individual" (MEI). The MEI approach involves calculating the risk associated with an
individual residing for seventy years at the point of maximum ground level concentration
of the emissions just outside the incinerator facility property line. EPA's proposal for
regulating sewage sludge incinerators is based on ensuring that the increased ambient air
concentrations of metal pollutants emitted from sludge incinerators are below the
ambient air human health criteria. The increased ambient air concentrations for four
carcinogenic metals, arsenic, chromium, cadmium, and nickel, are expressed as annual
averages. The concentrations are identified in the proposed regulations as Risk Specific
Concentrations (RSC). Both nickel and chromium emissions from sludge incinerators
presented a specific problem in establishing RSCs, because unknown portions of the
emissions of these metals are in forms which are harmful to human health. In
performing the risk calculations, EPA assumed that 1% of the emissions of chromium
from the sludge incinerators is in the most toxic form, hexavalent chromium. For nickel,
1-1
-------
EPA assumed that 100% of the nickel emissions are in the most toxic form, nickel
subsulfide.
Chromium is likely to be emitted in either the highly carcinogenic hexavalent state
(Cr+6) or in the noncarcinogenic trivalent state (Cr+3). Trivalent chromium has not been
shown to be carcinogenic and is toxic only at levels higher than those normally found in
sewage sludge incinerator emissions. Although hexavalent chromium (as the most
oxidized form) could be reasonably expected to result from combustion processes,
investigators speculate that most of the chromium is likely to be emitted in the trivalent
state.9 This speculation is based on hexavalent chromium being highly reactive, and thus
likely to react with reducing agents to form trivalent chromium.
Studies have been conducted to determine the potential for chromium in sewage
sludge to be converted to the hexavalent form. Analysis of laboratory combusted sludges
dosed with various levels of lime and ferric chloride revealed that the hexavalent to total
chromium ratio increased with lime dosage.9 One-hundred percent conversion of
trivalent chromium to hexavalent chromium was observed in several of the tests.9 These
tests indicate that when lime and ferric chloride are used as sludge conditioners, high
ratios of hexavalent to total chromium may be formed under certain incinerator
operating conditions. EPA has previously sponsored emission testing studies for
measurement of hexavalent chromium at two full-scale sludge incinerators.10'11 For one
site, the hexavalent chromium concentrations were below the analytical detection limit;
for the other site, a hexavalent-to-total chromium ratio of 13% was calculated. The 1%
value chosen for the draft regulations may seem low. This is the result, however, of
weighting various values to give the most credible ones more influence. With this
approach, lower values were assigned a stronger contribution.
The lack of a substantial data base on hexavalent chromium emissions prompted
the following statement in the EPA's Technical Support Document for the Incineration
of Sewage Sludge: "EPA plans to perform additional tests of sewage sludge incinerator
emissions for hexavalent chromium before this proposed rule is finalized. The additional
data should allow the Agency to better understand how hexavalent chromium is
generated in sewage sludge incinerators." As previously stated, EPA assumed that 100%
1-2
-------
of the nickel emissions are in the subsulfide form to calculate a RSC. Since the Agency
had no nickel subsulfide emission data from sewage sludge incinerators, it took the most
conservative approach in conducting the nickel risk analysis and assumed that all emitted
nickel compounds cause the same health effects as nickel subsulfide. Again, the
Technical Support Document stated: "As additional data become available on the form
of chromium and nickel emissions from combustion sources, the Agency will consider
what changes, if any, would be appropriate for these proposed regulations."
To collect additional data, a comprehensive test program was developed to
determine the ratios of hexavalent-to-total chromium and nickel subsulfide-to-total nickel
for a typical sewage sludge incinerator under normal combustion conditions (higher
concentrations of carbon monoxide and total hydrocarbons) and improved combustion
conditions (lower concentrations of carbon monoxide and total hydrocarbons).
Seven secondary objectives also beneficial to the overall test program were
established.
(1) Implement sampling and analytical procedures for chromium and nickel in
uncontrolled and controlled flue gas emissions from municipal sewage
sludge incinerators.
(2) Compare the ratios of emissions of (1) hexavalent-to-total chromium and
(2) nickel subsulfide-to-total nickel for various types of municipal sewage
sludge incinerators and for different operating conditions.
(3) Compare the emission results for chromium and nickel subspecies
determined by different analytical procedures.
(4) Gather data on other metals and organic and inorganic gaseous
components (as cited in the Federal Register. Volume 54, No. 23, February
6, 1989) in uncontrolled and controlled incinerator emissions to obtain
background data on the effect of operating conditions on these emissions.
(5) Evaluate application of a wet electrostatic precipitator as a retrofit control
system on existing facilities to meet the new sewage sludge emission
regulations.
1-3
-------
Five municipal wastewater sludge incinerators (designated Sites 5, 6, 7, 8, and 9)
were selected for testing. (Four incinerators tested, Sites 1, 2, 3, and 4, under a previous
project conducted by Radian Corporation are included in the Site numbering convention
used here). Methods evaluation for the hexavalent chromium methods were conducted
at Site 5. Continuous emissions monitoring evaluations for total hydrocarbon (THC) and
carbon monoxide (CO) were conducted at Site 7. The full scale testing for metals,
chromium and nickel speciation, and organics were conducted at Sites 6, 8, and 9. The
general characteristics of Site 6, 8, and 9 are summarized in Table 1.
At the start of the test program, no published EPA emission measurement test
methods for the sampling and analysis of hexavalent chromium or nickel subsulfide were
available. Testing conducted at Site 5 was used primarily to develop a satisfactory test
method for hexavalent chromium. Previous test methods for hexavalent chromium were
shown to result in significant and unquantified conversion of chromium from the
hexavalent to trivalent state. The hexavalent chromium test method developed and used
for this test program minimized the hexavalent to trivalent conversion and conversion
that occurred was quantified.
The test program conducted at Site 7 was intended only, to evaluate continuous
emission monitoring systems (CEMSs) for carbon monoxide (CO) and total hydrocarbon
(THC) and investigate if a correlation between CO and THC emissions exist at
municipal wastewater sludge incinerators. CO and THC were also measured at Sites 6,
8, and 9 during the comprehensive testing for hexavalent chromium, nickel species,
metals, and volatile and semivolatile organics.
At Site 6, combined thickened sludge was dewatered with four filter presses.
Lime slurry and ferric chloride solution are used to condition the sludge for dewatering.
The incinerator tested was one of the two identical multiple (eight) hearth furnaces. The
air pollution control system associated with this furnace consists of an afterburner (which
was not used during the test program), a water injection venturi, and an impingement
tray scrubber.
At Site 8, approximately 15 to 17 tons of solids were dewatered by one filter press
and fed to the fluidized bed incinerator. The air pollution control system associated with
1-4
-------
TABLE 1. CHARACTERISTIC OF INCINERATORS AND SLUDGES AT THE FIVE SITES
Site 6 - Normal Site 6 - Low CO Site 8 - Normal Site 9 - Normal Site 9 - Improved
Furnace type
Control device
Multiple Hearth Multiple Hearth Fluidized Bed Multiple Hearth Multiple Hearth
Venturi Scrubber Venturi Scrubber Venturi Scrubber/ Venturi. Scrubber/ Venturi Scrubber/
Pilot Wet ESP Wet ESP Wet ESP
Sludge feed rate (//hr) 3733
Inlet' stack gas parameters:
Gas temperature (°C) 469
Gas oxygen (%) . 13.0
Gas flow (dscmm) 468
3460
555
11.5
430
Midpoint' stack gas parameters:
No midpoint location at Site 6
Gas temperature (°C)
Gas oxygen (%)
Gas flow (dBcrtun)
Outlet' stack gas parameters:
Gas temperature (°C) 63
Gas oxygen (%) 14.1
Gas flow (dBcmm) 513
Sludge characteristics:
% solids 26
% volatiles 59
Heating value (btu//) 6094
65
13.3
531
27
56
5481
4968
604
8.3
2947
35
7.8
2982
28
7.8
1405
20
68
8299
7482
433
12.9
4191
27
11.2
3691
52
14.9
5530
21
63
8481
7460
688
10.4
4169
33
10.4
4169
68
12.4
5175
22
78
8601
""Inlet" and "Outlet" refer to inlet and outlet of the pollution control Bystems, "MidpoLnt" Ls between Wet
ESP and Scrubber Systems.
-------
this incinerator consists of a water injection venturi, and an impingement tray scrubber.
A pilot-scale wet electrostatic precipitator was evaluated during testing at Site 8.
At Site 9, the sludge incinerator is a multiple (seven) hearth furnace. At the
present time, the sludge is polymer conditioned and dewatered by two belt presses and
then deposited onto a series of inclined and horizontal conveyor belts for feeding into
the incinerator. The testing at Site 9 was to evaluate a furnace that did not use lime
conditioning for the sludge filtration. It was discovered at the completion of the program
that some of the Site 9 sludge that was trucked in had some lime added. Also a small
amount of lime was added to the wastewater entering the plant facilities. The total
amount of lime present was about 2.5% of the solids in the sludge, which is less than a
typical amount of lime used for sludge conditioning but more than normally found in
sludge at a typical plant that does not use lime. The air pollution control system
associated with this incinerator consists of an adjustable throat venturi scrubber and
three plate tray scrubber with a Chevron mist eliminator. A full-scale upflow, wet
electrostatic precipitator was evaluated during the test program at Site 9.
This document is labelled Volume I in a series of nine volumes. Volume I
presents a summary of the results and testing procedures from all five test sites (Sites 5,
6, 7, 8, and 9). Test data is presented in the Results and Discussion (Section 2.0) and
briefly summarized in the Conclusions (Section 4.0). Volumes II through IX document
results and procedures from each individual test site. (Volumes IV, VII, and IX are the
appendices to Reports for Sites 6, 8, and 9, respectively).
1-6
-------
2.0 SAMPLING AND ANALYTICAL PROCEDURE
At Site 5, tests were only conducted for hexavalent chromium methods
development purposes. At Site 6, emissions were measured at the inlet and outlet of the
control device. At Site 7, an evaluation of CO and THC CEMSs was performed. At
Sites 8 and 9, emissions were measured at the inlet of the venturi scrubber, at the
midpoint located between the venturi scrubber and the wet ESP, and at the outlet of the
wet ESP. For Sites 6, 8, and 9, midpoint and outlet air emission samples were collected
and analyzed for particulate matter, for metals, for polychlorinated dibenzodioxins and
furans (PCDD/PCDFs), volatile and semivolatile compounds (except Site 6), and for
hexavalent chromium and nickel species listed in Table 2. Inlet samples were collected
and analyzed for metals, chromium, and nickel species. Due to the difficult sampling
conditions at the inlet locations, only the concentrations of metals in the collected
particulate matter and the ratios of hexavalent chromium to total chromium and nickel
species to total nickel were determined. Continuous emission monitoring was conducted
for 02, CO,, CO, SOz, and NO, at the control system inlet and Oz (except Site 6), C02
(except Sites 6 and 9), CO, S02 (except Sites 6 and 9), NO, (except Sites 6 and 9), and
THC at the control system outlet stack. The monitoring data were used principally to
determine process and control equipment operating conditions during the chromium and
nickel speciation tests. Process samples consisting of sludge feed, scrubber inlet and
discharge water, and bottom ash (except Site 8) were collected. Process samples were
analyzed for the metals listed in Table 2 and were subjected to ultimate and proximate
analysis. The heating value of the sludge feed was calculated from the carbon and
hydrogen content.
Particulate matter and metals sampling was conducted following the procedures in
the draft EPA method, "Methodology for the Determination of Trace Metals Emissions
2-1
-------
TABLE 2. SPECIFIC ELEMENTS AND COMPOUNDS OF INTEREST
I. Metal Speciation
A. Trivalent Chromium
B. Hexavalent Chromium
C. Soluble Nickel
D. Sulfidic Nickel
E. Oxidic Nickel
IV. PCDDs/PCDFS
PCDDs
A. Mono-CDD
B. Di-CDD
C. Tri-CDD
D. 2378-TCDD
E. Other TCDD
F. 12378-PCDD
G. Other PCDD
H. 123478-HxCDD
I. 123678-HxCDD
J. 123789-HxCDD
K. Other HxCDD
L. 1234678-HpCDD
M. Other HpCDD
N. Octa-CDD
II. Total Metals
A. Arsenic
B. Beryllium
C. Cadmium
D. Chromium
E. Lead
F. Mercury
G. Nickel
III. Combustion Gases and
Criteria Pollutants
A. 02
B. C02
C. CO
D. S02
E. NO,
F. THC
PCDFs
O. Mono-CDF
P. Di-CDF
Q. Tri-CDF
R. 2378-TCDF
S. Other TCDF
T. 12378-PCDF
U. 2378-PCDF
V. Other-PCDF
W. 123478-HxCDF
X. 123678-HxCDF
Y. 234678-HxCDF
Z. 123789-HxCDF
AA. Other HxCDF
BB. 1234678-HpCDF
CC. 1234789-HpCDF
DD. Other HpCDF
EE. Octa-CDF
V. Semivolatile Organics VI.
A. Bis (2-ethylhexyl)phthalate
B. 1,2-Dichlorobenzene
C. 1,3-Dichlorobenzene
D. 1,4-Dichlorobenzene
E. Phenol
F. Naphthalene
Volatile Organics
A. Acrylonitrile H.
B. Benzene I.
C. Carbon tetrachloride J.
D. Chlorobenzene K.
E. Chloroform L.
F. 1,2-dichloroethane M.
G. Transl,2-dichloroethaneN.
Ethylbenzene
Methylene chloride
Tetrachloroethane
Toluene
1,1,1-Trichloroe thane
Trichloroethane
Vinyl chloride
2-2
-------
in Exhaust Gases from Stationary Source Combustion Processes." A diagram of the
multiple metals sampling train used in this test program is shown in Figure 1 and a copy
of the draft method is reproduced in Appendix B found in Volume IX: Site 9 Draft Test
Report, Appendices. The sampling train and procedures used are similar to those for
EPA Method 5 (40 CFR Part 60) with the following exceptions:
• A glass or quartz nozzle and probe liner are used;
• A Teflon filter support is used;
• A low metals background quartz fiber filter is used;
• 5% nitric acid/10% hydrogen peroxide solution replaced water in the first
two impingers and KMn04 in the third impinger;
• The glassware is cleaned according to the procedure in the draft method;
and
• The sample is recovered as shown in Figure 2.
After gravimetric analysis of the front half portion of the train, the samples were
digested according to the procedure and total metals determined using inductively-
coupled argon plasma spectroscopy and atomic absorption spectroscopy for total Cr, Ni,
As, Pb, Cd, and Be. A sample preparation and analytical flow chart is presented in
Figure 3. Mercury sampling was included in Sites 8 and 9 testing program using the
multiple metals train. It was later determined by EPA that the sample should be filtered
and the solids be digested. Since the need for this procedure was not know at the time
of the mercury sample preparation and analysis, the mercury results are considered
invalid and are not presented.
Flue gas sampling and analysis for hexavalent chromium followed the procedures
in the draft EPA method, "Determination of Hexavalent Chromium from Stationary
Sources." Either quadruplicate or duplicate sampling trains were employed. A diagram
of the recirculating reagent sampling train, shown in Figure 4, was used at the midpoint
and outlet locations for Sites 6, 8, and 9. The draft method is reproduced in Appendix B
of Volume IX: Site 9 Draft Test Report, Appendices. This procedure is based on EPA
Method 5 with the following modifications:
2-3
-------
All glass sample exposed surface to here
(Except when Tellon filler support Is used
Thermometer
Glass
Filter
Holder
Thermocouple
Glass
Probe
Glass probe liner
Iniplngers with
Absorbing Solutions
Heated Area
Ice Bath
Pilot
Manometer
Silica Gel
Empty (Optional Knockout)
5%HNO o /10% H0 O
4% KMnO 4/10%
SO,
Bypass
Valve
Vacuum
Line —
Vacuum
Gauge
Thermocouples
Orllice
Main
Valve
flTtl
Alr-Tlght
Hump
Oiy Gas
Motor
Figure 1. Schematic of multiple metals/particulate sampling train.
-------
Probe Liner
Front Half of
FiIter
Filter Support
1st Impinger
2nd & 3rd
4th & 5th
Last Impinger
Carefully
remove filter
from support
with Teflon-
coated tweezers
and place in
petri dish
Rinse three
times with
nitric acid
Empty
contents
into
container
KMnO,
BH
AR
Discard
Weigh for
moi sture
Rinse with
acetone
Measure
impinger
contents
Measure
impinger
contents
Measure
impinger
contents
Seal pctri
dish with
tape
Brush loose
part iculate
onto filter
Empty
contents
into
container
Empty
contents
into
container
Remove any
residue with
Rinse three
times with
nitric acid
Rinse three
times with
nitric acid
Rinse three
times with
permanganate
reagent
Rinse three
times with
nitric acid
Rinse three
times with
Brush with
nonmetalIic
brush and
rinse with
acetone
Brush line
with non-
metal I ic
brush and
rinse with
acetone
Check liner
to see if
part iculate
removed: if
not repeat
step above
(3)*
AR
(2)
F
(1)
BH
(«)
KMnO,
(5)
SF
(6)
* Number in parentheses indicates container number.
Figure 2. Sample recovery procedures for multiple metals train.
-------
Container 2
HN03 Probe Rinse
(Labeled FH)
Container 1
Acetone Probe Rinse
(Labeled AR)
Container 3
FiIter
(Labeled F)
Container 4 Container 5
Knockout & Permanganate Impingers
HN03/H202 Impingers (Labeled KHn04)
Solubilize residue
with conc. HN03
Reduce volune to
near dryness and
digest with HF and
conc. HN03 using
microwave digestion
Desiccate to
constant weight
Acidify to pH2
with conc. HN03
Reduce to dryness
in a tared beaker
Determine residue
weight in beaker
Determine filter
particulate weight
Reduce volume
to near
dryness and
digest with
HN03 and H202
Acidify half
of remaining
sample to pH
conc. HN03
Fraction 2A
Digest with conc.
HF and HN03 using
pressure relief
microwave digestion
procedure
Digest with acid
and permanganate
and analyze
for Hg by CVAAS
Fraction 3
Analyze for
As and Pb by GFAAS
(Not conducted)
F i Iter and dilute
to known volume
Fraction 1
Analyze by ICP for
target metals
Fraction 1A
Figure 3. Sample preparation and analysis scheme for multiple metals trains.
2-6
-------
GLASS
IMPINGER
TEFLON
T-UNION
1
NOZZLE
K»
I
—J
RECIRCULATING
LIQUID
TEFLON IMPINGERS
TEFLON
LINES
PERISTALTIC
PUMP
150 ml
0 1 NKOH
SILICA
GEL
75 ml
0.1 N KOH
75 ml
0.1 NKOH
EMPTY
WATER AND ICE BATH
TO
METHOD 5-TYPE
METERBOX
Figure 4. Schematic of recirculating reagent impinger train for hexavalent chromium.
-------
• The train does not have a filter section;
• The reagents are continuously recirculated from the first impinger back to
the nozzle to provide a flow of reagents through the probe, and thus
preventing hexavalent chromium in the probe drying out and possibly
converting to another valence state;
• 0.1 N KOH replaces water in the impingers;
• The entire surface exposed to sample is constructed of Teflon and/or glass;
• The glassware cleaning procedure includes a 10 hour soak in 10% HN03;
and
• The sample is recovered as shown in Figure 5.
Sampling for hexavalent chromium at the inlet locations was performed essentially
with a Method 5-type train at Site 6. At Site 8, the train was modified by eliminating the
filter and collecting the sample directly in a 1.0 N KOH solution. At Site 9, a peristaltic
pump was used to spray a 1.0 N KOH solution into the probe directly behind the
sampling nozzle during sampling. A schematic of the inlet hexavalent sampling train
employed at Site 9 is shown in Figure 6.
Hexavalent chromium samples were analyzed by ion chromatography coupled to a
diphenylhydrazine post-column reaction (IC/PCR) on the filtered impinger samples. To
determine the recovery of the radioactive hexavalent chromium spike, 0.5 ml fractions of
the IC/PCR discharge were collected at regular time intervals during the IC/PCR
analysis, and the gamma emissions measured for each fraction. For samples with the
slCr spike, the gamma emissions from the filter residue and the HN03 rinses were
measured before combining them for digestion and total Cr analysis. A sample
preparation and analytical flow diagram is presented in Figure 7.
Flue gas sampling and analysis for nickel species followed the draft EPA method,
"Methodology for the Determination of Nickel Compound Emissions from Stationary
Sources." A diagram of the nickel sampling train is shown in Figure 8 and the method
description is presented in Appendix B found in Volume IX: Site 9 Draft Test Report,
Appendices. Typically the sample trains collected for nickel speciation were paired with
2-8
-------
1 Combine filtrate
in sample
container
Combine rinses in
sample container
F iIter
Label sample
Label sample
Store and ship
to laboratory
Store and ship
to laboratory
Rinse impinger
nitric acid
Recover impinger
contents and
filter. Rinse
impinger train
with 0.1 N KOH
& filter sample
through 0.45 jim
Teflon fiIter
Nitrogen Purge of Train
Figure 5. Sample recovery scheme for hexavalent chromium impinger train.
2-9
-------
Quartz Glass
Probe Liner /
Bullon Hook Nozzlo
tJ
Roverse-Type
Pilot Tube
Temperaturo
Sensor
1/8" Tullon Line
(Reaching lo rear o(
bullon hook nozzlo)
Pilol
Manometer
or
Differential
Pressure
Gauge(s)
2-Liter
Impinger
Thermometers
Peristaltic
Pump
hormometer
mpingers
Check
Valve
>~~~«
>~~~«
>~~~<
~ ~~
100 mL
200 mL
Empty *~ Silica
Gel
Ice Bath
Bypass
Valve
Vacuum Vacuum
Gauge Line
Orifice
Main
Valvo
l Dry Gas
\ Metor I
Figure 6. Schematic of inlet location recirculating reagent impinger train for hexavalent chromium.
-------
Recirculatory Train A
with 53Cr+6 Spike
Recirculatory Trains B, C, and D
with 51Cr+6
¦ Residue*-
- Residui
FiItrate
Fi Itrate
Acid Digest :
Total Cr
Ana Iys i s
Cr and 53Cr
Analysi s
by I CP/MS
IC/PCR Analysis
for Cr+6
IC/PCR Analysis
for Cr+6
Ganma Count
of Residue
for 51Cr
Gamma Count
of Residue
for 51Cr
Filter through
0.45 Micron
Teflon fiIter
Filter through
0.45 Micron
Teflon fiIter
1 Contoination
of Residue and
HN03 Solution
for Total Cr
Preconcentrate
for 53Cr+6
Analysis by I CP/MS
Reci rculatory
Sampling Train
Impinger Sol.
and D.I. Rinse
Reci rculatory
Sampling Train
Impinger Sol.
and 0.1. R inse
Gamma Count
of IC Fractions
for Speciation
of 51Cr
Reci rculatory
Sampling Train
HN03 Rinse
Reci rculatory
Sampling Train
HN03 Rinse
Figure 7. Analytical protocol for quadruplicate recirculatory
train hexavalent chromium sampling at midpoint and
outlet locations.
2-11
-------
Glass
Filter Holder
Thermomeler
Quartz
Filter
Thermocouple Check
T Valve
Thermocouple
Glass Nozzle
Glass Probe
Reverse-Type]
Pilot Tube
Heated Area
Implngers
Pilot
Manometer
Ice Bath
Silica Gel
Water
Empty
Bypass Vacuum
vacuum
Gauge
Valve
Thermocouples
Orifice
Main
Valve
Air-Tight
Pump
Dry Gas
l Meter I
Figure 8. Schematic of nickel/nickcl subsulfide sampling train.
-------
a metals sampling train or in a quadruplicate arrangement with one metals sampling
train.
The sampling train and procedures are identical to those of EPA Method 5 (40
CFR Part 60) with the following exceptions:
• A glass or quartz nozzle and probe liner are used;
• A low metals background quartz fiber filter is used;
• The glassware cleaning procedure includes a 10 hour soak in 10% HN03;
and
• The sample is recovered as shown in Figure 9.
Each day the filters to be analyzed were stored in a desiccator under a dry nitrogen
atmosphere and sent to the analytical laboratory at the conclusion of test. The dry
nitrogen atmosphere was used because past experience has shown that oxidation of
nickel compounds can occur over a several week period.
Analysis of the nickel speciation samples was performed following the NiPERA
sequential leaching method. The ratios of sulfidic nickel species, nickel subsulfide
(Ni3S2) and nickel sulfide (NiS), to total Ni were determined. The method is not
capable of speciating between Ni3S2 and NiS. Individual nickel phases are extracted out
from the solid sample by sequential leachings using a series of solutions with increasing
oxidation strength. Four nickel phase groups are determined:
Nickel Groups
1) soluble nickel
2) sulfidic nickel
3) metallic nickel
4) oxidic nickel
Types of Nickel
water soluble nickel salts;
besides Ni3S2 and NiS, also dissolved
are arsenides NiAs and Ni,,As8, and
selenide NiSe;
Leaching Solution
0.1M ammonium acetate
peroxide-citrate
free or alloyed with iron (ferronickel); methanol-bromine
refractory nickel oxide;
nitric/perchloric acid
2-13
-------
FiIter and
Cyclone
Particulate
Hatter
(Fraction F)
Acetone
Front Half
Rinse
(Fraction AR)
0.1 N Nitric
Front Half
Rinse
(Discarded)
Back Half
Components
(Di scarded)
Recover impinger
solution, measure
volune and discard
so Iut i on
Brush and rinse
nozzle, probe,
cyclone, and
front half of
fiIter holder 3
times with acetone
Brush and rinse
nozzle, probe,
cyclone, and
front half of
filter holder 3
nitric solution
and discard
; Combine rinses in
I sample container
Label Sample
Archive in cool
dry area
Recover silica
gel, weigh, and
discard
Recover filter
and cyclone sample
dry with brush
Filter acetone
rinses through
particulate
For Zatka sample,
place in vacuum
fiItration device"
Rinse back half
components with
discard
Recover acetone
and store in labeled
container. Analyze
1 for every 8 samples
Recover particulate
into labeled petri
dish and store in
desiccator under
dry nitrogen untiI
analysis
Note: Inlet samples were not acetone rinsed.
Figure 9. Schematic of sample recovery procedures for nickel train.
2-14
-------
EPA Methods 1, 2, 3, and 4 were used in conjunction with the sampling
procedures described above. Method 3 samples were collected as backup for 02 and
C02 determination should the CEMSs data be unavailable.
Flue gas sampling for PCDD/PCDFs and semivolatile organic compounds
followed procedures in SW-846 Method 0010, except that a final toluene rinse was
conducted and analyzed separately for PCDD/PCDF. The samples were analyzed for
PCDD/PCDF using SW-846 Method 8290 and for other semivolatile organic compounds
using a combination of SW-846 Methods 3540, 3550, 3510, 3520, and 8270. A schematic
of the MM5 sampling train is shown in Figure 10 and copies of the relevant SW-846
methods are reproduced in Volume IX: Site 9 Draft Emission Test Report, Appendices.
The sample recovery scheme is presented in Figure 11 and the sample extraction scheme
is presented in Figure 12.
Flue gas sampling for volatile organic compounds employed the volatile organic
sampling train (VOST) shown in Figure 13 in accordance with SW-846 Method 0030,
which is reproduced in Volume IX: Site 9 Draft Emission Test Report, Appendices.
The CEMSs used to measure concentrations of CO, C02, 02, NO„, SO,, total
hydrocarbons (THC as propane) followed the EPA instrumental Methods 10, 3A, 7E,
6C, and 25A, respectively. The primary intent of the continuous monitoring effort was
to: (1) determine concentrations of these compounds, and (2) provide a real-time
indication of combustion conditions. The continuous monitoring systems were calibrated
daily, but no attempt was made to certify the monitors using the EPA instrumental test
methods.
The dewatered sludge samples were analyzed for the target metals after
determination of their moisture and ash content, heating value, and proximate and
ultimate analyses following ASTM Methods D3174, D3175, D3177, D3178, D3179, and
D2361 (not reproduced in Appendices because they are standard methods). A dried
portion of the sludge sample was subjected to microwave HN03/HF digestion in a
pressure relief vessel identical to the flue gas particulate samples described above. This
digestion procedure was chosen to provide for comparison of the metals in the sludge
with the flue gas samples and the bottom ash samples (see below). The digestion
2-15
-------
Nozzle
Pi tot tube ljt30jl Temperature
dr„h. ihiisS Indicator
1 .9-2.5 cm
I 9-2 5 cm
Heated Probe
Thermocouple (behind)
Stack Wall
5" Type
Pitot Tube
(Thermocouple^
Magnehelic Gauges
(Cooli ng Water)
Thermometer
©
(t her mo mete r^-^
rbent Trap
iniwia
Meter
Ice Both
Console
(OotiTnaUO) Water Water
Silica Gel
Flow Control Valves
Thermometer pine
Yacuum
Gauge
Calibrated orfice
Coarse
Dry Gas
Meter
Magnehelic Gauges
Figure 10. Schematic of the MM5 train for semivolatile organics and PCDD/PCDF.
2-16
-------
NOZZLE. CYCLONC AND
ni Fll TFfl HOUSING'
PnOBE LINER
BRUSH/RINSE
Willi ACLTONC
(3.)
BRUSH/RINSE WITH
tIEXANE
P»)
SUAL INSPECTION
K)
i
»—*
-J
ATTACH 250 ml
FLASK lO BALL
JOINT
BRUSH/RINSE
WITH ACETONE
(3x)
EMPTY FLASK
BRUSHfllNSE WITH
HEXANE
(3.)
EMPTY FLASK
VISUAL INSPECTION
Fll TER
REMOVE WITH
TWEEZEHS lO
PnECLEANED
ALUMINUM FOIL
OMUSII LOOSE
PARTICULA1L
ONTO FILTER
TRANSPORT IN
ORIGINAL GLASS
PETRI DISH
XAD
MOIXJI E
CONDENSER. FILTER SUPTOHIS.
OH FILILH HOUSING '
IMPINGEHS SILICA GEL
REMOVE
AND CAP
RINSE
WITH ACETONE
<3»)
RINSE WITH
IILXANE
(3<)
MEASURE
VOLUML
GAN
EMPTY CONTENTS
INTO SAMPLE
CONfAINfH
RINSE WTH
Dl WATER
(3«)
WEIGH
DISCARD
WRAP THE
MODULI IN
ALUMINUM FOIL
PR
SM
C4I
" Final loluene riitoe (3i)
Figure 11. Semivolatile organic train sample recovery scheme.
\
-------
PR CR SM F IR
Concentrate
Spike
Place in
Soxhlett
Liquid/Liquid
Extraction w/ MeCI
Spike
Extraction
w/ MeCI 2
Extraction
w/ Toluene
Combine
Split 1:1
Split 1:1
Combine
Archive
Extract
Concentrate
Analyze Aliquot
lor Semivolatiles;
Archive Remaining Extract
Solids
Toluene
Rinse
Extract A
(MeCI 2)
Extract C
(MeCI 2)
Extract B
(Toluene)
Extract D
(MeCI 2)
Extract E
(MeCl2 /Toluene)
XAD Resin Trap
and Filter
Impinger Contents
and Dl H2 O Rinses
Acetone/Hexane Rinses of
Probe Liner, Nozzle, and
Front Hall of Filter Housing
Combined with Acetone/Hexane
Rinses o( Condenser and
Back Hall ol Filter Housing
Concentrate
Hexane Exchange for
Me&2 /Toluene
Extract G
PCDD/PCDF
Cleanup
Extract H
Analyze Aliquot
lor PCDD/PCDF;
Archive Remaining Extract
Figure 12. Extraction for semivolatile organic samples,
2-18
-------
HEAT 3 . WAY
SAMPLE VALVE
I'HOHfc
ICE WATER
CONDENSER
ICE WATER
CONDENSER
TENAX / CHARCOAL
CARTRIDGE
N>
I
TENAX
CARTRIDGE
VALVE
DRY
GAS METER
SILICA GEI
DRYING
TUBE
ROTAMETER
PUMP
CONDENSING
IMPINGER
Figure 13. Schematic of volatile organic sampling train.
-------
solution was analyzed by ICAP following the procedures described for the flue gas
samples and archived for possible GFAAS analysis, however, GFAAS analyses were not
required.
Portions of the scrubber water samples were analyzed for metals by acidifying
with HN03 and then reducing to near dryness on a hot plate. Because the venturi
scrubber discharge water samples had a high solids content, the solids were subjected to
the microwave HNO3/HF digestion described above. The digested solutions were
analyzed by ICP for all the target metals except Hg following the procedures described
for the flue gas samples. A portion of each solution was archived for possible graphite
furnace atomic absorption spectroscopy (GFAAS) analysis, but the GFAAS analyses
were not required.
After determination of the moisture content following ASTM D3174, incinerator
bottom ash samples were analyzed for the target metals, including Hg, using the same
procedures as described above for the sludge samples.
Incinerator and control system operating parameters were monitored during all
manual test runs to characterize the system operations. The parameters typically
monitored are presented in Table 3.
2-20
-------
TABLE 3. TYPICAL PROCESS MONITORING DATA
Frequency of
Parameter
Readings
Source of Readings
Incinerator Operating Data
Wind Box Temperatures
60 minutes
Plant operating log
Bed Temperatures
60 minutes
Plant operating log
Freeboard Temperatures
60 minutes
Plant operating log
Heat Exchanger Inlet Temp
60 minutes
Plant operating log
Heat Exchanger Outlet Temp
60 minutes
Plant operating log
Incinerator Outlet O,
Continuous
Entropy CEMSs
Auxiliary fuel usage
As used
Plant operating log
Sludge Feed Rate
60 Minutes
Plant operating log
Sludge Feed Characteristics
Moisture (wt %)
Once per run
Entropy analysis
Volatiles (wt %)
Once per run
Entropy analysis
Heating Value
Once per run
Entropy analysis
Scrubber Svstem Operating Data
Differential Pressure (in. H20)
60 minutes
Plant operating log
Scrubber Inlet Temp (°F)
60 minutes
Plant operating log
Scrubber Outlet Temp (°F)
60 minutes
Plant operating log
2-21
-------
3.0 RESULTS AND DISCUSSION
The primary objective of this study was to determine ratios of hexavalent-to-total
chromium and nickel subsulfide-to-total nickel for a typical sewage sludge incinerator
under normal combustion conditions (higher concentrations of carbon monoxide and
total hydrocarbons) and improved combustion conditions (lower concentrations of carbon
monoxide and total hydrocarbons). The two combustion conditions (normal and
improved) were tested at Sites 6 and 9, which were multiple hearth furnaces. At Site 8,
a fluidized bed furnace, only the normal combustion condition was investigated because
furnace operating techniques are limited and these furnaces generally have good
combustion.
3.1 Metals and Particulate
Metals and sludge feed rates to the incinerators are shown in Table 4 for both
normal and low CO (improved combustion) conditions. Chromium, lead, and nickel
consistently had the highest metals feed rate to the incinerators. At Site 6, chromium
had the highest feed rate (48-58 g/hr) due to the contamination in the ferric chloride
used to condition the sludge at this site. At Site 6, lead had the second highest feed rate
(11 g/hr). At Site 8 and Site 9, lead had the highest feed rates (39 g/hr and 189-228
g/hr, respectively). Chromium had the second highest feed rate (30 g/hr) at Site 8,
followed by nickel (19 g/hr). At Site 9, nickel had the second highest feed rate (120-152
g/hr), followed by chromium (76-85 g/hr).
In Table 5, particulate matter and metals emissions factors from the control
device outlet are shown for normal and low CO conditions. The particulate matter
emission factor represents the mass of particulate emitted per mass of dry sludge fed.
3-1
-------
TABLE 4. FEED RATES FOR METALS IN THE SLUDGE (g/hr).
Site 6
Site 6
Site 8
Site 9
Site 9
Normal
Low CO
Normal
Normal
Low CO
METALS
Arsenic
ND
ND
ND
<100
<100
Beryllium
0.05
0.05
0.27
<0.8
<0.8
Cadmium
0.82
0.78
2.20
8.60
8.75
Chromium
57.9
48.0
30.0
75.7
85.3
Lead
11.4
11.6
39.0
189
228
Nickel
4.41
3.42
19.0
120
152
Total Sludge
Feed Rate
(lbs/hr)
3733
3460
4966
7482
7460
Dry Solids
Feed Rate
(lbs/hr)
971
934
979
1571
1641
TABLE 5. PARTICULATE AND METALS STACK EMISSION FACTORS FROM
SCRUBBER OUTLET FOR STEADY STATE (LOW CO) AND
NORMAL OPERATION.
EMISSION FACTORS (g metal emitted/g metal fed)
SITE 6
SITE 6
SITE 8
SITE 9
SITE 9
Pollutant
NORMAL
LOW CO
NORMAL
NORMAL
LOW CO
PM (g/kg dry
sludge feed)
0.28
0.39
0.01
0.21
0.31
Arsenic
ND
ND
ND
<0.011
<0.013
Beryllium
< 0.069
0.059
<0.0001
ND
ND
Cadmium
0.917
0.908
0.0003
0.336
<0.008
Chromium
0.011
0.005
0.0001
0.036
0.001
Lead
0.123
0.136
<0.0001
0.101
0.006
Nickel
0.030
0.013
<0.0001
0.004
0.0004
ND - Not detected, all sample measurements were below the analytical detection limit.
< - Outlet samples were below analytical detection limit, calculated ratio is less than
value shown.
3-2
-------
The metals emission factors represent the mass of metals emitted per gram of metal fed
to the incinerator in the sludge. For particulate matter, the emission factors were 0.37
g/kg and 0.45 g/kg for Site 6 at normal and low CO conditions, respectively, and 0.040
g/kg without the wet ESP and 0.0055 with the wet ESP for Site 8. For Site 9, the
particulate emission factors for normal combustion without the wet ESP was 0.36 g/kg
and for low CO combustion and with the wet ESP was 0.04 g/kg. Cadmium had the
highest emission factor of all the metals for each of the sites, ranging from 0.003 g/g for
Site 8 to 0.917 g/g for Site 6 with normal combustion. At Site 6, the lead emission
factor increased from 0.123 to 0.136 with improved combustion (higher hearth
temperatures). At Site 9, the addition of the wet ESP lowered the cadmium emission
rate from 0.101 to 0.006 g/g even with improved combustion and higher hearth
temperatures. Due to the high collection efficiency of the venturi scrubber/impingement
tray scrubber, metal emission factors were considerably lower for the fluidized bed
incinerator at Site 8 compared to the multiple hearth incinerators at Site 6 and at Site 9
without the wet ESP. With the wet ESP, the emission factors from Site 9 were
comparable to both Site 8 and Site 3, a fluidized bed incinerator in this study and tested
by Radian, respectively, the latter during the preliminary studies on sludge incineration.
The lower emission factors seen for the fluidized bed incinerators may have been due to
less volatilization and/or better removal in the venturi/scrubber system. The large
amounts of inert material discharged from a fluidized bed incinerator to the pollution
control device may provide condensation sites for the volatile metals allowing their
removal with larger particles.
At Sites 6, 8, and 9, the metals were measured at the venturi/scrubber inlet and
outlet and at the outlet of the wet ESP for Sites 8 and 9. The removal efficiencies were
calculated and are summarized in Table 6 (Arsenic and beryllium were essentially not
detectable at all three sites). For Sites 6 and 9 with multiple hearth furnaces and
venturi/scrubbers, only chromium and nickel had removal efficiencies within 10 % of the
particulate matter removal efficiency. At Site 6, cadmium and lead had removal
efficiencies of about 71% at normal combustion conditions and about 77% at low CO
conditions compared to about 98% particulate matter removal efficiency. For Site 9,
-------
TABLE 6. METALS AND PARTICULATE REMOVAL EFFICIENCY ACROSS
THE VARIOUS CONTROL DEVICES
Site 6
Site 6
Site 8
Site 8
Site 9
Site 9
Site 9
Site 9
Normal
Low CO
Normal
Normal
Normal
Normal
Low CO
Low O
Metal
Scrubber
Scrubber
Scrubber
Wet ESP
Scrubber
Wet ESP
Scrubber
Wet E!
(%)
(%)
(%)
(%)
(%)
(%)
(%)
Arsenic
NA
NA
>99.5
NA
NA
NA
NA
NA
Beryllium >85.5
>86.7
>99.95
NA
—
—
-
--
Cadmium 71.9
77.3
99.82
71.0
0.0
NA
45.0
>98.0
Chromium99.3
99.4
99.92
62.0
0.0
NA
89.0
88.0
Lead
71.3
78.1
99.91
>96.0
5.0
NA
54.0
96.0
Nickel
93.4
94.5
99.89
81.0
89.0
NA
96.0
90.0
Particulate
Matter
98.5
97.6
99.99
78.0
85.0
NA
95.0
87.0
NA - Not Appli cable
cadmium and lead had even lower removal efficiencies of 45% and 54%, respectively at low
CO conditions compared to 95% particulate matter removal efficiency. For Site 8 with a
fluidized bed and venturi/scrubber, cadmium, chromium, lead, nickel, and particulate
matter had similar removal efficiencies of >99%. The pilot-scale wet ESP at Site 8
removed an additional 62-96% of the metals and particulate matter emitted from the
venturi/scrubber. At Site 9, the full-scale ESP removed an additional 87-98% of the metals
and particulate matter emitted from the venturi/scrubber.
The ratios of individual metals to particulate matter are summarized in Table 7 for
Sites 6, 8, and 9 for normal and low CO conditions. For Sites 6 (normal and low CO
conditions) and 9 (normal conditions only), lead, at 28-32 mg/g and 12 mg/g , respectively,
followed by cadmium, at 1.6-2.0 mg/g and 2.0 mg/g, respectively, had the highest metals to
particulate matter ratios. In contrast the metals to particulate matter for chromium was 1.1
mg/g and for cadmium was 0.4 mg/g at Site 8. Generally the ratio of metals to particulate
matter was lower for the fluidized bed incinerator at Site 8 than for the multiple hearth
incinerators at Sites 6 and 9. At Site 6, the ratio of metal to particulate matter for
cadmium, chromium, and nickel decreased and increased for lead from the normal
3-4
-------
TABLE 7. RATIO OF METALS TO PARTICULATE MATTER EMISSIONS UNDER
STEADY STATE (LOW CO) AND NORMAL OPERATION
Metal
SITE 6 SITE 6 SITE 8 SITE 9 SITE 9
Normal Low CO Normal Normal Low CO
(ng metal/g particulate)
Arsenic <870
Beryllium <3.78
<622
<2.71
ND
ND
417
1094
<179
<1027
<6383
Cadmium 2056
Chromium 161
Lead 28173
Nickel 132
31689
61
1645
68
<1006
1964
1626
12552
336
<553
596
9191
468
ND - Not detected, all sample measurements were below the analytical detection limit.
condition to the low CO condition. At Site 9, the wet ESP at the low CO condition had
lower ratios for cadmium, chromium, and lead and a higher ratio for
nickel compared to the normal condition without the wet ESP. The ratio of lead to
particulate matter was lower at Site 8 compared to Site 6 even though the feed rate of
lead at Site 8 was 15-20 times higher than Site 6 (see Table 4).
3.2 Hexavalent Chromium
A major accomplishment of this test program was the sampling of hexavalent
chromium without artifact formation and the analyzing of the resulting samples
specifically for hexavalent chromium at low concentrations. Sampling activities
conducted at Site 5 were dedicated to developing a suitable measurement method for
hexavalent chromium in emissions from incineration of municipal wastewater sludge.
Hexavalent chromium sampling at the venturi/scrubber outlets at Site 6, 8, and 9
followed the same procedures of the draft EPA method. A new sampling technique was
developed for this program where the impinger reagent is constantly recirculated to the
inlet end of the sampling probe. A key element in sampling technique utilized for this
program was the use of a hexavalent chromium radioactive isotope, SICr+6, incorporated
-------
into each sampling train. With the recirculating train design, the surrogate added to the
impinger solution at the start of the test was exposed to the same conditions within the
train as the native hexavalent chromium. The ''Cr*"6 surrogate measured the degree of
conversion of hexavalent chromium to trivalent chromium occurring during the sampling
and the handling of samples prior to analysis. The surrogate recoveries for Sites 6, 8,
and 9 at both midpoint and wet ESP outlets and the ratio of hexavalent to total
chromium measured with the recirculating train are shown in Table 8.
TABLE 8. HEXAVALENT CHROMIUM SAMPLING RESULTS
SITE 6
SITE 6
SITE 8
SITE 9
SITE 9
Normal
Low CO
Normal
Normal
Low CO
Venturi/Scrubber Outlet
Surrogate recovery,% 90.5
Hexavalent to
95.6
66.8
84.3
90.5
total Cr ratio,%
1.9
4.4
<1.8
11.9
7.9
Wet ESP Outlet
Surrogate recovery,%
Hexavalent to
NA
NA
81.5
90.1
93.1
total Cr ratio,%
NA
NA
<1.4
29.9
42.5
NA = Not applicable, testing was not conducted.
Surrogate recoveries ranged from 67-96% during sampling at the venturi/scrubber
outlet at Sites 6, 8, and 9. For the sampling at the wet ESP outlets at Site 8 and 9,
surrogate recoveries ranged from 82-91%. The ratio of hexavalent chromium to total
chromium measured by the recirculating train at the venturi/scrubber outlets ranged
from < 1.8 - 11.9%, and at the wet ESP outlets, the ratio ranged from < 1.4 - 42.5%.
(The hexavalent to total chromium ratios were not corrected for surrogate recovery).
At Site 6, the hexavalent to total chromium ratio increased from 1.9% to 8.3%
between the normal combustion conditions and the low CO (improved combustion)
conditions. An explanation for this observation, shown graphically in Figure 14, is that
the higher hearth temperatures and excess oxygen levels recorded during the improved
3-6
-------
Outlet emissions data (excludes Run 7)
Run 11
Run 9
r = -0.98
Run 3
Run 13
60 80 100
CO to C02 Ratio (ppm to %)
120
Figure 14. Correlation of combustion efficiency and hexavalent to total chromium ratio.
3-7
-------
combustion condition favors the formation of hexavalent chromium. This effect was not
seen at Site 9. The fluidized bed incinerator at Site 8 had the lowest ratio of hexavalent
to total chromium in the venturi/scrubber emissions. A possible explanation for this low
ratio at Site 8 is the lower residence time that occurs in the fluidized bed incinerator
(seconds) compared to the higher residence time that occurs in multiple hearth
incinerators (hours). (It should be noted that the multiple hearth design was originally
used to roast ores such as chromium).
3.3 Nickel Speciation
The major objective of the nickel speciation testing was to determine the percent of
the nickel emissions in the form of nickel subsulfide. It was anticipated that the
nickel subsulfide emissions from multiple hearth incinerators would constitute less than
1% of the total nickel emissions, because these incinerators typically operate with high
excess air which is not favorable for the formation of nickel subsulfide. The results of
the sequential leaching nickel analysis indicate that within the detection limit of the wet
chemical method, no nickel subsulfide was present in the samples. Based on the
detection limits, the nickel subsulfide to total nickel ratio at Sites 6 and 8 is less than
12% for the inlet emissions and less than 10% for the outlet emissions. Samples
analyzed from the same runs by X-ray absorption near-edge structure (XANES) and
extended X-ray absorption fine structure (EXAFS) indicated that no nickel subsulfide
was detected within the instrumental detection limit of 10% of the total nickel. For Site
9 the ratio of nickel subsulfide to total nickel in the inlet emissions is less than 2 % and
in the midpoint emissions is less than 1%. (The reduction in the analytical detection
limit was due to the higher amounts of total nickel present in the emissions).
3.4 PCDD/PCDFs and Semivolatile and Volatile Compounds
Sampling for polychlorinated dibenzodioxins and furans (PCDD/PCDFs) was
performed at the venturi/scrubber outlet at Sites 8 and 9 and at the wet ESP outlet at
3-8
-------
Site 9. Sampling at Site 9 was conducted at both normal and low CO conditions. The
results for the PCDD/PCDF sampling are shown in Table 9. Total tetra-octa
chlorinated dibenzodioxins and furans (CDD + CDFs) were the highest (102 ng/dscm)
at the venturi/scrubber outlet at Site 9 during normal conditions. Improved combustion
at Site 9 lowered the total tetra-octa CDD + CDF emissions from 102 ng/dscm to 8.7
ng/dscm. The wet ESP at Site 9 reduced the total tetra-octa CDD + CDF emissions
from 102 ng/dscm to 15.6 ng/dscm under normal conditions and from 8.7 ng/dscm to 2.8
ng/dscm under the low CO condition. Total tetra-octa CDD + CDF emissions at the
venturi/scrubber outlet at Site 8 were 2.1 ng/dscm compared to 102 and 8.7 ng/dscm for
Site 9 at normal and low CO conditions, respectively.
At Site 9, sampling for semivolatile organic compounds was performed at the
venturi/scrubber outlet and the wet ESP outlet under both normal and low CO
conditions. The results for the semivolatile organic compound sampling are shown in
Table 10. Several compounds were found above the minimum detection limit at both
the midpoint and outlet locations under runs both normal and low CO incinerator
operations. The concentrations and number of the semivolatile compounds detected
were typically less under the low CO combustion conditions. For the normal combustion
conditions, eleven semivolatile compounds were detected for both runs:
1,4-dichlorobenzene, benzyl alcohol, 1,2-dichlorobenzene, 2-nitrophenol, benzoic acid,
1,2,4-trichlorobenzene, naphthalene, 2-methylnaphthalene, dibenzofuran, phenanthrene,
and bis(2-ethylhexyl)phthalate. For the low CO combustion conditions five semivolatile
compounds were detected for both sample runs: phenol, benzyl alcohol, 4-methylphenol,
benzoic acid, and 4-nitrophenol. Bis(2-ethylhexyl)phthalate was found in the sample
blank and the emission results are likely due to contamination.
The concentration of the volatile organics in the flue gas are presented in Table 11.
At Site 8 five of the target compounds were below the analytical detection limit during
all three test runs: acrylonitrile, vinyl chloride, 1,2-dichloroethane, and chlorobenzene.
The other eight target compounds were detected in all three test runs and averaged:
methylene chloride - 110 ug/dscm, chloroform - 17 ug/dscm, 1,1,1-trichloroethane - 6.8
3-9
-------
TABLE 9. PCDD/PCDF EMISSIONS SUMMARY
Concentration
(ng/DSCM1)
SITE 8
SITE 9
SITE 9
SITE 9
SITE 9
Normal
Normal
Normal
Low CO
Low CO
Isomer
Outlet
Outlet
Mid-Pt
Outlet
Mid-Pt
2378-TCDD
0 . 007
ND
ND
ND
ND
Other TCDD
0 . 120
1. 14
7 . 02
0. 15
0 . 14
12378-PeCDD
0 . 061
ND
ND
ND
ND
Other PeCDD
0. 053
0. 05
0.22
ND
ND
123478-HxCDD
0 . 002
ND
ND
ND
ND
12 3 67 8-HxCDD
0. 005
ND
ND
ND
ND
123789-HxCDD
0. 006
0. 03
ND
ND
ND
Other HxCDD
0. 036
0. 13
0 . 48
ND
0. 04
12 3 4 67 8-HpCDD
0. 048
0.29
1.73
ND
ND
Other HpCDD
0. 023
0.25
1.50
0.05
ND
OCDD
0. 359
1.35
9.24
0.48
1.45
Total Tetra-
Octa CDD
0.721
3 . 2
20.2
0.7
1.6
2378-TCDF
0.019
1.39
7.76
0.28
1. 12
Other TCDF
0 . 507
4 . 55
28 . 9
0. 10
3 . 19
12378-PeCDF
0.037
0.25
1. 69
0. 03
0. 01
23478-PeCDF
0.024
1. 17
7 .25
0 . 12
0 . 44
Other PeCDF
0.361
3 . 62
27. 1
0.58
1.87
12 3 4 78—HxCDF
0 . 044
0.26
1.79
0 . 02
ND
123678-HxCDF
0.022
0. 09
0.38
ND
0. 04
23 4 678-HxCDF
0 .018
0. 19
1. 14
ND
0. 06
12 3 7 89-HxCDF
0 . 001
ND
ND
ND
ND
Other HxCDF
0. 134
0.51
3 . 12
0 . 04
0. 17
12 3 467 8-HpCDF
0.071
0. 11
ND
0.03
0. 06
123 4789—HpCDF
0.010
0.01
ND
ND
ND
Other HpCDF
0. 054
0. 08
0 . 60
0.01
0. 02
OCDF
0 . 108
0. 06
1. 67
ND
ND
Total Tetra-
Octa CDF
1.41
12 . 4
81.9
2 . 1
7 . 1
Total Tetra-
Octa CDD+CDF
2 . 13
15.6
102
2 . 8
8.7
1 = 68 Deg. F — 29.92 inches Hg
bND = Reported as not detected or estimated maximum possible
concentration; both expressed as zero (0) in calculating totals
and averages.
Note: PCDD/PCDF emissions testing not conducted at Site 6.
3-10
-------
TABLE 10. SEMIVOLATILE EMISSIONS SUMMARY FOR OUTLET AND
MIDPOINT AT SITE 9
Concentration
(Mg/DSCM1)
Analyte
OUT-7A
MID-7A
OUT-7C MID
-7C
Phenol
ND
ND
176
162
bis(2-Chloroethyl)ether
ND
ND
ND
ND
2-Chlorophenol
ND
ND
ND
ND
1,3-Dichlorobenzene
ND
ND
ND
ND
1,4-Dichlorobenzene
30.8
33.4
ND
ND
Benzyl alcohol
800
1120
4100
3930
1,2-Dichlorobenzene
25.6
26.7
ND
ND
2-Methylphenol
ND
ND
ND
ND
bis(2-Chloroisopropyl)ether
ND
ND
ND
ND
4-Methylphenol
ND
ND
21.2
20.6
N-Nitroso-di-n-propylamine
ND
ND
ND
ND
Hexachloroethane
ND
ND
ND
ND
Nitrobenzene
ND
ND
ND
ND
Isophorone
ND
ND
ND
ND
2-Nitrophenol
196
284
43 . 1
76 . 4
2,4-Dimethylphenol
ND
ND
ND
ND
Benzoic acid
2850
3220
5090
4240
bis(2-Chloroethoxy)methane
ND
ND
ND
ND
2,4-Dichlorophenol
ND
ND
ND
ND
1,2,4-Trichlorobenzene
699
768
ND
ND
Naphthalene
976
864
ND
ND
4-Chloroaniline
ND
ND
ND
ND
Hexachlorobutadiene
ND
ND
ND
ND
4-Chloro-3-methylphenol
ND
ND
ND
ND
2-Methylnaphthalene
43.4
45.5
ND
ND
Hexachlorocyclopentadiene
ND
ND
ND
ND
2,4,6-Trichlorophenol
ND
ND
ND
ND
2,4,5-Trichlorophenol
ND
ND
ND
ND
2-Chloronaphthalene
ND
ND
ND
ND
2-Nitroaniline
ND
ND
ND
ND
Dimethylphthalate
ND
ND
ND
ND
Acenaphthylene
ND
ND
ND
ND
3-Nitroaniline
ND
ND
ND
ND
Acenaphthene
ND
ND
ND
ND
2,4-Dinitrophenol
ND
ND
ND
ND
4-Nitrophenol
ND
ND
97.4
1440
Dibenzofuran
45.2
44 . 7
ND
ND
2,4-Dinitrotoluene
ND
ND
ND
ND
2,6-Dinitrotoluene
ND
ND
ND
ND
Diethylphthalate
ND
ND
ND
ND
4-Chlorophenyl-phenylether
ND
ND
ND
ND
1 = 68 Deg. f — 29.92 inches Hg.
ND = Not detected or EV catches; used as zero (0).
3-11
-------
TABLE 10. (Continued)
Concentration
(Mg/DSCM1)
Analyte
OUT-7A
MID-7A OUT
-7 C MID-
7C
Fluorene
ND
ND
ND
ND
4-Nitroaniline
ND
ND
ND
ND
4,6-Dinitro-2-methylphenol
ND
ND
ND
ND
N-Nitrosodiphenylamine(1)
ND
ND
ND
ND
4-Bromophenyl-phenylether
ND
ND
ND
ND
Hexachlorobenzene
ND
ND
ND
ND
Pentachlorophenol
ND
ND
ND
ND
Phenanthrene
44.9
33 . 4
13 .7
ND
Anthracene
ND
ND
ND
ND
Di-n-butylphthalate
ND
ND
ND
ND
Fluoranthene
ND
13 . 3
ND
ND
Pyrene
ND
ND
ND
ND
Butylbenzylphthalate
ND
ND
ND
ND
3,3'-Dichlorobenzidine
ND
ND
ND
ND
Benzo(a)anthracene
ND
ND
ND
ND
Chrysene
ND
ND
ND
ND
bis(2-Ethylhexyl)phthalate
29 . 2
26.1
ND
71. 62
Di-n-octylphthalate
ND
ND
ND
ND
Benzo(b)fluoranthene
ND
ND
ND
ND
Benzo(k)fluoranthene
ND
ND
ND
ND
Benzo(a)pyrene
ND
ND
ND
ND
Indeno(1,2,3-cd)pyrene
ND
ND
ND
ND
Dibenz(a,h)anthracene
ND
ND
ND
ND
Benzo(g,h,i)perylene
ND
ND
ND
ND
1 68 Deg. f — 29.92 inches Hg.
2 Results are likely due to sample contamination.
ND = Not detected or EV catches; used as zero (0).
3-12
-------
TABLE 11. VOLATILE ORGANICS EMISSIONS SUMMARY
SITE 8
SITE 9
VENTURI/SCRUBBER
WET ESP
OUTLET
OUTLET
voc
Concentration,
ug/dscm a
Acrylonitrile
ND
1060
Vinyl Chloride
ND
66.2
Methylene Chloride (m/z = 86)
108
38 . 3
Chloroform
16 . 8
24 . 1
1,2-Dichloroethane
ND
ND
1,1,1-Trichloroethane
6.8
17.5
Carbon Tetrachloride
ND
ND
Trichloroethene
5,2
24 . 6
Benzene
6.2
6390
Tetrachloroethene
9.4
29. 0
Toluene
7 . 7
4080
Chlorobenzene
ND
55 . 5
Ethylbenzene
2 . 6
100
a = 68 Deg. F — 29.92 inches Hg
bND = Reported as not detected or estimated values; both
expressed as zero (0) in calculating totals and averages.
3-13
-------
ug/dscm, trichloroethene - 5.2 ug/dscm, benzene - 6.2 ug/dscm, tetrachloroethene - 9.4
ug/dscm, toluene - 7.7 ug/dscm, and ethylbenzene - 2.6 ug/dscm.
At Site 9, two of the target compounds were below the minimum detection limit
during all three test runs: 1,2-dichloroethane and carbon tetrachloride. Vinyl chloride
was measured in only two of the tube pairs. The other ten target compounds were
detected for all three test runs and averaged: acrylonitrile - 1060 ug/dscm, methylene
chloride - 38.3 ug/dscm, chloroform - 24.1 ug/dscm, 1,1,1-trichloroethane - 17.5 ug/dscm,
trichloroethene - 24.6 ug/dscm, benzene - 6390 ug/dscm, tetrachloroethene - 29.0
ug/dscm, toluene - 4080 ug/dscm, chlorobenzene 55.5 ug/dscm, and ethylbenzene - 100
ug/dscm.
3.5 Carbon Monoxide and Total Hydrocarbon Monitoring
At Sites 6 and Site 9, a positive correlation between carbon monoxide emissions
and total hydrocarbon (THC) emissions was observed. This relationship is shown
graphically for Site 6 and Site 9 in Figures 15 and 16, respectively. At Site 8, both the
CO and THC emissions were significantly low and a correlation could not be seen.
At Sites 6 and 9, the concentrations of THC and CO were reduced by about 75%
during the improved combustion conditions.
3-14
-------
Outlet emissions data (excluding Run 5)
30
20 -
26 -
E
24 -
o.
Q. 22
(/)
20
C
o
18 -
n
CO
16 -
{>
14 -
o
"O
1? -
>
X
10
"O
8 -
o
o
6 -
4 -
2 -
0 -
r = 0.97
200 400 600
Carbon Monoxide (ppm)
800
Figure 15. Total hydrocarbon emissions versus carbon monoxide emissions, Site 6.
3-15
-------
150
140
-
Run 5 ¦
130
-
120
110
/
100
Run 4 ¦
90
80
-
r = 0.93
s'
70
-
y
60
-
50
40
_
Run 3 .r
Run 2 ¦
30
-
20
Runs 8, 9, 10, 11,-12, and 13
10
¦ 1 "
0
i
-10
i iii
i i i
0.2 0.4 0 6 0.8 1 1.2 1.4
(Thousands)
CO, ppm
Figure 16. Total hydrocarbon emissions versus carbon monoxide emissions, Site 9.
3-16
-------
4.0 CONCLUSIONS
4.1 Metals and Particulates
The total metals and particulate results from Sites 6, 8, and 9 added significantly
to the OW data base especially with respect to the use of the addition of a Wet ESP as a
retrofit to existing systems or as part of the overall control system at a new facility. Of
the metals measured, chromium, lead, and nickel consistently had the highest feed rate
to the incinerators. Cadmium and lead had the highest emission factors of the metals
fed to the incinerators. The emission control devices at the multiple hearth incinerators
had similar removal efficiencies for particulate matter, chromium, and nickel, with lead
and cadmium removal efficiencies being less than particulate matter. At the fluidized
bed incinerator, the venturi/scrubber had the highest removal efficiency without
discriminating between metals and particulate matter. The wet ESPs were effective in
further removal of the metals and particulate matter emitted form the venturi/scrubbers.
4.2 Hexavalent Chromium
At the beginning of this study, EPA did not have a published test method that
gave acceptable results at combustion sources. Through the efforts in this study along
with concurrent work with EPA's Quality Assurance Division, Atmospheric Research and
Exposure Assessment Lab, the hexavalent chromium test method developed for this
program provided acceptable results for the measurement of hexavalent chromium
without artifact formation at the outlet locations. At Site 6, the ratio of hexavalent
chromium to total chromium was high when lime was used for sludge conditioning,
during good combustion conditions, and with the long residence times required for
combustion of sludge in a multiple hearth incinerator. At Site 8, the ratio of hexavalent
4-1
-------
chromium to total chromium in the emissions was very low from a fluidized bed
incinerator (despite relatively high total chromium levels), probably due to the short
sludge retention time in the incinerator and the absence of alkaline material in the
sludge. At Site 9, the ratio of hexavalent chromium to total chromium was significantly
higher than had been anticipated. The facility was selected because it does not use lime
for sludge conditioning. The high hexavalent chromium to total chromium ratio was
discussed with facility representatives and it was determined that some of the sludge that
is transported to the facility contains lime. Also some lime is used at the facility. The
archived sludge digested samples were analyzed for calcium. The sludge solids were
determined to contain 2 to 3% lime by weight. This percentage of lime may be the
reason for the higher than anticipated ratio of hexavalent chromium to total chromium.
4.3 Nickel Subsulfide
Prior to the program EPA did not have a published method for nickel speciation.
Based on new instrumental techniques developed by Brigham Young University and the
continued wet chemical techniques developed by Dr. Vladimar Zatka, it was
demonstrated that nickel subsulfide is not emitted from sewage sludge incinerators above
the level of detection for both analytical techniques. At Site 6, the ratio of nickel
subsulfide to total nickel was extremely low under both normal combustion and
improved combustion conditions. At Site 8, the ratio of nickel subsulfide to total nickel
in the emissions was extremely low, with the nickel sulfide/subsulfide species measured
at the inlet and midpoint being less than the detection limit. The ratio of sulfidic nickel
to total nickel in the emissions from Site 9 is extremely low at Site 9, with the reduced
nickel species being measured at less than detection limit (about 1 to 2% of the total
nickel).
4-2
-------
4.4 Continuous Emission Monitoring of CO and THC
The combustion efficiency at both multiple hearth incinerators was improved
during the test programs. The improved combustion furnace operating conditions
established for the second series of test runs at Site 6 and Site 9 reduced the
concentrations of CO and THC emissions by about 75%. A good correlation between
CO emissions and the THC emissions was seen at the sources that a wide range of CO
and THC was measured.
4.5 Semivolatile Organics
Compared to Site 3, a fluidized bed incinerator where the only semi-volatile
organic compound detected was bis(2-ethylhexyl)phthalate, several additional
semivolatiles were found in the emissions at Site 8. These were 1,2-dichlorobenzene, 1,4-
dichlorobenzene, benzyl alcohol, benzoic acid, and naphthalene.
Only two semivolatile organic compounds, benzyl alcohol and benzoic acid, were
found under normal and improved combustion conditions at Site 9. This number was
less than at Site 2, a multiple hearth incinerator where seven semi-volatile compounds,
phenol, naphthalene, bis(2-ethylhexyl)phthalate, 1,2,-dichlorobenzene, 1,3,-
dichlorobenzene, 1,4,-dichlorobenzene, and 2-nitrophenol were detected. Several
additional compounds were found in the Site 9 emissions for the normal or improved
combustion conditions; these compounds were 1,4-dichlorobenzene, 1,2-dichlorobenzene,
2-nitrophenol, 1,2,4-Trichlorobenzene, naphthalene, 2-methylnaphthalene, dibenzofuran,
phenanthrene, bis(2-ethylhexvl)phthalate, phenol. 4-methylphenol, and 4-nitrophenol.
4.6 Volatile Organics
The volatile organic compound emission results for Site 8 were consistent with the
results for Site 3 (another fluidized-bed incinerator). Carbon tetrachloride and
4-3
-------
chlorobenzene, reported in the emissions at Site 3, were not found in the emissions from
Site 8.
The volatile organic compounds detected in the Site 9 multiple hearth incinerator
emissions were similar to the compounds reported for Sites 1, 2, and 4 (other multiple
hearth incinerator tested). Carbon tetrachloride, reported in the emissions at the other
three sites, was not found in the emissions from Site 9,
4.7 Overall Conclusions from the Study
The primary purpose of the test program at Site 6 was to determine the effect of
lime conditioning and excess combustion air on the formation of hexavalent chromium
emissions. The Entropy sampling and analytical method for hexavalent chromium
worked extremely well in demonstrating the relationship of combustion conditions to the
formation of hexavalent chromium during lime conditioning. A correlation was
demonstrated between CO and THC emissions which was of interest to OW. The nickel
subsulfide emissions were demonstrated to be extremely low. It was also demonstrated
that the CO and THC concentrations could be reduced by about 75% with better
combustion conditions. This reduction of the CO and THC concentrations was not an
intent of the program but was a benefit to the OW data base.
An evaluation of CO and THC monitors was conducted at Site 7. The data
showed that "cold" and "hot" THC monitors give the same results. The application of
the monitors at Site 7 allowed the operators to adjust their incinerator conditions and
significantly reduce the CO and THC concentrations. The trends between CO and THC
concentration were very comparable.
The test program at Site 8 was designed to assess hexavalent chromium and nickel
subsulfide emission from a fluidized bed incinerator. During the planning stages, it was
decided that a pilot-scale wet ESP would be added to the control system at the site and
sampling would be conducted at the inlet, midpoint, and outlet of the control system.
Also sampling and analysis for dioxins/furans, semivolative organics, and volatile
organics were added to the program. Levels of hexavalent chromium, nickel subsulfide,
4-4
-------
CO, and THC were shown to be extremely low. The pilot-scale wet ESP demonstrated
significant collection efficiency at extremely low concentrations of particulate
and metals emissions. The organic emissions were found to be extremely low and were
principally the same compounds previously measured for sewage sludge incinerators.
Entropy conducted the final test program at Site 9 which included a full-scale wet
ESP. The pollutants measured at Site 9 parallelled those at Site 8. The Entropy
recirculating train method for sampling and analysis of hexavalent chromium yielded
consistent data and documented a higher level of chromium than had been anticipated.
It was demonstrated that less than 0.5% of the nickel emissions were nickel subsulfide at
the scrubber discharge. The correlation between CO and THC concentrations was
demonstrated once again as well the reduction in these emissions by about 75% by using
good combustion conditions. It was demonstrated that the full-scale wet ESP could
reduce concentrations of particulates and metals by about 90%. Dioxins/furan were
reduced by about 75% by using good combustion conditions and the wet ESP collected
an additional 75% of the dioxins/furan emissions.
The accomplishments of the study were far greater than could have been
anticipated at the outset of the program. Specifically, the following has been
accomplished.
1. Documented hexavalent chromium emissions from sewage sludge
incinerators.
2. Documented nickel subsulfide emissions from sewage sludge incinerators.
3. Developed a hexavalent chromium sampling and analytical method.
4. Developed a nickel speciation sampling and analytical method.
5. Provided additional metals data.
6. Provided additional trace organics data.
7. Documented a correlation between CO and THC.
8. Documented that CO and THC concentrations can be reduced when the
plant has a CO and/or THC monitor to improve combustion conditions.
9. Demonstrated that the use of a wet ESP is a viable retrofit option for
significantly reducing particulate and metals emissions.
4-5
-------
REFERENCES
1. Entropy Environmentalist, Inc. and DEECO, Inc., Emissions of Metals. Chromium
and Nickel Species, and Organics from Municipal Wastewater Sludge Incinerators
Volume II: Site 5 Test Report - Hexavalent Chromium Method Evaluation. Work
Assignment No. 0-5, EPA Contract No. 68-CO-0027, September 1991.
2. Entropy Environmentalist, Inc. and DEECO, Inc., Emissions of Metals. Chromium
and Nickel Species, and Organics from Municipal Wastewater Sludge Incinerators
Volume III: Site 6 Test Report. Work Assignment No. 0-5, EPA Contract No. 68-
CO-0027, September 1991.
3. Entropy Environmentalist, Inc. and DEECO, Inc., Emissions of Metals. Chromium
and Nickel Species, and Organics from Municipal Wastewater Sludge Incinerators
Volume IV: Site 6 Test Report - Appendices. Work Assignment No. 0-5, EPA
Contract No. 68-CO-0027, September 1991.
4. Entropy Environmentalist, Inc., Emissions of Metals. Chromium and Nickel Species,
and Organics from Municipal Wastewater Sludge Incinerators Volume V: Site 7 Test
Report - CEMs Evaluation . Work Assignment No. 0-5, EPA Contract No. 68-CO-
0027, September 1991.
5. Entropy Environmentalist, Inc. and DEECO, Inc., Emissions of Metals. Chromium
and Nickel Species, and Organics from Municipal Wastewater Sludge Incinerators
Volume VI: Site 8 Test Report. Work Assignment No. 0-5, EPA Contract No. 68-
CO-0027, September 1991.
6. Entropy Environmentalist, Inc. and DEECO, Inc., Emissions of Metals. Chromium
and Nickel Species, and Organics from Municipal Wastewater Sludge Incinerators
Volume VII: Site 8 Test Report - Appendices. Work Assignment No. 0-5, EPA
Contract No. 68-CO-0027, September 1991.
7. Entropy Environmentalist, Inc. and DEECO, Inc., Emissions of Metals. Chromium
and Nickel Species, and Organics from Municipal Wastewater Sludge Incinerators
Volume VIII: Site 9 Test Report. Work Assignment No. 0-5, EPA Contract No. 68-
CO-0027, September 1991.
8. Entropy Environmentalist, Inc. and DEECO, Inc., Emissions of Metals. Chromium
and Nickel Species, and Organics from Municipal Wastewater Sludge Incinerators
Volume IX: Site 9 Test Report - Appendices. Work Assignment No. 0-5, EPA
Contract No. 68-CO-0027, September 1991.
5-1
-------
9. Drees, L.M. Effect of Lime and Other Precipitants or Sludge Conditioners on
Conversion of Chromium to the Hexavalent State When Sludge is Incinerated. Final
Report. EPA Contract No. 68-03-3346, WA 05. 1988.
10. Majiam, T.T. Kasakura, N. Naruse and M. Hiraoka. 1977. Studies of Pyrolvsis
Process of Sewage Sludge. Prog. Wat. Tech. Vol. 9,381-396. Great Britain: Pergamon
Press.
11. Umashima, T., M. Naruse and T. Nasakura. 1975. Behavior of Cr + 6 in Incinerator
Process of Sewage Sludge. Paper presented at the 12th Annual Meeting of the
Association of Japan Sewage Works.
5-2
-------
TECHNICAL REPORT DATA
(Please resd lnttruci:cns on Ike reverse before compter
1. REPORT NO.
EPA/600/R-92/003a
PB92-151554
TITLE AN0 SUBTITLE
EMISSIONS OF METALS, CHROMIUM AND NICKEL SPECIES, AND
QRGANICS FROM MUNICIPAL WASTEWATER SLUDGE INCINERATORS
VOLUME I: SUMMARY REPORT
5. REPORT OATE
March 1992
6. PERFORMING organization cooe
7, AuTHORiSi
William G. DeWees, Robin R. Segall
F. Michael Lewis
8, PERFORMING ORGANIZATION REPOR' NO
9. PERFORMING ORGANIZATION NAME AND aooress
Entropy Environmentalists, Inc.
Research Triangle Park
North Carolina, 27709
10. PROGRAM ELEMENT NO,
11. CONTRACT/GRANT NO,
Contract No, 68-CO-QQ27
Work Assignment No. 0-5
12. SPONSORING AGENCY NAME AND AOORESS
Risk Reduction Engineering Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
13. type of report a,no PERIOO COVERED
Final Report 1989 - 91
14. SPONSORING AGENCY coqe
.EPA/600/14
15. supplementary notes
EPA Technical- Contact:
16. abstract
Dr. Harry E. Bostian, (513) 569-7619, FTS: 684-7619
." At Site 5 (continuing a numbering system initiated in a previous 4-aite
project) tests were only conducted for methods development purposes. At Site 6,
emissions were measured at the inlet and outlet of the control device. At Site 7,
evaluation of CO and THC CEMSa was performed. At Sites 8 and 9, emissions were
measured at the inlet of the venturi scrubber, at the midpoint located between the
venturi scrubber and the wet ESP, and at the outlet of the wet ESP. For Sites 6, 8,
and 9, midpoint and outlet air emission samples were collected and analyzed for
particulate matter, metals, PCDD/PCDFs, volatile and semivolatils compounds (except
Site 6), and hexavalent chromium and nickel subsulfide species. Continuous emission
monitoring (CEM) for 02, C02, CO, SOjy and NO,*- at the control system inlet and Oj'?
(except Site 6), C0j-(except Sites 6 and 9), CO, SOa (except Sites 6 and 9}, NO,-
(except Sites 6 and 9), and THC was conducted at the control system outlet stack.
The metals found in the greatest concentration in the sludge were lead,
chromium and nickel. The need for sampling of hexavalent chromium without artifact
formation and analysis of the resulting samples specifically for hexavalent chromium
at low concentrations was a major accomplishment of this test program. The results
of the nickel sampling and analysis indicate that within the detection limit of the
wet chemical method, no nickel subsulfide was present in the air emissions.
At Site 6 and Site 9, a positive correlation between carbon monoxide emissions
and total hydrocarbon (THC) emissions was observed. At Site 7, a positive
correlation was demonstrated between the "Hot" and "Cold" THC CEMs.
17.
KEY WOROS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSati Field,'G:oup
Wastewater, sludge disposal,
incinerators, combustion products
Emissions
chromium compounds
nickel compounds
total hydrocarbons
dioxin/furans
organic compounds
ia, distribution statement
19. SECURITY CLASS jTMs Report)
UNCLASSIFIED
21. NO. OF PAGES
fin
J RELEASE TO PUBLIC
20. SECURITY CLASS/Tlus page)
UNCLASSIFIED
22. PRICE
EPa Fotm 2320—1 (R«*. 4-7?) prbyioui f aition i j pb»olett
------- |