EPA
United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/2-78-198
December 1978
Research and Development
The Codisposal of
Sewage Sludge and
Refuse in the
PUROX System
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U-S. Environmental
The nine series are: ;
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research ;
4, Environmental Monitorinig
5. Socioeconomic Environmental Studies
6 Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports ,
9, Miscellaneous Reports \
..
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virgimk 22161.
-------
EPA-600/2-78-198
December 1978
THE CODISPOSAL OF SEWAGE SLUDGE AND
REFUSE IN THE PUROX SYSTEM
by
Union Carbide Corporation
Linde Division
Tonawanda, New York 14150
Grant No. S803769-01-3
Project Officer
Gerald Stern
Wastewater Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
This study was conducted
in cooperation with the
Sanitary Board of the City of
South Charleston, West Virginia
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory U.S. Environmental Protection Agency and approved for publica-
tton. Approval does not signify that the contents necessarily reflect the views
and policies of the U. S. Environmental Protection Agency, nor does mention of
tradenames or commercial products constitute endorsement or recommendation
for use.
XI
-------
FOREWORD
_ _ The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health and
welfare of the American people. Noxious air, foul water, and spoiled land are
tragic testimony to the deterioration of our natural environment. The com-
plexity of that environment and the interplay between its components require a
concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem so-
lution and it involves defining the problem, measuring its impact and
searching for solutions. The Municipal Environmental Research Laboratory
develops new and improved technology and systems for the prevention treat-
ment and management of wastewater and solid and hazardous waste pollutant
discharges from municipal and community sources, for the preservation and
treatment of public drinking water supplies, and to minimize the adverse
economic, social, health, and aesthetic effects of pollution. This publication
is one of the products of that research; a most vital communications link
between the researcher and the user community.
The codisposal of sewage sludge and municipal refuse in an environ-
mentally sound fashion provides a synergistic solution to a difficult waste
disposal problem with the added benefit of energy recovery that has previously
been lost. In light of current environmental concerns and energy limitations
new technologies such as reported here have taken on increased significance.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
111
-------
PREFACE
This report is intended to formally document the codisposal test
program conducted in South Charleston, West Virginia during the^spring of
1977. It consists of the final report On the operation that was submitted to
the EPA as well as a supplementary section containing information that was
not transmitted to the EPA. This supplementary section is for internal UCC
use onSTit documents the type and!extent of operational problems encountered
during the codisposal program as well as organizing the mass of data collected
into a set of useful tables for reference.
The EPA report and supplement are each presented as separate
entities. A table of contents for the supplement is presented here for reference.
IV
-------
ABSTRACT
This test program was conducted with three overall objectives:
1. To establish the technical feasibility of codisposal of filtered munic-
ipal sewage sludge of various types with mixed municipal refuse in the
PUROX System.
2. To determine the environmental effect of such a codisposal process.
3. To estimate the economic advantage of this codisposal process.
The PUROX System demonstration plant in South Charleston, West
Virginia was modified to mix filtered sewage sludge cake with shredded, mag-
netically separated refuse prior to feeding into the PUROX System. The modi-
fied system included the standard PUROX System equipment: a shredder, mag-
netic separator, oxygen blown converter, slag quench tank, offgas scrubber,
electrostatic precipitator, condenser and product gas combustor. A sludge re-
ceiving and storage station, a metering hopper, and a transfer conveyor were
added to provide the sludge processing capability. Raw primary sludge filter
cake with 25 percent solids and mixed primary and secondary sludges (mixed sludge)
filter cake with 20 percent solids were processed. In the process the refuse/sludge
mixture was dried, pyrolyzed, combusted, and the inorganic residue was slagged to
produce a fuel gas, an environmentally inert slag, and a wastewater stream.
Results of the tests indicate that sludge can be processed with refuse
in an efficient, environmentally sound manner. System performance at dry-
sludge to refuse ratios up to 0.075 was excellent. Values for key operating
variables such as oxygen consumption, gas production, and gas heating value
were within the range of variation experienced with refuse only. Use of 80
percent moisture sludges in the composition range studied resulted in no sig-
nificant variation in overall performance. Although maximum operating capacity
was not tested due to the limited supply of refuse, the substitution of wet
sludge for refuse on a unit-for-unit basis should not affect the total operating
rate when defined as total units per day fed. Cleanly, the yield of product gas
per unit of material fed will decrease at higher moisture levels. To process
higher dry sludge-to-refuse ratios, minor modifications of the refuse-sludge
blending operation may be required.
v
-------
The effect of effluent streams from the process on the environment falls
well within Federal guidelines for emissions. Heavy metals found in both the
sludge and refuse were effectively contained in the slag. Samples of slag were
tested for leachability, and results of' the tests indicated no significant leaching
from the slag. The particulate content of the product gas presents no apparent
problem when compared to regulation si on emissions from a stationary source.
Bench scale treatability studies of wastewater during this program and previous
studies with refuse-only indicate that PUROX System wastewater can be treated
to municipal sewer discharge standards.
The economics of this codispoisal process are very sensitive to the dry
sludge-to-refuse ratio, sludge moisture content, and overall facility size.
The estimated net sludge disposal cost for processing 63.5 dry Mg (70 dry tons)
per day of sludge at a sludge (dry solids) to refuse ratio of 0.05 is approximately
$110 per dry Mg ($100 per dry ton) using the basic process as reported here.
A modification of the process to provide improved filtration of sludge and pre-
drying of sludge filter cake can result in a reduction in the estimated sludge
disposal cost by 80 percent to approximately $22 per dry Mg ($20 per dry ton).
This report was submitted in fulfillment of Grant #3-803769 by Union
Carbide Corporation under the sponsorship of the Sanitary Board of the City of
South Charleston, West Virginia and the U. S. Environmental Protection Agency.
This report covers the period June 24f 1975 to June 16, 1977 and work was
completed as of January, 1978.
-------
CONTENTS
Foreword m
Preface ซ . j_v
Abstract i ...".!.* 1 *. v
Figures
Tables x
Abbreviations and Symbols xiii
Acknowledgment xv
1. Introduction 1
2. Conclusions 3
3. Recommendations 5
4. PUROX System Facility Process Description 6
.5. PUROX System Facility Equipment Description 22
6. Equipment Modifications for Codisposal of
Sewage Sludge and Refuse 28
7. Sludge-Refuse Codisposal Test Plan 34
8. Data Collection and Sample Analysis , 39
9 . Summary of Test Period Results 44
10. Codisposal Process Observations 52
Gas Production Rate . 52
Gas Heating Value 54
Gas Particulate Loading 59
11. Projection of Results to Case of Particulate Recycle ... 62
12. Environmental Effects of the Codisposal Process 70
13. Projected Economics of the Codisposal Process 81
References 98
Appendices
A.
B.
C.
D.
E.
F.
G.
H.
Analytical Procedures 99
Test Period 5. RefuseOnly - May 25-June 1, 1977
104
Test Period 1,
Test Period 2,
Test Period 6,
Test Period 7,
Test Period 3,
Primary Sludge - April 19-25, 1977 . . . . 115
Primary Sludge - April 27-30, 1977 . . . . 123
Primary Sludge - June 2-5, 1977 130
Primary Sludge - June 6-9, 1977 138
Mixed Sludge - May 2-5, 1977 . ' 146
Test Period 4. Mixed Sludge - May 23-25, 1977 154
VII
-------
FIGURES
Number
Page
1 Typical PUROX system mass balance 7
'2 Typical PUROX system energy balance . 8
3 PUROX system facility S. Charleston, West Virginia ...... 9
4 Packer truck delivering refuse . 10
5 Refuse inventory in storage building .,,..11
6 Front loader on refuse weigh scale ,, . . 12
7 Refuse on apron conveyor to hammermill . . 13
8 Vertical shaft hammermill. . . 14
9 Drum magnetic separator . .( 15
10 Slag conveyor ; 18
11 Converter and gas cleaning system 19
12 PUROX system facility for sludge/refuse codisposal
South Charleston, West Virginia 23
13 Sludge cake at dumping station 29
14 Front loader scooping sludge cake 30
15 Metering hopper augers . 31
16 Sludge cake on troughing conveyor 32
17 Metering hopper and troughing conveyor 33
18 Product gas heating value - refuse only vs operating rate . . . .56
19 Product gas heating value - raw primary sludge vs
operating rate 57
20 Product gas heating value - mixed primary and secondary
operating rate 58
21 Estimated net sludge disposal cost at 0.05 sludge
(ds)/refuse with refuse tip fee of $16.50/Mg ........ 86
Vlll
-------
FIGURES (Continued)
Number Page
22 Estimated net sludge disposal cost at 0.05 sludge
(ds)/refuse with refuse tip fee of $15/ton 87
23 Estimated net sludge disposal cost at 0.05 sludge
(ds)/refuse with refuse tip fee of $5.50/Mg 88
24 Estimated net sludge disposal cost at 0.05 sludge
(ds)/refuse with refuse tip fee of $5/ton 89
25 Estimated net sludge disposal cost at 0.30 sludge
(ds)/refuse with refuse tip fee of $16.50/Mg 90
26 Estimated net sludge disposal cost at 0.30 sludge
(ds)/refuse with refuse tip fee of $15/ton 91
27 Estimated net sludge disposal cost at 0.05 sludge
(ds)/refuse with refuse tip fee of $5.50/Mg 92
28 Estimated net sludge disposal cost at 0.05 sludge
(ds)/refuse with refuse tip fee of $5/ton 93
29 Estimated net sludge disposal cost at 0.60 sludge
(ds)/refuse with refuse tip fee of $16.50/Mg 94
30 Estimated net sludge disposal cost at 0.60 sludge
(ds)/refuse with refuse tip fee of $15/ton 95
31 : Estimated net sludge disposal cost at 0.60 sludge
(ds)/refuse with refuse tip fee of $5. 50/Mg 96
32 Estimated net sludge disposal cost at 0.60 sludge
(ds)/refuse with refuse tip fee of $5/ton 97
-------
TABLES
Number
1 Component Analysis of Refuse
2 Ultimate Analysis of Refuse;
3 Analysis of Typical PUROX System Residue . . . .. . . .
4 Analysis of Typical PUROX System Product Gas , . . . .
5 Analysis of Typical PUROX System Wastewater . . . . .
6 Original Sludge Codisposal Test Plan .
7 Sludge Codisposal Test Series
8 Overall Mass Balance Assumptions
9 Summary of Overall Mass Balances .
10 Summary of Product Gas Analyses ...........
11 Overall Heat Balance Assumptions . ..........
12 Summary of Overall Heat Balances
13 Summary of Slag Analyses
14 Gas Production Measured by Helium Trace
15 Gas Production Measured by Pitot Tube. ........
16 Gas Production Based on Gasifiable Material ......
17 Product Gas Higher Heating Value (HHV) Comparison . . .
18 Statistical Comparison of Product Gas Heating Values . .
19 Product Gas Particulate Levels
20 Calculated Particulate Levels in PUROX System Gas
Combustion Products at 12 Percent CO2 . . . . .
21 Particulate Levels in Product Gas Based on Sludge Rate. .
22 Material Balances with Particulate Recycle (Metric) . . .
23 Material Balance Results with Particulate Recycle (English)
24 Oxygen Consumption Expected with Particulate Recycle. .
25 Oxygen Consumption Rangejs with Various Feeds
26 Gas Production Expected with Particulate Recycle ....
27 Energy Balances with Particulate Recycle (Metric) ....
28 Energy Balances with Particulate Recycle (English). . . .
29 Slag Leachate Analyses. .;....
30 Concentration of Metal in PUROX System Gas
31 Grams of Metal Emitted in PUROX System Gas
32 PUROX System Wastewater Treatability Data ......
33 Product Stream Trace Metal Concentration with
Particulate Recycle
16
16
17
20
21
35
37
44
46
47
48
49
50
52
53
53
55
59
60
61
61
64
65
66
66
67
68
69
71
73
73
74
76
x
-------
TABLES (Continued)
Number
34
35
36
37
38
39
A-l
A-2
B-l
B-2
B-3
B-4
B-5
B-6
B-7
B-8
C-l
C-2
C-3
C-4
C-5
C-6
D-l
D-2
D-3
D-4
D-5
D-6
E-l
E-2
E-3
E-4
E-5
E-6
F-l
F-2
F-3
Trace Metal Concentrations in Product Gas Combustion
Products
Trace Metal Vapor Pressures and Maximum Gas
Concentrations - . . . .
Wastewater Metal Concentration from Refuse Only
Operation with Particulate Recycle
Trace Metal Concentrations in Slag .
Trace Metal Disposition in Process Effluent Streams . .
Estimated Economics - PUROX Codisposal Facility
Detection Limits for Trace Elements .
Analytical Instruments \ \
PUROX System Mass Balance, Test Period 5 (Metric). .
PUROX System Mass Balance, Test Period 5 (English) .
Assumptions for Component Mass Balances
Average Product Gas Analysis, Test Period 5 . . . .
Assumptions for Trace Metal Balances
Trace Metal Balance, Test Period 5 (Metric)
Trace Metal Balance, Test Period 5 (English)
PUROX System Heat Balance, Test Period 5
Mass and Component Balance, Test Period 1 (Metric) .
Mass and Component Balance, Test Period 1 (English) .
Average Product Gas Analysis, Test Period 1 . . . . .
Trace Metal Balance, Test Period 1 (Metric)
Trace Metal Balance, Test Period 1 (English)
PUROX System Heat Balance, Test Period 1
Mass and Component Balance, Test Period 2
Mass and Component Balance, Test Period 2
Average Product Gas Analysis, Test Period 2
Trace Metal Balance, Test Period 2 (Metric) .
Trace Metal Balance, Test Period 2 (English) .
PUROX System Heat Balance, Test Period 2
Mass and Component Balances, Test Period 6
Mass and Component Balances, Test Period 6
Average Product Gas Analysis, Test Period 6
Trace Metal Balance, Test Period 6 (Metric)
Trace Metal Balance, Test Period 6 (English)
PUROX System Heat Balance, Test Period 6
Mass and Component Balances, Test Period 7 (Metric).
Mass and Component Balances, Test Period 7 (English)
Average Product Gas Analysis, Test Period 7
(Metric)
(English)
(Metric) .
(English)
76
78
78
80
80
83
100
103
105
106
107
108
109
112
113
114
116
117
118
119
120
122
124
125
126
127
128
129
132
133
134
135
136
137
140
141
142
-------
TABIฃS (Continued)
Number Paffi
F-4 Trace Metal Balances, Test Period 7 (Metric) ......... 143
F-5 Trace Metal Balance, Test Period 7 (English) ........ 144
F~6 PUROX System Heat Balance, Test Period 7 ........ 145
G-l Mass and Component Balances, Test Period 3 (Metric) .... 148
G-2 Mass and Component Balances, Test Period 3 (English) . . . .149
G-3 Average Product Gas Analysis, Test Period 3 ......... 150
G-4 Trace Metal Balance, Test Period 3 (Metric) ........ 151
G-5 Trace Metal Balance, Test Period 3 (English) ........ 152
G-6 PUROX System Heat Balance, Test Period 3 153
H-l Mass and Component Balance, Test Period 4 (Metric)' 155
H-2 Mass and Component Balances, Test Period 4 (English) .... 156
H-3 Average Product Gas Analysis, Test Period 4 . . . 157
H-4 Trace Metal Balance, Test Period 4 (Metric) 158
H-5 Trace Metal Balance, Test Period 4 (English) ........ 15#
H-6 PUROX System Heat Balance, Test Period 4 ......... 160
XII
-------
LIST OF ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
Btu
C
d
ft
F
gal.
gr
g
GJ
HHV
hr
I
I
kg
kj
kw
Ib
m
m3
mg
Mg
/"g
MJ
MLVSS
Nm3
Pa
pH
psig
Sft3
SBOD5
SCOD
se
So
TBOD5
TCOD
British thermal unit
Celsius
day
foot
cubic feet
Fahrenheit
gallon
grain
gram g
Gigajoule =10 Joules
Higher heating value
hour
Isokinetic variation (%)
Joule
kilogram
kilojoule = 103 Joules
kilowatt
pound
meter
cubic meter
milligram =10" grams
Megagram =10" grams
Micrograms = 10~6 grams
Megajoule = 10$ Joules
Mixed liquor volatile suspended solids
Normal cubic meter at 0 degrees C, 1 atm
Pascal (pressure)
Negative logarithm of hydrogen ion concentration
pound per square inch gage
Standard cubic feet at 60 degrees F, 1 atm
Soluble biological oxygen demand
Soluble chemical oxygen demand
wastewater reactor effluent
waste water reactor input
Total biological oxygen demand
Total chemical oxygen demand
Retention time
Xlll
-------
LIST OF ABBREVIATIONS AND, SYMBOLS (continued)
SYMBOLS
Ag
Ar
As
Au
A12ฐ3
& o
Ba
BaO
Be
C
Cd
Or
Cu
CaO
Cr2ฐ3
Cl
CHgOH
C2H2 -
CHgCOOH
C2H4 -
f~l TJ __
C2H6
f w _
ฐ3H6
r* w
C3H8
ฐ6H6 ~
CO
co2 -
F
Fe
FeO
Silver
Argon
Arsenic
Gold
Alumina
Barium
Barium oxide
Beryllium
Carbon
Cadmium
Chromium
Copper
Calcium oxide
Chromic oxide
Chlorine
Methane
Methanol
Acetylene
Acetic acid
Ethylene
Ethane
Propylene
Propane
Benzene
Carbon monoxide
Carbon dioxide
Fluorine
Iron
Iron oxide
H
HC1
HF
H2
HNO3
H2S
H2Sฐ4
Hg
KI
KMnO4
MgO
Mn
MnO
NaOH
Ni
NiO
O
ฐ2
P
P2ฐ5
Pb
Pt
Si00
2
SO
S03
TiO
L*
TLV
TWA
Zn
Hydrogen (atomic)
Hydrochloric acid
Hydrofluoric acid
Hydrogen (molecular)
Nitric acid
Hydrogen sulfide
Sulfuric acid
Mercury
Potassium Iodide
Potassium permanganate
Magnesium oxide
Manganese
Manganese oxide
Sodium hydroxide
Nickel
Nickel oxide
Oxygen (atomic)
Oxygen (molecular)
Phosphorus
Phosphorus pentoxide
Lead
Platinium
Silicon dioxide
Sulfur dioxide
Sulfur trioxide
Titanium dioxide
Threshold level value
Time weighted average
Zinc
xiv
-------
ACKNOWLEDGMENTS
The Sanitary Board of the City of South Charleston, managed by
Lynn Cavendish under the direction of chairman Richard A. Robb has con-
tributed significantly to the overall success of this project. The assistance
of Mr. Carl Karlson and Mr. Jessie Hall at South Charleston Wastewater
Treatment Plant is similarly appreciated.
The willing cooperation of the personnel of the City of Huntington,
West Virginia, Wastewater Treatment Plant headed by Mr. J. P. Rardin under
the supervision of Mr. G. R. Weekley, Director of Public Works is also
acknowledged.
Similarly, the assistance of personnel at the Alleghany County Sanitary
Authority Waste Treatment Plant in Pittsburgh, Pennsylvania, under the
direction of Mr. William Trefz, Chief Engineer, is appreciated.
Finally, the cooperation and guidance provided by Mr. Gerald Stern,
Project Officer, and Dr. J. B. Farrell, Municipal Environmental Research
Laboratory, U. S. Environmental Protection Agency are acknowledged.
xv
-------
-------
SECTION 1
INTRODUCTION
The disposal of waste materials in an environmentally sound fashion
has been a major concern for many years. In view of the increasing scarcity
of energy, the disposal of waste materials such as mixed municipal refuse and
sewage sludge by the conventional techniques of landfilling or incineration
represents both a waste of valuable energy resource and an environmental
hazard. The PUROX System has been developed as an environmentally accept-
able conversion technology for the recovery of usable energy from solid waste
materials. The key element in this process is the converter which is an
oxygen-blown shaft furnace in which the solid waste materials are dried, py-
rolyzed, combusted, and the inorganic residues slagged. The use of oxygen
in the converter results in a very small gas volume to be cleaned, low exhaust
gas temperatures from the converter, minimum dilution of the fuel gas with
nitrogen, and maximum consolidation of inorganics in the residue. This con-
solidation is a result of the high temperature slagging conditions in the hearth
of the converter. The conversion process produces a useful fuel gas, waste-
water and an inert slag stream. The operation is supported by several sub-
systems: a feed material storage and preparation system, a gas cleaning
system, a wastewater treatment system, and an oxygen generating system,
and utilities and services .
The codisposal of sewage sludge and refuse in the PUROX System is
viewed as desirable for several reasons:
(a) Codisposal of refuse and sewage sludge permits disposal of sludge and
refuse with a net production rather than a net consumption of energy as
in conventional incineration techniques.
(b) The refuse bed in the shaft furnace provides efficient heat transfer from
ascending gas to dry the solids prior to pyrolysis and combustion as
well as acting as an effective heat sink to cool the exhaust gas to a
low temperature before it exits from the converter. This low exhaust
temperature ensures that no significant amount of heavy metals will be
present as vapor in the gas stream.
-------
(c) The glassy slag from the refuse provides an ideal matrix in which the
trace metal contaminants from the sludge are trapped by physico-
chemical bonding in the silicate matrix of the slag.
This program was devised to investigate the overall technical feasibility
and environmental effect of this new concept for disposing of sludge with
refuse in a slagging system. Theieconomics of conducting this codisposal
process on a commercial scale are also briefly discussed.
-------
SECTION 2
CONCLUSIONS
The codisposal of both raw primary and mixed primary and secondary
sewage sludge (mixed sludge) filter cake with municipal refuse in the PUROX
System has been demonstrated during a 64 day test period of 24 hour per day
operation. Over the range of feed compositions investigated, (0.024 to 0.074
sludge dry solids to refuse), sludge was dried, pyrolyzed, combusted, and
slagged in combination with municipal refuse in the PUROX System to produce
a useful fuel-gas, an environmentally inert slag, and wastewater which could
be treated to acceptable discharge standards.
Of particular interest is the deposition of heavy metals in the process
streams leaving the PUROX System. Projecting the actual test results to include
particulate recycle (not tested with refuse-sludge mixtures because of mechanical
problems in the recycle stream) it is projected that only Hg, Zn and Pb would
show less than 90% retention in the inert slag (however, over 75% of these three
elements would end up in the slag). The gas-cleaning system is, however,
known to be effective for removing these elements.
The environmental effect of this codisposal technique appeared to fall
well within federal guidelines for environmental emissions. The capture of
heavy metals in a non-leachable form in the slag as well as the recovery of a
large portion of the sludge heating value makes the use of this technique
particularly attractive.
A projection of the economics of commercially applying this process
shows a strong sensitivity to sludge moisture content, refuse-to-sludge ratio,
and overall facility size. The economics of the process are closely tied to
case-specific factors particularly the allowable refuse tipping, fee. At the dry
sludge-to-refuse ratio of 0.05, a net sludge disposal cost of about $110 per
dry Mg ($100 per dry ton) of sludge is projected based on a refuse tipping fee
of $14.30/Mg ($13 per ton) utilizing a codisposal plant processing 63.5 dry
Mg (70 dry tons) per day of sludge and 1270 Mg (1400 tons) per day of refuse.
At a sludge (dry solids) to refuse ratio of 0.074 using 23% solids raw
primary sludge cake, one Mg (0.91 ton) of refuse/sludge mixture could be
converted to 0.59 Mg (0.53 ton) of product gas, 0.19 Mg (0.17 ton) of slag,
and 0.43'Mg (0.39 ton) of wastewater using 0.21 Mg (0.19 ton) of oxygen.
3
-------
Over the range of feed compositions studied, the system performance
parameters in the codisposal mode were quite similar to the performance
processing refuse alone. Oxygen consumption and offgas quality were both
within the range of variation observed with refuse alone. Based on measured
values, the overall conversion of the energy in the feed material to useful gas
in the 12.8 - 14.8 MJ/Nm3 (325 - 375 Btu/Sft3) range is expected to exceed
70 percent when the particulate streams are recycled.
At the lowest sludge solids content (high moisture sludge), the blending
of the sludge and refuse was not uniform. Equipment specified for this task
requires careful selection to ensure proper operation of the PUROX System if
sludge moisture contents greater than 75% are to be used. Equipment changes
from that described here would be required to process less than 25 percent
solids material.
Mechanical problems with the particulate recycle system prevented
collection of a complete set of representative samples during the tests. How-
ever, additional internal stream samples were collected to permit the complete
evaluation of the operation and to allow reasonable projections from the actual
operating conditions to design operating conditions. Subsequent to the Co-
disposal tests, data was collected with the particulate recycle system operating
while processing refuse only. This data supports the reasonableness of these
projections.
-------
SECTION 3
RECOMMENDATIONS
Codisposal of sewage sludge and refuse in the PUROX System should
continue to be evaluated. Follow-on studies should be conducted to evaluate
alternate sludge/refuse blending devices which would permit operation at
higher sludge-to-refuse ratios. The feasibility of integrating the wastewater
treatment and refuse disposal facilities should be examined.
The low dry sludge-to-refuse ratios investigated in this study are
appropriate to meet the needs of municipalities with relatively small amounts
of sludge to dispose of and high allowable refuse disposal costs. For com-
munities with larger amounts of sludge and less favorable allowable refuse
disposal costs, higher ratios of dry sludge-to-refuse are appropriate.
Combination of these higher ratios with integration of refuse processing and
wastewater treatment plant warrants study. Codewatering of sludge and slurry
from the scrubber may significantly reduce costs over separate filtration.
Similarly, predrying the char/sludge filter cake prior to charging into the PUROX
System converter should provide economical and efficient utilization of oxygen
within the PUROX System and hence reduce overall sludge disposal costs.
-------
SECTION 4
PUROX SYSTEM FACILITY PROCESS DESCRIPTION
The PUROX System codisposal facility as designed by Union Carbide
consists of the following components: a front end refuse receiving and prepa-
ration system, a PUROX System consisting of, but not limited to a. refuse
gasifying converter, an offgas cleaning system, and auxiliary systems in-
cluding oxygen generating, wastewater treatment, and product gas handling
systems. Figures 1 and 2 present itypical mass and energy balances on the
181 Mg (200 ton) per day demonstration plant in South Charleston, West
Virginia. Figures 3-11 present photographs of the operating system.
Processing of refuse begins in the refuse receiving and storage building.
(Figure 3). This building provides one day's inventory of refuse at the system's
rated capacity. Refuse is brought ;to the plant in conventional packer trucks
and discharged directly onto the building floor (Figures 4 and 5). A front
loader manages the refuse inventory and loads refuse into the conveying system
for processing.
The front loader holds about 1.1 Mg (one ton) of refuse in its bucket,
carries the refuse to a scale for weighing, and discharges the refuse into the
processing system (Figure 6). The refuse entering the process has a composi-
tion similar to that listed in Tables 1 and 2. Refuse is dumped onto an apron
conveyor for transport to a shredder (Figure 7). The shredder rated at 13.6 Mg
(15 tons) per hour shredding capacity has significant overcapacity to allow for
mechanical problems in the feed system and downtime for routine shredder
maintenance (Figure 8).
Shredded refuse is discharged to a high speed belt conveyor for trans-
port to the feed conveyor. This refuse passes under a drum magnetic separator
where ferrous metals are removed and discharged onto a third conveyor
(Figure 9). The magnetically separated refuse drops from the belt conveyor
onto the feed conveyor to the feeder. The refuse feeder forms the shredded
refuse into pellets and introduces them into the converter. These pellets while
in the feeder provide a gas seal for the converter.
In the converter the refuse is contacted countercurrently with hot gases
from combustion in the hearth; in the upper portion of the converter free
6
-------
1
LI
(J
e
ป-
*L
'
i i- co ro
' (D CM
) < 0
S 0 >-
)
i
! s
LU - ro
1 Q
0 (0
1 p a
< s
O. ' CU
' ^ to
?=; l^1
CJ w
yj u i
1- LU O JD
& ao^ฐ:
0 0
{_) "ป '
il '-: '' _j " 2
" " 3
O^
* 1
^-^ ^-x r*-i
O (/) O
Q LU sf CM
Z (J CM CM _l CM O
LU >- CM LU O
X 0 -5 0
(- O O *-* IL O *~
L^
งLue>P
LU D ^
:U LL 0 ซ
CO LU iJU -
-------
r
CO CD
x
Q
o o in
cc
Q_ Oi 00
co
u
0)
CD
o
in
o o
QL
I11
vO
O
X
CO O
CM ซ-i
(S
8
in
CD
ro vo
u o
X
ro
ro
o
CD
zle? x
ro
8
o o
UJ
in
u
L_
Q
UJ
2 ~) H
LU ID CD
%ฐ.ฐ.
O O O
CD
O
x
M] eg a\
UJ
td
O O CQ
'CD
co
e>
ro oo
CD
vO
uj D i^- in
U U.
CO UJ UJ +H o
< cr or r1 ซ-ซ
CO
O
C
10
II
(0
B1
co
C
CD
CO
4->
M
>.
w
ง
(0
o
CSI
-i-4
P-.
-------
(0
-II
'Si
M
CO
0)
c
o
-M
CO
0
II
S3
.n
O
ra
o
(0
<4-t
a
CD
4->
CO
>1
CQ
X
o
CO
0)
S3
tn
-------
Figure 4. Packer truck delivering refuse.
10
-------
*
-^ .-e, ,
Figure 5. Refuse inventory in storage building.
11
-------
Figure 6. Front loader on refuse weigh scale.
12
-------
Figure 7. Refuse on apron conveyor to hanmermill.
13
-------
Figure 8. Vertical shaft hammermill.
14
-------
Figure 9. Drum magnetic separator.
15
-------
TABLE 1. COMPONENT ANALYSIS OF REFUSE*
Wet Basis
wt. %
Range (Wt. %)
Paper
Food
Yard
Wood
Plastic
Textile
Rubber & Leather
Glass
Ferrous Metal
Non -Ferrous Metal
Dirt & Ash
38
20
13
3
1
1
1
11
7
1
4
25-60
10-30
10-20
2-4
0.5-2
0.5-4
0.5-3
5-25
5-9
0.2-1
1-6
100%
Moisture included in above 26%
*(1) C. T. Moses, J. R. Rivero; "Design and Operation of the PUROX System
Demonstration Plant, "Fifth National Congress on Waste Management
Technology and Resource Recovery, Dallas, Texas, 1976.
TABLE 2 . ULTIMATE ANALYSIS OF REFUSE*
Component
Weight %
H20
C
H
O
N
S
Cl
Metal
Glass
Ash
26.0
25.9'
3.6
19.9
0.47
0.10
0.13
8.0
11.0
4.0
100.0
Higher Heating Value 11.61 MJ/kg (4,992 Btu/lb)
*(1) C. T. Moses, J. R. Rivero; '"Design and Operation of the PUROX System
Demonstration Plant, "Fifth National Congress on Waste Management
^Technology and Resource Recovery, Dallas, Texas, 1976.
16
-------
moisture is driven from the refuse. As. refuse descends down the shaft, it
contacts high temperature gas and the organic material in the refuse begins to
pyrolyze. In the pyrolysis reactions, cellulosic material is broken down into
smaller molecules which volatilize into the hot gas stream. Coincidentally,
hot gas produced by combustion of char in the hearth undergoes water-gas shift
and carbonization reactions with the refuse. The complex mixture of gas and
refuse undergoes a variety of reactions which result in the final offgas
composition. Gas typically leaves the converter between 93 and 316 degrees G
(200 and 600 degrees F) with a wet bulb temperature between 77 and 82
degrees C (170 and 180 degrees F). .
Oxygen blown into the hearth reacts with the char residue from
pyrolysis. The hearth operates at a temperature of about 1,650 degrees C
(3,000 degrees F) in order to slag the inorganic portion of the refuse. Molton
slag pours from the hearth into a water quench tank. Cooled slag is removed
from the bottom of the tank by a drag conveyor and deposited in a dumpster for
removal (Figure 10). A typical slag composition from, refuse operation is shown
in Table 3. It is similar to the compositiop of soda-lime or bottle glass which
is its main constituent. Most of the inorganic residue from the refuse is bound
in the glassy slag.
TABLE 3. ANALYSIS OF TYPICAL PUROX SYSTEM RESIDUE
Silicon
Aluminum
Calcium
Sodium
Iron
Magnesium
Potassium
Other
Weight %,
59.7
10.5
10.3
8.0
6.2
2.2
1.0
2.1
Expressed as Oxide
Range
57-62
9-13
9-12
7-10
1-8
1-4
-
-
Fuel gas is used in heating torches at the slag tap to maintain slag
fluidity at the pouring point from the converter. It is also used in the tuyeres
to preheat:the oxygen at the injection point to preclude local chilling by cold
oxygen. The energy required to preheat the oxygen to the hearth temperature of
1,650 degrees C (3,000 degrees F) from 15 degrees C (60 degrees F) is 988kJ/
kg (425 Btu/lb) of oxygen. This is equivalent to 0.009 kg (0.02 Ib) of natural
gas or 0.036 kg (0.08 Ib) of product gas per kg of oxygen. In a commercial
plant, product gas would be recycled to provide this energy. Although the net
17
-------
Figure 10. Slag Conveyor
18
-------
Figure 11. Converter and gas cleaning system.
19
-------
quantity of gas produced per Mg (ton) of feed might be slightly reduced with
recycle, the overall efficiency of conversion of the input energy to gas would
remain above 70 percent since a significant part of this recycle energy is
utilized in gasifying feed material. ,
The organic fraction of the refuse after conversion to a gas, exits the
converter and is processed in the gas cleaning system. The gas is cleaned by
water scrubbing followed by electrostatic precipitation. After cleaning, the
product gas is cooled in a condenser to remove moisture prior to use of the gas,
The converter, gas cleaning train, and combustor are shown in Figure 11. A
typical product gas analysis is shown in Table 4. In the demonstration plant
at South Charleston, West Virginia, the gas is flared.
TABLE 4. ANALYSIS OF TYPICAL PUROX SYSTEM PRODUCT GAS
Volume % - Dry Basis
Typical
Range
H2
CO
C2H2
C2H4
C2H6
Higher Hydrocarbons
C0
H2S
CH OH
o
Organic Vapors *
Water Content at 38 degrees C (100 degrees F), volume % = 6
23
38
5.0
0.7
2.1
0.3
0.6
0.5
0.4
0.5
27
1.5
0.05
0.1
0.2
21-32
29-42
4-7
0.2-1.5
1-3
0.1-0.5
0.2-1.3
0.1-0.8
0.1-0.6
0.25-1.4
20-34
0-2 .0
0.02-0.06
0.05-0.15
0.1-0.4
Heating Value, Dry Basis, MJ/Nm3 (Btu/S ft3)
Higher
Lower
14.6 (370)
13.6 (345)
11.8-15.4 (300-390)
11.0-14.2 (280-360)
*Higher alcohols, aldehydes, ketones, organic acids
20
-------
The scrubber water-particulate slurry is separated from the gas stream
and sent to a solid/liquid separation system. The system consists of a slurry
tank, a vacuum filter, a filtrate tank, and pumps and piping. Filtrate from the
system is recycled for reuse in the scrubber. The particulate is fed to the:,
hearth for disposal.
Typically approximately 0.28 Mg (0.31 ton) of waste water is produced
during the conversion of 1.0 Mg (1.1 .tons) of municipal refuse into product gas,
slag, and wastewater. The properties of this wastewater as-produced, and
after treatment in an oxygen activated sludge process are summarized in
Table 5. Dilution by a factor of 20:1 with fresh water or municipal sewage
is used during biological treatment. ซ
TABLE 5. ANALYSIS OF TYPICAL PUROX SYSTEM WASTEWATER
After Oxygen Activated
BOD
o
COD
Soluble Carbon
pH
As-Produced
52,000 ppm
77,000 ppm
22 ,000 ppm
3.7
Sludge Treatment*
195 ppm
381 ppm
99 ppm
7
*Assumes wastewater is then discharged into municipal sewer.
The operating rate of the converter is changed by varying the rate of
oxygen injection into the hearth. The hearth operates at a fuel-rich condition
to minimize oxygen consumption and maximize fuel gas production. Therefore
the rate at which refuse is. dried, pyrolyzed, combusted, and slagged is set by
the rate of oxygen which supplies the necessary energy for the drying, pyrolysis,
and slagging steps. To maintain the fuel-rich hearth condition, refuse is fed
into the converter periodically with the intent of maintaining as constant a bed
heighth as possible since this provides maximum heat utilization and hence
conversion efficiency. The slag in the hearth is maintained at a fluid condition
by the high temperature combustion of fuel with oxygen. The fluid slag drains
continuously from the tap into the quench tank. The basic control parameters
are setting the oxygen flowrate for a desired refuse processing rate and operating
the front end to maintain a full converter condition at the desired rate.
21
-------
SECTION 5
PUROX SYSTEM FACILITY EQUIPMENT DESCRIPTION
The PUROX System demonstration plant in South Charleston, West
Virginia, was built in 1974 to demonstrate oxygen pyrolysis of mixed munici-
pal refuse. It has been modified since original construction. The plant is
shown in its present configuration in Figure 12. The number following each
piece of equipment refers to Figure 12 .
The shredder (D) is a Heil (Tollemache) vertical shaft hammermill driven
by a 149 kw (200 horsepower) electric motor. It has a rated capacity of
13.6 Mg (15 tons) per hour. Because the PUROX System operates continuously,
the hammers must be resurfaced every few days and the interior wear surfaces
must be built up periodically. As a result of these requirements, the demon-
stration plant shredder requires 2 or 3 hours of maintenance each day.
The shredder is equipped with an overload protection circuit which
stops the shredder feed conveyor (E) when shredder motor current reaches a
predetermined value. This allows the hammermill to continue grinding without
increasing its load. When motor current drops below the set point, the
shredder feed conveyor restarts. The shredder is also equipped with an inter-
locked starting circuit which prevents starting the shredder unless the magnetic
separator, the shredder discharge conveyor (E), and the shredder motor cooling
fan are operating. This prevents (clogging of the shredder with discharging
material and overheating of the shredder drive . The shredder is automatically
stopped if any of these components malfunctions. The shredder is controlled
by the feed system operator who feeds in response to the converter demands
for refuse.
The magnetic separator (F) is an Eriez drum type magnetic separator.
It is suspended 0.25 m (10 inches) above the shredder discharge conveyor and
discharges to a belt conveyor (G) installed perpendicular to the shredder dis-
charge conveyor. The magnet is contained in a 1.2 m (four foot) drum
operating at 480 volts. The magnetic separator-and metals discharge con-
veyor (G) are controlled by the feed operator.
The feeder feed conveyor (l) is an apron type conveyor which transports
refuse from the shredder discharge conveyor to the refuse feeder. The refuse
22
-------
1
S ซ 2
If Si HH.
M * S? . 2 o ois
Iltlffflfl!
3 ฃ tt
ฉฉฉฉฉฉฉฉฉ
5 i
s I
? 5
ง 2 K I I
1 5 S at I'
'if I'll.
>s
.000ฉ ฉฉฉฉฉฉ
ฉ
tn ^
3 c
w 2
>, ^
S O
--I
0 ฃ
5
W '
s_ ro
K^ rr,
x S.
i!
CU O
-------
leveler is a belt conveyor installed above the feeder feed conveyor. The height
of the leveler above the feed conveyor can be adjusted between 0.03 and 0.25 m
(1 and 10 inches). The leveler maintains an even burden on the refuse feeder
by keeping the level of refuse on the feeder feed conveyor constant. The leveler
is controlled by the feed system operator, and starts whenever the feeder feed
conveyor is started.
The refuse feeder (2) has been designed by Union Carbide Corporation
to perform three functions. It compacts shredded refuse into a pellet, intro-
duces the pellet into the converter, and provides a gas seal for the converter.
The feeder contains nine principal parts; the refuse hopper, vane, two
cylinders, two rams, two restrictors and the hydraulic power package. The
refuse hopper receives refuse from the feeder feed conveyor. Attached to the
refuse hopper is a Drexelbrook capacitance type level detector which actuates
a feeder overload prevention system. When the level in the refuse hopper be-
comes too high, the feeder feed conveyor is stopped. This prevents refuse
from overflowing to the surrounding area. The vane directs incoming refuse to
one of two cylinders. The rams force the refuse in their cylinder through a
restrictor and into the converter. The restrictors open and close to maintain
the hydraulic pressure on the ram within a band which results in the formation
of pellets which maintain their integrity in the converter. The hydraulic power
unit consists of hydraulic pumps, valves and controls which automatically
operate the feeder in the correct sequence. The gas seal is formed by pellets
in the feeder which have not entered the converter and by a nitrogen purge
through the cylinders. The feeder can be operated remotely by the feed system
operator or locally at the feeder.
The refuse converter (3) is a vertical shaft furnace approximately 3 m
(10 feet) in diameter and 13.4 m (44 feet) high. The upper section of the con-
verter is a cylindrical carbon steel|shell lined with 0.10 m (4 inches) of re-
fractory. Pyrolysis gas exits from the top into the gas scrubber. The converter
is equipped with 9 nuclear sensing devices which detect the presence of
refuse at different levels. Ports for insertion of a poker to verify the refuse
level are also installed. The nuclear detector at the 3.7m (12 foot) level
actuates a low bed level alarm when no refuse is detected. A thermocouple is
installed at the top of the converter to measure the offgas temperature.
I
The lower section, or hearth, is a refractory lined cone. Fuel gas and
oxygen for combustion of carbonaceous material are introduced by water-cooled
nozzles called tuyeres. Slag forms a pool on the hearth floor and is continu-
ously tapped through an opening in;the hearth wall. Cooling water is supplied
to hearth components and to the hearth walls.
The quench tank (4A) is a water filled tank installed to provide a 1.9 m
(76 inches) deep water seal below the slag tap. It has two functions, cooling
24
-------
the molten slag and providing a pressure seal for the converter. Arecirculation
system replaces water in the slag duct with water from the main volume of the
tank to suppress steam formation and also to control tank level. The tank is
also provided with a temperature controller to maintain correct tank tempera-
ture. Slag is removed from the quench tank by a drag conveyor. A low pres-
sure alarm is installed on the recirculation pump discharge.
The slag conveyor (4B) is a drag type conveyor manufactured by Taunton
Engineering. It is equipped with a heavy duty chain and is driven by a 2 .2 kw
(3 horsepower) electric motor . The conveyor operates at a speed of 6.7 m
(22 feet) per minute. It is equipped with a protective trip which stops the con-
veyor and sounds an alarm if motor current exceeds a predetermined level. The
trip protects the conveyor should it become jammed by slag. The conveyor is
reversible to facilitate clearing jams.
The product gas scrubber (5) is a water spray scrubber consisting of
spray nozzles installed in the product gas discharge line. The nozzles are
designed to permit removal of an individual nozzle during operation. The
nominal scrubber water flow is 6.31 x 10"^ m^/sec (IQQ gallons per minute).
The purpose of the product gas scrubber is two-fold. It removes
90-95 percent of the particulate from the product gas. It also cools the product
gas from its exit conditions to near its wet bulb temperature.
The offgas line terminates in a gas-liquid separator (6) 1.83 m (6 feet)
high and 1.4m (54 inches) in diameter. Liquid is withdrawn from the bottom
through two 0.25 m (10 inch) lines. Gas exits horizontally into a 0.61 m
(24 inch) line and goes to the electrostatic precipitator inlet.
The electrostatic precipitator (ESP) (7) is a Research Cottrell tube type
precipitator. It is a standard unit containing 60 tubes which provide approxi-
mately 10.5 square meters (1,131 square feet) of collection area. Nominal
gas flow rate is 2.95 cubic meters per second (6,250 cubic feet per minute) at
design conditions.
The ESP is used to remove any fine particulate that was not removed
during the scrubbing process and to collect the "fine oil mist that is produced
as the gas is cooled to its wet bulb temperature.
The ESP is supplied with 450 volt alternating current (AC) which is
stepped up and rectified to supply direct current to the precipitator discharge
electrodes. An automatic control system adjusts the AC input to maintain
optimum precipitation conditions. The control system is also equipped with a
low voltage alarm and an overload protection circuit. The low voltage alarm
actuates if primary voltage is below 130 volt AC for a set period of time. The
overload protection circuit protects the rectifier by de-energizing the high
25
-------
voltage transformer if a portion of jthe circuit is overloaded. The ESP shell is
equipped with a capacitance type joil level sensor which actuates an ESP high
level alarm. A nuclear level deteptor is also installed.
i
I
Six high pressure steam nozzles are installed in the ESP base directed
at the pumpout trough. They are used to heat the accumulated tar and oil for
pumping. A Moyno progressing cavity pump is used to remove the oil. Steam
is also supplied to the ESP shell for heating. A nozzle installed on the top of
the ESP is used to spray cleaning solvent onto the tubes.
The product gas condenser (8) is a vertically mounted shell and tube
heat exchanger. It cools the gas to approximately 27 degrees C (80 degrees F)
to condense wastewater following gas cleaning in the ESP. Product gas enters
through a connection at the top of the condenser and exits from the side of the
condenser. Condensate is withdrawn from the bottom of the condenser. Cooling
water is applied to the shell side* The condenser is equipped with a capaci-
tance type level sensor which actuates an alarm and operates a solenoid valve
and pump through a timer. The condensate is normally returned to the scrubber
water slurry tank.
The combustor (9) is a National Air Oil (NAO) natural draft combustion
chamber. It is equipped with 12 NAO jet mix vortex burners supplied from a
common 0.3m (12 inch) header. The burners are supported in a tall chamber.
The walls of the 3.4 m by 1.2 m by 6.1 m (11 foot by 4 foot by 20 foot)
chamber are lined with O.lm (4 inches) of refractory, and are supported by
legs at each corner. A pilot burner, supplied with 0.1 MPa (15 p.s.i.g.)
natural gas, is located at each corner. The entire unit is surrounded by an
acoustical fence. The combustor is designed for a nominal flow of.l .0 cubic
meters per second (2,200 cubic feet per minute) at standard conditions of
15 degrees C (60 degrees F) and 0.1 MPa (one atmosphere) pressure.
The slurry and filtrate tanks (10A) are formed by a tank to which
legs and a baffle dividing the tank into halves have been added. The capacity
of each half is 9.5 cubic meters (2,500 gallons). A 2 .2 kw (3 horsepower)
"Lightnin" mixer is installed on the slurry side of the tank.
The vacuum filter (10B) As a rotary drum type vacuum filter supplied by
Ametek Engineers. The drum is 1.5 m. (5 feet) in diameter and 2.1m. (7 feet)
long. Polypropylene monofilament filter cloth is used as the filter medium.
The vacuum pumps (IOC) are Nash L-7 and CL-702 models and are
operated in parallel. Filtrate is collected in two vacuum receivers (10D)
0.61 m (24 inches) in diameter and 1.5 m (60 inches) tall, and then pumped
to the filtrate tank. One vacuum receiver was purchased from Ametek and the
second was fabricated locally. The filtrate pump is a self-rpriming Durco pump
rated at 0.011 cubic meters per second (180 gallons per minute) at a 15.2 m
26
-------
(50 foot) head, and it is driven by a 3.7 kw (5 horsepower) motor. The slurry
tank, vacuum filter, and vacuum receivers are equipped with level control
systems. Level is sensed in the tanks by Drexelbrook level detectors.
The particulate conveyors (11) are 0.15 m (6 inch) diameter screw con-
veyors manufactured by Jeffrey Manufacturing. The screws have 0.15m (6 inch)
pitch and are driven by 3.7 kw (5 horsepower) motor. The total length is
20.7 m (68 feet). The feeder is a pneumatically operated ram feeder.
27
-------
SECTION 6
EQUIPMENT MODIFICATIONS FOR
CODISPOSAL OF SEWAGE SLUDGE AND REFUSE
Processing sludge filter cake required the installation of additional
materials handling equipment at the facility. The new equipment is shown in
Figure ]2 . The major additions are a dumping station, a small front loader,
a metering hopper and a rubber belt conveyor.
Sludge cake is hauled to the site in dump trucks and is weighed at the
plant truck scales prior to dumping. The sludge is then dumped on a concrete
pad adjacent to the east wall of the shredder building (Figure 13). The pad is
surrounded on three sides by a 1.5 m (5 foot) wall and has a roof. A small front
loader is used to scoop up sludge cake for loading into the metering hopper
(Figure 14). The metering hopper delivers a metered flow of sludge onto the
sludge conveyor. Sludge is discharged from the sludge conveyor onto the
feeder feed conveyor where it is a
-------
-
Figure 13. Sludge cake at dumping station.
29
-------
-------
CD
Cn
fe
a
a
Cn
C
-i I
b
1-1
cp
0)
i-i
31
-------
Figure 16. Sludge cake on troughing conveyor.
32
-------
Figure 17. Metering hopper and troughing conveyor.
33
-------
SECTION 7
SLUDGE-REFUSE CO DISPOSAL TEST PLAN
The test program was planned to consist of a series of eight (8) test
periods with a total duration of 64 days of 24 hour per day operation (see Table
6). Each test was to consist of four (4) days of codisposal followed by three
days of refuse-only operation at reduced rate to flush the system of residues
from the preceeding sludge test and to accumulate material for the following
test. The intention was to begin the test series at low levels of sludge cake-
to-refuse and to increase the ratio in successive tests. In addition, a test
operating at the rated capacity of the unit was desired to determine performance
at capacity. At the time the test plan was prepared, it was recognized that
operation at capacity for an extended test would be difficult for two reasons:
the available supply of refuse in the Charleston area would not permit operation
above 136 Mg (150 tons) per day for more than one day without forcing a shut-
down of the unit over the weekend when refuse was unavailable and the single-
train design of the front end; i.e. the shredder and feeding system provided no
redundancy to support sustained capacity operation. For these reasons, the
operation was judged to be a high risk operation and hence scheduled as the
last test.
The Charleston area was unable to supply sufficient quantities of
sludge since much of the available supply was digested. However, an ade-
quate supply of primary filter cake was located in Huntington, West Virginia,
a distance of 60 miles from Charleston. This material was raw, primary
sludge filter cake with a total solids content of 24 percent and a volatile
solids content of about 45 percent. The volatile solids content was rather low
in this sludge due to the mixing of industrial wastes from a metal alloying
company with municipal waste. Sludge was available from Huntington
on an as-needed basis.
Small quantities of raw primary filter cake were available from the
South Charleston Waste Treatment Plant. This cake had a solids content of
about 30 percent and was used in preliminary equipment tests and shakedown
operation of the unit. Unfortunately, the material was not available in enough
quantity to carry out a complete test phase.
34
-------
*
12
a
OH
EH
CO
W
ฃH
^
<1
z
CO
1 1
p
0
O
w
O
Q
ID
1-1
CO
_
> r2 0
!> co ec;
* v
c
^ B
(0 ""'
O Ljl
r- . ^>
0 O
0 Ti '.
< 13 D>
CO ^
, .
G
0 O
CQ .ij.
iฃ$ .
0 vS1
oi &
CO
"""*
0 O
CO
0
CO EH
0 0
cn O
T3 fa
JH O
CO CO
W '
0 O
EH S
CM
O
0
O
0
^-^
o
11
|s^
en
^~.
CO
^_^
o
0
1 1
11
CD
0
CO
1
a
CO
to
O
CO
11
*>f
o
o
1 1
0
0
[^
t 1
11
CD
O
1 ]
j^
11
LO
r 1
^^
CO
O
11
11
CD
vf
CM
!? ง"
S ^ 6
a (S H.
i
Cn
_g
4->
G c
ifl 2
CM
CM
O
0
O
! 1
O
LO
11
co
ii
! (
11
r-1
O
11
""?
r 1
1 1
CO
O
11
CD
CM
Primary &
secondary
m
w
4-1
OH"
CO
^
O
o
o
CM
ซ
o
LO
CM
i 1
CO
1 1
1 1
I 1
CD
i 1
^_^
O
1 1
^
CD
0
CM
Primary &
secondary
t
CO
4-1
'1-4
OH
*r
CD
0
o
o
CO
o
LO
CM
I 1
CO
I 1
1 1
~
CM_
CD
CM
to
c^
CO
0
CM
Primary &
secondary
m
CO
4-1
.1-4
OH
LO
CO
0
o
0
o
LO
CM
i 1
co
1 1
r-4
to
co
CO
co
CD
co
o
CO
o
CM
Primary &
secondary
t
CO
OH"
to
o
i i
o
0
LO
o
LO
CM
I 1
co
1 1
II
CM
2!-
CO
CO
CO
CO
LO
c^
o
Primary &
secondary
CO
4-1
-I 1
OH
t-
CM
0
O
O
0
CO
II
CM-
CO
CD
i 1
c*?
11
CM
il
'o'
CD
ฃi
il
co
r 1
0
CO
(0
$ _S
(0 s-,
OH &
CO
(0
si
O
CO
CO
A*
0
0
^
0
4-J
4
0
rO
0
O
^"1
(0
'ฃ
c
0
o
G
CM
1 1
1 1
i4
4->
G
io
T3
G
O
2
c
o
o
G
CM
a
o
_j
0
:ion conduc
fO
0
a
o
r 1
(0
CQ
O
a
CO
-r-t
O
O
*
35
-------
The test plan placed a heavy emphasis upon operation with secondary
sludge. Unfortunately, the nearesj: substantial supply of secondary sludge cake
was the Allegheny County Sanitary Authority Waste Treatment Plant in Pittsburgh,
Pennsylvania, which produced a mixed sludge filter cake that was 60 percent
raw primary and 40 percent secondary sludge. The hauling distance from Pitts-
burgh to Charleston complicated the logistics and increased the cost of mixed
sludge testing.
The early operation with primary sludge went well in terms of both
equipment and system performance. However, operation with mixed primary
and secondary sludge did not proceed as well. The equipment used with these
mixtures had not been mechanically verified prior to startup due to unavailability
of the sludge. Equipment problems prevented processing of the high moisture
mixed primary and secondary sludge over the full range of blends projected in
the test plan.
In particular, the peculiar properties of high moisture content (75 per-
cent) mixed sludge filter cake led to equipment and materials handling problems
in the PUROX System feeding equipment. These handling problems were equip-
ment specific related to the inadequate blending of sludge and refuse prior to
pellet forming. With appropriate equipment selection and design, based upon
the operating experience gained with this material, effective processing of high
moisture sludge blended with refuse should be possible.
In light of the factors discussed above, the test program as executed
differed from the original plan. The program as shown in Table 7 consisted of
a set of seven test periods, six were with codisposal of sewage sludge and
refuse and one was for collection of baseline, refuse only data. Tlie codisposal
tests included two with mixed sludge and four with raw primary sludge. The
range of parameters studied included tests at wet sludge to refuse ratios from
0.06 - 0.23 corresponding to dry sludge to refuse ratios from 0.03 - 0.074;
throughput rates from 59 to 106 Mg (65 to 117 tons) per day; and total moisture
content of the refuse/sludge mixture up to 35 percent by weight.
In comparing the test program as executed with the original test plan,
the ratios of sludge dry solids-to-refuse above 0.074 were not achieved. The
use of mixed secondary sludge was: limited to a sludge dry solids-to-refuse
ratio of 0.049 due to the equipment problem stated previously. The capacity
test was not conducted due to a desire to obtain as much operating data at high
sludge ratios as possible and of course the limited refuse supply.
The impact of these modifications in the tests upon the basic goals of
the test program is not serious. Operation was achieved at substantial sludge
cake-to-refuse ratios. Detailed operating data was collected for the test periods,
Useful operational experience was obtained in handling and processing sludge
36
-------
CO
w
s
w
CO
H
CO
w
^
CO
/~\
GE CODISPC
Q
1=3
CO
J^
H
CQ
0
CQ
S >"
0 A
Pi f-i
o>
ly-J
J3 0)
CO >
r i
4->
O
T3
\
o -ft
4-j CD
*""*' ^
^ -
7X Q)
)i ง
.2 ^
pi ^
T3
0
0
P-, 0
CQ
0
Pi
CQ
'O
i 1
o
CO
55
0
ฃ
j3
CO
0
^
0
o
o
CO
*
4-1
0
H
CD
CO
o
*
0
CD
CO
0
e
o
,. ,.
CD
o
S
CM
CM
CO
r s
co
LO
t 1
LO
o
LO
oo
I 1
CO
CO
co
Pi
r-i
to
co
0
0
oo
CD
0
o
,.
r i
i-H
2-
t^
CM
oo
r s
1 1
00
^
!>-
o
CO
00
CO
LO
o-
^
ฃ
PH
ง
CM
CD
^p
o
ซ
o
CD
^
t 1
O
* X
CO
0
IN.
c^
1 1
o
o
[V.
"5
n
0
r-l
t J
o
CD
co
CD
r i
CD
r 1
CD
i 1
CM
CO
CO
LO
co
0
*
0
LO
CO
I 1
o
^1^
CD
1^,
22-
LO
CO
00
CD
1 1
1 1
LO
o
HH
o
CD
CO
CD
CO
LO
CM
o
CM
CO
1
^J,
1
1
f ^
0
1 1
CO
LO
CO
[N
1
o
I 1
CO
LO
CO
1
1
LO
"vf
CM
o
o
.1
CO
o
o
^ ^
ฃ^
"^f
CD
t^
CO
LO
^ ^
t-^
co
^
CO
0
r 1
CD
CO
LO
LO
co
CM
CD
CM
P-.
ง
CD
vf
IN.
o
o
vt*
CO
CM
0
^^
CM
^^
I-H
I-l
CO
CD
o
r 1
^
CM
CM
CM_
C-H
0
CM
o
LO
-CD-
CM
CD
CO
CM
CO
CM
CM
PH
r-.
(0
-r-l
.s
Cn
CQ
r
C
Q
tj!
G
-r 1
G
3
ffi
s
o
41
0
:3
i i
CQ
&
1
&
<
(0
Pi
1
P-.
W
f]\ .
VJ
4-1
ro
S
-i l
X
Q
fO
C
"*"'
0
Di
-a
r 1
CO
Ss,
rt
fO
s
I
5
-I-l
?
Tl
0
X
-r-l
a
fO
-r-l
G
(0
h, Pennsylv
econdary
Dl W
3 c?ง
X! o
CQ ^-H
4-1
-r-l
p^ >,
s s
o _a
0 ^
"B ฃ
^ o
i-i co
(0 i^
O O
O j3
1
CO
lง
oT
37
-------
cake through the PUROX System. Equipment limitations were identified and
design changes and concepts have bpeen identified which should provide improved
performance for the codisposal process. A modified process (which will be dis-
cussed in more detail in Section 13)'has been proposed in large part as a result
of the operating experience gained during this program.
Equipment malfunctions with the previously untested participate recycle
system prevented recycle of particulate material throughout the test period.
However, enough once-through scrubber water was sampled and analyzed to
determine its composition and to determine its treatability. These analyses
were used in computing the material balances.
The material and heat balances presented later are computed from
direct measurement of quantities wh'enever possible. Projected results for
system operation under standard operating conditions are also included. The
disposition of trace metals under steady-state conditions with particulate
recycle has been projected as follows. The results of trace metal balances
indicate that the bulk of the metals [exit in the slag. (See Section 12 for
details). The scrubbing system is effective in removing the metals from
the gas, and metals which do not return to the system in the particulate
recycle loop will enter the wastewater stream where they would be concentrated
in a waste sludge which could return to the PUROX System for eventual disposal
in the slag or be disposed of by an alternate technique. Subsequent tests
with refuse-only confirmed these projections.
The balances report particulate material as a product although it is
normally recycled as a liquid (oil) and a solid (char) to the converter. In the
material balances where the results projected for particulate recycle are
presented, these streams have been converted into offgas and slag based on a
model of the gasification and slagging process in the hearth of the converter
which is discussed in Section 11. The following sections present a detailed
discussion of the data collection and analysis, and of each test period.
38
-------
.SECTION 8
DATA COLLECTION AND SAMPLE ANALYSIS
Process rate data was obtained for the major process streams during
each test period. Operating personnel measured the refuse, sludge, oxygen,
fuel gas, metal, and slag rates as part of their normal operating procedures.
The rate of the product gas stream was calculated from gas analysis results
using helium as a tracer gas.
Each bucket of refuse entering the system was weighed by the front
loader operator prior to dumping. The net weight of each load was recorded on
daily tally sheets which were the source of refuse rate data. For each sample
period, the cumulative weight of refuse processed was plotted as a function of
time. A best fit straight line was drawn, and the slope of the line was
computed. The slope was equal to, the refuse rate of feed. This method of
handling the data was used because refuse was fed intermittently in response
to converter requirements and rates based on daily totals were erroneous. In
addition, no method existed for measuring the converter refuse inventory.
These two operating characteristics caused substantial errors when daily totals
were used. Use of the graphical method minimized the chance of error, and
also minimize the effect of feed system stoppages.
Each sludge load was weighed before it was added to the sludge hopper.
The net weight of each load was recorded on tally sheets and totaled for each
day. The same technique used to determine the refuse rate was used to obtain
the sludge rate during each period.
Oxygen supplied to the converter was measured by several methods.
An orifice plate provided a total flow signal to an oxygen flow integrator.
Rotameters also provided flow indication for the hearth components supplied
with oxygen. When oxygen flowrate was changed, the flows were calculated
arid recorded using both methods of indication. Oxygen consumed as a
function of time was determined by reviewing all available oxygen flow
measurements and using the best available data in the following order of
preference: (l) oxygen totalizer, (2) total flow rotameter integrated over
time, (3) individual component rotameters integrated over time and summed.
The resulting data was plotted using the method used to determine refuse and
sludge rate. The oxygen rate was equal to the slope of the best straight line.
39
-------
Rotameters measured the flow of fuel gas to the hearth and the flow was
integrated to determine quantity of fuel gas used. The plot of gas used versus
time resulted in the fuel gas rate.
Slag fell from the slag conveyor into the slag dumpster which rested on
a scale. The weight of slag in the dumpster was recorded prior to dumping.
Slag production rate was determined from this data by applying the same method
used for other streams. The weight of slag produced was plotted versus time
and a straight line drawn to fit the data. The slope of the line was equal to the
slag production rate.
Ferrous metal removed from the shredded refuse was conveyed to a
dumpster for recovered metal. The dumpsters were weighed and the weight of
metal was recorded. The ferrous metal recovery rate was determined using the
same plotting method used for refuse, sludge, arid slag.
Product gas flow rate per unit of feed was determined by using helium
tracer gas. When gas samples were taken for component analysis a metered
helium flow was injected into the hearth. The gas flow rate was calculated
from the helium concentration in the product gas. Raw sample results were
converted to a dry, nitrogen-free basis before calculating the gas flow rate.
These corrections were made due to nitrogen purges of converter components
near the gas sample port which caused a high concentration of nitrogen in the
samples. Samples with helium concentration less than one volume percent
were not used for flow calculations. The gas production in cubic meters at
15 degrees C (cubic feet at 60 degrees F) and one atmosphere per Mg (ton) of
material fed was determined for each gas sample using the following method.
The gas production rate from the helium trace was divided by the total feed
rate at the time of sampling (generally determined from the measured oxygen
rate and the oxygen consumption ratio - oxygen/feed for that period).
Additional gas flow rate data was obtained from pitot tube measurements
of gas velocity taken in conjunction with gas particulate samples. These
measurements were taken downstream of the product gas condenser with a
S-type pitot tube. Static and velocity heads were recorded every ten minutes
during a two hour sample period, and probe location was changed every ten
minutes to traverse the gas pipeline. Assuming plug flow and correcting for
natural gas added to purge the ESP insulator compartments the volumetric flow
rate was calculated from velocity measurements. Gas velocities typically
varied 15 to 20 percent over any ten minute period. Average gas flow rates
were calculated and gas production rates in cubic meters at 15 degrees C
(cubic feet at 60 degree F) and one atmosphere per Mg (ton) of feed were
determined using the same procedure described with the helium trace data.
40
-------
Pitot tube measurements do include water and nitrogen from converter purges
neither of which are reflected in the helium trace measurements. No gas
analysis was available with the pitot tube data to correct for nitrogen and
water.
The product gas was sampled for particulate material downstream of the
gas condenser. One particulate sample was taken during each test period
using EPA method 5, "Determination of Particulate Emissions from Stationary
Sources"*. A Research Appliance Company "Stacksamplr" was used for
sampling. The sample was analyzed immediately after completing the
sampling process.
Char production was measured during two periods of recycle operation
of scrubber water in the gas cleaning system. Measurements were taken at
four different converter rates with refuse only between sludge codisposal tests.
Char was collected in drums and the time to fill the drums was recorded. The
drums were weighed and the weight of char recorded. Char moisture content was
determined analytically. Using this data, the production rate of dry char in
kilograms (pounds) per hour was determined. A least-square curve-fit of the
char production data as a function converter rate resulted in a linear equation
predicting char rate. The equation predicted no char production at an operating
rate of 51.7 Mg/d (57 tons per day (TPD)) which was consistent with previous
South Charleston experience. This equation was used to estimate the average
char production during each test period.
Wastewater flowrate was calculated by difference between the entering
streams and the established exit streams. Due to mechanical problems,
operation of the gas cleaning s'ystem in a closed cycle was impossible during
most of the test. Scrubber operation with once-through water resulted in
several effluent streams and the net wastewater production was not
measureable.
Samples of the major streams were taken during each test period.
Sampling was performed by a data collection team. Analysis of the samples
was performed off site by several laboratories. The analytical techniques used
are described in Appendix A.
Refuse samples were obtained during three test periods. Grab samples
were taken in buckets at the end of the shredder discharge conveyor, and
represented the shredded, magnetically separated refuse fed to the converter.
Between 2 and 5 samples were taken during each test period and after com-
pletion of data collection, the samples were composited and reduced to a
single 0.02 cubic meter (5 gallon) sample. The composite sample was
delivered to Standard Laboratories, Inc. of Charleston, West Virginia for
analysis.
*(2) Federal Register. Vol. 36, No. 247, December 23, 1971.
41
-------
Analyses of the refuse for moisture, ultimate analysis, heating value
and trace elements were made. ;
A 0.004 cubic meter (one gallon) sample of sludge was taken from each
truckload of sludge delivered. The sludge samples taken during each test
period were composited and reduced to a 0.004 cubic meter (one gallon) sample.
The sludge samples were also delivered to Standard Laboratories for analysis.
Analtyical results include percent!solids, percent volatile solids, ultimate
analysis, heating value and trace elements.
Product gas samples taken from the top of the converter were obtained
twice daily. A typical gas sample consisted of three gas sample tubes filled
sequentially. Samples were obtained using a probe inserted into the converter.
The gas passed through a glass wool filter to remove particulates prior to
entering the sample tubes. A metered helium flow was injected into the con-
verter hearth to act as a tracer gas for calculation of the gas flow rate.
Gas samples were taken using the following procedure. Sample tubes
were purged with product gas for three minutes. After purging,the helium flow
was set and run for three minutes to reach steady state. Three consecutive
gas samples were then taken. Each sample tube was purged with product gas
for two minutes prior to isolating the sample. After completion of sampling,
the helium flow was stopped, the probe removed, and the gas samples were
taken immediately to the laboratory for analysis. The samples were repre-
sentative of the product gas downstream of the condenser since particulate was
removed by the glass wool filter and water was condensed in the sample tubes.
Gas samples were analyzed by Union Carbide Corporation, South Charleston
Laboratories by emission mass spectroscopy.
Slag samples were taken in 0.004 cubic meter (one gallon) cans placed
under the slag conveyor discharge. Up to 4 samples were taken on the final
day of each sampling period. After completion of sampling, a composite slag
sample was made. One quart of the slag composite was withdrawn and shipped
to Union Carbide Corporation, Tarrytown Laboratories. The remaining slag was
retained for leachate testing. Analysis of the slag was performed using a
variety of methods including chemical, atomic absorption, x-ray fluorescence
and induction coupled plasma techniques. Analyses were made for carbon,
major constituents, and trace metals.
Wastewater samples were taken from the slurry tank. Due to mech-
anical problems, the gas cleaning system was operated on once-through
scrubber water during the tests. Because of this mode of operation, the water
samples were very dilute in trace metals and the resultant concentrations are
often below detection limits. During subsequent periods of refuse only
operation with particulate recycle; samples of filtrate were taken which were
more representative of the wastewater.
42
-------
Two wastewater samples were obtained during each day of the sample
period. The samples were composited and the composite was delivered to
Union Carbide Corporation, South Charleston Laboratories for analysis.
Analyses were performed to determine grease and oil, chemical oxygen demand,
biological oxygen demand, total organic carbon, suspended solids, and trace
elements.
One set of samples were taken of both the oxygen and the fuel gas
supplied to the converter. Samples were taken from the tuyere supply lines for
each stream. They were analyzed by Union Carbide Corporation, South
Charleston Laboratories using emission mass spectroscopy.
The particulate collected from the product gas during three test periods
was shipped to Union Carbide Corporation, Tarrytown Laboratories for analysis.
The induction coupled plasma spectroscopy was used to analyze the particulate
for metals content.
t
Samples of slag leachate were obtained by the data collection team and
were delivered to Union Carbide Corporation, South Charleston and Tarrytown
Laboratories for analysis. The samples were prepared by placing 375 grams of
composited, as-sampled slag in a beaker with 1.5 x 10"^ cubic meters of.
distilled water. The beaker was gently agitated by air bubbles for at least 48
hours. The leachate and a distilled water blank were sent to Union Carbide
Corporation, South Charleston Laboratories for analysis. Analysis was for
biological oxygen demand, (BOD5), chemical oxygen demand, (COD), total
organic carbon, (TOC) , pH, chloride ion, and trace metals. Several leachate
samples were also sent to Union Carbide Corporation, Tarrytown Laboratories
for trace metal analysis .
43
-------
SECTION 9
SUMMARY OF TEST PERIOD RESULTS
In this section summaries and comparisons for the various test periods
are presented. Details of the calculation techniques and individual com-
ponent results are collected in App'endices B - H. Some introductory comments
on the calculation approach are also provided for background.
The unreliable operation of the particulate recycle system made
measurement of wastewater production impossible during the test periods.
This required computation of wastewater by difference using other measured
or assigned quantities as detailed in Appendix B and summarized in Table 8.
Essentially the overall material balance consists of four entering streams
(refuse/ sludge, fuel gas, and oxygen) and four exiting streams (ferrous metal,
product gas, slag, and wastewater). The unreliable operation of the particulate
recycle system resulted in two additional exit streams that are internal recycle
streams under normal operation ' char from the scrubber and oil from the ESP.
Operational data was collected to determine refuse, sludge, fuel gas, oxygen,
ferrous metal, product gas, and slag rates. .Char rates were measured peri-
odically. Oil rates were not measured, but were estimated from previous ex-
perimental work since their small magnitude would have little effect on the
material balances. Wastewater generation was computed by difference in the
overall balance.
TABLE 8. OVERALL MASS BALANCE ASSUMPTIONS
(1) Product gas saturated with water at 21ฐC (70ฐF).
(2) Nitrogen and argon entering with oxygen leave in product gas.
(3) Char production computed from rate equation based on
measurements over a range of operating rates.
C= -526.8 + 9.132 x (TPD)
C = Ib/hr of dry char
TPD = operating rate in tons per day of feed
(4) ESP oil rate is 2% of operating rate
44
-------
Component balances using measured data were attempted,but were found
to be completely unsatisfactory. Balances could not be achieved on either the
inorganic components or the carbon. It was apparent that the source of the problem
was the refuse sampling prior to analysis. Clearly, a heterogeneous material
such as refuse is extremely difficult to sample representatively. The test plan
called for sampling on a limited basis only. The results of the component balances
using this data indicated that the samples were not representative. To proceed
with component balances required an additional set of assumptions (see Table B-3)
which are detailed along with the component balances in Appendices B-H).
Table 9 presents a summary of the overall mass balance results for the
seven test periods. Recall period 5 is a baseline, refuse only period and
should be used for comparison. In general these results show little variation
between operating in the codisposal mode and operating with refuse only.
Gas composition for each of the test periods are compared in Table 10.
These compositions are all well within the range reported in Table 4 for refuse-
only and display no significant departures from refuse-only compositions.
Overall heat balances for each test were computed to evaluate the
efficiency of conversion of incoming materials into useful product gas.
Obviously, the uncertainty in the refuse composition is also reflected as an
uncertainty in the heating value of the incoming refuse. When the heating
values measured for the actual refuse samples analyzed are used, a significant
portion of the incoming energy is unaccounted for in the output streams. In
order to obtain a reasonable measure of conversion efficiency, the assumptions
detailed in Table 11 were used to compute conversion efficiencies based on
total energy leaving the system. Table 12 presents a summary of the overall
heat balances for the various test periods.
Table 13 summarizes the slag compositions obtained during the various
test periods. The basic composition of the slag appears to remain fairly
uniform within the previously noted range in Table 3. The trace components
will be discussed in more detail in Section 12; however, it is interesting to
note the substantial increase in nickel in periods 1, 2, 6, and 7 when raw
primary sludge was used. This is believed to be related to the presence of
industrial nickel alloying waste in the sludge from Huntington, West Virginia.
Tables 9, 10, 12, and 13 summarize the codisposal process. Com-
parisions between refuse-only (test 5), raw primary sludge (tests 1, 2, 6, and
7), and mixed sludge (tests 3 and 4) can be made. In general, the material
balances, slag compositions, and gas compositions are quite similar in all
three instances. The variation in conversion efficiency to gas shows, in one
case with mixed sludge, an indication of less efficient operation. This poor
efficiency suggests that inefficient heat transfer in the converter (channeling)
was occurring. (See Section 10 for more detail) .
45
-------
to to co
CO CO i-t
Cn GO CO
1
SI
t-t r-C C-.
co co co
^r "ool
fl
i-H O> il
O r-H CO |
rt in
-<
O> CO CO
co en co
CtJ
c ^j
O ฃ
o
21 S
46
-------
TABLE 10. SUMMARY OF PRODUCT GAS ANALYSES
Com- ;
ponent
CO
C02
CH4
C2
C0
2
c
2
c
3
c
3
C.
4
Cr
5
C_
6
H2
H
4
H
6
H
6
H
8
*
+
H
6
HHC*
H2
S
CHJDH
O0
2
Ar
*
6
C. is
Test. 1
Vol. %
31.25
36.24
23.52
4.46
0.25
1.32
0.33
0.27
0.25
0.47
6.35
0.17
0.32
0.00
0.14
0.05
0.62
Test 2 .
Vol. %
29.62
36.80
23.85
5.02
0.41
1.68
0.32
0.23
0.18
0.34
0.27
0.25
0.29
0.04
0.08
0.03
0.59
Mole Percents
Test 3 Test 4 Test 5
Vol. % Vol. % Vol. %
30
31
28
4
0
0
0
0
0
0
0
0
0
0
0
0
,6
a composite of butanes,
.32
.56
.86
.82
.48
.68
.23
.20
.02
.23
.17
.22
.12
.01
.12
.24
.74
25.30
36.72
25.01
6.48
0.60
2.26
0.46
0.39
0.21
0.44
0.31
0.26
0.23
0.02
0.11
0.26
0..95
28
37
23
6
0
1
0
0
0
0
0
0
0
0
0
0
0
butenes, butadiene,
.14
.28
.07
.21
.46
.90
.44
.32
.23
.40
.29
.24
.26
.02
.11
.03
.59
vinyl
Test 6
Vol. %
29.
33.
25.
5.
0.
1.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
64
74
58
89
52
91
34
27
12
28
22
26
17
02
04
04
61
acetylene
Test 7
Vol. %
26.81
34.19
26.85
6.12
0.72
2.19
0.40
0.28
0.19
0.40
0.36
0.34
0.37
0.03
0.09
0.04
0.63
and
diacetylene
+
*
C5 is
HHC
a composite of pentanes , pentenes, isoprene
are higher
hydrocarbons
C6 and
above
, and
cyclopenetadiene
excluding benzene
47
-------
TABLE 11. OVERALL HEAT BALANCE ASSUMPTIONS
1. Refuse heating value (excluding heat from metal oxidation) is the
average of measurements for test program 18.21 MJ/kg (7,831 Btu/lb)
on dry basis. ;
2. Refuse moisture content taken from component mass balance.
3. -Sludge higher heating value measured for representative sample from,
test period.
3 3
4. Fuel gas higher heating value is 46.78 MJ/Nm (1,188 Btu/Sft ) based
on measured gas analysis.
5. Slag higher heating value is zero.
6. Slag exits hearth at 1,650ฐC (3,000ฐF).
7. Slag specific heat capacity is 1.21 kJ/kg-ฐG (0.29 Btu/lb--ฐF).
8. Product gas higher heating value is the mean of all samples taken
during each period.
9. Char higher heating value is 21.76 MJ/kg (9 ,355 Btu/lb) on a dry basis.
10. ESP oil higher heating value is 32.5 MJ/kg (14,000 Btu/lb).
fi '
11. Hearth cooling loss is 1.4 GJ/hr (1.3x10 Btu/hr) based on
measurements.
12. Converter shell heat loss is 0.35 GJ/hr (3.3 x 105 Btu/hr) based on
calculations.
13. Heat lost in condensation during gas cleaning is 2.53 MJ/kg of waste-
water (1,086 Btu/lb).
48
-------
o
'!
o
u.
O
i
O
1Z CO
c
CM CM
^. o
^ tO
cn
CO O
CM m
cr
r-t
CO r-i
f- 0
CO T-H
iฐ- *_,
to (^
CO O
t-l r-t
.,
3
U)
CO
t*-
rH
r-<
ซ
L^
CO
to
0
CO
s
(0
CM
r-
CO
to
o
CM
cn
"co
^p
r*ซ
cn
m
CO
CM
o
r-t
-a-
^r
CM
r^
CO
^3"
to
t^
T
CD
S
CO
CO
(O
o
10
ft
0
cn
CO
CO
V
o
ฃ
13
I
1
1
1
1
1
'
.
1
1
1
,
Ferrous Metal
"in
^r
0
to
CO
r*ป
CO
10
to
0
i-!
3-
CM
CO
CO
^r
i-ซ
r-t
CO
T
to
r-.
o
to
to
rป*
o
LO
in
in
CO
to
r^
r>-
cn
CO
to
cn
*r
S
to
m
^r -
CO
*r
m
U)
CO
t>.
v
VO
cn
cn
*r
Product Gas
CO
CO
CO
C-*
in
CO
CO
CO
CO
cn
r-i
in
r-t
CO
CM
CO
CO
"?
CM
CO
CM
tp
CO
cn
CO
CM
CM
in
CM
o"
to
CO
o
CO
CO
CO
r*
CO
*r
cn
CO
O>
c
CO
^
uo
CM
i-i
CO
OJ
CO
r^.
CO
t*v
cn
.^r
CM
in
CO
tn-
in
,^r
to
CO
CO
o
cn
o
CO
^r
T
in
V
^
t^
o
in
r>.
"S
tn
t>*
o
\n
r*
V,
0
JC
O
.CM
CO
r^
in
CM
CO
CM
T
^r
to
(O
^r
v
^
tn
^
r^
m
o"
CO
to
t*.
tv.
o
tn
CO
CO
in
CM
to
to
^r
to
CM
r~i
to
(O
T
to
C2
O
o^
o
r*.
cn
CO
C-v
^
to
CO
*aซ
CO
CO
^
to
CO
^p
CO
CO
CM
CO
in
to
in
CO
^"
T
CO
r-
^*
o
CM
^*
CO
*r
^r
S
O)
*T
CO
*r
T
Wastewater
r*. to
CM *a*
in to
rH 0
cn CM
1-4 CO
to og
r-i r-t
in t^.
r^. in
cn CM
to
CO (O
CM Q
o to
rH tD
rH 0)
o in
o m
r-4 C>.
to in
in c-x
o cn
r-t C*.
t^T o
co r*^
CM t-ป
-1 CO
CO r-i
tO LO
CO CM
-t CO
to cn
t- r>.
n r^
ft tO
0 CM
^r m
CM *-*
r-i Px
CO IO
r^ CM
O ro
-4 CO
CM T
CO CO
r-< r^.
i i CO
p-ซ CM
o m
*-ซ cn
i-* rv
CM 0
to cn
r-1 CO
> CO
Heat tosses
Total Out
cฃ
CO
to
m
3S
to
in
to
&s
to
CO
to
as
CO
t*-.
in
a*
r-v
CO
in
a<
cn
to
ae
m
cn
m
*
T (
^
6?
CO
ป
r^
69
CO
*r
c^
6?
o
t-%
t-.
ss
^r
^r
t~.
as
CM
1^
5?
cป
to
Jt^
tx
\n
r^
+
CM
&
O
a
ฃ
o
8 ฃ
2 2
I
CJป Dl
2 2
c c
o o
-------
TABLE 13. SUMMARY OF SLAG ANALYSES
Major
Components
C (wt. %)
sio2
M2ฐ3
C 0
a
FeO
MgO
P2ฐ5
BaO
MnO
Cr2ฐ3
TiO.
ฃ,
N10
Trace
Components
Cd (ppm)
Cr
Cu
Fe
Hg
Mn
Ni
Pb
Zn
' ' - ' ?
Test Test
I
1
60
10
12
8
I
0
0
0
0
p
0
9?
10
891
4300
55000
0.01
1200
2700
119
455
. f.
2
,0 0,8,
.0 59.8
.0 11,9
.0 10.5
.1 8.1
.9 ,0.9
.9 0.6
,'2 0.2
.2 0.1
.05 0.1
,51 0,5?
.3 , 0.2
He 93,72
10
645
4000
53000
0.02
1700
1300
119
482
' 'i
- - 41
Test
'3'
'0.2
64,. P
lil.O
12.1
'6.7
1,2
0.6
0.1
0.3
0.05
P,5&
'0.05
96.86
6
363
3800
53000
0.01
1400
145
128
456
Test
4
0.9
66,5
10.5
9.9
6.3
1,6
0,8
0,1
0,2
0.5
P. 6
0,05
97^95
5
3420
2860
488QO
0.05
1140
15P
^2Q
3.5Q
'ป ,i | ,.ii ni'T,w i r | if! u
i ,- . -i , . -,,
Test
5
1,8
61.5
UT6
9.8
8.6
0,9
0.ง
Qfi
0,3
0.5
Of7
0.05
. 95,75
' '5
3420
4QOO
61970 11
0,03
16PP
3P5
"94
3Q9
' "'''V| "".'I'""1'1 '
50
P J
Test
6 ;
0.8
56.1
9.7
.8.6
If.e
1,9
p. 4
0.1
0,3
. 0.5
0.6
:P,05
93 , ง5
'5
3420
3400
0.01
1300
959 .
75;
404 ;
,T , i , F.r.. r- .j
1 "*
7
0
56
9
11
8
1
0
0
0
9
0
0
*1"ii"M.np
5
3420
33,00
643PO
0.1?
1400
1500
243
npo
TJ"
S|
,4
^
,3
,3
1 T
t3
,o
T>
T-2
,.?
f5
t&
e
,,1
i|i^H;
--
-------
Improper blending of the wet sludge with refuse results in poor pellet
formation during the feeding process using the existing equipment. Instead of
being intimately mixed with the refuse to form a cohesive, stable particle, the
slug of wet sludge acted as a lubricant for the pellet thus interfering in the
bonding process which normally holds pellets together. This resulted in
feeding loose, shredded refuse into the converter where its variable porosity
and packing characteristics produced channeling of the hot hearth gas up through
the bed without proper heat exchange with the charge. The loss of energy due
to offgas temperatures in excess of 427 degrees C (800 degrees F) produced the
observed decrease in conversion efficiency in these cases.
I'he detailed observations and calculation for each of the test periods
are contained in Appendices B - H. Section 10 presents a discussion of general
process-related observations. Section 11 details the calculational technique
and projected results for operation in the normal mode with particulate recycle
(char and ESP oil recycled to the hearth for gasification). Section 12 discusses
the environmental emissions observed during the testing with particular
emphasis upon the disposition of trace heavy metals. Section 13 presents
some projected economics for the codisposal process as-practiced and also
more favorable economics for a modified form of the process.
51
-------
SECTION 10
CODISPOSAL PROCESS OBSERVATIONS
GAS PRODUCTION RATE
The methods used to calculate the gas production rate were described
earlier. The gas production rates calculated per Mg (ton) of material fed are
shown in Tables 14 and 15. Comparison of helium trace and pitot: tube results
shows that for all but one test period helium trace results were lower than
pitot tube measurements. The inclusion of nitrogen and water vapor in the
pitot tube results contributes to part of this difference. The gas production
measured by pitot tube at ten minute intervals varied significantly. The
standard deviation of the pitot tube results for different test periods varied
between 53.15 and 97.46 Nm3 per Mg (1,800 and 3,300 Sft3 per ton) of feed.
The standard deviation of helium trace results varied between 23.63 and 59.06
Nm3 per Mg (800 and 2000 Sft per ton).
TABLE 14. GAS PRODUCTION MEASURED BY HELIUM TRACE
Gas Production
Test Number of Per Unit of Feed
Period Samples Nm3/Mg S ft3/Ton
Standard Deviation
Nm3/Mg s ft3/Ton
5
1
2
6
7
3
4
6
3
6
9
14
12
6
437.
402
440
490
396:
452:
396
14,800
13,600
14,900
16,600
13,400
15,300
13,400
24
38
53
59
41
53
44
800
1,300
1,800
2,000
1,400
1,800
1,500
52
-------
TABLE 15. GAS PRODUCTION MEASURED BY PITOT TUBE
Gas Production
Test
Period
5
1
2
6
7
3
4
Number of
Samples
12
No samples
10
12
12
12
12
Per Unit of
Nm /Mg
325
taken
478
499
526
543
449
Feed
SftYTon
11,000
16,200
16,900
17,800
18,400
15,200
Standard
Nm /Mg
97
53
38
77
50
77
Deviation
Sft3/Ton
3,300
1,800
1,300
2,600
1,700
2,600
The fuel gas production rate will vary substantially as the gasifiable
composition of the feed varies. High ash or water content will significantly
reduce the gasifiable material available in a unit of feed. Comparisons of gas
production rates based on the total feed can be misleading because of this
effect. Comparison on a dry, ash free basis would be more meaningful. How-
ever, it was evident from the mass balances performed that a representative
composition for the refuse was not obtained. Consequently, without an accurate
set of refuse analyses to determine the gasifiable fraction of the feed, a reliable
comparison could not be made. The gas production per Mg (ton) of gasifiable
feed (organics on a dry, ash-free basis) is shown in Table 16 using the back-
calculated refuse compositions to determine moisture and ash content. The
results show variation similar to the gas samples per Mg (ton) of total feed.
TABLE 16. GAS PRODUCTION BASED ON GASIFIABLE MATERIAL
Gas Production
Per Unit of Feed
Test Period
5 (refuse only)
1 (refuse + primary)
2 (refuse + primary)
6 (refuse + primary)
7 (refuse + primary)
3 (refuse + mixed)
4 (refuse + mixed)
Nm /Mg
916
859
865
1051
913
1116
927
Sft3/Ton
31,000
29,100
29,300
35,600
30,900
37,800
31,400
53
-------
GAS HEATING VALUE
Product gas higher and lower heating values were calculated for each
sample from the product gas analysis on a water, oxygen, and inert free basis0
(Inerts include nitrogen, argon, and helium.) C4 and higher hydrocarbons,
except benzene, were lumped together as 04*3, C^'s and higher hydrocarbons
(HHC) in the calculation. The mean gas higher heating value and standard
deviation were calculated for each [period, and these mean heating values
were used in subsequent heat balances.
Gas heating value results were also reviewed, using several methods,
for dependence on process variables. Sample results were divided into five
groups, one group of samples from refuse-only operation (test period 5), two
groups of samples taken during raw primary sludge processing (test period 2 was
group 2 and test periods' 1,6, and,7 were group 3), and two groups of samples
during mixed sludge processing (te|st period 3 was group 4 and test period 4 was
group 5). Mean gas heating value;and its standard deviation were calculated for
each group. Two sample groups (groups number 2 and 4 in Table 17) exhibited
significantly lower heating values than the other three. Investigation of operation
during the periods associated with these two groups revealed the occurrence of
significant periods of high offgas temperature suggesting refuse, bed channeling
which is believed to be caused by poor blending of high moisture sludge with
refuse during feeding. To quantify the extent of this effect, the fraction of
operating,time the, converter offgas was within different temperature bands was
determined and plotted. Offgas temperature above 427 degrees C (800 degrees
F) is considered evidence of channeling, and temperature above 538 degrees C
(1000 degrees F) is evidence of serious channeling. The temperature records
support the contention that channeling did occur at least 20% of the time in one
case and 35% of the time in the other. Samples obtained during high offgas
temperature periods exhibit lower mean heating value and higher standard
deviation within a sample group than the other sets in Table 17. These two groups
were judged unsuitable for use in subsequent study of heat value behavior.
For each sample group unaffected by high offgas temperature, higher
heating value was plotted as a funption of converter rate. The 9 5 percent
confidence interval on the mean ofjall samples in the group was superimposed
on each plot. , The graphs are shown by Figures 18, 19, and 20 for refuse
only, primary sludge, and mixed sludge respectively. Results indicate a
trend toward higher gas heating values at higher converter rates. Though
insufficient samples were taken at low rates to confirm this conclusion, such
a trend is consistent with predicted converter behavior. Since converter heat
losses are relatively constant, the portion of refuse combusted to cover
these losses is constant as rate changes. This would predict more combustion
in the hearth and a resultant lower product gas heating value at lower operating
rates. More specific conclusions could not be substantiated with the
available data.
54
-------
TABLE 17. PRODUCT GAS HIGHER HEATING VALUE (HHV) COMPARISON
Mean Higher Standard 95% Confidence Interval
Heating Value Deviation Lower Value Upper Value
Sample ' Number of
Group . Feed Samples
1
2
3
4
5
Refuse 15
Refuse + 12
Primary Sludge
Refuse + 24
Primary Sludge
Refuse + 11
Mixed Sludge
^Refuse + 9
Mixed Sludge
MJ/Nm3 '
(Btu/Sft3)
15.04
(381.9) >
14.11
(358.3)
14.78
(375.3) :
12.78
(324.4)
15.22
(386.6)
MJ/Nm3
(Btu/Sft3)
0.43
(10.8)
0.82
(20.8)
0.78
(19.7)
1.98
(50.4)
0.39
(18.5)
MJ/Nm3
(Btu/Sft3)
14.8
(376.8)
13.7
(347.2)
14.5
(368.1)
10.6
(269.9)
14.8
(375.1)
MJ/Nm3
(Btu/Sft3)
15.2
(387.0)
14.5
(369.3)
15.1
(382.5)
13.7
(349.0)
15.7
(398.1)
Samples taken at approximately the same rate, but while processing
different feeds, were compared statistically. The comparisons between groups
were made using the Student-t method assuming equal true means and equal true
standard deviations. The t statistic was computed for comparing groups such as
a and b. The t statistic is defined for such a comparison as t = (Xg-}^)/
S [(l/na) + (l/nb)]i;*. The arithmetic mean of group a consisting of na observa-
tions is )fe . The square root of the pooled variance estimate is S which is
given by S2 = [(na-l) S2a + (nb-l) S2b ] / [na + nfo-2 ] where Sa is the standard
deviation computed for group a. If the calculated t statistic is greater than
the tabulated t statistic for a desired confidence level, then the assumption
of equal true means is invalid. In this case where comparisons are being made
between the mean heating values for various groups, if the computed t is less
than the t value for the desired confidence level then the groups are not
statistically different. These tests could not confirm a difference between any
sample groups with a 97 percent confidence. Results indicate that at the sludge
ratios tested, product gas higher heating value was unaffected by sludge
processing. A summary of the statistical results is shown in Table 18.
55
-------
to
i-l
(0
(-1
O
ฉ3
ง
a
o
CO
M-(
S
3
4-1
to
s
to
to
60
O
o
irt
r*
13
O
PW
oo
(U
3
if
fa
56
-------
c
ffl ฉ <3E>
1 O
ฃ ' ~ x^-S
^ 0(333
o ro
0 0)
So C
" CD O
i ;
'
'.
i,
r.
0ฉ
0
SX
9
ฉ0
E>ฉ
ro
Z 0 0
> \
.; I ~) ID 0
I 2 -< -
u
H
irl
JJ
s
1
O dJ
2 ^
?
!-l
1
M
O.
'O S
"&o ^
1
S 3
- |
I
^ -u
O ccj
ID 0)
CO
to
00
a
13
o
<
a>
-------
(0
o
0)
O (0
O Q)
in fi
en O
Q
o
Q
in
o
58
-------
TABLE 18. STATISTICAL COMPARISON OF PRODUCT GAS HEATING VALUES
ซ
Sample Group
Refuse Only (1)
Refuse +
Primary Sludge (3)
Refuse +
Mixed Sludge (5)
Number of
Samples
6
15
8
Mean Higher
Heating Value
MJ/Nm3
(Btu/Sft3)
15.3
(389.5)
14.7
(374.2)
15.2
(386.6)
Standard
Deviation
MJ/Nm3
(Btu/Sft3)
0.39
(9.9)
0.72
(18.3)
0.78
(19.8)
Groups Compared
Calculated t
t for 97.5 Percent
Confidence Level
1 - 3
1-5
3-5
1.89
0.31
1.50
2 . 09 3
2.093
2.093
GAS PARTICULATE LOADING
The result from gas particulate sampling of the product gas after the
condenser are shown in Table 19. The particulate loading and isokinetic
variation (I = 100% at isokinetic flow condition) for each sample are listed.
The particulate sample calculations have been corrected to remove the natural
gas used to purge the ESP insulator compartments. The isokinetic variation is a
measure of how close to isokinetic flow the sample was obtained. Isokinetic
variation between 90 and 110 indicates an acceptable sample.
The particulate levels measured were all very low. The highest
particulate level was 0.0449 g/Nm3 (0.018 gr/Sft3) measured during primary
sludge processing at 106.3 Mg (117.2 tons) per day. Two samples had
unacceptable values of I. For both samples I was below 90 which should
cause too high a calculated particulate loading.
59
-------
TABLE 19. PRODUCT GAS PARTICULATE LEVELS
Test
Period
Type
Feed
Particulate Loading
g/Nm3 gr/Sft3
Isokinetic
Variation
No Sample Taken
5 Refuse 0.0176 0.007
1 Refuse +
Primary Sludge
2 Refuse + 0.0372 0.015
Primary sludge
6 Refuse + 0.0223 0.009
Primary Sludge
7 Refuse + 0.0449 0.018
Primary Sludge
3 Refuse + 0.0188 0.008
Mixed Sludge
4 Refuse + 0.0201 0.008
Mixed Sludge
58.13
102.5
92.26
86.93
91.98
91.70
The regulated particulate level for a stationary source is 0.18 g/Nm
(0.08 gr/Sft3) corrected to 12% CO2 .* The regulated particulate level for a
sludge incinerator is 0.65 g/kg of dry sludge input (1.3 Ib/ton of dry sludge)+.
These levels are applied to combustion products, so to compare requires putting
the product gas numbers on a combustion product basis. This can be done by
calculation to give the results indicated in Table 20. Table 21 presents the
particulate loading based on the sludge incineration standard. Although two of
the test periods (2 and 6) exceed the emission regulation, it should be recalled
that this particulate is measured in the gas prior to combustion. If as little as
10% of the particulate is combustible, all the tests would fall below the regu-
lated limit. The relatively high loading of particulate per unit of sludge
processed is the result of the small fraction of sludge in the feed. At higher
sludge-to-refuse ratios, the particulate loading calculated in this manner should
be well below the regulated value|.
*(2) Federal Register, Vol. 36, No. 247, December 23, 1971.
I
+(3) "Background Information for New Source Performance Standards" (Vol. 3)
EPA Report 450/2-74-003, APTP-1352 C (Feb. 1974).
60
-------
TABLE 20. CALCULATED PARTICULATE LEVELS IN PUROX SYSTEM
GAS COMBUSTION PRODUCTS AT 12 PERCENT CO2*
Test Period
Particulate Level
g/Nm3 gr/Sft3
5 (Refuse)
1 (Primary)
2 (Primary)
6 (Primary)
7 (Primary)
3 (Secondary)
4 (Secondary)
0.0031
0.0056
0.0034
0.0063
0.0031
0.0029
0.0013
0.0023
0.0014
0.0026
0.0013
0.0012
*EPA regulation is 0.18 g/Nm3 (0.08 gr/Sft3) at 12% CO,
TABLE 21. PARTICULATE LEVELS IN PRODUCT GAS BASED ON SLUDGE RATE*
Test Period
5
1
2
6
7
3
4
g/kg Sludge (ds)
_
-
0.666+
0.719
0.421
0.312
0.331
Lb/ton Sludge (ds)
_
-
1.332+
1.438
0.842
0.624
0.662
If particulate is assumed to consist of at least 10% combustible material,
these values will both be below the regulated limit.
* EPA Regulation is 0.65 g/kg (1.3 Ib/ton) of dry sludge.
61
-------
SECTION 11
PROJECTION OF RESULTS TO CASE OF PARTICULATE RECYCLE
The system performance expected with the recycle of particulate material
and oil into the hearth combustion zone can be calculated from a heat balance on
the hearth. The key assumption is that the additional energy required for the
gasification of the recycled material is that needed to heat the material to 1650
degrees C (3000 degrees F),vaporize the water, and slag the ash. Pyrolysis
reactions of the organic portion of the char and oil are assumed to have negligible
heat requirements. The material is treated as a mixture of carbon, hydrogen,
oxygen, water and ash. Since the amounts of oxygen and hydrogen in the re-
cycled particulate are very small, they will be neglected in the model. Gasi-
fication of this material is carried out via the following reactions.
I C + CO9 = 2CO 23.618 MJ/kg (10,154 Btu/lb) of carbon
II C 4- 1/2 O sCO - 9.187 MJ/kg (-3950 Btu/lb) of carbon
ฃ*
in
C + H0O = CO + EL 10.96 MJ/kg (4711 Btu/lb) of carbon
For simplicity, all the carbon in the particulate is assumed to be pyrolysis
residue which will be gasified via Equations I, II or III. This will result in a
higher level of carbon to be gasified than would actually occur because some
carbon will be gasified as hydrocarbon in gaseous pyrolysis products.
An examination of the relative kinetics of reactions I, II and III is also
required by this analysis. The carbon-oxygen reaction occurs rapidly at 1650
degrees C (3000 degrees F). The steam-carbon reaction, though slower than
carbon-oxygen reaction II, is significantly faster than the carbon CO2 reaction.
With these rate considerations in mind, the process of gasification can be
modeled as follows. Carbon will react via equations II and III. As long as
sufficient water is available for reaction, carbon will react via equation III
with heat required for reaction supplied by equation II. If insufficient water is
available to consume all the carbon by reactions II and III, then reaction I will
occur. To minimize the oxygen consumed, the quantity of carbon gasified by
equation II will be just sufficient to heat the recycle stream and drive the al-
ternate gasification reactions. The offgas composition is implicitly assumed to
be similar to the gas measured.
62
-------
The amounts of carbon gasified by reactions I, n and III can be com-
puted from simple mass and heat balances on the recycled material. For the
first step, assume no reaction via equation I. The carbon and heat balances
are as follows:
c
n
cm - Tc
+ (H)A +
(4711) =
(3950)
TC is total carbon; W is water; A is ash; H is heat to vaporize water; H is
heat to melt ash. After solving for C and (J the moles of C are compared
to the moles of water available. If U is greater than W, then reaction I will
occur and a different set of balances must be solved.
If reaction I is allowed to occur, assume reaction III goes to completion
and all water available is consumed. The amount of carbon consumed by
reaction III can then be related by stoichiometry to the water available as
follows:
cm = lf(w>
The carbon and heat balances can now be reduced as follows:
CII +
7f (W) = T
(HJA + f| (W)
(10,154) =
(3950)
The solution of these balances allows the calculation of the additional
gaseous product, additional oxygen consumed', and additional moisture in the
product gas, if any. By applying these techniques to the material balances
presented earlier, the effects of particulate recycle on process parameters can
be predicted. Table 22 and 23 show the results of this analysis on the overall
mass balances for each test period.
The oxygen consumption calculated for each test period is shown in
Table 24. The oxygen use during test period 3 when offgas temperature was
high, as a result of channeling, is significantly higher than that of the other
tests. Oxygen consumption without particulate recycle is about 0.02-0.03
units of 0_ per unit of feed less than the results in Table 24. The oxygen
consumptions during other tests are in fairly good .agreement.
63
-------
W
pa
B
W
O
w
CO
CM
ta
en
0
CO
w
0
VI
0
ca
0
CM
W
0
cn
0
LO
0)
EH
0
CQ
LO.
CM
CO
CO
CD
LO
co
8
OS
CO
(Nl r-l
ซ
CD o
CO CM
CO ^
LO CO
LO
co co
CD Cv.
CO
H CO
co 'i-
CM 'CO
CM cp
>flซ ;CM
CO CM
i-H l-H CJJ CO M*
t^. Lri rrf CQ -|Q
ฃป, CO
w
3 TJ
BS
00
co
LO
CM
CD
0
LO
CQ
CO Cx ซvT
OO LO CO
i-H CO O
CM LO
CM CO CO
co co
CM O
CO CM
LO CO CO
CO
LO
CM CO CM
LO CO
cn
fr, CO
co co
CO
LO CM CM
CO r-H CO
rt! CM CM
(0
CO
CD
CM
CO CM
CD CM
O LO
*3< CM
CM CO
0
11
ro
o
CQ
0
0
1-)
Cn
0
co
0
0 CQ
X! fO
w
c X
0
0 0
"O
0
i<
a
0
CQ
rQ
-O
0
o
0
0
O
(0
0
o
fO
0
o
a,
LO
CO
0
a
3
w
CQ
ro 2S C
Cn a 0
_, 0 Cn
0 ** IS1
30 /$
Pin X! O
64
-------
<""">
ffi
CO
I-H
1-1
0
^
H
pq
O
S
pq
K
Pq
EH
JD
Q
l-H
EH.
PC;
Q_! *^^*
ffi <ฃ>
> >-<
> 0
CO
EH CO
1-3 &
ID o
co c.
S
O
%$
< O
Cฃ CO PM O EH
LO IX *^f CO
* CO CM CD
CO CM -sT
co c^ ^F co
CM i ! i 1 CO
^ i-H CM
CO CO CO
ป
CO CD CD CD
^F i~H CM
CM CM CO LO
vf CO O CD
LO CM CO
CO CD CD CO
^ i i co
CO CD CO CD
"st1 LO CM CD
LO CM CM
CO IX 'vF CD
ซ^" i-H CO O
LO CM CO
-H W S-,
in fd o)
S CD O> 43
0 tH jo jo
^ O co ^
CM
LO
^
i-H
CM
ซ
CM
CO
O
CM
O
i)
t-
O
CM
r-H
I 1
CD
O
<"!
.x.
CM
r 1
i-H
CD
0
i 1
i-H
i 1
fd
8
EH
r-H
3
O
n
\^s
t
PM
co
0
0
S-,
0
t3
O
0
o
CO
O
1 i
G
0
CD
X
O
0
rd
+J
-M
rd
0
fli
S_i
a
o
+J
oxygen
0
,c~{
+ป
rC
"1
!!
rd
0
^
0
+5
o
-M
-C!
i i
a
a
CO
CO
rH
fd
Cn
r-H
0
*
CO
fd
4?
m
S
Q)
CO
^i
CO
PUROX
+4
rt
O
a
td
S
CM
+J
0
i i
fd
"3
CJ1
0
_i
fd
_Q*
^
"O
0
O
fd
ซ 1
a
0
S*i
0
0
Cn
g
"H
M-H
O
0
o
s
fd
4U
S
^
o
S1
M
Ix
r 1
en
O
S
en
"*"*
0
D
rd
i-H
rd
&
0
^N
rH
ซ
a
4-*
C
O
S-i
0
a
LO
r,^
CD
-f-J
fd
T3
0
1 1
a
a
ฃ3
en
cn
p.
0
Cn
IX
X
O
65
-------
TABLE 24. OXYGEN CONSUMPTION EXPECTED
WITH PARTICULATE RECYCLE
Test Period
Mg O2/Mg Total Feed
5
1
2
6
7
3
4
0.224
0.197
0.214
0.228
0.213
' 0.266
0.208
Table 25 presents a summary of the range of predicted oxygen consump-
tions measured when processing various types of feeds. Period number 3 was
excluded due to the feeder problems discussed previously.
TABLE 25. OXYGEN CONSUMPTION RANGES
WITH VARIOUS FEEDS
Type of Feed
Oxygen Consumption
(Mg O9/Mg Feed)
Refuse only
Refuse + primary sludge
Refuse + mixed sludgjs
0.20 - 0.22
0.20 - 0.23
0.21
The conclusion from these results is that over the range of sludge compositions
and sludge fractions tested, there was no significant change in oxygen con-
sumption.
The fuel gas production that was calculated assuming the model as
described is shown in Table 26. Comparison with Table 14 indicates increases
in gas production with particulate recycle of from 10-25 percent. The breadth
of this range is a result of the range of operating rates and associated particulate
production rates. Particulate recycle represents a significant portion of the
total energy production from the PUROX System.
66
-------
TABLE 26. GAS PRODUCTION EXPECTED
WITH PARTICULATE RECYCLE
Test Period
5
1
2
6
7
3
4
Gas Production
3 "3
Nm /Mg Sft /Ton
514
487
526
543
502
526
490
17,400
16,500
17,800
18,400
17,000
17,800
16,600
The energy balance projections for particulate recycle operation are
shown in Tables 27 and 28. As discussed earlier, the measured energy content of
incoming refuse was high when compared to typical results. The variation in
refuse composition and heating value was not adequately defined by the refuse
samples taken. To compensate for the error induced by the unrepresentative
refuse analysis, the efficiency of energy conversion was calculated based on
the total energy leaving the process. The calculated efficiencies are all
between 70 percent and 80 percent.
67
-------
s
w
!ง
*""''
w
O
s
M
W
EH
r3
O
EH _
*
CM
w
HH
CQ
EH
,
CO
0
EH
CO
CO
0
EH
C-.
CO
0
EH
CO
co
0
EH
CM
4J
CO
0
EH
rH
4-*
CO
0
EH
LO
CO
0
EH
CM ^ CM CO CO CM CM CO CO O "31
CM IX OO CM OOt^COCOLO t~- '**
OCOLOCD LOCOLOCOCO O r^
CD CD CD r 1 OO r )
LO"3t'3lOO COCOCOOCM rH C>J
C^ CO CO *sj* ^J* CO LO ^J1 rH CO HH
CDCMCQCQ OOCM-vfCMCO OO SX
CD C^ ^f rH CD
:
OOOIXLO OCOrHrHLO O 00
t^COCMCO ^OOrHLO OO ^t*
CMLOCDlX O^IXCDC^ CD IX
O rH OO rH O
rH rH * rH
^x
CDt^-C^O CDCDCDCDO O CO
CDOCMOO CV-OC^CMCD ** '^f
rHiHCDCM C^CMCOOCO 00 IX
CD t-- "vf rH CD
sง
O) ^f ^f t^ CO C**** OO C*sl OO "^ i-O
,_i ,_H CO CD t^ O i 1 CO OO OO <*D
00 ^^ CO CO ^^* 'sf ^f ^l ^f C^ C^*1*
CD O CO rH 00 CM
rH
^
LO^f^CO CDrHCDCMrH CM rH
OCDOOOO -sfCMrHCDLO CO LO
CO rH CD ^J1 CO *^ ^f rH CO ^J* IX
CD O CD rH CO CM
CDOC^-CO rHCMSfCDCO CO O
Cป-OCMO; LOLOCOLOCM co ix
CM CDCM CDCOCOOt^- "^ X.
CO CD LO rH X. rH
U 0 C *>>
e . . I a. 1 111 1 ||
(Tjcooi I-H O 4J..J-H 0>;r!
fn3T3rH(a St3im(5(0 'wets
S's^S-s g js ^ | -g ^rsK
coccJcoP-tEH cuco^ffiEH RO
3
O
4->
(0
0
c
0
0
co
rd
PQ
*
68
-------
w
o
ฃ
H,
W
O
B
w
OS
w
t""'
1-5 ,-.
7-"\ CtJ
8 Q
EH X
OS j}
CL, CQ
ffi "o
M cn
> o
CO 3
w "-a
o g
LOOOi-H COCOCOLOi-H CD [ป
CD I-H r^. I-H o
t _.J
55
COi-HCDCO OLOCOLOCD t^- 00
c-xoc^-Lo cocDLorvio CD ^
CO i 1 CO CO LO i 1 CO CD O C^- C^
LO co ^ co
Q
C^COi-HrH OOCOCDCOCO OO OO
CD CD CD ^ CO OO CO IX CD ""st1 CID
COCMCDCO i-HCOCOOCD CO C^,
CD CD CD i-H C r-H
v^
"^"vFcoco cocDrx-i-Ho co I-H
OCO^CO COCDCDOCO O LO
r-Hr-HCDCD ฃ^.COCOi-HCD CO IX
CD CD LO i 1 IX CM
C-- CDCD o "vf LO i 1 O CD O
^st* f~~) IX CM *sP CO *^ CO C*3 (~"i IX
oo oocv. CDCOCOOCO -sf1 tx
ts CO LO i-H IX i-H
. w *
w CD ^ m C n*
<0 ฃ cn ฎ O H
w CD (OcnTj y^S
e 0 o 5 s x | a 3 ง |-ง
'gS'O-'S g & ซ 13 ซ ^ E^g
S^^ง-S gjS^|^ ^r9K
COOSCOPnEH DHCO>PHEH QO
4-*
0
4-1
rd
0
*-
o
T3
0
cn
CQ
*
69
-------
SECTION 12
ENVIRONMENTAL EFFECTS OF THE CODISPOSAL PROCESS
The environmental effects of refuse and sludge codisposal were
estimated and compared with refuse-only operation. The environmental effect
of either type of operation is minimal. The estimate was based on the
substances emitted to the environment from slag and in PUROX System gas
combustion products.
Simple slag leaching tests were conducted using the procedure
described in Section 8. The slag leachate obtained was analyzed by Union
Carbide Corporation South Charleston Laboratories. Later samples were also
analyzed by Union Carbide Corporation Tarrytown Laboratories because they
could obtain lower detection limits,for some analyses.
Since no standard currently exists for leachate quality, the analyses
of the leachate samples by themselves have little meaning. However, for
purposes of comparison, EPA standards for public drinking water supplies have
been used to document the general overall superior quality of the leachate.
Although the samples do slightly exceed some of the limits set down in the
public drinking water standard, its very proximity suggests a minimal environ-
mental effect upon the ground water. Earlier tests using slag from, refuse-only
operation indicate that the soil itself leaches substantially more material than
the slag*.
The leachate samples prepared from slag obtained during the tests
were analyzed by both laboratories. As can be seen from Table 29 the
concentration of the trace components are essentially at or below the
detection limits of the instrumentation used (see Appendix A for limits). The
general proximity of the leachate to the public drinking water standard is
strong evidence for the bound nature of the trace components in the slag.
*(1) C. T. Moses, J. R. Rivero; "Design and Operation of the PUROX System
Demonstration Plant" , The Fifth National Congress on Waste Management
Technology and Resource Recovery, Dallas, Texas, 1976.
70
-------
*
CO
H
CO
, 1
is
-"
0 ^ n - r_^ *1 | 3 r 5 r 1 r -1 h1 t |I| )-J] [ "] | "| . Q
^o2 *ฐQQ .QQQ .QQQQ . 33
m
O -I-*
CD
PI
I C -S
ffil pH cQfd0^(h'r^hฐ10X!i^;^GCJi 0
CQ
0
"s-.
o
"rd
tn k
0 R
*d/ Q
rH r~J
s-i fd
ฐ ^
fd G LO
^ O tx
O ป-; CD
rQ W r-H
fd 0
'~' j^ j _^*
ill
fd o <ฐ
EH CO g
-11ฐ
TJ fd fd ฐฐ
-2 0 0 CM
jr; a a
0 t-l S-H *
a o o o
^_, O O &
S 0 0 -
4_> I"Q *^j CD
CP rQ ,Q
f^ U J-i .
iH fO fO r 1
IOQ.O
13 G G
G 0 0 J
^ 'c "c ฎ
^ ID ID w
S --H
!>i J>ปj CJ^
0 X! X) 0
a CQ CQ
S-rH TH 1 1
. CQ CQ fd
<0 >>, >. Si
co A? ^-T 0
_. fO (8 T3
0 G C 0
< < PM
JtJ _j_ _| [_ ^Z.
71
-------
The product gas particulate level was measured during all test periods
except one, and the levels are shown in Table 19. Particulate levels were low
during all periods and ranged from 0.0176 g/Nm3 (0.007 gr/Sft3) to 0.0449
g/Nm3 (0.018 gr/Sft3). The isokinetic variation of two samples was below the
acceptable range of 90-110 percent and these two samples are not reliable.
The measured particulate level was, converted to particulate in the gas com-
bustion products on a 12 percent CC>2 basis and the calculated particulate
levels are shown in Table 20. The dry catch obtained during test periods 5, 6
and 7 were analyzed for trace metals and the results are shown in Table 30.
The product gas particulate sample collected during the refuse only
test (period 5) produced an isokinetic factor (I) of 58.13. This value of I was
well below the acceptable level, but no other data for this test period was
available. The particulate level was 0.017 g/Nm3 (0.007 gr/Sft3) in the
product gas which was equivalent to 0.0031 g/Nm3 (0.0013 gr/Sft3) in the gas
combustion products. The low value of I was caused by a procedural error
during sample collection. Particulate samples collected during an earlier
refuse only run resulted in a particulate level of 0.023 g/Nm3 (0.0094 gr/Sft3)
in the product gas.
Three particulate samples were taken during mixed sludge tests
(periods f2, 6 & 7). Particulate loadings of 0.036, 0.022, 0.044 g/Nm3
(0.015, 0.009 and 0.018 gr/Sft3) were measured, and they are equivalent to
0.0056, 0.0034, and 0.0063 g/Nm3 (0.0023, 0.0014 and 0.0026 gr/Sft3) in
combustion products at 12 percent CO2ซ T^e samples were taken at system
rates between 77.1 and 112.4 Mg (85 and 102 tons) per day. The third sample,
taken at 90.7 Mg (100 tons) per day, has a calculated isokinetic factor of 86.9
which was slightly below the acceptable range. The isokinetic factor was
satisfactory for the other samples .
Particulate measurements were obtained during mixed sludge
processing at 78.9 and 91.6 Mg (87 and 101 tons) per day. The measured
particulate concentration was 0.019 g/Nm3' (0.008 gr/Sft3) during both period
3 and,4. The calculated combustion product particulate level was 0.0031
g/Nm (0.0013 gr/Sft3) for the 78.9 Mg (87 tons) per day sample and 0.0029
g/Nm3 (0.0012 gr/Sft3) for the 91.6 Mg (101 tons) per day sample.
The results of the particulate sample analyses are shown in Table 30.
The reported results have been converted to a mean metals concentration in the
product gas, expressed in//g of metal per cubic meter of gas at standard
conditions. The equivalent concentration in gas combustion products is also
tabulated. Using the results,the metal emitted in the product gas during one
day of operation was estimated. These estimates are summarized in Table 31.
Only the metal which was contained in the particulate is included in this
calculation.
72
-------
TABLE 30. CONCENTRATION OF METAL IN PUROX SYSTEM GAS
Component
Product Gas
3^ **
Combustion Products
m3 @ 12 percent COJ**
As
Ba
Be
Cd
Cr
Ca
Fe
Hg*
Mn
Ni
Pb
Zn
4.6
0.65
0.28
0.46
5.4
74
229
-
2.8
14.6
12.4
72
0.60
0.09
0.04
0.06
0.71
9.7
30.0
-
0.37
1.9
1;6
9.4
insufficient amount of material available to analyze.
** 1 g/NmS = 4.13 x 10~7 gr/Sft3
TABLE 31. GRAMS OF METAL EMITTED IN PUROX SYSTEM GAS
Test Period
Metal
As
Ba
Cd
Cr
Pb
Basis: 1 day
5
0.10
0.31
0.04
0.38
0.76
6
0.21
0.02
0.02
0.20
0.47
7
0.28
0.02
0.03
0.19
0.38
System Rate Mg (tons) Per Day 52.6 (58)
77.1 (85)
90.7 (100)
73
-------
As noted earlier, recirculation of scrubber water was not achieved
during the codisposal tests as a result of mechanical problems with the
particulate recycle system. This resulted in a significant dilution of the
wastewater stream due to the continuous addition of about 0.38 rn^ per minute
(100 gallons per minute) of makeup to the scrubbing system or a dilution
factor at a 90.7 Mg per day (100 tons per day) operating rate of '-^O to 1.
Since this was the only wastewater available, treatability tests were con-
ducted with this material using two liter, continuous flow reactors to determine
the kinetics of activated sludge treatment of this wastewater. The general
conclusion of the treatability work was that 95% reductions in biological
oxygen demand (BOD) can be achieved readily at this dilution. Table 32
presents a summary of typical wastewater properties and reactor operating
parameters from the treatability tests.
TABLE 32. PUROX SYSTEM WASTEWATER TREATABILITY DATA
Wastewater Analysis
Reactor Operating Conditions
TBOD 1350 ppm
D
SBOD- 1321 ppm
o
TCOD 3078 ppm
SCOD 3012 ppm
COD/BOD5 2.28
F/MBOjVday
MLVSS
td/ days
MLVSS, ppm
1.18
1684
So (TBOD,.), ppm 1350
O
Se (SBOD ), ppm 59
3
These results with wastewater from sludge codisposal confirmed
earlier treatability studies which generally showed no technical problem in
treating wastewater from the PUROX System. It has been recognized that
significant economies will be realized by using municipal sewage as a source
of dilution water since its relatively low BOD (200 ppm) and significant
content of nutrients will enhance (treatment of PUROX System wastewater in a
commercial installation. The treated waste stream could then be discharged
back into the sewer main for final disposal. Removal of heavy metals from
the PUROX System wastewater stream would be achieved by precipitation
prior to treatment if that is necessary.
After the conclusion of the codisposal program in June 1977, operation
of the PUROX System plant using refuse only continued throughout the summer.
During this time significant improvements were made in the mechanical
74
-------
equipment utilized for particulate (char) recycle resulting in successful opera-
tion of the system. Data similar to that described in. this codisposal program
was collected and analyzed to evaluate overall system performance under
these conditions.
In terms of overall process performance, the heat and material balances
are in general agreement with the projected performance for refuse-only
discussed in Section 11. Oxygen consumption and product gas generation are
in good agreement with the calculated results as would be expected. These
results suggest that the codisposal results with particulate recycle would also
agree well with the projected results discussed earlier.
The question of environmental effects during particulate recycle,
especially the ultimate disposition of heavy metals, was not addressed during
the earlier discussion since it was felt this could only be satisfactorily
answered by experimental measurement. These experimental measurements
were made during the particulate recycle testing with refuse only subsequent to
the codisposal program. The similarity of the refuse-only experimental
particulate recycle results to those projected for refuse-only indicates a high
probability that the projections for codisposal with particulate recycle would
agree well with the actual results. In particular, the environmental effects
reported here for refuse-only with particulate recycle should be expected
also in the codisposal case. The trace metal balances for the various test
periods in Appendices B-H indicate that the actual quantity of metals in the
sludge is of the same order as that in the refuse. This suggests that test data
from refuse only periods would provide considerable insight into metal disposition
with sludge/refuse mixtures.
During the refuse-only, particulate recycle test, samples of the three
effluent process streams (product gas, slag, and wastewater) were collected
and analyzed for trace metals since it was felt that the handling of these
materials represents the greatest environmental challenge to any disposal
process. During the period the data in Table 33 were collected, operating
problems with the scrubber and electrostatic precipitator resulted in
performance at non-optimum levels,with particulate levels an order of
magnitude higher than those reported in Table 19 (although still well below
the Environmental Protection Agency regulation). In spite of this problem,
no undesirable metal concentrations were found in the combustion products
when combusted at the 12% CO_ level. Table 34 presents a comparison
between the trace metal concentrations in the combustion products-of
product gas obtained during particulate recycle with the time-weighted-
average (TWA) threshold level values (TLV) currently regulated. The regulated
levels in Table 34 are taken from the American Conference of Industrial Govern-
mental Hygienists and represent the current status of regulation for trace metals
in the work place. They are not strictly applicable to stack emissions.
75
-------
TABLE 33. PRODUCT STREAM TRACE METAL CONCENTRATIONS
WITH PARTICIPATE RECYCLE
Metal
Effluent Stream Concentration
Product Gas* Slag
Wastewater
Cd
Cr
Cu
Fe
Hg
Mn
Ni
Pb
Zn
0.017 ppm
0.069
0.60
12.4
0.064 '
0.091
0.44
0.29
3.6
9 ppm
322
3600
57000
0.88
3500
176
204
962
0 . 22 ppm
0.07
0.12
39.9
0.02
19.9
0.11
43.0
,83.7
*Computed for metal content measured on isokinetically collected particles.
TABLE 34. TRACE METAL CONCENTRATIONS IN PRODUCT GAS
COMBUSTION PRODUCTS
Combustion Product Concentration ( mg/ m"'
dry @ 12 percent COJ*
ฃป
Metal
Refuse +
Refuse + Sludge
(~s5% sludge (ds)/refuse)
TLV-TWA #
Cd
Cr
Cu
Fe
Hg
Mn
Ni
Pb
Zn
3
0.0022 mg/m
0.0090
0.079
1.62
0.0084
0.012
0.058
0.038
0.47
0.0001 mg/m
0.0007
0.0097
0.03
below detection limit
0.0004
0.002
0.002
0.009
0
0.05 mg/m
0.5
0.2
5
0.05
5
1
0.15
5
^Computed from metal content in isokinetically collected particulate in the
product gas.
+Refuse-only with particulate recycle and non-optimum gas cleaning.
^Based on average of measurements using primary and secondary sludges with
more optimum gas cleaning performance.
#ACGIH, National Safety News, pp. 83-93, September, 1977.
76
-------
As can be seen from Table 34, even under non-standard operating conditions,
the net emission of metals is well below the regulated levels.
Trace metals in the product gas have been detected on the
small amount of entrained particulate that remains in the gas following
the gas cleaning step. Vaporized metals in the gas can be neglected
based on the following considerations. The low vapor pressures of
the trace metals at the exit conditions of 38 degrees C and 10,133 Pa
(100 degrees F and 1 atm) are indicated in Table 35. Vapor pressures are
listed for the reduced metals and some of the chloride salts. If the partial
pressure is taken as the vapor pressure (generally the partial pressure is less
than the vapor pressure), maximum gas phase concentrations can be estimated
as shown in Table 35. The levels shown are essentially all below detection
limits except for mercury and mercuric chloride (HgCl_). Although direct
measurements have not been made, an estimate of the concentration can be
obtained by considering the total amount of mercury entering with the refuse
feed0 Measurements made during 1977 testing indicate a maximum mercury
concentration in the refuse of 0005 ppm. If all this mercury were to end up
in the product gas as vapor (an unlikely event as later discussion will indicate)
the resulting concentration of mercury vapor in the product gas prior to com-
bustion would be less than 0.08 ppm. After combustion at standard conditions
of 12% CO,.,, this translates to about 0.01 ppm in the combustion products
which is well within the regulated value. An upper limit on mercury
concentration in the feed can be estimated using the TLV value of 0.05 Mg/m .
Based on this TLV, the maximum allowable mercury level in the feed to the
PUROX System would be 0.22 ppm if the disposition is assumed to follow a
similar pattern of all ending up in the gas (see Table 33). Based on these
considerations, the trace metal content of the particulate was judged to be
the only significant emission of metals in the gas phase.
The wastewater contains a significant amount of trace metals as can be
seen from Table 36. Since the organic content of this stream is also signi-
ficant, it will be processed through an activated sludge wastewater treatment
plant to reduce the organic content to an acceptable discharge level. Pre-
treatment regulations for wastewater treatment plants are currently being
prepared by regulating agencies. Although these pretreatment regulations
have not been formally issued, the pretreatment requirements shown in Table
36 represent current thinking regarding acceptable heavy metal concentrations.
These requirements can be readily met through the use of a simple chemical
precipitation technique such as CaO addition. The metals are then essentially
removed from the water phase and concentrated in a small amount of chemical
sludge as insoluble hydroxides. This small quantity of sludge can be either
recycled to the hearth of the converter where it will become part of the slag or
it can be disposed of in a approved chemical landfill depending upon the
relative cost of either approach. As can be seen from Table 36, chemical
precipitation essentially reduces the heavy metal content of the wastewater to
the analytical detection limits based upon solubility product calculations.
77
-------
TABLE 35. TRACE METAL VAPOR PRESSURES AND MAXIMUM
GAS CONCENTRATIONS .
Component
Vapor Pressure*
38 degrees C
(100 degrees F)
Maximum Gas Vapor Concentration
38 degrees C, 10,133 Pa
(100 degrees F, 1 atm.)
Cd
Cr
Cu
Fe
Hg
Mn
Ni
Pb
Zn
Cu2Cl2
FeCl2
Hg2Cl2
HgCl2
PbCl2
<10~J^atm.
-------
The slag concentration of trace metals with particulate recycle is
compared in Table 37 with the slag obtained using refuse-only without
particulate recycle and slag from refuse plus sludge tests. Significantly, the
Pb, Zn, Cd and Hg all show substantial increases following particulate recycle
to the hearth adding support to the mechanism of heavy metal capture in the
slag phase. Although leaching tests were not conducted using the slag from
the particulate recycle period, there is no reason to expect that its excellent
non-leaching properties would differ significantly from the results previously
reported.
Table 38 indicates that certain volatile metals (Cd, Pb, Zn, and Hg) are
less effectively retained in the slag than other less volatile metals (Cr, Cu, Fe
and Ni). However, the relatively low product gas temperature combined with the
efficient gas cleaning in the scrubber and ESP results in a very small amount of
these metals being lost into the product gas. The combination of efficient
particulate removals from the product gas and the approximately ten-fold dilution
of the gas during combustion diminishes the stack emission levels to a point far
below that for typical incineration devices. The net result is a minimal potential
hazard from stack emission of trace metals.
The mechanism of capture of these volatile metals in the slag as
indicated in the data here is believed to be related to the formation of
relatively stable inorganic salts which become physico-chemically bonded
within the silicate matrix of the slag. Of the metals not in the slag,
all but Hg are essentially in the wastewater where they can be easily
separated. The actual amount of Hg is so very low that the 15% of the total
that does leave with the gas in the particulate results in a combustion product
emission that is an order of magnitude below the regulated limit. Further
cleaning of the gas could remove most of the residual amount if it is required.
The overall conclusion from this initial examination of data from
refuse-only operation with particulate recycle indicates an environmentally
sound handling of the trace metals in the PUROX System.
79
-------
TABLE 37. TRACE METAL CONCENTRATIONS IN SLAG
1
Metal
Cd
Cr
Cu
Fe
Hg
Mn
Ni
Pb
Zn
Refuse
5.4 ppm
3400
4000
61935
0.3
1600
304
95
310
Refuse +
primary sludge
[~5% sludge (ds)/refuse]
7 ppm
2081
3771
71672
5
1399
1614
141
609
Range
(5-9) ppm
(650-3400)
(3300-4300)
(5300b-H4000)
(0.02-10)
(1200-1700)
(960-2700)
(80-240)
(400-1100)
Refuse + mixed
sludge [<~'5% sludge
(ds)/refuse]
8 ppm
1853
3338
51053
0.03
1274
174
118
403
Refuse
[Parti culate
Recycle]
9 ppm
322
3600
57000
0.88
3500
176
204
962
TABLE 38. TRACE METAL DISPOSITION IN PROCESS EFFLUENT STREAMS
Metal
Percent of Total
Metal Emitted in
Product Gas
Before Combustion
Percent of
Total Metal
Emitted in Slag
Percent of Total
Metal Emitted
In Wastewater*
Cd
Cr
Cu
Fe
Pb
Mn
Ni
Zn
Hg
0 . 50 percent
0.05
0.04
0.05
0.30
-
0.60
0.80
15.0
96.4 percent
, 99.92
99.95
99.85
77.60
99.20
99.30
88.70
82.50
3. 10 percent
0.03
0.01
0.10
22.10
0.80
0.10
10.50
2.50
* Trace metals would be precipitated prior to wastewater treatment.
80
-------
SECTION 13
PROJECTED ECONOMICS OF THE CODISPOSAL PROCESS
The test program conducted at the South Charleston PUROX facility
determined the performance of the PUROX System while processing relatively
dry sludge to refuse ratios (about 0.075) and modest solids contents (15 to 30%)
in the sludge. These conditions are appropriate for a community with a small
amount of sludge to dispose of 45-63 Mg (50 to 70 tons) per day of dry sludge
and access to a substantial supply of refuse. Under these conditions of
operation, the PUROX facility economics are basically determined by the
alternatives open to the community for refuse disposal. Only if the refuse
disposal economics are attractive will the sludge disposal be attractive.
For the economics to become less dependent on the allowable cost of
refuse disposal the ratio of dry sludge to refuse must be increased. Under
these conditions the economics become more sensitive to the economics of
sludge disposal. It further has the advantage that larger quantities of sludge
can be disposed of in a facility of reasonable size. Less refuse is needed-and
the allowable cost of disposal of refuse becomes less critical to the overall
economic attractiveness.
Projections of sludge disposal costs are provided for facilities
processing four different sludge solids contents (cases A-D). Case A is
based on sludge dewatered to 20 percent solids and delivery of the 20 percent
sludge cake to the PUROX facility. Case B is based on sludge dewatered to
25 percent solids and delivery of the sludge cake to the PUROX facility. For
these cases, the PUROX facility is operated in a similar manner to the test
program in which separately dewatered sludge is fed to the converter along
with municipal refuse.
Economics and practical considerations favor removing as much moisture
as is reasonable in a combination of dewatering and drying steps before intro-
duction of the sludge with refuse into the PUROX converter. This minimizes
use of the PUROX converter and gas cleaning capacity for processing moisture.
It maximizes the use of the PUROX System capacity for the critical task of
producing energy and disposing of the sludge solids. Cases C and D are
included in the projections of sludge disposal costs to" reflect improvements in
the codisposal process, as compared to that which practiced at South Charleston,
81
-------
Case C is based on delivery of 3 percent sludge solids to the PUROX facility.
At the facility, the 3 percent sludge solids would be combined with the PUROX
System scrubber slurry and jointly dewatered to 35 percent solids with a belt
filter press. The scrubber slurry contains char, produced from the pyrolysis
of refuse and sludge, which enhances the dewatering characteristics of sludge.
Bench scale evaluations of the dewaterability of the char and sludge mixture
have shown significantly reduced polymer requirements from that of separate
sludge dewatering and substantially higher cake solids following the filtering
process.
Case D is based on codewatering of sludge and char to 35 percent as in
Case C and drying the 35 percent cake solids in a separate dryer to approxi-
mately 70 percent solids. In cases C and D the char/sludge mixture, after
dewatering (and drying in Case D), is fed into the converter together with
municipal refuse. To properly compare these cases, all costs associated with
dewatering from 3 percent solids and ultimately disposing of the sludge have
been included. Dewatering costs were taken from published literature as $55
to $66 per dry Mg ($50 to $60 per dry ton) for dewatering from 3 percent to
25 percent solids.*
Table 39 is provided to identify: capital equipment and costs, amorti-
zation of capital cost, annual operating costs and revenues, and the net sludge
disposal cost for a facility designed to codewater, dry and process 363 dry Mg
(400 dry tons) of sludge per day (Case D). The ratio of dry sludge to as-
received refuse is 0.30. The capital and operating costs are estimates based
on recent proposals submitted by Union Carbide and recent price quotes from
filter press and dryer equipment manufacturers.
The estimated net sludge disposal cost of $44/dry Mg ($40/dry ton) is
based on assumed facility revenues from the sale of fuel gas at $2.23/GJ
($2.35/MM Btuj", the sale of ferrous metal at $44/Mg ($40/ton), and a refuse
disposal fee of $16.50/Mg ($15.00/ton) of as-received refuse.
The PUROX System includes the equipment required for gasification
of the feed material and gas cleaning. The oxygen plant includes the
equipment required to supply high purity oxygen to the PUROX System
converters. The wastewater pretreatment system includes equipment for
storing and adjusting the pH of the PUROX System wastewater prior to discharge.
The wastewater is assumed to be returned, based on a diurnal cycle, to the
wastewater treatment plant. Minimal costs are anticipated for the diurnal
treatment of the wastewater; therefore, no costs were allocated in the
projected economics.
*(7) Van Note, R.H., et al, "A Guide to the Selection of Cost Effective
Wastewater Treatment Systems," EPA-430/9-75-002 , July, 1975
82
-------
TABLE 39. ESTIMATED ECONOMICS - PUROX CODISPOSAL FACILITY
Design Capacity: 363 dry Mg of sludge/day
(400 dry tons of sludge/day)
dry sludge to refuse ratio of 0.30
Capital Costs (in thousands of dollars)
PUROX System, Oxygen Plant, Wastewater
Pretreatment System 57 ,000
Front End System 15,000
Site Facilities 13,000
Buildings and Site Improvements 4,000
Codewatering and Drying Equipment 22,000
TOTAL ' 111,000
Annual Operating Costs (in thousands of dollars)
Facility Operation and Maintenance 6,370
Facility Power and Miscellaneous Production Material 7,330
Levelized Amortization of Capital 10,500
TOTAL 24200
Annual Revenues (in thousands of dollars)
Refuse Disposal 7,870
Fuel Gas 9000
Ferrous'Metal 1470
TOTAL . 18340
Annual Sludge Disposal Cost (in thousands of dollars) 5860
Net Sludge Disposal Cost $44/Mg
of dry sludge
($40/Ton)
of dry sludge
83
-------
In this case of 363 Mg/day (400 tons/day) of dry sludge at a sludge
dry solids to refuse ratio of 0.30, 807.8 m3/day (213,400 gal/day) of waste-
water with a BODs load of about 30,000 ppm would be returned to the adjacent
wastewater treatment plant which was producing the 363 dry Mg (400 dry tons)
per day of sludge. If it is assumed that about 0.9 Mg (1.0 ton) of dry sludge is
produced from 3785 m3 (10 gallons) of wastewater, with an influent concentration
of 200 ppm BOD5, then the wastewater treatment plant will handle a flow of
1.514 x 106 m3/day (400 x 106 gal/day). The total BQD5 into the wastewater
plant (excluding the PUROX System flow is 302.8 Mg/day (333.8 tons/day).
The BOD_ from the PUROX System codisposal operation would add 24.2 Mg/day.
This represents an 8.0 percent increase in load on tiie wastewater treatment
plant. Since wastewater plants are sized with about 30 percent additional
capacity to handle diurnal variations, storage of the PUROX wastewater for
addition at appropriate times in the diurnal cycle could be accomplished with
minimal cost. .
The front end system includes equipment for refuse receiving, storing,
shredding, magnetic metal separation and delivery to the PUROX System. Site
facilities include the PUROX System control building, electrical power distri-
bution equipment, process and utility interconnecting piping, cooling water
system, a fuel gas compression system to compress the PUROX System fuel
gas to 207 KPa (30 psig), and a process steam boiler. Buildings and site
^improvements include personnel and administrative buildings and site
preparation prior to construction. The codewatering and drying equipment
include belt filter presses for codewatering the char sludge mixture to 35 per-
cent solids, dryers to dry the mixture to 70 percent solids, and sludge con-
veying equipment.
The facility annual operating costs include operating labor and super-
vision, maintenance, facility power at $.03/KWHR, miscellaneous production
materials and the amortization of the capital cost at 7% for 20 years.
Figures 21 through 32 are provided to illustrate estimated net sludge
disposal costs for various sludge solids contents, refuse disposal fees, and
facility throughputs 0 The net sludge disposal cost is the difference between
total facility capital and operating costs and facility revenues expressed in
dollars per dry Mg or dollars per dry ton. The cost includes dewatering
(from 3% clarifier solids), processing, and ultimately disposing of a dry Mg
or dry ton of municipal sludge at the PUROX facility. The estimated net sludge
disposal costs for all the figures were calculated in the same manner and with
the same assumptions of electric pdwer costs, and amortization rate as in
Table 39.
The figures are based on processing three different sludge to refuse
ratios; 0.05, 0.3, and 0.6. The net sludge disposal cost for each of the
three ratios is shown on the basis of refuse disposal fees of $5.50/Mg
84
-------
($5.00/ton) and $16.50/Mg ($15/ton) of as-received refuse. The revenues
from the fuel gas and ferrous metal sales are based on the assumptions used
in Table 39.
All cost estimates and economic projections of sludge disposal costs
(1977 dollars) are budgetary based on general design information. Site and
project-specific costs such as financing costs, capital escalation and
interest during project construction, land cost, insurance cost, consulting
fees and miscellaneous legal fees are not included.
The sensitivity of the sludge disposal costs to the allowable refuse
disposal fee is illustrated in Figures 21-32 At very low dry sludge to refuse
ratios refuse disposal economics control rather than those of sludge disposal0
If a refuse-only facility can be economically justified then sludge can be
disposed in the facility with very attractive disposal costs. If refuse-only
cannot be justified the relatively small amount of sludge represented by a
0.05 dry sludge to refuse ratio will not appreciably alter the overall economics.
This is illustrated by comparing the sludge disposal cost for allowable refuse
disposal costs of $5.50 and $16.50 per Mg ($5 and $15 per ton).
Figures 21-24 illustrate the economics of sludge disposal based on a
sludge-to-refuse ratio of 0.05 which is equivalent to the national per capita
ratio of sludge generation to refuse generation. Figures 25-32 illustrate the
economics of sludge disposal based on operating at a higher sludge (dry
solids)-to-refuse ratio than the national per capita. Many municipalities
may require a higher sludge-to-refuse processing ratio because of limited
refuse availability at a given site (generally below 1814 Mg (2000 tons) per
day) and large quantities of wastewater processed in one wastewater treatment
facility. Figures 25-28 are provided to indicate estimated sludge disposal
costs for a 0.30 ratio of sludge-(dry solids)-to-refuse. Figures 29-32 are
provided to indicate estimated sludge disposal costs for a 0.6 ratio of
sludge-(dry) solids)-to-refuse. At these ratios, large quantities of. sludge
can be processed with substantial reductions in the quantity of refuse required
relative to the results presented in Figures 21-24.
A study of the figures illustrates that the lowest net sludge disposal
cost, based on refuse revenues of $16.50/Mg ($15/ton), occurs at the lower
ratio of sludge to refuse (0.05). This results because at the low ratio the
PUROX facility economics are determined by the refuse disposal fee. For
refuse disposal fees of approximately $16.50/Mg ($15/ton), disposal of small
quantities of sludge will be attractive. The refuse disposal fee is specific
to the location and may be more or less than $16.5Q/Mg ($15/ton). Lower
refuse disposal fees, such as those figures provided on the basis of $5.50/Mg
or $5/ton of refuse, result in higher sludge to refuse ratios (0.60 case)
becoming relatively more favorable.
85
-------
o
(Nl
oo
o
o
to
fd
o
CD
0
D.
Cn
O
a
J3
en
0
-.
0
0
0
g
ฃ
o i-
o to
w
O
O
r <
ffl
U?
0
0)
0
w
CM
0
LO
CO
(Nl
jo
q.soo
86
-------
o
o
o
o
o
O
CO
o
O __ s
to >,
i i (0
a
a
o
a Q,
O r; O J2
to 5 o en
t! cxi 3
^ ซ-" O
fe
a
2
0)
ra
0
en
0
-a
0
o
0
S-H
O
cxi
O
o
I
0
o
LO
1 1
o
0
i ,
o
o
CO
1 J
0
CD
CXI
f '
0
o
I 1
1
o
o
CD
1 t
w
o
o
la
m
O
a
CO
CD
0
c
(0
S
-f-j
-M
CO
H
0
m
jo uo^ Aap/$) ^soo lesodsfp aBpnfs
87
-------
o
o
o
o
CO
O
o
to
CD
tn
Q.
CM
>-
_ง
C
Cn
CD
03
O
-------
0
o
4->
0
en
Q)
rt
-3 o
LO o
o
LO
o -co-
< CQ O
o o
o o
r-l O
CN1
o
CO
o
o
to
ro
c
O
(0
o
fe
c
o
Cn
O
s
030
"* | g
4->
0
Cn
ii
ra
CD
O
O
O
w
j-J
CO
O
O
r I
fO
cn
O
D,
as
Q)
w
^->
0
c
T3
0
"(0
a
-r-4
w
-------
o
p
.2
j->
ID
(0
m
O
a
en
tn
T3
CD
G
-0
CD
H
(0
6
-i-t
*j
CQ
H
LO
CM
CD
>-.
CD
-i-i
tn
o
o
-------
(0
CD
I
CD
S-H
s
CD
o
CO
C
O
<
O
o
LO
o
Q
o
o
LO
10
to
o
O
co
CO
CO
O
o
o
o
CM
^
(0
*"O
CD
a
c
^J
>1
3i
^j
I
CD
0
i __i
0
FT^
fO
-a
S-i
CD
Q.
C
O
o ii
o
o t;
-1 a
Cn
b
fj
0
w
a
0
r^ j_,
10
10 0
>
"0
O
0
I
rd
U)
O
a.
CO
^a
0
Cn
-a
i i
w
0
a
"O
(0
S
-r-1
W
P-l
CO
CSI
0
CO
W
o
o
CO
co
CO
PL,
O
O
CM
O
LO
O
O
CD
LO
(efipnjs jo ucq Ajp/$) isoo insodstp aBpnfs
91
-------
(0
t-4
0
CO
*0
2
Q)
D^
JS
co
0
M-l
O
" o
o LO
CO
LO
O -CO-
to
CO
CO
-------
O Q
CD
CD
Cn
o
o
LO
CD
CD
o
o
CO
CO
co
to
d
fe
a
c
o
o fc o "3
o n, o 5,
co o J3
S ""^ Cn
O -3
><
s
Cn
O
O
[>,
CD
(D
CD
CO
CD
i-H
73
(D
CD
0 O
tn a)
T3 i-
5 w
CO
(0
w
W
CXD
CM
CD
S-i
o
o
CO
CO
co
O
O
CM
O
LO
CD
O
O
UO
jo uo} Ajp/$) q.soo
93
-------
o
ID
CD
LO
CO
CO
CO
J-.
CD
G.
o
CO
o
LO
o.
a
tn
o
0
(U
o
CO
w
oo
oo
CO
o
o
I 1
(0
w
a
CO
'I I
T3
CD
w
-a
CD
+J
(0
w
-------
CO
i-l
ffl
CO
5
0)
Cn
o
10
CD
W
G
O
<
O
o
*
LO
II
CO-
o
LO
CO
CD
Cn
Ti
03
to
to
o
CD
O
CO
O
co
o
LO
o
CO
o
O3
(86prv[s jo no} Ajp/$) }soo jesodsip
oo
CO
LO
-a
o to
oo 73
S-H
CD
a
a
2
>.
6
o
LO
C-J
o
o
a
o
a
Cn
CD
CQ
CD
CD
-1-4
^->
ro
S
H
O
CO
CD
S-H
3
D)
95
-------
O
O
.2
4-1
a
0)
w
2
0)
o
to
o co
in co
CO LO
(0
T3
to
p.
0)
o *'-'
LO 4-1
I
Cn
1
3 8
I ซ3
ซ 0
O.
Cn
(D
CO
-------
p
o
-*-*
+->
(0
CD
I
2
CD
CD
en
S-i
M
O
C
O
4-1
o
o
LO
O
o
CO
to
o oo
LO co
'CO LO
^
>, C
ง-งงฃ
" fe ^^
a g,
c ^
o H1
a,
c> ^
LO en
~
CD
CO
CD CO
O CO
CXI CO
CD
O
CD
o
o
LO
o
LO
LO
jo uo^ Aip/$) isoo IB sod sip eBpnjs
8
r1
td
CO
a
CD
CD
C
T3
CD
4-ป
(0
W
w
CO
CD
97
-------
REFERENCES
1. Moses, C. T. and Rivero, J. R. "Design and Operation of the PUROX
System Demonstration Plant, " Fifth National Congress on Waste
Management Technology and Resource Recovery, Dallas, Texas.
2. Federal Register, Vol. 36, No. 247, December 23, 1971.
3. "Background Information for New Source Performance'Standards",
Vol. 3, EPA Report 450/2-74-003, APTD-13526, February, 1974.
4. Federal Register, Vol. 40, No. 248, December 24, 1975.
5. National Safety News, pp. 83-93, September, 1977.
6. Lange's Handbook of Chemistry, llth ed., McGraw-Hill (1973).
7. "A Guide to the Solution of Cost Effective Wastewater Treatment
Systems," EPA Report 430/9-75-002, July, 1975.
98
-------
APPENDIX A
ANALYTICAL PROCEDURES
All streams crossing the system boundaries were sampled. In general,
a number of samples of each stream were taken during each run (typically, one
week), and these were then composited for analysis. The streams sampled
are listed below, and each is described in more detail later.
Refuse
Sludge
Slag
Wastewater
Product gas
Char
Particulates from isokinetic samples
Slag Leachate
REFUSE
Refuse analyses were subcontracted to Standard Laboratories of
Charleston, WV. Ultimate and proximate analyses, as well as determinations
of heating value, were performed on each composite. These were done in
accordance with ASTM methods D3176-74, D3172-73 and D3286-73. Trace
elements (see listing in Table A--1) were determined (generally) by atomic
absorption.
SLUDGE
Sludge analyses were also subcontracted to Standard Laboratories.
Again, ultimate, proximate, heating value, and trace element analyses were
performed. Standard methods as noted above were used.
SLAG
Slag analyses were performed at the Union Carbide Corporation Contract
Scientific Laboratory in Tarrytown, NY.
Majors were determined by X-ray fluorescence. The samples were
ground to a fine powder, then fused in a Pt/Au crucible in a lanthanum oxide/
lithium tetraborate matrix. This was then poured into a crucible mold. A
polished disc was made, and the analysis was then performed.
99
-------
TABLE A-1. DETECTION LIMITS FOR TRACE ELEMENTS
Constituent
Zn
Cu
Cd
Ni
Fe
Hg
Pb
Cr
Mn
Cl
F
P
As
Be
Ba
Se
Ag
Typical Detection Limits ,
S. Charleston Lab
1.0
0.1
1.0
0.1
0.1
0.05
0.1
0.1
0.1
1
:
; i.o
0.1
1.0
1.0
-
0.1
ppm
Tarrytown
0.03
0.04
0.02
0.04
; 0.04
0.003*
0.05
0.04
0.02
0.01
0.05
0.1
0.1
0.03
0.03
0.05
0.03
*Lower limits can be attained if necessary.
100
-------
Trace metals were determined by induction coupled plasma (ICP). The
samples were dissolved in HF/HC1 in a Teflon bomb @ 240 degrees C. The
resulting solution was diluted with water, then aspirated into the sample
chamber. The plasma is formed in argon carrier gas. (ICP is similar to
emission spectroscopy, except that liquid samples can be analyzed. Also,
since sample is energized in a plasma, no electrodes are required.)
Carbon samples were burned in Oo at 1500 degrees C in induction
furnace using an iron accelerator. COo is captured on a molecular sieve. When
combustion is complete, the molecular sieve is heated very rapidly, and the
CC>2 is analyzed using a thermal conductivity sensor.
Sulfur usually determined by burning sample in induction furnace with
iron accelerator at 1500 degrees C. The 309/803 produced is absorbed in a
Kl/starch solution, then titrated with potassium iodate.
Chlorine, fluorine samples were ground, then pyrohydrolyzed at 1000
degrees C with steam, using vanadium oxide catalyst. The HF/HC1 formed
were distilled over, trapped in a water solution, and continuously titrated
with NaOH. At the NaOH end point, aliquots were taken. Chlorine was
determined by titrating potentiometrically with silver. Fluorine was determined
by buffering the solution, then using an ion-selective electrode.
Mercury samples are prepared by grinding to a fine powder, then
digesting in a Teflon bomb at 150 degrees C with HF/H2SO^/KMnO4 solution.
The permanganate is reduced with hydroxylamine hydrochloride (NH^OH* HC1),
then the sample is placed in the sample holder of the atomic absorption unit
and reduced in situ with stannous chloride. The measurement is made via a
cold vapor technique.
WASTEWATER
Wastewater was normally analyzed by personnel at the Union Carbide
Chemicals and Plastics facility in South Charleston. Certain samples were
run independently at the Central Scientific Lab in Tarrytown.
Grease and oil were separated by extraction, then determined
gravimetrically. Suspended solids (weight) determined by filtering sample,
then weighing. COD is determined by refluxing sample with dichromate/
sulfuric acid solution. Mercury/silver salt complex is added during digestion
to catalyze the reaction. Excess dichromate is titrated with ferrous ammonium
sulfate. BOD determination is a complex procedure. Complete description may
be found :in the latest edition of APHA "Standard Methods for Examination of
Water and Wastewater" . Essentially, the procedure calls for seeding the
sample with microorganisms and incubating for five days. Residual dissolved
oxygen can then be determined, normally by using the azide modification of
101
-------
the iodometric method. TOG is determined by instrumental means. The carbon
is burned in an oxygen stream, and the resulting CC>2 is then measured.
Mercury is determined in the same fashion as previously described, except
that the sample preparation steps are not necessary. Chlorine and fluorine are
determined by argentometric titration and ion-selective electrode, respectively.
pH is measured with a pH meter. Trace metals measured directly with ICP.
Suspended Solids
For trace metals samples were dried at 105 degrees C, dissolved using
the Teflon bomb technique, then the solution was run by ICP. Mercury and
sulfur were analyzed using the techniques described previously.
For C, H, N the samples are mixed with a catalyst and burned at 1000
degrees C. The gas, containing H2O, CO2 and NOX is passed through a tube
containing Cu metal at 600 degrees C to convert NOX to N2 . The gas then goes
into a gas chromatography column, and the components are analyzed via a
thermal conductivity detector.
PRODUCT GAS
Gas samples were not composited. Normally, several samples were
collected twice per day. Helium was metered into the gas upstream of the
sample point, prior to collection of each gas sample, to provide a method of
estimating the gas volumetric flow. The samples were analyzed by mass
spectrometer. Typically, all compounds in concentrations greater than 0.001
volume % were reported - this usually included up to about 35-40 compounds.
Analyses of On and natural gas were also done by mass spectrometer to
establish baseline conditions.
CHAR
Char analyses were subcontracted to Standard Laboratories,. Analyses
were the same as those for refuse and for sludge.
PARTICULATES
Environmental Protection Agency Method 5 was used to collect the
product gas. The isokinetic technique was used to calculate the gas
volumetric flow rate, and to collect particulates to determine grain loading
in the gas. The total paniculate catch was typically quite small, reflecting
very low grain loading in the gas . ,
Trace metals were determined by dissolving the sample in HC1/HNO3.
(If organics did not decompose, perchloric acid was added.) The solution was
then run by ICP. Where sample size permitted, mercury, sulfur carbon,
hydrogen and nitrogen analyses were run.
102
-------
SLAG LEACHATE
To ascertain whether PUROX slag is inert, samples of slag were
agitated with distilled water for up to two weeks. Analyses were similar to
those for the wastewater samples .
A list of analytical instruments that were used is given in Table A-2.
TABLE A- 2. ANALYTICAL INSTRU MENTS
X-Ray Fluorescence
Atomic Absorption
Carbon/Hydro g en/
Nigrogen Analyzer
Carbon Analyzer
Sulfur Analyzer
Pyrohydrolysis Unit
Induction Coupled Plasma
A Siemens SRS-1 sequential X-Ray fluorescence
spectrometer, with Siemens K-4 X-ray generator,
chromium X-ray tube, and PDP-11 minicomputer
was utilized.
A Perkin Elmer 460 atomic absorption spectro-
photometer was used.
An F & M Scientific #185 (Hewlett-Packard) C-H-N
analyzer (uses He carrier gas, oxygen generated
catalytically) was employed.
LEGO induction furnace and Angstrom Carbomatic
#523C (uses 02 for combustion and as carrier gas)
was used.
LEGO induction furnace #5213 was used.
Nickel tube in Lindberg Hevi-Duty Furnace #2000
was used to carry out the following analyses:
Fluorine analysis - Orion ion-selective electrode
in pH meter.
Chlorine - Orion silver sulfide electrode in pH meter.
Jarrell Ash Cat. No. 90975 (specially modified by
J. A. for Tarrytown Laboratory) was used.
103
-------
APPENDIX B
TEST PERIOD 5. REFUSE ONLY - 25 MAY - 1 JUNE 1977
During the period between 25 May and 1 June the PUROX System was
operated on refuse only to provide a baseline for comparison with sludge
processing. An overall material balance was calculated for the refuse only ''
baseline period to provide a reference for the sludge processing results.
The balance is shown in Table B-lซ The slag and metal streams were deter-
mined using the methods described in Section 8. The dry, inert free, gas flow
rate was calculated using the helium trace data. Argon was added to the gas
flow rate based on the average gas composition determined for the period.
Nitrogen equal to the mass of nitrogen in the oxygen supply was also added to
the product gas. The gas stream was then saturated with water at 21 degrees
C (70 degrees F). The sum of dry gas, nitrogen, argon, and water masses was
the total gas flow rate. The dry char production rate was correlated with the
converter operating rate to obtain a functional relationship between char
production rate and operating rate. Because the oil stream was not measured
it was assumed to be 2 percent by weight of the incoming refuse based on
previous measurements.
The water stream was calculated by difference between the entering
and exiting masses. As a check the effluent production in gallons per ton of
refuse processed was calculated and compared to previous balances performed
on the system. The calculated result was 0.346 cubic meter per Mg (83
gallons per ton) which compared well with previous measurements which fell
between 0.334 and 0.417 cubic meters per Mg (80 and 100 gallons per ton).
3
Gas production during this period was 437.1 Nm (normal cubic
meters, 0 degrees C, 101 KPa) [14,800 Sft (standard cubic feet, 60 degrees F,
one atmosphere)] of dry, inert free gas per Mg (ton) of refuse and fuel gas
consumed. The ratio is the mean of six samples which ranged between 410.5
and 466.6 Nm3 per Mg [13,900 and 15,800 Sft3/ton]. The pitot tube result
was 325.6 Nm3/ Mg (11,026 Sft3 per ton) of feed. Significant error is possible
in the calculated average gas flow,rate. The standard deviation of 12 pitot
tube measurements taken over a two hour period on 26 May was 97 Nm^/Mg
(3285 Sft3 per ton).
104
-------
"c
T
4-
ฃ
LO
Q
C
p:
OH
1
.
C
t^-
,
ft)
Q
i-H
*5i
~1""1 *"^ ^5
to v^
CQ
S
0^
JH 'g
"o S.
g
(0
0)
CO
[v.
1 LO 1 1 1 1 'I
r 1
CD t^.
1 ,_* 1 1 1 1 1 0*
CM
1 "~1 1 1 1 1 11
o
CD IS. ^
1 O ' ' ** 1 ' O
i-H i 1 CM
. CD CO ID
i ซl ซll 1 9
*-> O r-H
LO ^ tD
1 ID 1 i-l 1 1 1 ^
i-H f-H
LO C^s i-H CO CS3 <7)
CO 1 ' i-H LO O tD [V.
l>- ซ-) CD CO
CD
0 T3 %
CD CD CQ rfj
tl, 1 1 * ^ _j V^
fl) *- "V1
I11!?!! 1 i
K; co CM O H iง P-I
ts t~i
*? r-l ' '
i 1
r_|
1 1 1 -vl*
CM
1 ฐ. 1 ^
-H 0
CO <^ ^
,4100
1 CM i-H
' 0 ซ
o o o
CO
0
Di Jo to
_t~i 'Til fd
co 0 0 ^
1
1
1
1
1
CO
o
CD
4->
p
o
1 1
(d
4->
O
H
105
-------
^r
i
(0
H P
0 -* ^
! I 1 1 II
i-)
r-H OO
I > to OO OO
t 1 1 i 1 CD CD CO r-l
OO i-l CD **
CD
-i-l
CD ^ W
o -a to
CD Q) to O
- IM (0 p; C
Di PM CD H ง eu
106
CM r-l
CO* -! I '
i-H
1 1 1 CO '
1 | 1 "1 1
O
CO CM L1*1*
1 * * 1
r-l CM r-l
' O O O '
CO CO CM LO
O r-l i 1 O '
LO OO CO r-l CO
OO CM r-l OO CD
r-l CM CD
+J
O
s-i CD '
C7> (rj +J . fO
11 n X -^ ^
-------
The oxygen consumption was 0.201 Mg (tons) oxygen per Mg (tons) of
refuse plus fuel gas. The consumption ratio includes fuel gas in the denom-
inator because the ratio is a partial function of the fuel gas rate and because
this method is consistent with previous studies. The ferrous metals separation
ratio was 0.08 Mg (tons) of metal per Mg (ton) of refuse. This ratio is
primarily dependent on the metals content of the incoming refuse. Slag
production was 0.23 Mg (tons) of slag per Mg (ton) of refuse.
Component balances were performed on the system for major
constituents, carbon, hydrogen, oxygen, sulfur, and water. A balance was
also performed on the noncombustible portion of the incoming streams This
balance includes unseparated metal, glass, and ash; the component was named
"ash" . The assumptions for the component mass balance are given in Table
B-3.
TABLE B-3. ASSUMPTIONS FOR COMPONENT MASS BALANCES
(1) Wastewater stream COD is approximately 50,000 ppm.
(2) Soluble organic compounds in the wastewater have the C:H:O ratio of
acetic acid (CH,COOH).
o
(3) Water produced by combustion and pyrolysis in the hearth is 0.042
m3/Mg of feed (10 gal/ton).
(4) All SiC>2 in the slag entered as oxide. Other oxides present were
oxidized from metal in hearth.
(5) The ESP oil has a composition of 80% carbon by weight, 10%
hydrogen, and 10% oxygen.
Product gas composition and ultimate analysis were determined by
averaging all gas samples obtained during the test period. The result is
shown in Table B-4. The slag composition used was the sample result from
this test period. The oxygen balance was calculated from the oxide concen-
tration in the slag. Oxygen present in the slag as SiO2 was assumed to have
entered the process as glass and was included in the "ash" balance. The
char composition was the average of three samples taken during the period.
Oil was assumed to have the composition listed in Table B-3. The
composition of the water stream was determined by difference.
107
-------
TABLE B-4. AVERAGE PRODUCT GAS ANALYSIS
Test Period 5
Mole Percent
H2
CO
r-w
V /Xl ป
4
C2H2
C2H4
C2H6
C3H6
C3H8
C4
Q
5
f TT
ฐ6 6
HHC
H2S
CH-OH
3
CO0
2
Q
2
Ar
28.14
37.28
6.21
0.46
1.90
0.44
0.32
0.23
0.40
0.29
0.24
0.26
0.02
0.11
23.07
0.03
0.59
The first set of balances calculated used the average refuse com-
position determined from the three refuse samples analyzed. The mean
composition was used because thekhree samples exhibited substantial
variation in moisture, carbon, and ash content. The component balances
obtained using the mean refuse analysis were unsatisfactory. The incoming
ash was insufficient to account for the ash contained in the slag. The
entering carbon was much more than the carbon contained in the product gas,
char and oil. Assignment of all excess carbon to the effluent water resulted
108
-------
in a concentration which was inconsistent with previous experience and with
the results of wastewater samples obtained. Similar results were obtained
during all periods of sludge processing when the mean refuse analysis was
used.
Because use of the refuse analyses obtained resulted in poor
component balances a different method was chosen to calculate typical
balances. Several assumptions were made which specified the effluent
stream. The assumptions are listed in Table B-3. The composition of refuse
necessary to balance the system was then calculated and compared to typical
published refuse analyses. The component balances obtained using this
technique are shown alongside the overall balance in Tables B-l and B-2.
The hydrogen, oxygen, and water do not balance separately. This is
due to the production of water in the pyrolysis and combustion reactions. The
missing mass of hydrogen and oxygen can be found in the water column.
Balances were performed on 9 trace metals of particular interest. The
assumptions used in these balances are given in Table B-5. Analyses were
performed for several other trace components , but they were either below the
sensitivity of the particular analysis or were not of special interest. The
results are shown in Tables B-6 and B-7.
TABLE B-5. ASSUMPTIONS FOR TRACE METAL BALANCES
(1) Trace metal analyses are on a wet basis
(2) Char was 40% moisture by weight
(3) Scrubber flow was 0.38 m^/min (100 gal/min)
(4) Condensation in the scrubber is 0.023 m /min (6 gal/min)
The incoming quantities of the metals were determined using the
average concentration in the three refuse samples taken. Since the average
refuse analysis was unrepresentative for major components it was probably
also unrepresentative for trace elements. For this reason the balances are
probably inaccurate, and the distribution of the trace metal between slag,
char and water are of greater significance. The amount of trace metals in the
slag and wastewater exit streams were calculated from the analysis of
109
-------
samples taken during this test period. Unfortunately the dilution of the
wastewater by makeup water to the scrubber makes the metals content of the
wastewater stream uncertain. Char metals content was based on the average
of three samples obtained.
Due to dilution effects the concentrations of cadmium and mercury in
the effluent water were below the detection limits for their respective analyses.
Since significant quantities of metal could be contained in these streams below
the detection limit, no conclusion can be drawn from this test concerning the
ultimate distribution of these two metals. The char stream contained a large
fraction of the exiting amounts of lead and zinc. The distribution of these two
metals between the exit streams under particulate recycle conditions was not
known, but for both these metals the amount in the slag is several times the
amount in the wastewater. For the remaining five metals, chromium, copper,
iron, manganese, and nickel the majority of the metal exited in the slag. The
portion of each metal in the exiting streams is shown for comparison purposes.
The heat balance shown in Table B-8 was performed on the system to
evaluate the efficiency of energy conversion. Because of refuse heating
value uncertainty the calculated efficiency was based on the total heat
contained in the exit streams. The assumptions made in calculating the heat
balance are listed in Table 11. Bleating values are expressed as higher heating
value and sensible heats are referenced to 15 degrees C (60 degrees F). The
projection to the case of total pariiculate recycle is discussed in detail in
Section 11.
The measured refuse heating value excluding metal oxidation used was
18.2 MJ per kg (7831 Btu per pound) on a dry basis and was the average of three
samples. A small change in this value produces a substantial effect in the heat
balance. Product gas heating values were calculated for each sample from the
composition of the gas determined by sampling. The value used in the heat
balance was the average of all gas samples obtained during the test period.
Char heating value was the average of the values from samples. Oil and water
stream heating values were assumed based on previous PUR OX System tests
(see Table 11).
The heat lost was composed of three large losses: hearth cooling loss,
converter convective and radiation loss, and condenser cooling loss. The
hearth cooling loss was measured from temperature changes in the cooling
water and its rate and the value determined was applied to all test periods.
This was a good assumption since hearth temperatures were relatively constant
for all operating conditions. The;converter convective and radiation loss was
calculated in previous studies ancl this value was applied to all test periods
also. The condenser loss was estimated by taking the water stream from
nominal converter exit conditions|of 260 degrees C (500 degrees F) to condenser
outlet conditions of 21 degrees C! (70 degrees F).
110
-------
Based on the exit streams the efficiency of energy conversion was
63.6 percent. Under conditions of char recycle an efficiency of 77 percent
would be expected based on the analysis in Section 11.
Ill
-------
o
0)
to
P
O
g
OH
w
o
CQ
to
I
PQ
rd
en
rtJ
CQ
H-" Cn
CM
Cn
II
0
P-i
Cn
Cn
Tl Cn
(0
.p->
0
CO
CO
to
00
LO
CO
CO
00
oo
CO
CO
CM
to
CO
CO
oo
w
0
CO
i-l O
CO LO
CO
to
oo
LO
oo
CO
oo
oo
CO
CO
CM
CO
CO
CO
oo
LO
CD
-
co
CM
CO
LO
CD
LO
CD
1 > L
(0 Cn
S ฃ
H CO
CM
CO
CM
CM
LO
CM
CO
CO
CM
CO
CM
CM
CM
V
LO
CM
CO
OO
CD
0
LO
CD
LO
CD
CD
CM
CD
CM
CD
0
CD
CO
CD
CD
LO
CO
0 -H
O (0
co
LO
CD
CD
CD
LO
o
O
CD
CD
OO
LO
CD
CO
LO
CM
Cn
(0
ii
CO
LO
CD
CO
co
CD
co
CD
CM
CD
CO
CO
CM
(t)
6
CM
CD
CO
CD
0
o
CD
4-1
0
CD
CO
(0
X!
(0
ffi
HH
n
c
to
U
112
-------
^~.
w
rH
i 1
LO
Q
O
2
w
EH
23
EH
W
O
a
ง
CO
rJ
EH
H
O
<ฃ
ex
EH
^
1
CQ
W
i-J
PQ
EH
oo"^
0
CD rH
X
X!
i i
N ง
' Si
11
X! J3
in* &* ;~i
Q
i i
..
.2 CD X!
W &-i Z2-
CO
51
** 3"
o *
*r^*
O d
t i
(0
Q)
CD
ฐ^ I
OO
CM
ts!
oo '
oo
CO
0 '
rH
r^" -
rH
O
ป
to I
CM
CM
i
LO
CO
to
LO
^
to '
CO
CD
I
rH
0 CD
W D>
"CD 2,
fฃ CO
CD
CD
CO
CM
t^
CO
oo
oo
o
rH
I 1
rH
rH
0
CO
CM
CM
LO
CO
CO
i 1
LO
"^
CO
oo
CD
rH
1 1
. 1
rd
"o
EH
i i
I-H
t i
*
rH
rH
CO
rH
rH
CM
CD
LO
LO
OO
0
CD
CM
CM
0
O
OO
J- J
co
LO
CM
11
CD
rH
CJ>
rd
i
CO
rH
O
0
CO
CM
00
rH
LO
O
CM
LO
CM
LO
LO
CM
LO
rH
CO
rH
CM
LO
LO
0
i-t
rd
6
o
<0
V
co
LO
CM
0
CO
rH
CO
rH
O
O
OO
rH
CD
CO
rH
O
CO
rH
V
*
S-f
CD
Wastewat
CM
rH
rH
CO
CD
CD
i 1
O
CM
rH
IS.
LO
CO
o
OO
LO
CM
o
LO
CO
LO
CM
CM
to
0
rH
CO
t 1
CD
CO
O
MH
- o t;
o -go
-H CD _
rd O rd
O (DO
EH CU EH
0 1 1
0
1 1
CO If) t**
LO rH CM
CD
*^ C*^ CD
vt* CO rH
CD
0 O 0
O OO CM
CD
O CO ^
CO CO O
CD
LO CM CO
CD tD CO
CD
O OO CM
i-H OO O
CD
CO O CM
LO ^ O
CD
00 CN] ,
o 1
LO "xt*
CM C-.
S-H
M
1
CD
% <3 ซj
rd r1 rd
V 0 ฃ
w
4-1
ej
-i-4
r-H
C
_O
O
Q
4 I
CD
0
-a
CD
w
rd
0
rd
tn
K
T3
Cj
rd
73
O
"o
CQ
0
* Quantiti
113
-------
TABLE B-8. PUROX SYSTEM HEAT BALANCE. TEST PERIOD 5
'
Refuse*
Sludge
Fuel Gas
Oxygen
Total In
Metal
Gas
Slag
Char
Oil
Wastewater
Losses
Total Out
Efficiency (as operated)
Of Conversion (as projected) ',
GJ/Day
827.9
-
92.7
-
920.6
-
507.6
33.2
55.3
57.4
38.4
105.6
797.5
123.1
63 . 6 Percent
77.0 Percent
Btu x 10 /Day
784.7
-
87.9
-
872.6
-
481.1
31.5
52.4
54.4
36.4
100.1
755.9
116.7
*Value is average for entire program. Efficiency based on total out.
114
-------
APPENDIX C
TEST PERIOD 1. PRIMARY SLUDGE - April 19-25, 1977
The plant processed its first sludge during the period of 19 to 25 April
1977. Feeding of 32 .4 percent solid dewatered primary sludge cake from
Huntington, West Virginia, began at 2200 on 19 April 1977 at an approximate
operating rate of 77.1 Mg (85 tons) per day of refuse and 6.2 Mg (5.6 tons) per
day of sludge. The dry sludge-to-refuse ratio was 0.02. Operation was smooth
for two days though several minor problems occurred. The plant was auto-
matically shut down at 1055 on 21 April when an electrician accidentally shorted
the shutdown circuit while wiring the ESP recycle blower. A rapid startup was
conducted and the plant was operating by 1230. The vacuum filter was tested
on 22 April and operated for 3.5 hours on 23 April. The scrubber was operated
on recycle water during this time. Several troublesome problems occured in
the front end between 22 and 24 April but rate was only affected once until
shredder problems precluded feeding during the morning of 25 April. The test
period was ended when the shredder drive sheaves failed and the plant was
shut down.
The heat and mass balances calculated for the first sludge test period
are shown in Tables C-l through C-6. Sludge was fed to the converter as 6
percent of the feed stream on a wet basis. The balances were obtained by the
methods described in Appendix B. Similar problems to those which occurred
during calculation of the refuse only balance were encountered with these
balances.
115
-------
TABLE C-l MASS AND COMPONENT BALANCE. TEST PERIOD 1
STREAM
Refuse (as received)
(as fed)
Sludge
Fuel Gas
Oxygen
Total In
Metal
PUROX Gas
Slag
Char
Oil
Wastewater
Total Out
Total
(Mg)
77.1
-
5.1
1.3
14.1
97.6
4.2
38.7
19.9
3.4
1.6
29.8
97.6
Basis: 1 Day
C H
(Mg) (Mg)
_
16.9 2.1
0.5 0.1
1.0 0.3
-
-
-
14.5 1.7
0.2
1.9 0.1
1.3 0.2
0.5 0.1
-
o s
(Mg) (Mg)
-
13.4 0.2
0.2
-
13.6
-
-
21.2
2.2
0.1
0.2
0.8 0.1
-
H20
(Mg)
-
21.8
3.4
-
-
-
-
0.7
19.5
1.3
-
28.3
-
Ash
(Mg)
-
18.1
0.8
-
-
-
-
-
17.5
1.4
-
-
'
116
-------
f~\
CO
i-H
C
W
11
Q
Pi:
CU
H
CO
w
H
W
O
] O ^ '"S1
O ^f O LO CO CSl
ii ^1 r-t (M
^^
Q W ง 1 ซ rH CO 1. 1 1 CD 1
!-) -ii (MOO ' .rH
w
W """^ :
(0 52 ID tD r-l O C>J
^ O 00 O ซ-l ID O
ฑi i-i rH
s
"T^W r-j tO^HLOLO CD [> .CD
"OO LO LOi-HLOt^ St*COt-t
E-l OO i 1 C3 "vF CS
rM ' :
^___
Ti
4-> ^
^"* -^H
CD "g) ^^
C 0 TS
O 0) 0)
Q, s_, m
Sw
0 w w (0
rS ^2- vS- w O
dm <8a,S x
S 3 'OrsS'O .tjct'cn
S^ 5ง^^^|^
CO Pi CO PL, OH ,SiCL,CQ
in
1 1 1 1
CD
1 1 1 1
' CO
CO
r T I rH 1 1
O 0
O CM CO
O O Q
i 1 CM t-f ( 1
o p o
1-1 ^ to
(M n-> O
co oo rซ. LO
co . rH esi c^
co o
rH
S-, "*
fl) Qy
+J CQ jj
re) co 3
| 3 o
(3 _ " 1o 43
6 6 ^ S H
117
-------
TABLE C-3 AVERAGE PRODUCT GAS ANALYSIS, TEST PERIOD 1
H2
CO
CH4
C2 H2
C2H6
C4
HHC
H2S
CH3OH
CO 2
02,
Ar
Mole Percent
31.25
36.24
4.46
0 .25
1.32
0 .33
0 .27
0.47
. 0.35
' 0.17
0.32
0 .00
0.14
23.52
0 .05
0 .62
118
-------
o
I 1
S-j
P
H
i-H
P
O
2
H
O-.
EH
CO
W
EH
W
O
* 1
K *~" i-H r-l
[^ CO LO
"3 LO LO rH
ฃ v co r-~
N Ci r-l i-H
"g O LO LO
S " LO co co
CD CD
___^ rH i 1 CM
CJ tJ> ...
S > ' LO CO
!><
5 oo LO co
O "3 ...
J2 v; CM i 1 'St1
l~t PH CS' , | r-H
to
-/-)
W CO CO CD
(0 . .' .
CQ -5 rH CO CD
0 _V CM O CM
f i , O- ^}< CO - C~ป
vf CO D^
"S ...
3 S LO r^ CM
U O- CJ CM LO
CD LO LO
^ j? oo o co
O ซ rH r-H CM
^ rH CO CD
T3 51 OT. "~I ฐ.
O 0- 0 0 rH
^ KJ W CD ^H
M Q} 13 *"O fo
111 *?^
!ง CO Cd CO EH
O rH 1 1
CM O CM
O 0 0
O LO LO
ซ
CD rH O
"vt1 1 LO
rH rH
[^ co o
co o -xt*
LO rH CD
CD CO LO
co co i cv!
CM CM
vF CO tv.
r
^3 C^ f O^
CO CO
t>ซ ^* 1 1
.
CM LO CO
CD O 1 CD
O rH 1 1
rH rH
"vf* CM CD
. . .
LO O 1 LO
CO i 1 CD
l^. CM CD
t^ CM 1 CD
rH rH
CO CM o
. . 1 ,
O O CD
0
4_j
fO ^
, ! 2
O^ (Q CO fO
Co fH fo *^r
ra 0 ^ EH
CM 00
LO "vf
CD
O O
CD
CD rH
CO CD
CO rH
CD i 1
CD CO
CO rH
CD
-
0 CD 0
Q. CO i f
CO
-rH
rC
CD i-H
S CO rH
"G CO rH
13
JD SS
ro . .
3 CD CO
J) Q CO CD
D, .
ft "^^
03 -*-1
!-.
0 MH
I * Q
ป^ ^ ^^ fO
^i p|^ fO rl
* D-. CO O
119
-------
co
a.
co
O
LO
I
CQ
(0
Q
en
r4
ca
ffi
X!
Q) ฃ
O
O
T) X!
O cJ.
CM
CO
CO
CM
O
oo co
CM
co co
CD
CD O
LO CD
CM
CO
CO f- O
LO CM CO
CD
CO
CO D^
O rH
CM CM
f-. O C--.
CO CD CM
CM CO
Q) CD
vi
fO
co
CO
CD
CO
co
oo
(0
CM 00
LO -tf
CD
CD "sP
CD
CD iI
CO CD
CO i-t
CD iI
co co
CO rH
o
11
CO
CD
O
co
CN!
CD
CN1 00
co
CO
CD
CO
CD
00
CD
CO
co
-------
The gas production and gas higher heating value for this period were
401.6 Nm3 per Mg (13,600 Sft3/ton) of total feed and 14.19 MJ per Nm3
(360.4'Btu/Sft3) respectively. These results were lower than the values deter-
mined for the refuse test period. Only one set of three gas samples was ob-
tained during this test period, however, and the result may be in error. A gas
particulate sample was not taken at this time so pitot tube data was not avail-
able for comparison. The gas production with sludge was within a standard
deviation of the gas production observed during the refuse only test. The low
heating value may be the result of sampling during low rate operation. The
oxygen consumption was 0.17 Mg (tons) oxygen per Mg (ton) of combustible
which was lower than that observed with refuse only. Effluent production was
0.363 cubic meters per Mg (87 gallons per ton) slightly higher than the refuse
only result.
. The trace metal balances obtained were similar to the balances which
resulted from the refuse only tests. The amounts of copper and nickel entering
the system were greater than with refuse only due to the high concentration of
these metals in the sludge. No wastewater samples were taken during this
period, so the metal balances do not include the wastewater stream. The
fraction of each metal in the exiting streams is shown for comparison purposes.
The heat balance on the system is presented in Table C-6. The con-
version efficiency based on heat leaving the process was 59.5 percent. This
efficiency was lower than with refuse only. Lower efficiency was consistent
with predicted system behavior due to the additional moisture added in the
sludge. The low efficiency when compared to the refuse only test was caused
by the lower qas production per Mg (ton) of feed. Heat contents of the'-other
streams were similar to the refuse only result on a per Mg (ton) of material fed
basis. The efficiency for the case of particulate recycle has been projected
to be 75.1 percent.
121
-------
TABLE C-6 PUROX SYSTEM HEAT BALANCE. TEST PERIOD 1
Refuse*
Sludge
Fuel Gas
Oxygen
Total In
Metal
PUROX System Gas
Slag
Char
Oil
Wastewater
Losses
Total Out
A
GT/Day
960.5
19.4
68.4
1048.3
BTU x 10"6/Day
910.4
18.4
64.8
993.6
499.6
39.4
75.0
64.6
44.3
116.2
839.0
209.3
473.5
37.3
71.1
61.2
42.0
110.1
795.2
198.4
Efficiency (as operated) 59.5 percent
of Conversion (as projected) 75.1 percent
* Value is average for entire program. Efficiency based on total out.
122
-------
APPENDIX D
TEST PERIOD 2. PRIMARY SLUDGE 27 APRIL - 30 APRIL 1977
The second test period processing 27.5 percent solids primary sludge
at a dry sludge-to-refuse ratio of 0.036 was conducted between 27 - 30 April
1977. Heat and mass balances on the second sludge test period were also
similar to those obtained during refuse only operation. The sludge fraction was
slightly higher than during the first codisposal test, comprising 10 percent of
the total feed. The methods used calculating these balances were identical
to those used previously. The results are shown in Tables D-l through D-6.
Q
The gas production was 440 Nm per Mg (14,900 Sft3/tx>n) slightly
higher than the refuse only case, and the standard deviation of the gas produc-
tion from pitot tube data during steady operation was 52.3 Nm3 per Mg
(1772 Sft3 per ton). Oxygen consumption was 0.19 Mg/Mg (tons/ton) of feed
which fell between the values observed during the two test periods discussed
earlier. Product gas heating value was 14.24 MJ per Nm3 (361.6 Btu per
standard cubic foot) lower than the refuse only result, and similar to the first
sludge test.
Trace metal balances were also similar to those previously discussed.
The effect of Huntington sludge was noticed again in the large quantities of
copper and nickel exiting the process. The only significant difference between
this period and previous tests was the lower fraction of total nickel entering
which exited in the slag, 69 percent against 94 percent during other tests.
However, 6.5 times as much nickel entered the process during this period
than entered during the refuse only test. Additionally the nickel which exited
the system was only 27 percent of the nickel which entered. Since the sludge
contributed 96 percent of the entering nickel the total incoming nickel was
reasonably accurate and the error is probably in the slag or char streams. The
dilution of the wastewater by scrubber makeup rendered the wastewater metal
content uncertain. The fraction of each metal in the exiting stream is shown
for comparison purposes.
The heat balance for this test period resulted in a conversion efficiency
of 61.9 percent, quite close to that obtained during refuse only operation. The
heat content of all exit streams was nearly identical to other results when com-
pared on a per unit of material fed basis. The efficiency projected for the case
of particulate recycle was 76.8 percent.
123
-------
*ป
w
0
1
*L
& ฃ
S ฃ
o E
cu ^
S ^
0 C
p ฃ
^ Si
rfj A"
CO n_.
CQ g;
CO M
<; ^_
O S
CM^
K-
CQ ^
'B
o ง
^ "^
(0
P
'55
co
ro
CQ
of
(0 D1
ro
2
4->
CQ
O rH
C^ rH
rH
CO "st1 ฐO
co in o
CM
o
CD "f rH CO
CO O Lf5 ^
-H r-1 CM
CM rH CO CO
CM O O "-1
CD LO CD "-1
CO O O ^
i-H rH
CO CO CM CD "tf CM UO
LO C^ rH LO CD ^t* CM
C*ป rH CD "*
rl
o> s~
m ซ
Pi PM
co ซ S
^, -S- co O
ro c fi . .
Q) CD U C!) rH ><
3 ง* -* S "ซ 5 9
3 3 s & ซ o5 g
CD ^H 3 iS O *3 P
Pi CQ P-, O EH r2; P-,
CD rH
i-H
CO
CD
CM
rH rH
0 0
CO C^
CM O
rH CXJ f-1
O O O
CM CD nt LO
O rH rH o
rH ^ CD CD
CD CO rH CO
rH CM
-------
CQ
-tH
i 1
B
CM
P
O
2
w
P-.
H
CO
H
EH
W
O
<
<
oa
H
^S
2
s
o
0
p
CO
1
CM
1
P
H
03
* II 1
CO ^* , O. O.
1 LO 1 CO I.' LO
ป-l : . i-H ' , CM
, ^ rH CTJ , , , CD
,..
, LO LO Q 1 1 "^
* 1 I *
Or-! t^
CM rH
'
o rH CO CM (O to CD
OO ' OO t i t^ CD TS< to
OO : rH o ซ*
rH
(D
d) ^-^ CQ
0 T3 ; tO
0) 0) CQ Q
^ tw !0 c; C
CO jrj (Q Cy _ - ^) t ( jy ^^
^""^ ^^^ ~O *3Jป ^_ (0 4^ ry*
2 1 s g 1 1 .g
^ LO
rH
OO
1 ' 'CD
CM
1 rH 1 rH
LO C<1 00 '
CM' I '" '
I rt. M. 1
CM rH ^ CO
o
CM rH
rH 00 CO Si1.
i 1 CO rH rH
CM CO
0
CD :
J-. '-'
O> (0 CQ
JO Jg ^J 03
1
1
1
1
,
1
to
CD
O
i I
I-i
3
O
1 I
10
+j
o
EH
125
-------
TABLE D-3. AVERAGE PRODUCT GAS ANALYSIS. TEST PERIOD 2
Mole Percent
2
CO
CH4
C2H2
C2H4
C2H6
ฐ3H6
C3H8
Q
4
C
5
C6H6
HHC
H0S
2
CH0OH
3
C09
z
Q
2
Ar
29.62
36.80
5.02
.41
1.68
.32
.23
.18
.34
.27
.25
.29
.04
.08
23.85
.03
.59
126
-------
"c
t
CM
Q
c
P:
OH
EH
K
E-
V.
O
12
13
cc
E-
H
s
w
c
PC
EH
*
l
P
H
1-1
PQ
EH
. ^
h-i D^
HM -
c "B
1
^_
S ri4
ฃ 'B
J^ฃ.
Q tJl
j>t ^iS
fO
P
r (
*
^ CD '5)
W ["' t V
tO "
CQ
3 tn
v^ / k^'
!_ Ol
o ,*
a 1?
O &
e
(0
CD
S-i
CO
i-H CD
CD O
i-H O
CD CD
i-H CD
CD
i 1
CD i-l
CM
i-H
IX CD
SP LO
LO
*xP ^H
CM CM
i-H
CO CD
O
CD OO
i-< CO
St* CO
OO CO
"sP CO
CM LO
CD CO
CX LO
1 )
1 1 ^p
CD LO
O CD
CD CD
w tn
3 "T3
m -)
S. 3
&. CO
^
CD
I-H
LO
OO
CO
i 1
O
CM
i l
CD
9
0
CO
LO
sp
i i
tx
CD
i-H
oo
ฃซ*
CM
CO
CO
LO
i-H
i 1
0
EH
CO
CO
O
CM
CD
CD
N
LO
CM
CO
CO
CM
LO
I-H
CD
i-H
LO
CO
F*ซ*
"vP
CM
i-H
CO
1 1
o
tn
K
G
(0
13
o
'o
n
CD
1-4
' *
*Quantii
127
-------
to
W
Cu
EH
a
W
o
rd
P
en
td
CQ
CO
D)
ffi
PL,
CD
r -
O *2,
O
to
CO
CO
CO
CD O
0, EH
co
OD
co
co
LO
CO
CO
oo
sf
CD
CO
(0
1I
CO
CO
CM
oo
CM
CD
CO
CD
CM
CD
CO
OO
oo
LO
CO
LO
LO
CO
co
CD
a
6
CM
(M
CO
LO
co
.
CD
0)
*->
I
CD
+J
W
(0
O
cu
-a
0
CO
ro
0
10
o
(0
128
-------
TABLE D-6. PUROX SYSTEM HEAT BALANCE. TEST PERIOD 2
Refuse*
Sludge
Fuel Gas
Oxygen
Total In
Metal
PUROX System Gas
Slag
Char
Oil
Wastewater
Losses
Total Out
?
Efficiency (as operated)
Of Conversion (as projected)
GJ/Day
953.5
21.4
63.4
1038.3
-
543.4
38.0
75.0
64.6
44.3
113.2
878.4
159.8
61 .9 Percent
76.8 Percent
Btu x 10 ~ /Day
903.7
20.3
60.1
-
984.1
-
515.0
36.0
71.1
61.2
42.0
107.3
832.6
151.5
1
* Value is average for entire program. Efficiency based on total out.
129
-------
APPENDIX E
TEST PERIOD 6. PRIMARY SLUDGE, 2 JUNE - 5 JUNE 1977
The third test period with 26.2% solids primary sludge at a dry sludge-
to-refuse ratio of 0.024 took place between 2 and 5 June. This period of
testing with primary sludge produced heat and mass balances with several
differences from earlier results. The same techniques were used to calculate
the balances. The wet sludge ratio during this test was 6 percent, the same
ratio used during the first test, but the sludge had a higher moisture content.
The balances are shown in Tables E-l through E-6.
The trace metal balances obtained during this test were quite similar
to those obtained during the refuse only test and earlier primary sludge tests,
The distribution of metal among the three exit streams was essentially the
same as the distribution measured during other tests despite the lower total
slag rate. The dilution effect of the scrubber makeup water rendered the
wastewater metal content uncertain. The fraction of each metal in the
exiting streams is shown for comparison.
The slag production during this test was 0.18 Mg (ton) of slag, per Mg
(ton) of feed, much less than during other test periods. The lower slag
production was indicative of a change in the composition of the incoming feed.
Gas production was 490.2 Nm3 per Mg (16,600 Sft3 per ton), higher than
previously measured, and oxygen-to-feed ratio was 0.21, also higher than
previously measured. Pitot tube measurement of gas rate resulted in similar
gas production result of 498.1 Nm3 per Mg (16,866 Sft3 per ton),,, The
standard deviation of 12 gas production values obtained using pitot tube data
was 38.5 Nm3 per Mg (1302 Sft3 per ton).
High gas production and oxygen consumption probably resulted from
,a change in refuse composition. The gas heating value was 14.11 MJ per
Nm3 (358.3 Btu per standard cubic foot), lower than previous results.
Calculated wastewater production was 0.38 m3 per Mg (100 gal/ton), which'was
higher than refuse-only results. The higher effluent production probably was
caused by wetter sludge and refuse.
130
-------
The heat balance, shown in Table E-6 differed from refuse results in
the same manner as the mass balance The heat content of the slag stream
was lower, consistent with the lower slag production. The product gas con-
tained more energy per ton of feed and more of the total exit energy than the
refuse only result. This resulted in a conversion efficiency of 65.6 percent,
despite the low gas heating value. The increase in efficiency was probably
due to a change in incoming refuse rather than the effect of sludge processing.
The efficiency with particulate recycle has been projected to be 74.8 percent.
131
-------
o
CO
p
(X
8
EH
a
o
CQ
CO
cc
(0
p
en
-I
CO
fO
ffl.
CO Cn
O
0)
CO
LO
oo
oo
LO
LO
CD
01
oo
oo
oo
(M
CO
0) ..
O 13
CD 0)
(0 fd
CM
oo
OO
oo
CD
CO
CD
-------
CO
t-t
r 1
C
H
to
Q
O
2
H
Cu
EH
CO
W
H
LANCES .
_^
CO
CO g
+J
"to"
> O o
(0 -t->
Q *-
*
w
-1-1
rO "M"
CQ ffi 1
, .
en
n ฃ
-2
1) CQ
ro g
.P +j
tH -~^
H
(0
0
CO
^H to to ^
1 . . 1 1 III..
CD 0 CT> 0 1 1
0- !>. [>. CO
1 0 pg' ' ' ' ' 0ฐ 1 1 1 LO
CS1 CM
1 "1 1 II 1 1 1 1 1 1 "1
O CD
sf if t~~ OO CO r-J ^
'r-HCD 'CO ' 'Or-)'oo
l-l i-l CM
C"*ซ. ^1^ CO f~~t *~H
I.I .111.11..
r-l O <-H O O
tOCOLO i 1 r-l CO r-l LO
1 1 11 .ซ..
1 'sj'1 CD i^ *^ CD CD 'H CD
ii i)
O t>- CT5CM OO OO OO O O CO O)
e e. oe .ซ
i I ' CO i J ^P O CM C^ i 1 i 1 i 1 tO
CO i 1 OO CO i-l CM
T?
CD
"CD ^^ w &
O ""O (0 +j
CD CD m rn ro
a> *" """ a>> ^ c ฃ X ^
(0 fO Q^ J^l r"H fQ ^3 5ซt "*"*
3 N ' **-'' 'w 'TT? ^, fO .|_i Q^ C7> /rt CO
ง ISoEHSgSoSl
1
1
1
1
'
1
oo
.
o
oo
+j
O
1 1
(0
-M
o
133
-------
TABLE E-3. AVERAGE PRODUCT GAS ANALYSIS. TEST PERIOD 6
Mole Percent
H2 !
CO
CH4
C2H2
C2H4
C2H6
C3H6
C3H8
C4
C5
C6H6
HHC
H2S
CH3OH
CO2
ฐ2
Ar
29.94
33.74
5.89
.52
1.91
.34
.27
.12
.28
.22
.26
.17
.02
.04
25.58
.04
.61
134
-------
'c
S-
4-
d
*
CO
Q
0
5
fi:
O.
EH
CO
H
T*^;
fc
H
O
<ฃ
3
CO
rr^j
CJ
cS
(d X5 'B
Qn . v
^^ C^
rH
^ ^
>r_]
^ |j_i _S/
CQ ; -
5*t5
ฃ1
, -^j
o j;
T3 ^
o &
a
(0
CD
CO
CM
"*.
1
O
O
CM
r 1
^
CO
CD
O
''xj'
CM
CO
O
LO
CD
CO
sf
CO
I 1
rH
CO
rH
00
CD
0
Refuse
0
o
CO
*
o
CD
CO
rl
CD
0
"*.
o
C--
c^
CD
CSI
^*
0
CM
CO
0
0
CD
-C5
CO
CM
"*.
i 1
CO
o
CM
r-H
CO
CO
CM
CM
ซ
i-H
^H
co
CO
!>-
CN1
C^.
CO
co
csi
CM
1 1
LO
1 1
^
r-.
o
:_
G
O
EH
O
rH
O
O
CD
OO
0
CO
r-H
CO
O
0
oo
CO
rป|
rH
OO
e
CO
CO
CO
co
CO
LO
o
o
Cn
(0
i
CO
CO
0
0
CM
ซ
CO
"^
CM
0
i 1
CO
r-.
oo
e
f^.
CM
tv.
CM
co
e
o
CO
0
*
0
S-,
(0
6
CM
t~ป
CM
CM
O
CD
O
O
^
O
CD
CD
0
CO
. ซ
1 1
1 1
LO
0
o
CO
o
o
oo
LO
o
V
*
i-t
CD
4->
Wastewa
co
rH OO rH
e 1
0 CD CO
IS- CM
'^ IS. OO *^
e
ซ- CO CO CD
^ CO
co oo
CM OO CM 1
rH IS. CM
^ CO CD CO
ซ
** CD CO CM
rH CO
IS, CS( CO
00 CO O 1
CO
CO CD CO r-I
ซ
C-~ CD CM o
f^ CD
rH
i 1
LO
CD LO -st1 rH
CD CM IS. O
CO CO
LO CO IS.
=*< CO r-H '
co co
vj1 O O
rH
co i>x I
O CO CD
,
CD
[ t t 1 *fl
*1~* *r * j
33 ง
0 1ฐ -5
f i
-------
en
OS
W
CQ
co
(0
P
id
CQ
8,
(0
CU
CQ
CO
CO
LO
CD
CM
OO
LO
CD
CO
CM
CO
CM
CO
LO
CO
CO
CM
LO
CD
tn
CD
CO
CO
CD
CO
CD
CM
CD
CM
Q)
Oi
-a
CO
CO
CO
CO
CM
LO
CD
CD
CO
CM
CD
CM
CO
CD
CM
CO
CO
CM
CD
CO
(M
co
a
CO
CO
o
LO
CM
CO
CO
rd
ii
CO
o
CO
(M
CM
LO
CM
co
CO
-------
TABLE E-6. PUROX SYSTEM HEAT BALANCE. TEST PERIOD 6
,:*s ;
Refuse*
Sludge
Fuel Gas
Oxygen
Total In
Metal
PUROX System Gas
Slag
Char
Oil
Wastewater
Losses
Total Out
Efficiency (as operated)
1 Of Conversion (as projected)
GJ/Day
619.6
10.7
92.7
-
723.0
-
43302
1908
19.7
46.6
38.4
102.9
660.6
62.4
65.6 Percent
74.8 Percent
Btu x 10 6/Day
587.3
10.1
87.9
-
685.3
-
410.6
18.8
18.7
44.2
36.4
97.5
625.7
59.6
*Value Is average for entire program. Efficiency based on Total Out,
137
-------
APPENDIX F
TEST PERIOD 7. PRIMARY SLUDGE - JUNE 6-9, 1977
The final test with 22 .3 percent solids primary sludge at a dry sludge-
to-refuse ratio of 0.074 was conducted between 6 and 9 June. The test plan
was to operate at a high rate with as high a sludge ratio as possible. The heat
and mass balances calculated for the final period of primary sludge processing
are shown in Tables F-l through F-6. During this period, sludge was run at
the highest rate which the feed system could handle, and contributed 23 per-
cent of the total feed on a wet basis. The total converter rate during this test
was 106.1 Mg (117 tons) per day, the highest during any test period.
q 3
Gas production was 395.7 Nmฐ per Mg (13,400 Sft per ton), which was
the lowest during any test period. The low gas production reflected the large
quantity of water fed with the sludge. Pitot tube measurements resulted in a
higher gas production of 524.5 Nm3 per Mg (17,761 Sft3 per ton) with a
standard deviation of 78.3 (Nm3 per Mg (2,651 Sft3 per ton) for 12 samples*
Oxygen-to-feed ratio was 0.18 which was within the range of oxygen consump-
tion measurements determined for, sludge processing and below that obtained
during refuse only processing. Gas heating value was 15.10 MJ per Nm^
(383.5 Btu per standard cubic foot), very close to the refuse only result.
The heating value was based on 15 samples which exhibited a standard
deviation of 0.72 MJ per Nm3 (18.3 Btu per standard cubic .foot) was signifi-
cantly higher than heating values determined during other primary sludge test
periods. Slag production was 0.21 Mg (tons) per Mg (ton) of feed which
was within the range of slag production observed during other tests. Effluent
production was 0.405 cubic meters per Mg (107 gallons per ton) of feed which
was higher than the refuse only result and other sludge processing results.
The high effluent ratio was caused by the large quantity of sludge processed.
Trace metal balances obtained were similar to those determined for
other periods. The nickel balance was quite poor, exiting nickel was only
11 percent of the incoming nickel. As earlier, the dilution of the wastewater
by scrubber makeup rendered its metal content uncertain. Another difference
was the relatively large amount of mercury found in the slag, which was much
higher than during other periods. This finding was encouraging since the
capture of significant amounts of mercury in the slag had not been observed
earlier. The fraction of each metal in the exiting streams is shown for comparison.
138
-------
The heat balance was similar to the results measured during earlier refuse
only and codisposal tests. Higher production of char at the high operating rate
resulted in a larger fraction of the total heat out being represented by the char
stream. Also, the fraction of total heat out in the product gas stream was less
than with refuse only reflecting the high total moisture content in the feed due to
the increased wet sludge content. The conversion efficiency of the system for
this period was 56.8 percent which was lowest of any period. The low efficiency
was another result of the high moisture content of the sludge. High moisture
resulted in lower gas production per unit of feed because it contributed nothing
to the heating value of the gas, and it also increased the heat losses due to the
large quantity of water which was heated and then condensed. The efficiency
projected for the case of particulate recycle is 74.8 percent.
139
-------
o
w
O
CO
CO
p
CO
O
CM
W
CQ
D
If
CO
LO
(M
LD
LO
I CM
CM
CD
CO
0)
CD --,
O T3
to
(D
^ ~
O Q
CM
O
I 1
rt
140
-------
CO
^
o
w
P
0
2
H
CU
g^
CO
H
EH
CO
W
0
s
^
M
12
w
S3
0
a.
O
O
P
CO
CO
s
.
CNI
1
H
CQ
EH
CQ
. C!
w O
< -ii
ol
C-J O
ffi
CQ
G
ra .g
CO
O 0
ซii
!>i CQ
TJ ffi ง
_+-^
CQ
-i-l
CQ
ro
m
*^" *
CQ.
O o
2:'
"Jo's
4_J J^J
0 P
EH ^ "
s
(0
0
CO
CO CO LO to
IN- CNI IN. CM
1 r-H 11 | | r-H 111
CO CM O CM
CO CN r-H CT5
1 C-J i-l 1 1 1 1 1 1 ^ 1
r-1 r-. ,_,,_,
10011 1110101
CO^ 'vf COCMr-HCMCO
tOO O CD
0) ,N W to CQ
O T3 to ^30
0 0 CQ Q tO CQ
ฉWCQ<1>Qง1-1^|X! ง1-H
(0 (tj ^^ CTI *~~* n3 CD s_i "^~* t>
5"^"" '?'ฎ^'^0S (Oj-aro S
,-,
O
*N^
r-H
*tL
O
r-H
(3
141
-------
TABLE F-3 .
AVERAGE PRODUCT GAS ANALYSIS
Test Period 7
Mole %
EL
2
CO
GEL
4
C2H2
C2H4
C2H6
C3H6
r H
3 8
C.
4
Q
5
C H
6 6
HHC
H0S
2
CH0OH
3
co_
2
O0
2
Ar
26.81
34.19
6.12
0.72
2.19
0.40
0.28
0.19
0.40
0.36
0.34
0.37
0.03
0.09
26.85
0.04
0.63
142
-------
^
C
p;
^
^
!>.
Q
C
PL
CL,
EH
CO
pq
H
CO
H
O
s
5
tt
i-^
^
PL
w
O
i^
pj
E_j
^
J
Pn
a
CQ
^.
ฃ> xgj
(-M . ,
^_^^
tn
C _y*
N 0-
, ,
~* S1
s ^
^
G 0>
i2;
, s
XJ _ง"
p-l * '
>.
fO
T!
o ^.
- g
1 s
1 1
co
CM
LO
to
00
11
[-*
LO
I 1
co
to
*
^f
1-1
CM
S!
^*
CO
CO
CM
CO
o
CM
*
o
11
Refuse
IS.
-^
o
CO
*^
CM
co
CO
o
t~^
LO
1-1
to
oo
. 9
CD
>f
1 '
LO
CM
^P
oป
LO
CO
co
*
CM
0
CO
00
CO
CM
OO
O
11
CM
O
CD
0
CO
00
r-.
o
CO
CM
LO
ซ
LO
to
CD
I 1
11
CO
LO
c^
c^
c~-
0
co
a
wt
(0
CD
CO
CM
oo
CD
r 1
CD
CO
CM
i-l
LO
CM
=*<
LO
to
ซ
i"~i
CO
CD
LO
il
i 1
CO
CD
O
0
en
10
i
CO
oo
I 1
o
^*
CD
CM
CM
CO
11
"vf
to
CM
co
LO
oo
LO
00
11
0
00
11
CD
CO
^
LO
o
1
o
CM
c---
CM
V
CM
0
CO
O
O
CM
O
CO
O
CD
CO
CD
CO
O
CD
to
O
O
CD
LO
0
\r
*
b
j_j
Wastewa'
r-.
oo c^. co
CM CO CO
CD 1
"=3* CO CO 1-1
CD C^ CM O
tO CD
CM
r-l CO "vf
LO CD O
^t1 LO "st* |
t^. CM CM CO
il CD O O
CO 1^ CM
CO CO <3<
C^ t^ CM
LO CD 1
i 1 l>* CO I>x
CO LO CO O
tO CO r-1
CO
I 1
CO C^ CO
f^ CO CO
!> l>ป CM 1
0 00
LO 'vf CO
CO CD I
CO CO t^.
to
-^f in
o ,-H co i
u
O
1 1
o
E-I (_
*-* r\ "^
3 u tO
ฐ fi I
ฃ P % ro
o b ^ ฃ .5
EH CL, CO O ;>
CO
-1-1
-i-H
1
p;
O
-i
o
0)
o
-a
(U
en
o
^
0
(0
Cn
S
ฃH
fO
T3
O
m
O
CO
Q)
*Quantiti
143
-------
CO
i
w
CO
O
CO
s
tsi
CQ
1-1
CO
(D
O
O
O
CO
CM
co
eg
.CD
CO
co
If)
LO
co
CO
co
CM
CD
CO
CD
co
co
LO
co
LO
CD
CO
CO
CO
co
LO
CO
co
CM
co
co
LO
co
CM
LO
CD
CO
CD
CD
CO
co
LO
CO
CO
CO
co
CM
LO
CD
LO
CO
co
CD
LO
LO
LO
CO
CD
-SJ1
CXI
CO
CD
CO
CM
O
tn
rd
ii
CQ
co
CD
LO
CM
CM
CO
CD
CO
CM
1
O
o
co
V
CO
CD
co
CO
CD
LO
CO
r-1 CO
CD
CD
CO
CO
co
CO
co
V
CQ
.03
cn
CO
CM
co
i-l LO
CM O
CO
CO
co
O
I1
(0
O
CD
O
P-,
CO
CD
CO
co
CD
LO
CM
CD
CO
LO
CO
co
CD
tn
to
rI
CO
CD
CD
CM
CM
CM
O)
CO
co
co
co
(M
CO
LO
CO
rd
43
O
oo
CD
CD
O
O
co
O
CO
Q)
CD
CQ
(0
43
CD
i-,
fO
Cn
-K
O
co
0
fO
144
-------
TABLE F-6o PUROX SYSTEM HEAT BALANCE. TEST PERIOD 7
Refuse*
Sludge
Fuel Gas
Oxygen
Total In
GJ/day
1027.8
53.0
92.7
-
1173.5
Btu x 1
974.2
50.2
, 87.9
-
1112.3
0"6/day
Metal
PUROX System
Slag
Char
Oil
Wastewater
Losses
Total Out
A
Efficiency (as
Of Conversion
Gas 637.8
35.7
132.3
82.5
73.9
161.1
1123.2
60.3
operated) 56.8%
(as projected) 74.8%
604.5
33.8
125.4
78.2
70.0
152.7
1064.6
47.7
*Value is average for entire program. Efficiency based on total out.
145
-------
APPENDIX G
TEST PERIOD 3. MIXED SLUDGE - 2 May - 5 May 1977
The first test of 21 percent solids mixed primary and secondary sludge
at a dry sludge-to-refuse ratio of 0.049 was conducted between 2 and 5 May.
The test began with the feeding of Pittsburgh sludge on May 2. The test was
relatively free of mechanical breakdowns . High converter offgas temperature .
and low slag duct pressure, which are thought to be symptoms of bed channeling,
were present during much of the period. ' -.-'
The first test of mixed sludge processing resulted in the balances shown
in Tables G-l through G-6. The most noticeable difference between this period
and the refuse only period was the high oxygen consumption. The high measured
oxygen consumption could have been the result of bed channeling during the
test period. Channeling is believed to be caused by a deterioration in bed
porosity due to poorly formed pellets of sludge and refuse. The PUROX System
feeder has been specially designed to produce well formed pellets from shredded
refuse over a range of moisture levels. However, if the moisture level of the .
feed is high and refuse and sludge are not uniformly blended, the pellets will
be poorly formed. Poorly formed pellets can produce a variable porosity in the
refuse bed leading to channeling with resultant inefficient heat transfer which
produces high offgas temperatures and high oxygen consumption. The gas
heating value was also affected. The heating value result of 12.78 MJ per
Nm3 (324.4 Btu/Sft3 was much lower than the refuse only result and results
during other codisposal periods^. The standard deviation of the 9 heat value
samples was 1.97
other periods.
MJ per Nm3 (50 .1 Btu/Sft3) which was much higher than
'' O
Gas production was slightly higher than with refuse only at 451.8Nm
per/Mg (15,3008ft3 per ton) of feed. Effluent production was 0.467 cubic
meters per Mg (112 gallons per ton) of feed which was high, and resulted from
the high moisture content of the mixed sludge.
The trace metal balances were similar to those obtained for refuse only.
The mixed sludge did not contain a major fraction of any incoming metal. As a
result the incoming metal totals probably contain large errors due to the un-
certainty in refuse composition. Because no char samples were collected during
mixed sludge tests the char composition used to calculate the exiting nickel
146
-------
was that;determined from char samples obtained during the refuse only test.
This choice was made to prevent''the char collected during primary sludge pro-
cessing, which contained large concentrations of nickel, from causing errors
in the balance. The distribution of metal among the exit streams was similar
to the distribution observed for refuse only. The dilution of the wastewater by
scrubber makeup water limited the detection of metals. The fraction of each
metal in the exiting streams is shown for comparison purposes.
The system heat balance, shown in Table G-6, was similar to refuse
only results. Losses per ton of feed were higher due to the high sludge
moisture content and improper blinding. Conversion efficiency was 58.7
percent, and was lower than the refuse only test result primarily due to the
additional water processed. The .conversion efficiency projected for particulate
recycle was 71.2 percent. '
147
-------
o
CD
CO
Q
i
w
a,
a
D
O
CO
id
Q
ca
TM
w
O
CM
ffi
CO
(0
CO
I t-1 O I I I
CO CM
*
I i-H l>-
I I
OO O
CD iI CO
T-t O O
LO LO O
CD
to
1 I
O i-l I I
co
c-1 I>~ OO
1-1 oo
. 1I
C-- CNI
co t-i
co
a
o
O I!
*sf* OO CO
i-l (M OD
co co
CO
CO
3 ง
CD
1 ง
P-4 CO
^ < a
fi -rH ^
O O ^
CD
W CO
D
148
-------
co
--i
i i
fi
W
CO
Q
o
tf
H
H
CO
W
H
CO
W
O
3
m
EH
jeE^
H
S
O
O-,
)>j
0
D
Q
CO
CO
s,
(M
1
0
1-4
CO
CQ ID CO LO
^ g 1 .
~ 0 1 ;* 1-. 1 1 1 101
ffi CM
^
CQ
02 ^ 1 1 1 1 1 1 111
^ x
01
O S CD LO to ^ LO
O 1 ซ i * i i
_l 1 I 1 II'
ฃ- en o co co r-i
r-H (XI
fO --^
P 2
hrt G co r-i co CD
-1 *^ R i ...|i | . i
.. S^-- r i O O r-H
CQ
CQ
fO
oa ^^
J2 O CD rl LO O
r ") S-H
O 1 CD O r-H 1 | 1 TT 0
=> i 1 r-H
^0^ CO O-^CSICD COCMO
OP [>~IOrICDt^ OJr-H'vF
E-H >
ง 89"
O CO (D
g CQ CQ $
O vfS- ^2- CQ Q
O ^ c G
g 0 ฃOcSMr-H><
(3 w ^ _, tn < (OO
gi2 'B'a5>'-2'cu0::ง1
CO C< COpinOE-H SdnCO
CD
0 1 I 1 I
^
1 1 ^f I 1
CO
1 1 1 1 1
r-H CD
1 . . 1 1
O O
r-H r-H r-H
1 1
O O O
CO CO CD
a
r-H r-H O 1 1
CO LO i 1 (D
CNI r-l CD 1 IN.
CO CD
0) 0
4-> CO ,,
(0 CQ .3
1 ฐ o
1 -3 S 8 1
149
-------
H2
CO
CH4
C2H2
C2H4
C2H6
C3H6
C3H8
C4
C5 i
C6H6 ',
HHC
H2S
CH3OH
C02
02
Ar
Mole Percent
30.32
31.56
4.82
0.48
0.68
0.23
0.20
0.02
0.23
0.17
0.22
0.12
0.01
0.12
28.86
0.24
0.74
150
-------
-3
Q
iง
co
Q
O
3
w
CO
W
EH
N
W
O
^;
55
CM rH
fi k/l CO CM
LX] <= ,_)
-rH ^ "! ^
S * ' ^ (^
^ ฐ ฐฐ
C> tjl * *
>ฃ; j^ LO CO
*$ ' "* co
-* CJl * *
T-> A^ CD Cx.
O ^-* CM
^~. LO CO
K< ^ ^ CM '
V^ ' *" * 11
CO CD
Tl 'S *** 4^
v-x "* ^ CO CD
i-4 ^ 0 CD
- CM
*vP rH
CD CD CO
CM rH 00
i-H rH
r-H rH CO
CM CO CO
c*^ r^, to
LO CD
CM CO CM
CO 00 CD
c^a "st*
CO CO 'vP
CD CO rH
r 1
CM ^^* ^J^
CO r-i i i
tt
-H O O
"H lj
JS S1 ซ
o 5 .j?3
EH CO O
CM
1^.
CM
V
CD
LO
o
CM
o
LO
0
CM
o
"vP
CO
CD
LO
o
CD
o
o
CD
LO
0
V
*
f_
CD
+J
(d
Wastewi
CD
I 1
0
SP
II
CD
CM
CM
LO
ซ
o
CM
1 1
*
o
CM
co
to
Cx*
LO
^p"
LO
CM
LO
CO
CM
o
4->
o
la
EH
"3
0
r-H
(0
"o
EH
M-l
O
Percent
^
e
CO
co
co
CD
-------
w
w
^^^
CO
p
i
w
W
U
1
LO
pa
10
to
10
to
ffi
CO
o
X!
CD
O
O
O
!>- LO
xt* (M
*
CO O
co
CD
C- CD
CD
CO
0)
(M
CD
CM CO
CD CXI
(M LO
CM
LO CD
*tf r-t
t^- LO
t-l
LO
CD CO
(D 0)
ra
CNI
co
oo
CO
CD
LO
CO
CO
CM
CD
co i>.
~ LO
o co
co
CD CO .
CO CO
co co co
CD
4-J
I
CD
en Jg to
_S J3 J?
ra -O ^
vt"
LO
CS1
CT>
CM
LO
CO
CD
CM
LO
CO
O
r)
(0
O
i
(0
0
O
cu
co
CO 1-4
CD CO
co r--
CD CO
en
CO
CO
oo
CD
CO
CO
(Nl
fO
LO
CD =}<
"
CO iI
CO i-H
r-H 05
CO CD
C^ CM
LO LO
ii - ซ:
03 O
CT)
LO
CM
LO CO
< ,
CO r-I
ro
I
s
O
(D
(->
0)
-o
CD
W
rd
ft
0)
!-,
(0
CD
rd
O
w
0)
fO
3
152
-------
TABLE G-6 PUROX SYSTEM HEAT BALANCE. TEST PERIOD 3
Refuse*
Sludge
Fuel Gas
Oxygen
Total In
Metal
PUROX System Gas
Slag
Char
Oil
Wastewater
Losses
Total Out
A
GJ/Day
667.5
28.4
68.4
764.3
419.6
25.2
45.4
53.8
47.3
124.0
715.2
49.1
BTU x 10~6/Day
632.7
26.9
64.8
724.4
397.7
23.9
43.0
51.0
44.8
117.5
677.9
46.5
Efficiency (as operated) 58.7 percent
of Conversion (as projected) 71.2 percent
* Value is average for entire test. Efficiency based on total out.
153
-------
APPENDIX H
TEST PERIOD 4. MIXED SLUDGE - MAY 23-25, 1977
The final test using 20.3 percent solids mixed secondary and primary
sludge at a dry sludge-to-refuse rate of 0.035 was conducted between 23 and
25 May. The heat and mass balances for this period are shown in Tables H-l
through H-60 Mixed primary and secondary sludge was processed at a rate of
10.5 Mg (11.6 tons) per day with'78.0 Mg (86.0 tons) of refuse for a wet sludge-
to-refuse ratio of 0.135. The balances were calculated using the same methods
for the refuse only case.
Mass balance results were similar to the results obtained during the
refuse only test and quite different from the earlier mixed sludge test. Gas
production was 395.7 Nm3 per Mg (13,400 Sft3 per ton) of feed; lower than the
refuse only and earlier mixed sludge results. The oxygen-to-feed, ratio of 0.18
was consistent with the refuse only and primary sludge results and much lower
than the result for mixed sludge obtained during the 2 May - 6 May period.
Product gas heating value was 15.22 MJ per .Nm3 (386.6 Btu/Sft3), also con-
sistent with refuse only results.
The trace metal distribution was similar to the results from the other
test periods. The nickel contained in the char was again determined using the
nickel concentration of the refuse-only char sample. The dilution of the
wastewater by scrubber makeup limits wastewater metal content measurements.
The fraction of each metal in the exiting streams is shown for comparison.
The heat balances on this period was consistent with previous balances.
Losses were high due to the large quantity of water added by the sludge stream,
and the conversion efficiency was 57.9 percent during the test. The efficiency
projected for particulate recycle was 71.2 percent.
154
-------
^_^
o
g
w
Q
O
11
OH
En
CO
13
*.
M
O
a
2
CQ
S
H
0
O-.
s
O
O
P
S
CQ ง
o -3
CSl 'g
M
'5
CO |g
'oi
o >>!
<
to
T!
1 ' . , C3^
S
CQ '
CQ
(0
CQ
Cn
i i ^^
"Q ^
En
s
fO
CD
CO
r-i co co to
CO O LO il
1 ซ-* ' 1 1 1 1 I 1-i 1
Cx CO Cx r-l
"* CO O O
1 CSl 1 1 1 1 1 1
r-l r-l
1 0 I I | | I I | 0 |
"vt* ^f LO CO Cx CN1 CN!
CNI O LO CM i 1 O O
1 r-l 1 r-l I 1 CSl
i 1 r 1 CO Cx CM
1 J* CD CSl CO 'st*
CO O ซ-H LO O CSl r-l
1 r-l 1 1 1 r-l
O LOCxr-iCOCOCOcSICSlCO
CO Or-HtOtOCOi ICx-NPi 1
Cx I r-l r-l O -sP r-l
11
91
-f-t
ซn fO c C
1>CQCQ 91O 0'~'^-i!>_ d>--; rc)O ^si
ง*" ^I^S^g*^^
CD r=)3-I^O
-------
ffi
CQ
3
8
w
Q
CO
O
CQ
CQ
to
|H
co
(0
CQ
OJ O
CQ
o s
co
o
0
CQ
CO
CD
oo
Q)
CD
CO
0)
O
0
Pi
en
CM
CO
0
r-.
00 r-l
CM O
oo co LO
CO 01
CM CS3
CO
CD CO
CD i-H
CD CD
CM
CD
10 CM
O CM
O- CD
LO O5
CQ
(0 to
0
CQ
CO
(0
0
0
rd
H->
0
$
s1
Cti
CJ)
CD
OO
CM CD
CM CM
O O
CD CO
CM
i
w iS (0
^ .0 ฃ
156
-------
TABLE H-3. AVERAGE PRODUCT GAS ANALYSIS. TEST PERIOD 4
Mole %
H2
CO
CH4
C2H2
ฐ2H4
ฐ2H6
C3H6
ฐ3H8
C4
ฐ5
ฐ6H6
HHC
H2S
CH OH
O
Cฐ2
ฐ2
Ar
25.30
36.72
6.48
0.60
2.26
0.46
0.39
0.21
0.44
0.31
0.26
0.23
0.02
0.11
25.01
0.26
0.95
157
-------
O
w
PH
H
03
O
W
O
10
TJ
ca
0)
O
O
O
tn
Cn
(0
0
4^
CO
CM
00
LO
CO
co
CO
LO
CM
CO
CD
Q)
w
0
CD
O
CM
*
LO
co
co
LO
LO
CD
CO
CD
CD
LO
0
Cn
CO
CD
CM
CM
rH
CM
CO
CD
LO
CD
CM
CM
LO
00
LO
CM
CM
CD
CM
LO
CD
CO
CD
CD
co
xt*
r
CD
CD
CO
LO
CD
CM
CM
CO
CO
CO
CO
CM
CM
CM
CD
CO
rd
ft
O
CM
CO
oo
CM
co
CD
CD
LO
Q)
4->
(0
.(0
CO
CD
co
CO
CO
CM
CO
co
CO
co
CD
CM
CO
co
co
LO
o
o
0
o
cu
co
co
CO
LO
CD
CO
CO
CD
LO
CD
co
CD
CD
LO
CD
Cn
fO
**
CO
CM
co
CD
CD
LO
CO
co
CD
CO
CO
(0
J3
o
CO
CO
LO
CD
o
o
CD
+J
w
.to
o
cu
cu
CO
to
&
-------
ffl
CO
r-H
O
S
"JF
P
O
2
w
p^
CO
w
E-H
8
1
~H
,
^ CD'S
il PH -Ci-
CQ
-f-4
01
fO * N
CQ 0 rQ
O ci
s-i jQ
Ox ^- J
T3 ^Q
O ^
CO
to
^
CO
0
CD
CO
o
CM
i
co
CM
CO
1-4
il
0
co
LO
CO
CD
CD
CD
CD
LO
LO
CM
*^F
CM
CM
,
Refuse
CM
^
o
LO
f^
c*^
LO
r 1
CO
CM
i 1
LO
to
1 1
CO
00
to
1
LO
^f
i-H
1 1
11
CO
LO
o
LO
co
CO
to
- ซ=*
LO
co
1
to
^r
^
r-l
to
to
'I1
CO
to
r-H
CD
CD
LO
o
Cปx
to
CO
CO
ฃ
t 1
o
o
.. CD
1 1
CM
CO
i-H
to
t-ป
"*.
co
"=3*
i-H
^
o
CM
to
CO
i-H
o
CD
o
! 1
CM
CD
CM
rH
CM
O
(3
CO
^
CXI
0
LO
(V^
c^.
co
CO
o
CV.
CD
IS-
0
CO
CM
11
CO
CM
co
1^.
CM
0
to
CO
o
S-i
(0
o
o
0
to
V
co
11
1 1
o
to
o
LO
0
LO
CO
1 1
LO
t
0
CO
1 1
0
CO
11
V
*
(
ffi
_l_l
fO
Wastew
^
r-H
CM
0
CM
CD
CO
LO
CO
ts.
CO
LO
co
LO
CO
IX
1 1
CO
1 1
CM
CO
to
CO
1 1
CM
*
LO
co
.
0
1 1
3
0
r 1
0
EH
m
3 ฐ
o ฃ
r^ 0
(0 O
0 CD
H O-i
CO CM
IS. CM 1
CO r-H
vf co co
co co o
CD
LO LO
CD O
OO i i |
is. o co
O CO r-l
CO i-H
CD to LO
LO "vf O
CD
i 1 O CD
to co o
CO i-H
IS- O CO
999
CD O O
IS. CM
to 'vf
LO =!<
CD 1
0 0
CD CD |
CM CO
ฃH
0
(0
0
Cn ^ w
to J2 a
ฃ 0 ฃ
w
~e
-rH
i H
G
0
4-J
o
I t
0
13
o
rj-j
0
CQ
(0
0
(0
in
c
fO
-a
o
"o
C/)
0
H->
o
*
159
-------
TABLE H-6. PUROX SYSTEM HEAT BALANCE . TEST PERIOD 4
Refuse*
Sludge
Fuel Gas
Oxygen
Total In
Metal
PUROX Gas
Slag
Char
Oil
Wastewater
Losses
Total Out
A
GJ/Day
902.2
33.6
92.7
1028.5
-
535.5
34.2
90.8
71.7
56.1
136.8
925.1
103.4
Btu x 10~s/day
855.1
31.8
87.9
-
974.8
-
507.6
32.4
86.1
68.0
53.2
129.7
877.0
97.8
Efficiency (as operated) 57.9%
Of Conversion (as projected) 71.2%
*Value is average for entire program. Efficiency based on total out.
160
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-6 0 0/2-78-198
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE ;'
The Codisposal of Sewage Sludge and Refuse In
the PURQX System
5. REPORT DATE
December 1978 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
PUROX System Engineering
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Union Carbide Corporation
Linde Division
Tonawanda, New York 14150
10. PROGRAM ELEMENT NO.
1BC611. SOS #1. Task B/05
11. CONTRACT/GRANT NO.
S803769-01-3
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research LaboratoryGin.
Office of Research and Development
U. S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
,OJL
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: Gerald Stern 513/684-7654
16. ABSTRACT
This program was conducted to establish the technical feasibility of codisposal of
filtered sludge cake from primary and secondary wastewater treatment with municipal
refuse in the PUROX System; to determine the environmental effect of this process; and to
estimate the economics of sludge disposal in this fashion. At the PUROX System
demonstration plant in South Charleston, West Virginia, operation at dry sludge-to-refusc
ratios up to 0.075 was obtained. System performance was similar to that obtained with
refuse only. Environmental effects were within Federal guidelines for emissions.
Economics were found to be strongly influenced by site-specific factors with sludge
disposal costs varying from less than $20 per Mg to above $100 per Mg of dry sludge.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Sludge disposal
Refuse disposal
Gasification
Pyrolysis
Slagging
PUROX System
Sludge and refuse
codisposal
13B
18. DISTRIBUTION STATEMENT
Release to public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
177
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
161
* U.S. GOVERNMENT PRINTING OFflCE: 1979 -657-060/1548
-------
-------
-------
-------
-------
vo
oo
e
Z D
m
O -n
-0 O
H 1 CD
C < C
z > cn
~l "l ป
H
m
rn yj
^ rn
Tl -
r (fl
?8
rn ฐ
m
0)
0)
1
C/3
o m
< ป 7
- ^
1 S ง
o CD a
CD
:ฑ CD
O
C/5
0
03 O
O 2.
5? 10
O
S 1 ฐ o
CD i- CD
00 2' FT O
O
m
O
c g
Co c:
" "
M
CTl
oo
2 -a
0)'
o o
Z 0)
-n
CD o
rn
> >
O
CJ
a
H m
m 0)
O j=i
O
------- |