&EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/7 79-230
October 1979
Research and Development
A Prototype Mobile
System for Pyrolysis
of Agricultural and/or
Silvicultural Wastes
Interagency
Energy/Environment
R&D Program
Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
us EPA
and Chemi^ Ubraries
EPA West Bldg Room 3340 EPA-eoo/ 7-79-230
Maiicode 3404T October 1979
1301 Constitution Ave NW
Washington DC 20004
202-566-0556
A PROTOTYPE MOBILE SYSTEM FOR PYROLYSIS
OF AGRICULTURAL AND/OR SILVICULTURAL WASTES
by
J. W. Tatom, A. R. Colcord, W. M. Williams,
and K. R. Purdy
Engineering Experiment Station
Georgia Institute of Technology
Atlanta, Georgia 30332
and
J. J. Demeter, C. R. McCann, J. M. Ekmann,
k/^ and D. Bienstock
>& Pittsburgh Energy Research Center
^ U. S. Energy Research and Development Administration
r\ Pittsburgh, Pennsylvania 15213
Grant No. R 803430
Project Officer
W. W. Liberick, Jr.
Energy Systems Environmental Control Division
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
Repository Material
Permanent Collection
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory, U. S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U. S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
ii
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on our
health often require that new and increasingly more efficient pollution con-
trol methods "be used. The Industrial Environmental Research Laboratory-Cin-
cinnati (lERL-Ci) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and economically.
This project was designed to answer specific questions related to the
development of a prototype system for pyrolysis of agricultural and silvi-
cultural wastes into clean fuels. These questions were concerned with the
operating characteristics of the Georgia Tech Engineering Experiment Station
pyrolytic converter, the utility of pyrolysis gas as a fuel for a spark-
ignition internal combustion engine, and the combustion and emission charac-
teristics of pyrolysis char and oil. The research users will find the data
base of this report an adequate starting point for the evaluation of the
technical feasibility of the mobile pyrolysis concept. Further information
on the conversion of agricultural and forestry wastes to usable fuel products
can be obtained from the lERL-Cincinnati Fuels Technology Branch.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
iii
-------
ABSTRACT
This research program was initiated to investigate three elements of a
prototype mobile system for pyrolysis of agricultural and/or silvicultural
wastes into clean, transportable fuels: the pyrolytic converter itself,
a pyrolysis-gas-fueled internal combustion engine, and the combustion and
emission characteristics of pyrolytic char and oil.
An experimental study of the performance of the Georgia Tech Engineering
Experiment Station one-tonne-per-hour pyrolytic converter was conducted.
Peanut hulls and pine sawdust were used as representative agricultural and
silvicultural waste materials. Effects of converter capacity, feed material,
mechanical agitation, bed depth, and air-to-feed ratio on product yields
were determined. In addition, the performance of an integrated, mechanical-
agitation, process-air-supply system (Airgitator) designed to improve the
throughput of the converter was determined.
From these studies and an earlier study performed on a 0.5-tonne-per-hour
converter, it appears that feed material, converter capacity, mechanical
agitation, and "Airgitation" have little influence on product yields. Bed
depth, although not affecting the combined energy yield of the char and oil,
substantially influenced the relative amounts of char and oil produced.
The air-to-feed ratio was found to be the dominant variable. The combined
energy yield of the char and oil was a simple linear function of the
air-to-feed ratio.
A spark-ignition internal combustion engine fueled with dry simulated
pyrolysis gas was experimentally investigated to determine stable, full-
throttle operation of this engine and to compare brake power output with
that when the engine was fueled with gasoline. Excellent stability was
iv
-------
obtained and the brake power output was found to be 60 to 65 percent of
that when the engine was fueled with gasoline.
An experimental study of the combustion and emission characteristics of
powdered char was performed by the Pittsburgh Energy Research Center. Stable
combustion and satisfactory combustion efficiency were obtained in a 227
kg/hr pulverized-coal-fired, water-wall combustor fueled with powdered
high-volatile char alone and with 50-50 blends of high-volatile or low-
volatile pulverized char and coal. Excellent flame stability and carbon-
combustion efficiency were obtained when powdered char was mixed with
pyrolysis oil and No. 6 fuel oil and fired as a slurry in an oil-fired
boiler. Studies of the combustor and boiler flue gases showed significant
reductions in S0? emissions. The low sulfur and nitrogen content of the
char makes it an attractive fuel to mix with either high-sulfur coal or
oil to meet S07 and NO emission regulations without emission-control
£» X
devices other than a baghouse or an electrostatic precipitator for
particulate control.
This report was submitted in fulfillment of Grant No. R 803430 by Georgia
Institute of Technology under the sponsorship of the U. S. Environmental
Protection Agency. This report covers a period from May 1, 1975, to
July 31, 1977, and the work was completed as of July 31, 1977.
-------
CONTENTS
Foreword ill
Abstract iv
Figures viii
Tables x
Nomenclature xi
Acknowledgments ". xii
1. Project Goals 1
2. Conclusions and Recommendations 2
3. Introduction 5
4. Parametric Study of EES Pyrolysis System 10
5. Performance Study of Integrated Mechanical Agitation
—Air Supply System 40
6. Performance Study of Spark-Ignition Engine Fueled with
Simulated Pyrolysis Gas . 56
7. PERC Combustion and Emission Study of Pyrolysis Char
and Oil 66
References 87
Appendices
A. Laboratory procedure 89
B. Laboratory data 92
C. Listing of data reduction computer program 110
D. Approximate analysis of brake de-rating factor 124
vii
-------
FIGURES
Number Page
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
EES pyrolysis system process flow diagram
Fourth EES pyrolysis pilot plant
Close-up view of EES pyrolysis pilot plant
Close-up view of conveyer and input system — EES pyrolysis
pilot plant
Close-up view of cyclone and condenser system — EES pyrolysis
pilot plant
Close-up view of off-gas burner — EES pyrolysis pilot-plant . . .
Schematic of EES converter with rotating agitator
Percent available energy in char-oil mixture — parametric study
Percent available energy in char-oil mixture — parametric
Energy breakdown of pyrolysis products — parametric study ....
Oil yield variation with increasing bed depth
Effects of feed moisture on available energy from char-oil
mixture (Reference 1)
Schematic of EES converter with integrated mechanical
agitation — air supply system
Final design of Airgitator
Overview of Airgitator
Airgitator as installed
Percent available energy in char-oil mixture — Airgitator/
parametric study composite
Percent available energy in char-oil mixture — Airgitator/
parametric study/Reference 1 composite
Energy breakdown of pyrolysis products — Airgitator/Ref erence 1
composite
Heating value of noncondensible gas — Airgitator study
11
12
13
13
14
14
17
31
32
33
36
38
39
42
43
44
44
51
54
57
55
58
viii
-------
FIGURES (continued)
Number Page
23 Impco CA-300A mixer 58
24 Overview of gaseous fuel system 59
25 Gaseous fuel metering system 59
26 Wide-open-throttle performance of General Motors truck engine
for gasoline and simulated pyrolysis gas 62
27 View of 227 kg/hr (500 Ibm/hr) pulverized coal-fired-furnace . . 68
28 Schematic of 227 kg/hr (500 Ibm/hr) pulverized coal-fired
furnace 70
29 Half-section view of principal components of the combustion
system 71
30 Multifuel burner assembly 72
31 Worn Impact rotor disc and hammers after char pulverization ... 78
32 View of hammers before and after pulverization of char 79
33 View of 981 kW (100 HP) Firetube boiler used for coal-oil
slurry combustion studies 81
34 View of boiler, No. 6 fuel oil storage tanks, and slurry
mixing and feed tanks 82
35 Simplified flow diagram of the 981 kW (100 HP) coal-oil slurry
and char-oil slurry combustion test facility 83
ix
-------
TABLES
Number Page
1 Test Summary—Parametric Study 26
2 Summary of Transformed Data—Parametric Study 28
3 Test Summary—Airgitator Study 47
4 Summary of Transformed Data Airgitator Study 49
5 Optimum Engine Performance 65
6 Brake De-rating Factor—SPG/Gasoline 63
7 Heating Values of Stoichiometric Mixtures 64
8 Brake-De-rating Factors 65
9 Typical Analyses of Pittsburgh Coal, Char and Char Coal Blends . 74
10 Experimental Results of Combustion Tests with Pulverized Coal,
Char and Char Coal Blends 75
11 Sieve Analyses of Pulverized Coal, Char and Char Coal Blends . . 77
12 Typical Analyses of Oil and Coal-Oil Slurry and Char-Oil
Slurry 84
13 Operating Conditions for Selected Periods with No. 6 Fuel Oil,
Coal-Oil Slurry and Char-Oil Slurry 85
-------
NOMENCLATURE
Symbols
h
HV
L
M
w
Definition
enthalpy
Heating Value
Losses (see Equation 2)
Mass
weight fraction
Subscripts
a
c
ch
f
h
o
n
wi
wo
xch
xf
air
carbon
char
feed
off-gas
hydrogen
oil and oxygen
nitrogen
water in feed
water in off-gas
ash in char
ash in feed
xi
-------
ACKNOWLEDGMENTS
This effort was supported by the Industrial Environmental Research
Laboratory (IERL), U.S. Environmental Protection Agency, under Grant No.
R 803430. We wish to express our appreciation to Mr. Walter W. Liberick, Jr,
of IERL and Mr. Donald A. Oberacker of the Municipal Environmental Research
Laboratory for their contributions and suggestions.
The School of Mechanical Engineering, Georgia Institute of Technology,
provided the staff and facilities for conducting the internal combustion
engine study. We greatly appreciate Dr. Stothe P. Kezios1 cooperative
efforts in this regard. Special appreciation is expressed to Dr. Wendell M.
Williams for directing this study and to Mr. Louis A. Cavalli for his
assistance.
The Pittsburgh Energy Research Center, U.S. Energy Research and Development
Administration, performed the combustion and emission study. Mr. Daniel
Bienstock and Mr. Joseph Demeter directed this excellent investigation.
xii
-------
SECTION 1
PROJECT GOALS
This project was designed to answer specific questions related to the
development of a prototype system for pyrolysis of agricultural and
silvicultural wastes into clean fuels. These questions were concerned with
the operating characteristics of the Georgia Tech Engineering Experiment
Station (EES) pyrolytic converter, the utility of pyrolysis gas as a fuel
for a spark-ignition internal combustion engine, and the combustion and
emission characteristics of pyrolysis char and oil.
The specific project goals were:
1. To determine the influence of system capacity, feed material
mechanical agitation, air-to-feed ratio, and bed depth on the
product yields of the EES pyrolytic converter.
2. To determine the performance of an integrated mechanical
agitation and process-air supply system.
3. To determine the full-throttle performance of a spark-
ignition engine fueled with dry simulated pyrolysis gas.
4. To determine the combustion and emission characteristics of
powdered char, powdered char and high-sulfur coal, and a
slurry of powdered char, pyrolysis oil, and No. 6 fuel oil.
(This portion of the project was performed by ERDA's
Pittsburgh Energy Research Center through a separate
agreement with the US EPA's Industrial Environmental
Research Laboratory.)
-------
SECTION 2
CONCLUSIONS AND RECOMMENDATIONS
CONCLUSIONS
From the results of this project, the following conclusions can be drawn:
1. The effects of converter capacity, feed-material type, mechanical
agitation, and "airgitation" are all minor in comparison to that
of the air-to-feed ratio.
2. From the results of this project and earlier work, the available
energy in the char-oil mixture appears to be solely a function
of the air-to-feed ratio; all data are correlated by a single
linear relationship between the available energy and the air-to-
feed ratio.
3. Although the available energy in the char-oil mixture is only a
function of the air-to-feed ratio, the relative amount of char
and oil is dependent upon the bed depth.
4. Peanut hulls can be processed easily either with or without
mechanical agitation.
5. The char and oil yields are unaffected by the substitution of an
integrated mechanical agitator and process-air supply system
(Airgitator) for the fixed, water-cooled air tubes of the EES
converter. Except for an apparent increase in the off-gas-stream
particulate content and temperature, the Airgitator performed well.
6. The conversion of existing intermittent-duty, spark-ignition
gasoline engines to continuous-duty, pyrolysis-gas engines
appears to require only the development of an automatic fuel-air
mixer. The full-throttle brake-power output of a six cylinder
engine fueled with dry simulated pyrolysis gas was 60 to 65
percent of that when the engine was run on gasoline.
-------
7. Stable combustion and satisfactory combustion efficiency were
maintained in the Pittsburgh Energy Research Center (PERC) 227
kg/hr pulverized-coal-fired, water-wall combustor fueled with
high-volatile char alone, and with 50-50 blends of high-volatile
or low-volatile char and coal.
8. When pulverized char was mixed with pyrolysis oil and No. 6 fuel
oil and fired as a slurry in an oil fired boiler, excellent flame
stability was experienced and carbon combustion efficiency was
equal to that obtained with No. 6 fuel oil alone.
9. The low sulfur and nitrogen content of the char makes it an
attractive fuel additive for either high-sulfur coal or fuel oil.
In addition to extending fuel oil supplies, powdered char, added
in the proper proportions, will permit compliance with S0_ and
NO emission regulations without emission control devices other
3t
than a baghouse or an electrostatic precipitator for particulate
control.
RECOMMENDATIONS
Although the results of the study strongly support the technical feasibility
of the mobile pyrolytic converter concept by providing additional operating
data, by demonstrating operation of a spark-ignition engine on simulated
pyrolysis gas, and by demonstrating the attractive combustion and emission
characteristics of the char and oil, the following tasks are recommended
before a complete mobile system demonstration project is initiated:
1. An improved off-gas system should be developed that will permit
continuous, round-the-clock operation with at least 90 percent
uptime. (The seriousness of the off-gas system servicing problem
was not recognized until continuous operation was attempted at
the 45 tonne-per-day demonstration plant owned and operated by the
Tech-Air Corporation, the exclusive licensee for the process.)
2. An automatic fuel mixer for essentially particulate-free pyrolysis
gas should be developed and demonstrated on a spark-ignition
engine coupled to an electric generator.
-------
3. The Airgitator should be studied during long-term runs to
determine its ability to provide stable operation and its
effect on the particulate level of the off-gas stream. Also,
a more balanced configuration should be developed to reduce the
unsymmetrical forces on the drive system.
4. When tasks one and three have been successfully completed, a
full-scale mobile pyrolytic converter should be designed,
fabricated, and tested.
-------
SECTION 3
INTRODUCTION
GENERAL
This report describes an experimental program to develop a mobile pyrolysis
system for conversion of agricultural and silvicultural wastes at the site
of their production into clean and easily transportable fuels. The program
included a series of tests using peanut hulls and pine sawdust as feed
materials in the one tonne/hr Georgia Tech Engineering Experiment Station
(EES) pyrolytic converter pilot plant, and was a follow-on study to earlier
work [1,2,3,4] using wood waste as the feed material in a smaller, 227 kg/hr
(500 Ibm/hr) EES pilot plant.
RATIONALE FOR MOBILE PYROLYSIS CONCEPT
Agricultural wastes represent a huge potential source of energy for the U.S.,
but certain problems have limited their use as fuels in the past and must
be dealt with in any successful energy conversion system. These problems
include the following:
• Agricultural and silvicultural wastes (organic matter) typically
contain 30 to 70 percent water, and therefore, are relatively
low in heating value per kilogram. Since these waste materials
would be scattered all over the country-side, transportation
costs per megajoule to large thermal conversion plants would
be very high.
• Because of the moisture content of these waste materials, the
practicability of using existing thermal conversion equipment
is doubtful, at least at its rated capacity. Most likely, new
or modified facilities will be required. (The overall steam-
side efficiency of boilers utilizing wet organic fuels, such as
-------
bagasse and bark, is typically 60 to 65 percent. Thus, a
serious conversion penalty results from using these materials
as-received.)
• The particulate emissions from boilers operating on raw
organic fuels would likely require the installation of
expensive flue gas clean-up equipment.
• Agricultural wastes, with a few exceptions, are produced
seasonally, rather than continuously. Thus, a steady supply
of fuel from these wastes is not available; also, it would be
impractical to tie up costly equipment that cannot be used
year round.
• Associated with the construction of a waste conversion facility
dependent upon an adjacent, fixed supply of wastes over a long
time period are contractual problems between the producer of the
wastes and the waste utilizer. Although the waste producer might
initially be spending two to five dollars per tonne for disposal
of raw wastes, he might hesitate or refuse in a long-term contract
to give away or perhaps pay a charge for disposal of his wastes.
And clearly, once a facility for waste utilization has been
constructed, the waste producer, upon termination of the original
contract, would have the waste utilizer in an uncomfortable economic
position.
One solution to these problems is to utilize a mobile pyrolysis system that
could be transported to the site of waste production and there convert the
wastes into a char, an oil, and a low-quality gas. The gas could be used
to dry the wet feed and to operate the associated equipment, and the oil
and char could be sold as fuels. The reduced weight and associated trans-
portation costs thereby effected would be very substantial. A further
benefit would be the greater leverage provided the waste utilizer in contract
negotiations with the waste producer, since the unit could always be moved
to a new location. The portability feature would also guarantee greater
equipment utilization and, through proper scheduling between seasonal
agricultural wastes and continuously available silvicultural wastes, could
provide an almost constant supply of fuel. Finally, since the portable
-------
system could be assembled in factories using mass production techniques, it
would likely be less expensive than a comparable, fixed installation system.
The Engineering Experiment Station (EES) at Georgia Tech over the last eight
years has developed a simple, steady-flow, low-temperature, partial-
oxidation pyrolysis system which is completely self-sustaining. In the EES
design waste material is pyrolyzed in a vertical porous bed. This unit
requires no special front-end system, has very few moving parts, and depends
upon a relatively small blower to provide the air supply necessary to
maintain the partial oxidization of the feed. Typically, a tonne of as-
received wastes would be converted, using the EES process, to about 225 kg
(495 Ibm) of a powdered char-oil fuel, similar to coal, with a heating value
of 25.6 to 30.2 MJ/kg (11,000-13,000 Btu/lbm). Thus, depending upon the
feed moisture content (50 percent assumed), the energy available for use
at the central thermal conversion plant could be 64 to 76 percent of that
theoretically available from the original dry waste; and, at a boiler
conversion efficiency of 80 to 85 percent, the overall steam-side efficiency
of the process could be up to 65 percent. Hence, the percentage of useable
energy could be as great as that available with direct burning, but with
avoidance or significant reduction of the problems of:
• Transporting the wastes.
• Modification or construction of new facilities compatible
with fuels derived from organic wastes.
• Emissions resulting from unburned fuel particles.
The powdered char-oil fuel could be burned in either suspension-fired or in
stoker-fired boilers with essentially no modification. Also, it could be
blended with cheaper high-sulfur coal to an additional economic advantage.
Two additional elements, which make the concept even more attractive, have
recently come to light, i.e.:
• The application of the mobile pyrolysis concept to large
barges* moving on the thousands of miles of inland and
* The barge concept was developed by Mr. Kevin Everett of the Florida Resource
Recovery Council and is described in an unpublished paper [5].
-------
inter-coastal waterways appears to have great promise. This
would not only permit an increase in the size of the mobile
system, but also would allow its application to the municipal
wastes of smaller communities which presently cannot individually
justify or afford a large, economical waste conversion system,
but with other communities could successfully operate such a
system.
• The char-oil fuel produced by the mobile pyrolysis system [1]
was considered primarily as a coal substitute which could be
used in existing suspension or stoker-fired systems. It appears
now, from work with coal-oil slurries at Combustion Engineering
[6], General Motors [7], and at ERDA's Pittsburgh Energy Research
Center (PERC) [8], that firing of combinations of petroleum oil
and the char-oil mix in energy release ratios of up to 50 percent
may be practical in existing oil-fired boilers with minimal or no
modification. The low sulfur content and relatively low ash
content of the char-oil mixture make it highly desirable as a
fuel-oil extender, and presently no technical obstacles preventing
its use are anticipated. Because so many existing boilers are
oil-fired, this development may represent an important step away
from reliance on oil alone as a boiler fuel.
These two considerations should have relatively little influence on the
development of a portable system, but strengthen significantly the justifica-
tion for use of the portable concept for production of the char-oil fuel.
(Combustion and emission tests of char/coal and char-oil/fuel-oil mixtures,
performed by ERDA/PERC, form a part of this study and are reported in
Section 7.)
OBJECTIVES
The investigations, which were primarily experimental, had the following
objectives:
• To determine the effects of scale on pyrolytic converter
performance.
-------
• To determine the effects of changing feed material on pyrolytic
converter performance.
• To determine the effects of mechanical agitation on pyrolytic
converter performance.
• To determine the influence of air-to-feed ratio and bed depth
on product yields.
• To determine the performance of an integrated mechanical
agitation-process air supply system.
• To determine the full-throttle performance of a spark-ignition
engine fueled with simulated dry pyrolysis gas.
• To determine the combustion and emission characteristics of
powdered char and pyrolysis oil from the thermal conversion of
a 50/50-tnixture of pine sawdust and bark—char by itself, char
blended with powdered coal, and char blended with pyrolysis oil
and No. 6 fuel oil.
In the following sections a description of these studies is presented.
-------
SECTION 4
PARAMETRIC STUDY OF EES PYROLYSIS SYSTEM
GENERAL
This experimental program was conducted in the new, one tonne/hr EES pilot
plant. Peanut hulls were used as the feed material in a series of nine
tests and sawdust was used in two tests, for a total of 11 tests in the
complete study. All told, approximately 40 metric tons (44 tons) of feed
were used in this program. The tests involved investigation of the influences
of scale, feed, air-to-feed ratio, mechanical agitation, and bed depth on
product yields. This section presents a description of the test facilities,
the calibration and testing procedure, the laboratory procedure, the data
reduction methodology, and the results of this test program.
FACILITIES
A process flow diagram of the EES pilot plant is shown in Figure 1.
Photographs of this unit showing views of the separate components involved
are presented in Figures 2 through 6.
The system operates in the following manner: The peanut hulls (dried at
the sheller) are collected, weighed, and stored in drums. During a test,
the drums are emptied into a receiving bin which supplies a conveyor to the
pyrolysis unit with input feed. The pyrolysis unit is 5.5 meters (18 feet)
tall and 1.8 meters (6 feet) on each side. The inside of the unit is cylin-
drical, with a diameter of 1.2 meters (4 feet) and a depth of 2.4 meters
(8 feet). The feed enters the converter through a gate valve at the top
and passes down through the vertical bed. Process air tubes are located
in the lower portion of the bed. These water-cooled tubes supply enough
air to oxidize the feed in their immediate proximity, and thereby produce
10
-------
AIR
COMPRESSOR
c
DRY
FEED
FLOW METER
COOLING V-/ IN
WATER
OUT
HOT GAS TO
ATMOSPHERE
t
COOLING
AIR
J ORIFICE
CONVERTER
~i
HOT GAS AND
PARTICULATES
r
CHAR
CYCLONE
I
HOT
GAS
1
N| OUTT
FAN
CONDENSER
WASTE
1
_ WARM
J GAS
&_ SAMPLE
BURNER
DRAFT
FAN
PORT
OIL
PARTICLES
Figure 1. EES pyrolysis system process flow diagram.
-------
Figure 2. Fourth EES pyrolysis pilot plant
12
-------
Figure 3. Close-up view of EES pyrolysis pilot plant.
Figure 4. Close-up view of conveyer and input system—EES pyrolys-is
pilot plant.
13
-------
Figure 5. Close-up view of cyclone
and condenser system—
EES pyrolysis pilot
plant.
Figure 6. Close-up view of off-
gas burner—EES pyrolysis
pilot plant.
I-
-------
sufficient heat for pyrolysis of the remaining feed material. The char at
the bottom of the bed passes through a mechanical output system and into a
screw conveyor that transports it into receiving drums.
The gases produced during decomposition of the feed pass upward through
the downward-moving feed and leave the unit near its top. The gases then
pass through a cyclone where particulates are removed and then to an air-
cooled condenser which operates at a temperature above the dew point of the
mixture. The condenser removes the higher boiling point oils, which are
collected and weighed. The remainder of the uncondensed oils, the water
vapor, some condensed oil droplets, and the noncondensible gases pass
through the draft fan and into the burner which incinerates the mixture.
The amount of gas production is controlled by the bed temperature, which,
in turn, is controlled by the air-to-feed ratio.
The instrumentation used in the study included:
• An in situ calibrated orifice to measure process air flow rate.
• Scales to weigh the dry input feed, the char, and the oil yields.
• A water meter to measure total cooling water flow.
• Dial thermometers to measure inlet and exit cooling water temperatures.
• Various thermocouples to measure the pyrolysis gas temperature at
several points in the system, internal bed temperature, external
surface temperatures, and the burner temperature.
• A multiple channel recorder to provide continuous read-out of the
various thermocouples.
• A gas sampling system for laboratory analysis of off-gas composition.
The system is normally operated at a system pressure of a few centimeters
of water below ambient; thus, any leaks present generally result in the
introduction of air into the system. However, within the cavity between
the sliding plates of the gate valve, the displacement of the pyrolysis gas
by the input feed does result in some lost gas when the gate valve operates.
As the process rate of the unit increases, the gas production increases
and the system pressure tends to rise. To control the pressure, the draft
15
-------
fan speed can be varied within certain limits. The unit has pressure relief
doors which operate at about 25 centimeters (10 inches) of water. These
doors provide a safe means of relieving overpressure from any system
malfunction.
The process rate of the system is governed by the setting of the output
feed mechanism. A level indicator senses the need for additional feed, and
activates the gate valve and conveyor system to provide the necessary input.
Thus, the feed system is activated only upon demand, not continuously;
hence, the gases lost through the gate valve do not represent a significant
energy loss or pollution problem.
The condenser is of a relatively simple design having a series of air-cooled
vertical tubes through which the hot pyrolysis gases pass. The condenser
temperature is governed by a thermostatically operated fan which controls
cooling air flow. In all except the last tests, the condenser was operated
at about 93°C (200°F). It has been observed that oil droplets are frequently
carried in suspension through the off-gas system, past the draft fan, and
into the burner. This results in an error in the materials balance; however,
additional analysis of the gas stream provided data to correct for the error.
In many of the tests, a simple rotating mechanical agitation system was
utilized to enhance the flow of material through the waste converter and
to prevent the formation of bridges or arches which can obstruct the
downward-moving feed. A schematic view of the agitator used in these tests
is shown in Figure 7. The system was operated by a high torque gear drive
system. The maximum rotation speed of the agitator was about one revolution
per minute.
It should be noted that the off-gas flow rate was not measured directly
during the tests because of the presence of droplets of oil and moisture in
the stream which make conventional instrumentation techniques impractical.
Instead, analytical techniques involving nitrogen, carbon, hydrogen, and
oxygen balances were used to compute the flows of the various constituents
which make up the off-gas stream.
16
-------
NONCONDENSIBLE
GASES
FEED
CHAR
Figure 7. Schematic of EES converter with rotating agitator.
17
-------
CALIBRATION AND TEST PROCEDURE
Prior to conducting the tests, many elements of the system instrumentation
were carefully calibrated. The accuracy of some components, such as the
thermocouples, was not checked since the required degree of precision did
not demand temperature measurements of greater accuracy than the nominal
values of the manufactured wire. Also, the accuracy of the cooling water
meter was taken at face value from the name-plate data. Careful attention
was given to calibrating the process air orifice against a laminar flow
element. This ASME sharp-edged orifice was calibrated in situ to insure
accuracy. Tares were individually determined for all the drums in which
the dried feed was stored.
The procedure during the tests was relatively straightforward: The unit,
loaded with feed or char the previous day, was heated-up by use of an
electrical resistance heating element. When the temperature was sufficiently
elevated, process air was introduced slowly and the element was removed.
Once it was apparent that the system was operating in a self-sustaining
mode, the output system was activated and slowly brought up to the operating
capacity chosen for the test. Likewise, the process air feed rate was
adjusted to correspond to the desired air-to-feed ratio for the test. The
system was then allowed to come to a steady-state condition, which required
a nominal four hours. Constant checks and adjustments were made during
this period to insure that the actual operating conditions were those
desired. However, it was found that feed process rate and air-to-feed ratio
could be controlled only within limits of approximately plus or minus
10 percent.
Upon initiation of the test run, continuous records of time, feed input,
char output, oil output, orifice manometer readings, and the various
temperatures were made. In addition, a continuous sample of the pyrolysis
off-gases was taken. Every effort was made to insure that the unit remained
in a steady-state operating mode by continuous surveillance and adjustment
of the various instruments measuring and controlling the inputs of the
system. "Grab samples" of the feed from each drum were taken throughout
18
-------
the run. At the completion of a run, all of the char and oil produced were
collected and representative samples of each were obtained. The char
sample was obtained by use of a grain sampler. The oil was collected in a
large drum, mixed thoroughly, and a sample of about one-half liter (one
pint) taken. All of the feed grab samples were mixed and cut using a
riffle splitter to obtain a composite sample of about one kilogram.
LABORATORY TESTING
The laboratory determined feed and product characteristics and subsequently
analyzed the data. Thus, the work was checked carefully and every precaution
was made to insure the accuracy of the results. However, despite these
efforts, occasional inconsistencies did arise. Although inherent errors
associated with the specific test procedures themselves clearly contributed
to the problem, it is believed that the principal explanation for these
occasional inconsistencies lies in the difficulty of sampling. Frequently,
of necessity, a few grams sampled from a run were taken to represent the
entire production of the oil or char in some piece of sensitive, chemical
analysis laboratory equipment. Thus, even though several tests were usually
made, there were some occasional problems with repeatability of results.
Although these variations are predominantly less than one percent, the
overwhelming impression is of good repeatability. The presence, especially
in the CHNO analysis, of even small inconsistencies was found to have a
significant effect on the test results. Thus, while these data stand up
well by ordinary standards, the sensitivity of the overall test results to
some of these data make close scrutiny necessary. A review of the breadth
of the laboratory work done reveals a wide assortment of different analytical
procedures. These procedures include analysis of the:
1. Feed for:
• percent moisture
• percent ash
• percent acid-insoluble ash
• percent carbon
• percent hydrogen
• percent nitrogen
19
-------
• percent oxygen
• heating value
2. Char for:
• percent moisture
• percent ash
• percent acid-insoluble ash
• percent volatiles
• percent carbon
• percent hydrogen
• percent nitrogen
• percent oxygen
• heating value
3. Oils for:
• percent moisture
• percent carbon
• percent hydrogen
• percent nitrogen
• percent oxygen
The composition of the off-gas was determined by gas chromatography and
reported as:
• percent nitrogen
• percent carbon monoxide
• percent carbon dioxide
• percent hydrogen
• percent methane
• percent C9 components as C^H, and C~H,
• percent C_ components as C_H0 and C_H,
J J o J o
• percent C, components as C,H
Presented in Appendix A are brief descriptions of the laboratory procedures
followed to obtain all these data and estimates of the accuracy limits
intrinsic to the test themselves. The data are presented in Appendix B.
20
-------
DATA REDUCTION
General
The primary data obtained from the pilot plant testing, plus the laboratory
findings, provided a substantial body of information and a solid basis to
conduct complete energy, mass, and elemental balances for each test. In
fact, a redundancy in the available information provided the means for an
even more complete evaluation of the internal consistency of the data.
Presented in this section is a discussion of the rationale by which the
data were reduced, and additionally provided is a description of a sensi-
tivity analysis by which the influence on the overall balances of small
variations in the measured results is determined. Finally, a method by
which the initial data are transformed into a generally consistent set of
revised data which simultaneously satisfies the physical conservation
principles and the laboratory findings is presented.
Data Reduction Methodology
The data from the pilot plant testing included the mass of feed processed,
the corresponding char, recovered oil and aqueous yields, and an integrated
off-gas sample. Data regarding pyrolysis bed and off-gas temperatures,
cooling water flow and temperatures, and surface temperature completed the
information available from the testing. The laboratory findings, as
described previously, included percent moisture, ash, carbon, hydrogen,
nitrogen, oxygen, and heating values for the feed, char, and oil. In
addition, the composition of the noncondensible gas was provided. This
then allowed computation of the heating value of the gas.
Using part of these data and the laws of energy, mass, and elemental
conservation, a system of algebraic equations were written. These equations
have been solved on the computer and the calculated results have been
compared with the remaining observed data to obtain a measure of the
internal consistency of the entire set of data. The effects on internal
consistency of small variations in the values of the original data have
also been studied. It has been found that typically variations in specific
measured values of no more than a few percent are required to put all the
21
-------
into a generally consistent form. Since it must be recognized that all the
data are subject to some uncertainty, it has been assumed that, on the
average, the modified values (e.g., the original value plus the computed
variation) are likely superior to those actually measured or initially
computed and, therefore, these modified values have been used in the data
analysis and in the presentation of the results; study of the latter, as
presented in the following section, provides further justification for this
action since the revised data are generally consistent with earlier results
(1) and show an acceptable degree of scatter.
Analysis
The equations*used in the data analysis include:
Conservation of Mass:
M+M+M,+M=M.- + M+M. (1)
g o ch wo f a wi
Conservation of Energy:
(HV+h)M+(HV+h)M+(HV,+h,)M,+h M
g gg o o o ch ch ch wo wo
(HVf + hf) Mf + h M + h M . - [conduction and cooling-water losses]
/
By establishing ambient conditions as a reference, h,. and h can be set to
i a
zero. Since the sensible and latent heat terms involving h , h , h , , and
g o en
h . and the heat losses are generally small in comparison to the other terms,
it is convenient to combine these terms into a single expression
L=h M +h M +h,M,-h.M.+ [conduction and cooling-water
g g o o ch ch wi wi
losses]
and to rewrite the energy equation as:
(HV ) M + (HV ) M + (HV , ) M .+ h M = (HV.) M,. - L (2)
88 o o v ch/ ch wo wo f f
Since L is small compared with the other terms, approximate values can be
taken with little error in the resulting solution.
* A table of Nomenclature is presented on page xi.
22
-------
Conservation of Nitrogen:
wM+wM+w,M,=w,.Mr + w, M (3)
ng g no o nch ch nf f ha a v '
Conservation of Carbon:
wM+wM+w,M,=w,:M,. (4)
eg g co o cch ch cf f v '
Conservation of Hydrogen:
w, M + w, M + w, , M , + w, M = w, c M. + w, . M . (5)
hg g ho o hch ch hwo wo hf f hwi wi v '
Conservation of Oxygen:
w M+w M+w,M,+w M =* w . M,. + w M+w.M. (6)
og g oo o och ch owo wo of f oa a owi wi
In addition to these relations, the Dulong-Petit equation was used to
calculate the heating value of the oil:
HV = 14,500 w + 61,000 w. (7)
o co ho '
The CHNO analysis of the oil requires that:
w + w, + w + w = 1 (8)
co ho no oo
Likewise the CHNO analysis of the char and feed requires that:
w ,+w, ,+w ,+w ,=l-w , (9)
cch hch nch och xch
v f + w.f + v f + \g ,. = 1 - v c (10)
cf hf nf of xf x '
Correspondingly, a computed CHNO composition of the off-gas from the gas
chromatographic results requires that:
w +w, +w +w =1 (11)
eg hg ng og
These 11 equations represent a complete description of the applicable
conservation principles for the data, and, upon simultaneous solution and
comparison with laboratory data provide a redundant body of information
with which to check the internal consistency of the results.
The procedure followed in the data reduction has been to simultaneously solve
the first eight equations for the values of:
23
-------
M *, M *, M * HV +, w + w + w + and w +.
g o wo o co ho no oo
It has been assumed that the 26 terms: M,,M,,M,M.,HV,HV,HV ,
f ch a wi g' o ch
hwo' ™f L' Wng> Wnch' Whf Wna' Wcg' Wcch' Wcf> Whg' Vh' Whw' whf Wog> Woch'
wow* Wof' and Woa are known to wlthin a certain precision—generally less
than 10 percent (based on previous pilot plant and laboratory experience).
Once values of the eight unknowns were determined, a sensitivity analysis
was made to determine the effect of small changes in the 26 known coeffi-
cients on the eight unknowns. This was done with a computer program
(SENSAN—presented in Appendix C) which calculated the values of eight
unknowns (Mg, MQ, MWQ, HVQ, WCQ, w^, w^, and WOQ) for a plus 10 percent
and a minus 10 percent change in each of the 26 known coefficients. The
purpose of this analysis was to identify those coefficients which have a
major influence on the values of the unknowns, particularly the oil composi-
tion. Since the final object was to obtain a set of data as internally
consistent as possible, the next step was a least-squares procedure by
which variations between the measured and computed values of w , w, , w ,
co' ho' no
and WQO were minimized. This was accomplished by introducing combinations
of up to four of the major influencing coefficients previously determined
and by allowing the values to vary simultaneously about their "known" value,
usually within bounds of + 10 percent. The calculated oil composition
(ITERAT—presented in Appendix C) was compared with the laboratory analysis
in an attempt to find a combination of coefficients which gave the best agree-
ment between the calculated and measured oil composition. This generally results
in a complete set of transformed data which is very nearly internally consistent
and which represents an exact solution to the first eight equations.
Sample calculations for Test 1, which illustrate the output of the SENSAN and
ITERAT programs, are also presented in Appendix C.
* These three values could not be determined simply from the test results,
while Mf, Mch» Ma, and M^, could be measured directly.
+ The CHNO composition of the oil and its heating value have been chosen as
"unknowns" because it is believed there is greater uncertainty in the measured
oil composition and heating value than for the feed, char, or gas (which could
have just as easily been used) due to the presence of water.
24
-------
TEST RESULTS
Overview of Test Conditions
This experimental program involved a series of 11 tests, nine with peanut
hulls and two with sawdust. In addition, there were several unreported tests
at the beginning of the program to check out the procedures with peanut hulls
and the basic agitator used in this study. Two tests were found to have
defective off-gas compositions, apparently due to an air leak somewhere in
the system. Therefore, the primary basis for the results presented in this
section is the nine remaining tests.
Of the nine tests, eight were conducted using peanut hulls and one using
sawdust. There was one extended run of 12 hours using peanut hulls (Test 7),
but normally the runs lasted two to three hours. In the nine basic tests,
the influences of mechanical agitation, changing feed material, changing bed
depth, and changing air-to-feed ratio were studied.
Table 1 presents a summary of the test conditions, as well as some of the
observed data from the pilot plant tests. Basic agitation was involved in
eight of the 11 tests conducted, and three were completed without any form
of agitation.
Testing was conducted at two bed depths, 127-132 cm (50-52 inches) and 89 cm
(35 inches). The air-to-feed ratio was varied from 0.172 to 0.613, the
normal range of operation. Off-gas temperatures were generally in the range
of 77 to 88°C, except in the two tests with sawdust, which ran somewhat
hotter. Although not reported, the condenser thermostat temperature was
usually set in the range of 93 to 99°C.
The dry feed rates varied from slightly under 400 kg/hr (882 Ibm/hr) to
nearly 700 kg/hr (1,543 Ibm/hr). One puzzling result was the wide variation
in the recovered oil and aqueous phases from the condenser. Reference to
Appendix B reveals that sometimes the water content was quite significant,
and other times it was small. Apparently, minor variations in the off-gas
and condenser temperatures can produce significant changes in oil yields.
25
-------
TABLE 1. TEST SUMMARY-PARAMETRIC STUDY
to
Test *
1
2
3
4
5
6
7
9
10
11
12
Feed
Material
Peanut
Peanut
Hulls
Hulls
Pine Sawdust
Pine Sawdust
Peanut
Peanut
Peanut
Peanut
Peanut
Peanut
Peanut
Hulls
Hulls
Hulls
Hulls
Hulls
Hulls
Hulls
Feed
Rate
ke/hr
572
390
676
464
494
481
476
408
501
570
471
Char
(kg/kg)
0.217
0.239
0.266
0.249
0.288
0.321
0.229
0.400
0.249
0.270
0.284
Yields
Oil &
Aqueous
(kg/kg)
0.039
0.085
0.057
0.070
0.079
0.072
0.047
0.161
0.0453
0.234
0.178
Off-Gas
(kg/kg)
1.100
0.941
0.849
0.932
0.860
0.884
0.994
0.897
1.170
1.040
1.510
TOTAL FEED
1 Air-to-Feed Off-Gas 5
Ratio Temperature
(kg/kg) (°C)
0.364
0.265
0.172
0.251
0.227
0.277
0.270
0.458
0.464
0.539
0.613
PROCESSED - 40 Mg
111
93
113
140
86
85
88
78
88
87
83
Bed f
Temperature
(°C)
649
732
760
732
649
716
704
960
560
682
787
Bed
Depth
(cm)
132
132
132
132
132
132
132
89
89
89
89
Agitation
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
TOTAL OPERATING TIME - 107.5 hr
* Test runs were of two to three hours duration, except number 7, which was a 12-hour run.
+ Yields in mass of product per mass of dry feed.
} The "off-gas yield" (including moisture of combustion, uncondensed oil, oil in suspension and noncondensible gas) is determined by difference.
§ The "off-gas" temperature is that measured as the gas exits from the pyrolytic converter.
# The indicated temperatures correspond to the average maximum measured by the thermocouples in the lower bed of the converter. Since the
temperature of the bed varies three-dimensionally in space and also varies in time (due to variations in the environment near the sensing
element;, the quantitative significance of the specific indicated temperatures is doubtful. However, they are presented for completeness
and to indicate the range of temperatures encountered.
-------
Recovered yields (on a dry basis) are generally much smaller than computed
yields, as discussed in the following section.
In the course of the testing almost 40,000 kg (88,000 Ibm) of feed were
consumed, and the unit was operated a total of 107.5 hours.
Analysis of the Data
In addition to the data shown in Table 1, the laboratory analysis of the
feed, char, oil, and noncondensible off-gas are presented in Appendix B.
The data from these tables were transformed in the manner described in the
previous section to produce a generally consistent set of results, which
is believed to be, on the average, more accurate than the original raw data.
These transformed data are presented in Table 2 and constitute the basis for
all further discussion of the testing. Shown also in the table, in
parentheses, are the amounts the values were altered from the original.
Inspection reveals that only a few of the data were modified and the changes
were generally small.
Although many of the modifications appear to have been made randomly, there
is a rough pattern to some of the changes. For example, there appear to be
relatively frequent reductions in the order of eight percent of the off-gas
nitrogen composition and in the char carbon content required to make the
data more consistent. Likewise, there appear to be several cases in which
the carbon content of the feed and the heating value of the feed must be
increased about six percent to make the results internally consistent. An
explanation for the need for nitrogen reduction is the possibility that
some air may have leaked into the system. At present, no plausible
explanations can be offered regarding the three remaining changes.
An area of concern, at first glance, is the considerable variations present
in the computed oil heating values and also in the measured values tabulated
in Appendix B. Comparison shows frequent, substantial variations between
individual values of these two sets of numbers. These differences require
some explanation. Concerning the calculated values, since the computed oil
CHNO analysis is often somewhat different than the measured analysis, which
27
-------
TABLE 2. SUMMARY OF TRANSFORMED DATA-PARAMETRIC STUDY
NJ
00
Data
GAS.
N2
C
H2
o2a
HV
CHAR
N2
C
H2
*
°2,
HV
FEED
N,
£.
C
H2
Of
HV
Units
kg/kg
kg/kg
kg/kg
kg /kg
MJ/kg
kg/kg
kg/kg
kg/kg
kg/kg
MJ/kg
kg/kg
kg/kg
kg/kg
kg/kg
MJ/kg
Test 1
0.485
0.191
0.021
0.303
A. 184
0.025
0.721
(4%)
0.026
0.089
25.57
0.017
0.457
(6%)
0.061
0.437
19.46
Test 2
0.530
(-8%)
0.199
0.021
0.289
6.226
0.021
0.829
0.018
0.032
29.75
0.021
0.462
(2%)
0.058
0.452
18.41
Test 3
0.382
(-8%)
0.258
(-2%)
0.027
0.364
8.226
0.011
0.844
0.017
0.064
30.68
0.001
0.450
(2%)
0.054
0.488
17.97
(6%)
Test 6
0.442
0.194
0.028
0.336
6.393
0.029
0.724
0.017
0.165
28.36
(10%)
0.012
0.445
(6%)
0.057
0.457
18.99
Test 7
0.434
0.201
0.028
0.338
6.393
0.027
0.795
(-8%)
0.016
(5%)
0.121
29.29
(2%)
0.012
0.473
0.057
0.458
19.36
(-2%)
Test 9
0.517
(-8%)
0.199
0.017
0.306
5.577
0.027
0.677
(8%)
0.018
0.121
28.13
0.012
0.444
(8%)
0.059
0.446
19.99
(2%)
Test 10
.
0.574
0.163
0.019
0.244
5.510
0.008
0.808
(-8%)
0.015
0.103
27.66
0.012
0.464
(4%)
0.059
0.446
19.18
(6%)
Test 11
0.478
0.189
0.017
0,314
5.368
0.008
0.809
(-4%)
0.013
0.031
28.59
0.012
0.444
(8%)
0.059
0.446
18.36
(10%)
S . __-_J
Test 12
0.510
(-8%)
0.199
0.016
0.314
5.883
(-8%)
0.011
0.773
0.009
0.089
26.73
0.012
0.483
0.059
0.446
19.99
(-2%)
. J \
* Oxygen computed; 02 = 1 - N2 - C - H2.
+ Not ash free; on dry basis.
-------
TABLE 2 (continued). SUMMARY OF TRANSFORMED DATA-PARAMETRIC STUDY
ro
\o
Data Units
OIL
(Measured)
N2 kg/kg
C kg/kg
H2 kg/kg
02* kg/kg
OIL
(Computed)
N2 kg/kg
C kg/kg
H2 kg/kg
02 kg/kg
HV MJ/kg
AIR kg/kg
MOISTURE 4
(in) kg/kg.
CHAR kg/kg,
OIL kg/kg
OFF-GAS kg/kg^
MOISTURE ^
(out) kg/kg
ENERGY ={=
LOSSES MJ/kg
Test 1
0.040
0.657
0.071
0.242
0.034
0.650
0.043
0.269
28.13
0.364
0.046
0.217
0.291
0.577
0.325
1.26
Test 2
0.047
0.831
0.059
0.064
0.039
0.813
0.004
0.144
28.13
0.265
0.045
0.239
0.228
0.395
0.449
1.26
Test 3
0.016
0.758
0.067
0.145
0.024
0.670
0.001
0.306
22.69
0.172
0.053
0.266
0.207
0.333
0.420
1.26
Test 6
0.029
0.732
0.080
0.158
0.046
0.723
0.021
0.210
27.90
0.277
0.048
0.321
0.140
0.478
0.361
1.26
Test 7
0.078
0.687
0.081
0.155
0.078
0.723
0.024
0.175
27.20
0.270
0.048
0.229
0.279
0.442
0.367
1.26
Test 9
0.014
0.737
0.080
0.168
0.056
0.582
0.093
0.270
32.77
0.458
0.287
0.400
0.065
0.682
0.598
1.26
Test 10
0.015
0.725
0.084
0.176
0.028
0.743
0.013
0.215
26.96
0.464
0.287
0.249
0.214
0.634
0.654
1.26
Test 11
0.015
0.722
0.080
0.182
0.008
0.691
0.090
0.212
36.03
0.539
0.287
0.270
0.085
0.886
0.585
1.26
Test 12
0.017
0.712
0.075
0.197
0.043
0.679
0.097
0.181
36.73
0.613
0.287
0.284
0.113
0.940
0.564
1.26
Oxygen computed; 0., = 1 - N2 - C - H2
f Mass of material per mass of dry feed.
-------
in turn varies considerably, it is not surprising that the calculated
heating value, via the Dulong-Petit equation, varies also. Perhaps,
therefore, a more meaningful value would be an average, which is 31.01 MJ/kg
(13,335 Btu/lbm). Regarding the laboratory-reported heating values for the
indicated moisture contents, an average of the dry heating values is again
probably a more accurate value. It should be noted that the uncertainty
in the moisture percentage can be significant; thus, the corrected heating
value is also uncertain. However, upon adjusting the indicated values to
a dry basis and after computing an average value, the result obtained is
33.08 MJ/kg, which is 6.7 percent greater than the average of the computed
results. It is believed that the justification for working with this
average value is adequate, and that these two values are in sufficient
agreement to satisfy the accuracy requirements of the study. Several
informative graphs can be derived from the results presented in Table 2.
These are presented in the next six figures, which correlate closely with
corresponding figures in Reference 1.
Graphical Data Presentations
Perhaps the most important results of the entire program are those given in
Figure 8, which presents the percent available energy of the char and oil
(relative to that of the feed) as a function of the air-to-feed ratio. The
figure shows that for all the tests at various bed depths, with and without
agitation and with both sawdust and peanut hulls, the data correlate to a
single line. This line is identical to that reported in Reference 1 for
sawdust converted in a unit with half the capacity of the present unit.
In fact, when the data from the present program and those from the earlier
study are combined, the agreement is striking. This is illustrated in
Figure 9 for which the best-fit straight line is again identical to both
that in Figure 8 and that from Reference 1.
This suggests, therefore, that to an acceptable degree of engineering pre-
cision, the available energy fraction of the feed in the char-oil mixture
is independent of unit size, feed material, bed depth, and the presence of
mechanical agitation, and is a linear function of only the air-to-feed ratio.
Figure 10 presents an energy breakdown of the pyrolysis products as a
30
-------
o
o
Ul
3
90-
80-i
70-
60-
Z
Ul
o
z
!" 50-
u
Ul
X
o
40-
30-
20-
10 -
SYMBOL NOMENCLATURE
BED
DEPTH
(cm)
89
132
WITH
AGITATION
O
A
WITHOUT
AGITATION
I I 1 I I
0.1 0-2 0.3 0.4 0.5
AIR-TO-FEED RATIO, kg/kg
I
0.6
I
0.7
Figure 8. Percent available energy in char-oil mixture-
parametric study.
31
-------
o
o
90-
80-
70-
IU
O
60-
O
Z
UJ
50-
u>
CO
o
X
o
UJ
UJ
IL
40-
30-
20-
10-
SYMBOL NOMENCLATURE
PREVIOUS STUDY [l ]
ALL DATA +
PRESENT STUDY
BED
DEPTH
(cm)
89
132
WITH
AGITATION
O
£
WITHOUT
AGITATION
0.1
I
0.2
—r
0.3
°-4
0.5
0.6
0.7
AIR-TO-FEED RATIO, kg/kg
Figure 9. Percent available energy in char-oil mixture-
parametric study/Reference 1 composite.
32
-------
ui
3
_J
0
z
1-
<
UJ
2
0
X
0
HI
UJ
u.
L
^~
z
UJ
u
oc
UJ
a
•inn
90-
8O-
70-
6O-
5O-
4O-
3O-
2O-
1O-
O .
3 2JT 6 1^ g 10 11 12
• «-<-> • (^)(J W ^>
% AVAILABLE HEAT FROM CHAR
A
A 0 O
U O
A u
£s
.^
% AVAILABLE HEAT Q —O'"^"""
FROM OIL ^
q^ ^^
• ^J^"-'""^ ^
^-~-r~'
^JL——— ^ | % AVAILABLE HEAT IN OFF-GAS
A 1 A/V cp n O
1* -"-0 v^« w
* *
% LOSSES 4 LATENT HEAT IN OIL 4 MOISTURE
1 1 1 1 1 1 1
O O.1 O.2 O.3 O.4 O.5 O.6 O.7
AIR-TO-FEED RATIO, kg/kg
Figure 10. Energy breakdown of pyrolysis products—
parametric study.
33
-------
function of the air-to-feed ratio. Examination of the figure reveals the
relative consistency of the data and, as in Figure 8, suggests that the
dominant influencing variable is the air-to-feed ratio. Comparison of
similar results from Reference 1 shows generally good agreement with the
total of the sensible energy in the oil and water in the off-gas, and heat
lost by conduction and to the cooling water. Likewise, the energy in the
off-gas is almost identical to that reported in Reference 1. Finally, the
combined energy in the char-oil blend agrees very well with that reported in
Reference 1.
However, there is a significant difference in the way in which the separate
energies in the oil and char vary from those presented in Reference 1. An
explanation for this difference may shed considerable light on the physical
processes at work, and provide a means of varying the relative amounts of
oil and char produced at a given, fixed air-to-feed ratio.
In Reference 1, the char yields decreased linearly and the oil yields
increased linearly with increasing air-to-feed ratio, whereas in the present
study the char yields remain practically constant and independent of
air-to-feed ratio, and the oil yields decrease with increasing air-to-feed
ratio. However, in Reference 1 the pyrolysis off-gas temperatures were
always in the range of 150-175°C, whereas in the present study the off-gas
temperatures using peanut hulls were in the range of 75-95°C. This
difference in the off-gas temperature is very significant because, in the
latter case, the higher boiling point oils are condensing in the bed.
Laboratory experience has shown that when pyrolytic oils are heated, a
significant amount of carbonization occurs along with evaporation. Hence,
in the current study, once the oils condensed and were reheated in the
downward-moving feed, only a portion of the original oil evaporated and the
remainder was converted into solid carbon. The result was an almost constant
char yield and a diminishing oil yield with increasing air-to-feed ratio.
The reason the off-gas temperatures in the present study were so low compared
with those of Reference 1 is that the bed depth was generally near the
34
-------
maximum. The results from Reference 1, at a smaller scale, had suggested
that for maximum oil yields a greater bed depth was desirable; therefore,
in the present study the greater bed depths had been deliberately chosen
to obtain the greatest amounts of oil. It appears, however, that the bed
depths selected were considerably greater than the optimum for oil
production.
Physical reasoning suggests that for a given feed, for fixed values of
process air and feed rate, and for a very shallow bed depth, the off-gas
temperature approaches the temperature in the combustion zone and there
is little or no pyrolytic conversion of the feed. Under these conditions
breakdown of the oily products produces more gaseous constituents. For
increasing bed depth, pyrolytic conversion of the feed begins to occur and
the oil yields grow as the off-gas temperature decreases. However, as the
bed depth increases beyond some optimum point, significant amounts of
condensation occur in the bed and the oil yields are diminished. Clearly,
at some critical bed depth, moisture condensation occurs and above this
point the process becomes unstable. All this behavior is illustrated
graphically in Figure 11, which also shows the surmised operating zones for
the present study and for Reference 1. Taken together, this all suggests
that, while the sum of the energy in the char and oil is basically dependent
on the air-to-feed ratio, the distribution of the energy between the oil and
the char is a function of both the bed depth and the air-to-feed ratio.
Thus, a means to independently vary the relative amounts of oil and char in
the pyrolysis products for a fixed air-to-feed ratio exists. Conveniently,
over a range of bed depths the off-gas yields appear to be relatively
independent of the bed depth and only a function of air-to-feed ratio.*
In more specific terms, to maximize char yields the pyrolysis unit should
be operated at the greatest allowable bed depth. Conversely, to optimize
oil yields, the corresponding optimum bed depth should be determined and
the unit operated near this point. It should be recognized that when the
* This indicates that in this image the carbonization of the oil results in
a minor amount of oil gasification and, therefore, that the oils are broken
down into the more volatile fractions. Since the condenser temperature in
the testing was limited by moisture condensation considerations, this would
explain why the recovered oil yields were generally so small.
35
-------
\NV\
•- \Ss \ \ \ •.
• , \ -. ••. \ v-
\ \ \\
BED DEPTH
A OPERATING ZONE IN REFERENCE 1
B OPERATING ZONE IN PRESENT STUDY
C FORBIDDEN OPERATING ZONE DUE TO MOISTURE CONDENSATION
D OFF-GAS DEWPOINT
E MAXIMUM PRACTICAL BED DEPTH
Figure 11. Oil yield variation with increasing bed depth.
36
-------
char yields are maximized, a very large portion of the oil produced is
likely to be unrecoverable because its boiling point lies below the dew
point of the off-gas mixture. Thus, while the available energy in the
char-oil mixture is approximately constant (at a given air-to-feed ratio),
it may be more desirable in many situations to avoid a deep bed in order to
actually recover a maximum percentage of the oil in a useable form. There-
fore, it appears that for maximum recovery of both the char and the oil,
operation near the point of maximum oil production is indicated.
It should be noted that the presence of water in the feed acts effectively
to increase the bed depth, since greater amounts of energy are required to
pyrolyze the feed and, thus, the off-gas temperature tends to be reduced.
Therefore, if a maximum of both char and recoverable oil is desired, it
would be best to operate with as dry a feed as possible.
Figure 12 is a crossplot of computed data from Reference 1 and experimental
data from the present study. The figure provides a convenient means for
determining the required air-to-feed ratio for a given feed moisture
percentage; further, it allows computation of the available energy in the
char-oil mixture. The computation assumptions regarding the energy require-
ments to operate the portable unit are taken from Reference 1. To illustrate
the use of the figure, at a feed moisture percentage of 20 percent, the
required energy for drying and processing is 1.71 MJ/kg (735 Btu/lbm) dry
feed. At an air-to-feed ratio of 0.19, the available energy in the gas
is 1.71 MJ/kg (735 Btu/lbm) dry feed, and that available in the char-oil
is 15.3 MJ/kg (6,600 Btu/lbm). This establishes the relation between the
moisture content and the air-to-feed ratio.
Figure 13 presents a plot of the heating value of the noncondensible com-
ponent of the off-gas in MJ/m as a function of the air-to-feed ratio. As
before and as in Reference 1, there is a correlation with this parameter,
although the data scatter is greater than desired. The curve drawn through
the data lies within 5 to 10 percent of the corresponding curve from
Reference 1 and, thus, again establishes the close correlation of the data
from the two studies.
37
-------
. 12
O)
S 10
(0
° 8
O
E
UJ
_l
m
Ul
X
u
K
5 8
oc
UJ
10
1 12
IL
14
<
J
I 16
t-
i is
5000
'I'll I I I
0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1
AIR-TO-FEED RATIO, kg/kg
_L
I'll I L
10 20 30 40 50 60 70 80 90
FEED MOISTURE CONTENT, %
ASSUMPTIONS
1. GROSS HEAT
ENERGY REQUIRED
TO PROCESS DRY
FEED, 0.84 MJ/kg*
(360 Btu/lbm*)
2. GROSS HEAT
ENERGY REQUIRED
TO DRY THE FEED,
3.49 MJ/kgb
(1500 Btu/ibm")
1000
2000
3000
4000
5000
E
a
00
60OO
7000
• DRY FEED
b MOISTURE
12
s
10
en
<
O
O
oc
u.
6
a
UJ
oc
4 I
UJ
cc
UJ
i
Figure 12. Effects of feed moisture on available
energy from char-oil mixture (ref. 1)
38
-------
350 -
300-
-14
-12
-10
250-
3
ffl
- 200-
ui
O
- 150-
UJ
x
100-
50-
SYMBOL NOMENCLATURE
BED
DEPTH W|TH
(cm) AGITATION
WITHOUT
AGITATION
89
132
O
A
- 6
- 4
-2
I
0.1
O-2
O.3
0.4
O.5
0.6
AIR-TO-FEED RATIO, Kg/kg
Figure 13. Heating value of noncondensible gas—
parametric study.
39
-------
SECTION 5
PERFORMANCE STUDY OF INTEGRATED MECHANICAL
AGITATION—AIR SUPPLY SYSTEM
GENERAL
The present concept of the EES waste converter system operation involves the
addition of process air near the bottom of the vertical, gravity-fed porous
bed. This air allows combustion of a small fraction of the feed material
and, thus, provides the heat required for pyrolysis. The air is added by
means of several fixed, water-cooled air tubes. The presence of these air
tubes represents a hindrance to the flow of the feed material and is, thus,
partially responsible for the need for a mechanical agitation system to
enhance feed throughput. There is also the fact that, since the system
throughput is limited to a large extent by gravity, residence times are far
greater than required to pyrolyze the feed.
Thus, there appears to be considerable advantage in the use of an integrated
mechanical agitation-process air system, especially if the mechanical
agitation system is a requirement in any case to process bulky wastes. By
adding such a system, the principal hindrance to flow through the converter
is changed into a means for facilitating the flow. Such a system may also
allow the processing of somewhat wetter feed than the present EES waste
converter permits.
This section presents a description of a "first-generation" integrated
mechanical agitation-process air supply system, or "AIRGITATOR," and a
discussion of the tests conducted with it.
40
-------
SYSTEM DESIGN
There are conceptually a large number of possible configurations that the
system might have taken. However, it was decided at the outset that the
simplest configuration possible was to be selected. This was done in order
to minimize fabrication problems and to avoid, as much as practical, the
possibility of failure and the opportunity for leaks by minimizing the number
of welds. Thus an "L" shaped system was chosen.
The AIRGITATOR is presented schematically in Figure 14, and the final design
is shown in Figure 15. Its outer tube is made of 4130 high-strength alloy
steel 5.08 cm (2 inches) in outside diameter with 0.635-cm (0.25-inch) walls.
Designed to withstand the high-torsional loads and hostile environment
inside the reactor, the outer tube provides a passage for cooling water.
To introduce process air and circulate cooling water while the device is
being rotated, a three-passage union was required. A Deublin three-passage
union was used during the initial tests. However, because of the high
pressure drop across this union, sufficient process air could not be
introduced through the AIRGITATOR. To eliminate this problem: (1) a
three-passage union was designed, fabricated, and installed; and (2) the
diameter of the air discharge holes was increased.
In the initial design the horizontal portion of the unit extended to within
2.5 cm of the inside wall of the converter and the end was cut off squarely.
A later modification involved the removal of 2.5 cm from this horizontal
portion and the beveling of the end so that the end surface formed a sharp
edge which cut through the char. These modifications were made to avoid
the binding of feed between the walls and the end of the unit in situations
where, due to irregularities in the inner surface, the end approached the
wall too closely.
The final design of the AIRGITATOR, including the three-passage union, is
shown in Figure 15. A photograph of the unit, fabricated in the EES shop,
is presented in Figure 16. Figure 17 is a close-up view of the AIRGITATOR
41
-------
NONCONDENSIBLE
GASES
FEED
OIL
CHAR
Figure 14. Schematic of EES converter with integrated
mechanical agitation—air supply system.
42
-------
Figure 15. Final design of Airgitator.
43
-------
Figure 16. Overview of
Airgitator.
Figure 17. Airgitator
as installed.
44
-------
as installed. As can be seen, the installed system is not complex—it
involves only a drive system, the three-passage union, and the "L" shaped
AIRGITATOR.
FACILITIES
The facilities for this study were the same as those used in the parametric
study. Their description begins on page 10.
CALIBRATION AND TEST PROCEDURE
The calibration and test procedure was the same as that used in the para-
metric study. Its description begins on page 18.
LABORATORY TESTING
The laboratory testing was also the same as that used in the parametric
study. The description begins on page 19-
DATA REDUCTION
The data reduction procedure was the same as that used in the parametric
study. Its description begins on page 21.
TEST RESULTS
Overview of Test Conditions
This experimental program, with a total of 10 tests, involved two phases.
A series of four tests was conducted first to determine the feasibility of
combining agitation with process air injection. A series of six tests was
then designed to study the performance of a particular form of integrated
mechanical agitator—process air supply system (AIRGITATOR).
In the first phase peanut hulls were used as the feed material. Two of the
four tests in this phase were considered to be checkouts and no data were
taken. As proposed, peanut hulls were also to be used as the feed material;
however, peanut hulls were no longer available when the project go-ahead
45
-------
was received. Since earlier work [1] using sawdust as the feed material
could be used as a basis for comparison, pine sawdust was substituted for
the proposed peanut hulls.
The main objective of the phase-two tests was to operate the AIRGITATOR over
a range of air-to-feed ratios and determine the maximum processing rate.
Air-to-feed ratios of 0.14 to 0.34 were used and the maximum process rate
was over 800 kg/hr (1,760 Ibm/hr).
Table 3 presents a summary of the test conditions and selected observed data
from these tests. Data are given for two checkouts and two peanut-hull runs
(phase one) and the six sawdust runs. Two bed depths were used—127 cm with
the peanut hulls and 89 cm with the sawdust. The off-gas temperatures ranged
from 121 to 260°C and the processing rate varied from 324 kg/hr (714 Ibm/hr)
for peanut hulls to a maximum of 833 kg/hr (1,836 Ibm/hr) for the pine
sawdust.
System Testing
The checkout tests of the AIRGITATOR resulted in almost disastrous results.
The main bearings supporting the unit failed after several hours of testing,
apparently as a result of very large torques that were occasionally required
to rotate the system. It was concluded that the tip binding, described
earlier, was the cause. The tip was modified and the complete drive system
was strengthened substantially. This modified unit was then tested and no
problems were encountered. Apparently these improvements were sufficient to
overcome the difficulty.
One important feature in Hater tests was the use of two wall-mounted air
tubes in the start-up of the unit and also occasionally to stabilize the hot
char bed during normal operation. The extra depth to the hot char bed
provided by these two tubes not only made it possible to establish a stable
hot char zone initially, but also provided a cushion against "losing the
char bed" in anomalous circumstances in which the instantaneous feed rate
exceeded the charring rate and threatened the loss of the hot char which
sustains the bed operation.
46
-------
TABLE 3. TEST SUMMARY-AIRGITATOR STUDY
Test
Number
8
13
14
15
16
17
18
19
20
21
Feed
Material
Peanut Hulls
Peanut Hulls
Peanut Hulls
Peanut Hulls
Sawdust
Sawdust
Sawdust
Sawdust
Sawdust
Sawdust
'Feed
Rate
(kg/hr)
Char
(kg/kg)
Yields*
Oil &
Aqueous
(kg/kg)
Off-Gas+
(kg/kg)
Air-to-Feed
Ratio
(kg/kg)
Off-Gas ,
Temperature*
CO
Bed
Temperature^
<°C)
Bed
Depth
(cm)
CHECK OUT "AIRGITATOR"
CHECK OUT MODIFIED "AIRGITATOR"
490
324
416
569
570
833
597
463
0.414
0.283
0.460
0.389
0.420
0.247
0.405
0.240
0.035
0.262
0.164
0.111
0.096
0.114
0.178
0.087
0.691
0.645
0.624
0.699
0.752
0.818
0.714
1.100
0.140
0.190
0.245
0.199
0.268
0.179
0.297
0.337
174
226
121
149
177
160
149
260
471
471
538
371
510
482
510
482
127
127
89
89
89
89
89
89
* Yields in mass of product per mass of dry feed.
+ The'"off-gas yield" (including moisture of combustion, uncondensed oil, oil in suspension and noncondensible gas) is determined by
difference.
^ The "off-gas" temperature is that measured as the gas exits from the pyrolytic converter.
5 The indicated temperatures correspond to the average maximum measured by the thermocouples in the lower bed of the converter. Since
the temperature of the bed varies three-dimensionally in space and also varies in time (due to variations in the environment near the
sensing element), the quantitative significance of the specific indicated temperatures is doubtful. However, they are presented for
completeness and to indicate the range of temperatures encountered.
-------
An additional problem apparently occurred at shallow bed depths. The
rotating vertical shaft of the AIRGITATOR appears to have provided a lower
resistance path to a portion of the off-gases than the porous bed of
pyrolyzing feed material, especially to those combustion products generated
by introduction of the process air near the axis of rotation. Such a short
circuit of the off-gas flow field in the bed could cause abnormally high
off-gas temperatures and particulate levels. If this in fact occurred, then
two possible solutions are (1) to construct baffles on the rotating shaft
to increase the gas flow resistance or (2) to redesign the AIRGITATOR to
permit its introduction through the bottom of the converter.
Analysis of Data
In addition to the data presented in Table 3, the laboratory analysis of the
feed, char, oil, and noncondensible off-gas is presented in Appendix B,
Tables B-12 through B-17. These data were transformed in the manner
previously described (pages 21-23) to produce a generally consistent set
of results. The transformed data are presented in Table 4 and they form
the basis for all further discussion of the data.
The numbers shown in parentheses in Table 4 are the percentage changes which
were made in the original data. As in the parametric study, these changes
were generally less than 10 percent. In general, the corrections have been
to the analysis of the feed material. Since a very small sample is used in
the CHNO analysis to represent as much as several thousand kilograms of feed
material, these corrections appear to be within the degree of precision with
which the feed material can be representatively sampled.
Graphs similar to those for the parametric study were derived from the data
of Table 4.
Graphical Data Presentations
Figure 18 shows the energy content of the char-oil mixture, in terms of the
energy content of the input feed, as a function of the air-to-feed ratio.
Data are presented for both the AIRGITATOR study and the parametric study.
These data are well correlated by a single straight line, even though three
48
-------
TABLE 4. SUMMARY OF TRANSFORMED DATA-AIRGITATOR STUDY
Data
GAS
N2
C
H2
°2+
HV
CHAR
N2
C
H2
°2
HV+
FEED
N2
C
H2
°2
HV
Units
kg/kg
kg/kg
kg/kg
kg/kg
MJ/kg
kg/kg
kg/kg
kg/kg
kg/kg
MJ/kg
kg/kg
kg/kg
kg/kg
kg/kg
MJ/kg
Test 14
0.396
0.216
0.018
0.369
5.531
0.011
0.393
(-50%)
0.018
0.115
29.05
0.007
0.304
(40%)
0.061
0.427
19.78
Test 15
0.351
0.218
0.011
0.422
3.582
0.007
0.818
(-4%)
0.014
0.091
28.82
0.007
0.466
(8%)
0.061
0.427
19.78
Test 16
0.432
0.206
0.014
0.349
4.795
0.000
0.750
0.036
0.180
27.95
0.001
0.482
(1.9%)
0.060
0.438
(-6%)
19.92
Test 17
0.421
0.205
0.012
0.363
4.247
0.000
0.727
0.037
0.209
27.43
0.004
0.483
0.054
(-10%)
0.444
(2%)
19.47
Test 18
0.454
0.197
0.012
0.336
4.544
0.001
0.739
0.036
0.180
27.84
0.004
0.468
(2%)
0.058
0.450
(-6%)
19.15
Test 19 Test 20*
0.331
0.234
0.013
0.420
4.736
0.005
0.812
0.029
0.126
31.30
0.000
0.442
(-8%)
0.061
(3%)
0.489
(6%)
19.13
Test 21
0.386
0.221
0.012
0.381
4.736
0.003
0.827
0.028
0.115
31.68
0.001
0.477
0.054
(8%)
0.463
19.64
* Error In collected data prevented a meaningful analysis.
+ Oxygen computed; 02 = 1 - N2 - C - H2 .
f Not ash free; on dry basis .
(continued)
-------
TABLE 4 (continued). SUMMARY OF TRANSFORMED DATA-AIRGITATOR STUDY
en
O
Data
OIL
(Measured)
N2
C
\
°2
(Computed)
N2
C
H2
°2
HV
AIR
MOISTURE
(in)
CHAR
OIL
OFF-GAS
MOISTURE
(out)
ENERGY
LOSSES
Units
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
MJ/kg
kg/kg+
kg/kg+
kg/kg+
kg/kg+
kg/kg+
kg/kg+
MJ/kg+
Test 14
0.012
0.703
0.077
0.208
0.087
0.660
0.102
0.152
36.73
0.140
0.065
0.414
0.124
0.275
0.392
1.26
Test 15
0.012
0.694
0.077
0.217
0.111
0.676
0.106
0.107
37.89
0.190
0.065
0.283
0.209
0.424
0.339
1.26
Test 16
0.009
0.608
0.073
0.322
0.045
0.564
0.100
0.291
33.13
0.245
0.078
0.460
0.115
0.437
0.311
1.26
Test 17
0.011
0.719
0.029
0.241
0.038
0.684
0.037
0.242
28.24
0.199
0.049
0.389
0.182
0.369
0.308
1.26
Test 18
0.007
0.614
0.058
0.321
0.013
0.609
0.066
0.316
29.37
0.268
0.050
0.420
0.110
0.461
0.327
1.26
Test 19 Test 20
0.009 0.010
0.612 0.611
0.062 0.061
0.317 0.318
0.012
0.635
0.071
0.282
31.59
0.179
0.036
0.247
0.230
0.406
0.332
1.26
Test 21
0.011
0.595
0.060
0.334
0.003
0.588
0.075
0.334
30.47
0.337
0.082
0.240
0.223
0.667
0.289
1.26
* Oxygen computed; 02 • 1 - N2 - C - H2.
+ Mass of material per mass of dry feed.
-------
o
o
o
oc
Ul
z
Ul
90-
8O-
7O_
C3
O
SL
6O-
5O-
a
Q
Ul
Ul
u.
SYMBOL NOMENCLATURE
3O-
BED
DEPTH
(cm) AGITATION
AIRGITATION NEITHER
20-
89
127
132
0
©
1O -
T
O.1
I I I
O.2 O.3 O.4
0.5 0.6
O.7
AIR-TO-FEED RATIO, kg/kg
Figure 18. Percent available energy in char-oil mixture—
Airgitator/parametric study composite.
51
-------
bed depths and two feed materials were used. The apparent generality of this
correlation is even more convincing when the data in Figure 18 are combined
with those in Reference 1, as shown in Figure 19. Thus the available char-
oil mixture energy seems to be practically independent of the unit site
(cross-sectional area), feed material, bed depth, mechanical agitation, and
"airgitation;" it seemingly depends only on the air-to-feed ratio.
Figure 20 displays the data in the manner of Figure 10 of the parametric
study. The open-dotted symbols for tests 14 through 21 use the nomenclature
of Figures 18 and 19. The filled symbols represent data of Reference 1,
which were also obtained for pyrolysis of pine sawdust, but in a smaller
converter which had only fixed air tubes. Comparing Figures 10 and 20, it
appears that the losses (e.g., sensible heat of the off-gas steam and
cooling-water stream, latent heat of the water in the off-gas stream and
the condensed oil, heat losses to the atmosphere) and the available heat
(of combustion) from the off-gas stream are practically the same. However,
even though the total available heat (of combustion) of the char-oil mixture
is the same (see Figure 18), the manner in which the separate energies in
the oil and char vary with air-to-feed ratio is considerably different.
The AIRGITATOR test data apparently agree with those from Reference 1, but
the peanut hull data (Figure 10) have quite a different nature—the available
heat from the char is practically independent of the air-to-feed ratio. This
feature, as discussed in Section 4 is apparently due to condensation of the
higher temperature oils in the bed since the off-gas temperatures were
quite low (78 to 96°C). In the AIRGITATOR tests the off-gas temperatures
ranged from 121 to 260°C, even though peanut hulls were used as the feed
material in two of the runs.
Figure 21 presents the heating value of the noncondensible gas as a function
of the air-to-feed ratio. For comparison, the AIRGITATOR data are shown
together with the data from the parametric study. There are apparently
two separate correlations—one for the parametric study data and one for
the AIRGITATOR data. In spite of this difference, when the product of the
heating value and the mass of gas per kilogram of dry feed is formed, a single
correlation results, as shown in Figure 20.
52
-------
90-
80-
o 70-
o
ui
> 60
O
50-
O
40-
30-
20-
10-
SYMBOL NOMENCLATURE
PREVIOUS STUDY fl ]
ALL DATA ^
PRESENT STUDY
BED
DEPTH
(cm) AGITATION
O
89
127
132
AIRGITATION
0
NEITHER
A
I
0-1
I
I
I
O-4
0.2 0.3
AIR-TO-FEED RATIO, kg/Kg
l
O.5
f
0.6
O-7
Figure 19. Percent available energy in char-oil mixture—
Airgitator/parametric study/Reference 1 composite.
53
-------
Ul
a
ui
x
100-
90-
80-
70-
70-
£ son
o
0 40-
Ul
Ul
H 30^
ui
U
? 20-
10-
B ©03 m 0 © 0
15
©
©
% AVAILABLE HEAT FROM OIL AND LOST CARBON
% AVAILABLE HEAT FROM CHAR
__ SENSIBLE HEATX
* COOLING WATER AND CONVECTION LOSSES^
% AVAILABLE HEAT FROM OFF-GAS
Ol4 Q.'S
AIR-TO-FEEO RATIO, kg/kg
0-6
0.7
08
Figure 20, Energy breakdown of pyrolysis products—Airgitator/
Reference 1 composite.
-------
350 -
300-
- 14
- 12
250-
3
ffi
- 200-
3
O
5 150-
UJ
z
100-
50-
A
3
- 10
SYMBOL NOMENCLATURE
BED
DEPTH
89
127
132
AGITATION AIRGITATION
O ©
NEITHER
I
0.1
I
0.3
I
0.4
I
0.2 0-3 0.4 0.5
AIR-TO-FEED RATIO, kg/kg
0.6
— ft
- 6
-4
- 2
Figure 21. Heating value of noncondensible gas—
Airgitator study.
55
-------
SECTION 6
PERFORMANCE STUDY OF SPARK-IGNITION ENGINE
FUELED WITH SIMULATED PYROLYSIS GAS
SUMMARY
The energy-conversion efficiency and self-sufficiency of a portable pyrolysis
system will be enhanced if operating power can be derived from that portion
of the pyrolytic off-gases which is burned in the atmosphere or otherwise
not utilized. It was accordingly decided, as a first step in the development
of a power system, to test a spark-ignition engine fueled with a simulated
pyrolysis gas for feasibility and performance.
3 3
A 3,867-cm (236-in ), 6-cylinder, General Motors gasoline truck engine of
7.5:1 compression ratio and other familiar characteristics was used. The
simulated pyrolysis gas was constituted on a volume-fraction basis as:
• Carbon Monoxide 0.24
• Hydrogen 0.12
• Methane 0.07
• Nitrogen 0.57
3
The higher heating value of this gas was approximately 6.71 MJ/m
(180 Btu/ft3) in contrast to the 37.3 MJ/m3 (1,000 Btu/ft3) typical of
common gaseous fuels (e.g., natural gas).
The engine ran smoothly on the low-energy fuel despite prior pessimism
expressed from most sources. The wide-open-throttle output was from 60 to
65 percent of that obtained for gasoline within the operating speed range
of 1,500 to 3,000 rpm. Fortuitiously, the level of power reduction is
consistent with that normally recommended for continuous operation of small
industrial engines. The risk of operating above-rate power is thus
56
-------
eliminated along with any need to increase power by supercharging or other
means.
Manual control of the fuel delivery for optimum performance was necessary
since no automatic metering valve is commercially available for such
unusual fuel. The need is thus foreseen to develop an appropriate delivery-
pressure control system to compensate for variations in gas composition and
load characteristics.
The overall successful experience with simulated pyrolysis gas increases the
likelihood that actual process gas can be used to fuel a spark-ignition
engine.
BASIC TEST EQUIPMENT
The General Motors truck engine utilized in these tests is located in the
School of Mechanical Engineering and is rated at 8.58 kW (115 hp) at 3,700
4pm. (See Figure 22.) The distributor spark-advance was manually controlled,
and inlet air flow was monitored by a Meriam 50MC2-4PF laminar-flow element
and a manometer. An Impco CA-300A mixer for dual-fuel operation was employed
for its convertibility to gasoline (Figure 23). Modification was necessary
to the extent of eliminating the internal metering by removal of the Impco
3 3
gas valve assembly which had been calibrated for 37.3-MJ/m (1,000-Btu/ft )
3 3
fuel rather than for the 6.71-MJ/m (180-Btu/ft ) gas of interest. Load
control was accomplished by means of a Taylor Model D-31, 179-kW (240-hp)
water dynamometer (Figure 22).
GASEOUS FUEL SYSTEM
Fuel was manifolded from four size-lA gas bottles initially at 12.4 MPa
(1,800 psig); each set of bottles provided fuel for 10-15 minutes of running-
time at wide-open throttle. (See Figure 24.) A high-volume Matheson Gas
Products Model 3052350 regulating valve delivered the fuel at approximately
138 kPa (20 psig), through a 1.91-cm (0.75-in) O.D. copper tube, to a
succession of elements (Figure 25) consisting of:
57
-------
Figure 22. Overview of GM truck engine.
Figure 23. Impco CA-300A mixer.
58
-------
Figure 24. Overview of gaseous fuel system.
Figure 25. Gaseous fuel metering system.
59
-------
• A Gould Type Q2 general-purpose, solenoid-
operated shut-off valve;
• A Rockwell Model 243-8 HP pressure regulator
with variable delivery pressure ranges from
300 Pa (12 inches of water) to 45 kPa (6.5 psig);
• A 3.8-cm (1.5-inch) diameter butterfly valve;
• A 100-kPa (15-psig) pressure gage; and
• A Meriam 50MC2-2S laminar-flow element.
Gas delivery rate to the Impco mixer was principally controlled by the manual
setting of the Rockwell pressure regulator. All components were selected
in accordance with the need to deliver the low-energy gas at approximately
four times the normal volume flow rate required for any given power level.
PROCEDURE
The engine was started on gasoline for convenience and for conservation of
the simulated pyrolysis gas. Upon reaching a stable operating condition at
a fast idle speed of 1,200 to 1,500 rpin, the gasoline delivery was interrupted
by the solenoid cut-off valve in the gasoline fuel line. Operation continued
for up to one minute because of the gasoline present in the carburetor fuel
bowl. At the first sign of hesitation the solenoid and butterfly valves in
the simulated pyrolysis gas fuel line were opened. It was also necessary
at this instant to open the throttle further to let the engine breath more
of this less energetic air-fuel mixture. Reestablishment of smooth engine
output usually followed with minimal stumble or stalling.
The throttle was then advanced to wide-open operation and the dynamometer
load was adjusted to produce the desired operating speed. Fuel delivery
pressure was then manually adjusted for maximum power at the desired engine
speed, necessitating minor adjustments of the dynamometer load. Also, the
spark advance was adjusted for maximum power at the desired engine speed,
necessitating further minor adjustments of the dynamometer load and the
fuel delivery pressure.
60
-------
RESULTS
Optimum performance under specific conditions was as given in Table 5. Using
data for the brake power output of the engine at specific engine speeds, as
given by the smoothed presentation of these data in Figure 26, the de-rated
TABLE 5. OPTIMUM ENGINE PERFORMANCE
Fuel
Gasoline
Gasoline
Gasoline
Gasoline
SPG+
SPG
SPG
SPG
Speed
(rpm)
1,490
2,000
2,500
3,000
1,525
2,025
2,525
3,000
Brake
Power*
(kW)
31.4
42.5
50.8
59.7
21.2
25.2
31.5
37.1
Torque
(N-m)
201
203
194
190
133
119
119
118
Spark
Advance
(degrees)
18
23
27
40
30
35
40
4QT
* Brake power is the net power output of the engine
+ Simulated pyrolysis gas
^ Optimum spark advance was not determined
brake power output of the engine operating on simulated pyrolysis gas, as
compared with gasoline, is given in Table 6. Defining the brake de-rating
factor as the ratio of the brake power output of the engine when fueled with
simulated pyrolysis gas to that when it is fueled with gasoline, the brake
de-rating factor varies between 0.60 and 0.65 (Table 6, Figure 26).
Stable operation of the engine beyond 3,000 rpm was not possible, in part
because the maximum delivery rate of the chosen pressure regulator was being
reached, and apparently because the maximum flow that could be stably
accommodated by the modified Impco CA-300A gas inlet-was also being exceeded.
The expensiveness of the SPG and the limited interest in higher speeds
mitigated against further modifications of the inlet gas-flow capacity.
61
-------
70
60
50
cc
ui
o
o.
O
UJ
DC
O
40
30
20
10
0
1000
GASOLINE
SIMULATED
PYROLYSIS GAS
J_
_L
J_
1500
2000 2500
ENGINE SPEED, rpm
3000
0.7
0.6
0.5
0.4
0.3
0.2
0-1
3500
CC
O
O
z
oc
I
DC
m
Figure 26. Wide-open-throttle performance of General
Motors truck engine for gasoline and
simulated pyrolysis gas.
62
-------
TABLE 6. BRAKE DE-RATING FACTOR—SPG /GASOLINE
Speed
(rpm)
1,500
2,000
2,500
3,000
Gasoline
(kW)
32.1
42.0
51.3
59.7
SPG
(kW)
20.9
25.2
30.7
37.1
Brake De-rating
Factor
(SPG/Gasoline)
0.65
0.60
0.60
0.62
* Simulated pyrolysis gas.
Brake power output correction factors for atmospheric conditions were not
applied to these data because the operating conditions were consistently in
a range for which such corrections would have been minimal in comparison to
the data tolerance.
DISCUSSION OF RESULTS
The objectives of this exploratory study were (1) to determine whether it is
possible to run a spark-ignition engine on simulated pyrolysis gas and
(2) to determine the brake de-rating factor in case stable operation is
possible. Not only did the engine run smoothly on this "low-energy" fuel,
despite prior pessimism expressed from most sources, but also the brake
de-rating factor was found to be ideal for placing an intermittent-duty
engine into continuous duty—the brake de-rating factor was about 0.6 over
the normal engine-speed range.
The "low-energy" nature of this type of fuel warrants further comment.
Low-Btu pyrolysis gas typically has a higher heating value of about 180 Btu
per cubic foot of gas, whereas natural gas has one of about 1,000 Btu per
cubic foot of gas and gasoline (C_H1Q) has one of about 6,000 Btu per cubic
O J-O
foot of gas. On the surface, it might appear that such pyrolysis gas is
energy-deficient when compared with natural gas and gasoline. That this is
not the case is evident from study of the data of Table 3. Since a spark-
ignition engine operating at wide-open throttle breathes a given volume of
63
-------
a near Stoichiometric air-fuel mixture at approximately atmospheric pressure
and temperature, the only fair way to compare these fuels is on the basis of
the heating value of a given volume of Stoichiometric mixture of the fuel
and air. The last two columns in Table 7 show that gasoline provides the
most energetic mixture and pyrolysis gas the least. However, when pyrolysis
gas is compared with gasoline and then with natural gas, it is seen that
its indicated de-rating factor is about 0.72 for gasoline and about 0.80 for
natural gas, when the indicated de-rating factor is defined as the ratio
of the lower heating value of the pyrolysis gas per cubic meter of mixture
to that of the reference fuel. From this it can be concluded that pyrolysis
gas from a partial oxidation process is a practical fuel substitute for
either natural gas or gasoline for a spark-ignition engine.
TABLE 7. HEATING VALUES OF STOICHIOMETRIC MIXTURES
Fuel
Gasoline
Natural Gas
Pyrolysis Gas
Phase
Gas
Gas
Gas
HHV*
(MJ/kg)+
48.258
55.501
6.855
i
LHV*
(MJ/kg)
44.789
50.014
6.373
Air-Fuel
Ratio
(kg/kg)9
15.15
17.27
1.84
HHV „
(MJ/mV
3.705
3.445
2.661
LHV_
(MJ/m )
3.439
3.101
2.474
* Higher heating value.
+ Energy per kg of fuel.
^ Lower heating value.
§ Mass of air per kg of fuel.
# Energy per m-* of mixture of fuel and Stoichiometric air at 1 atm and 25°C.
The indicated de-rating factor of 0.72 is the upper limit of what might
have been expected for the GM truck engine tested. When it is noted that
the friction power requirement of the engine is essentially independent of
the fuel being used, it is clear that the brake power output of the less
energetic air-fuel mixture will be proportionately less than that of the
more energetic air-fuel mixture, the lower limit being the case where the
air-fuel mixture is only energetic enough to supply the friction power and
the effective brake de-rating factor is zero.
64
-------
An approximate analysis of the brake de-rating factor is presented in
Appendix D. Using the relationship derived for the brake de-rating factor
in terms of the indicated de-rating factor and the friction power of the
engine tested (Equation (D-4), page 125), the brake de-rating factor for
this engine fueled with natural gas instead of gasoline was computed; the
results, together with the experimentally determined brake de-rating factors
for simulated pyrolysis gas, are presented in Table 8. The engine brake
power when running on SPG was about 70 percent of that computed for running
on natural gas (Table D-3, page 128).
TABLE 8. BRAKE DE-RATING FACTORS
Speed
(rpm)
1,500
2,000
2,500
3,000
Brake De-rating
Factor*
(Natural Gas/Gasoline)
0.88
0.86
0.86
0.86
Brake De-rating
Factor"1"
(SPGT/Gasoline)
0.65
0.60
0.60
0.62
* Computed (See Table D-2, page 127).
+ Experimental (See Table 6).
f Simulated pyrolysis gas.
From these observations it may be concluded that "low-Btu" pyrolysis gas
should serve as an acceptable substitute for either gasoline or natural gas
as a fuel for a spark-ignition engine. The brake de-rating factor of about
60 percent is ideal for using an SI engine designed for intermittent duty as
a continuous-duty power source.
65
-------
SECTION 7
PERC COMBUSTION AND EMISSION STUDY
OF PYROLYSIS CHAR AND OIL
SUMMARY
The combustion and handling characteristics of char from pyrolyzed wood
wastes were determined in a 227 kg/hr (500 Ibm/hr) pulverized-coal-fired
(PCF) combustion test facility, and as a slurry with No. 6 fuel oil in a
981 kW (100 HP) oil-fired boiler. In the PCF combustor, tests were also
run with a 50-50 blend of Pittsburgh-seam, high-volatile coal with a high-
volatile pyrolytic char, as well as with a low-volatile pyrolytic char.
Stable combustion could be maintained with a secondary air-preheat tempera-
ture of 316°C (600°F), the temperature generally used when firing coal, at
a carbon combustion efficiency of 97.3 to 98.6 percent. With the low
nitrogen content of the char, nitrogen oxide emissions were very much lower
than those obtained from coal (0.25 compared to 0.80 Ibm NO /10 Btu) at the
same firing conditions. The NO emissions obtained with the 50-50 blends
X
appeared to be an average of the values obtained for the fuels separately.
Similarly, S02 emissions were low with the char alone (0.18 Ibm S02/10 Btu),
and with the blends were an average of the values obtained with the fuels
separately.
A 60-40 blend of pulverized char and char-oil, combined with No. 6 fuel oil
to produce a slurry containing 30 percent char, performed well in a 981 kW
(100 HP) oil-fired firetube boiler modified to fire coal-oil slurries.
Excellent flame stability was experienced, and the carbon-combustion
efficiency was similar to that obtained with No. 6 fuel-oil and coal-oil
slurry. Nitrogen oxide emissions were significantly lower than those
obtained when firing coal-oil slurry, and S0_ emissions were about 50 percent
lower. Some fouling of the small ports in the burner nozzle occurred as a
66
-------
result of the accumulation of small fibers passing through the filter
screen.
INTRODUCTION
Combustion and emissions tests with chars from pyrolyzed wood wastes were
conducted in both a 227 kg/hr (500 Ibm/hr) pulverized fuel combustion test
facility and a 981 kW (100 HP) oil-fired boiler modified to fire coal-oil
slurries at the Pittsburgh Energy Research Center of the U. S. Energy
Research and Development Administration, in cooperation with the Industrial
Environmental Research Laboratory of EPA.
A considerable amount of work concerned with the combustion of low-volatile
char produced as a by-product in coal conversion processes had been performed
earlier at the Pittsburgh Energy Research Center [9, 10]. Tests revealed
that a low-volatile (5 percent) char such, as that produced in the COED
process could be successfully burned in a dry-bottom, water-walled PCF
combustor by preheating the pulverized-char/primary air mixture to 232°C
(450°F), or by using an auxiliary fuel (such as natural gas or coal) to
provide flame stability.
PULVERIZED SOLID FUEL COMBUSTION/EMISSION STUDY
Description of Facility
The solid-fuel combustion tests were conducted in an experimental,
pulverized-coal-fired furnace designed to simulate the performance of an
industrial steam-generating unit used in commercial power generating plants.
The wall-fired, dry-bottom furnace was capable of burning 227 kg (500 Ibm)
of coal per hour, with an exit gas temperature of 1,093°C (2,000°F). Heat
3 3
release rate was about 670 MJ/hr-m (18,000 Btu/hr-ft ). A photograph of
the furnace is shown in Figure 27. The furnace had water-cooled walls with
refractory applied in the burner zone to provide flame stability by preventing
excessive heat transfer to the walls in the vicinity of the burner, and to
obtain an exit gas temperature of 1,093°C (2,000°F).
67
-------
Figure 27. View of 227 kg/hr (500 Ibm/hr) pulverized coal-fired furnace.
68
-------
Coal was burned in a direct-fired system through four burners in the front
wall of the 2.13-m (7-ft) wide, 3.66-m (12-ft) high and 1.52-m (5-ft) deep
rectangular furnace. A simplified flow sheet of the combustion system is
shown in Figure 28. Provision was made to preheat secondary air and to vary
distribution of combustion air between the primary and secondary air streams.
Variations in coal feed rate can result in pockets deficient in either fuel
or oxygen, producing fluctuations in the fuel-air ratio. Consequently, a
recycle loop was provided in the primary air-coal transport line to obtain a
more uniform coal feed rate to the burners. Figure 29 shows the principal
components of the combustion train in half section. Combustion products
exited the furnace at 1,093°C (2,000°F); flowed through the convective
heat-transfer section where the gas temperature was lowered to 538°C
(1,000°F); through the air heater for preheating the secondary air; then
through a baghouse filter. The cross-section of a multifuel burner is shown
in Figure 30. The flame profile could be continuously varied from a short,
bushy pattern to a relatively long, narrow pattern by adjustment of swirl
induced in the secondary-air stream.
Operation of Facility
The four front-wall burners were designed to fire natural gas and/or
pulverized solid fuel. Prior to each test period, the experimental furnace
was fired with natural gas to preheat the refractory and to provide a source
of preheat for the secondary air. During the preheat period, combustion air
flows were established, and secondary-air swirl adjustments were made to
provide flames that were attached to the burners, but not drawn into the
burner tubes or impinging on the side walls. Preheating was continued until
the secondary-air temperature was about 288°C (550°F). Natural gas flow was
then reduced by 50 percent, and pulverized char, char-coal blend, or coal
feed was started at a rate of 113 kg/hr (250 Ibm/hr). With the oxygen
content of the flue gas used as a monitoring guide, the natural gas to each
burner was decreased as the solid fuel rate was increased to maintain a
constant oxygen level in the flue gas.
Natural gas fed to each burner was then gradually decreased to the minimum
amount necessary to produce stable flames, as determined by visual
69
-------
Stack
Char
hopper
Char- air
recycle
loop
Natural
gas
Char preheaters
Secondary air
L-13867
Figure 28. Schematic of 227 kg/hr OOO Ibra/hr) pulverized-coal-fired furnace.
-------
Recuperative
air preheater-
Convective heat transfer section
Figure 29. Half-section view of principal components of the combustion system.
-------
N5
Pilot
qas
Auxiliary
gas
Secondary
air
Primary
air-cool
mzs&rr
Pilot gas
spark igniter
Secondary air
swirl vanes
Primary air
swirl inducer
Figure 30. Multifuel burner assembly.
-------
observation. In the present study, auxiliary gas was not required to
maintain stable flames with the char-coal blend or the high-volatile char
alone. After the stabilizing gas flow was eliminated, final adjustments
were made on the flow of char or char-coal blends, and on secondary air to
provide the desired excess-air level for the test period.
A base combustion test was first conducted with the Pittsburgh-seam coal
used later for blending with both the low- and high-volatile char. The
firing rate with the coal was 227 kg/hr (500 Ibm/hr); the firing rate with
the char and the char-coal blends was adjusted to give a heat rate equivalent
to 227 kg of coal per hour (500 Ibm/hr). Combustion tests were conducted
with a high-volatile char alone and 50-50 blends by weight of coal with both
the high-volatile and low-volatile chars. The coal-char blends were pre-
pared in a "V" type blender with 25 mm (1/4-inch) X 0 crushed coal and the
char as received. The char and coal-char blend were transferred into and
from the blender with a vacuum transfer system to avoid a dust nuisance.
Each 91 kg (200 Ibm) coal-char charge was blended for about an hour; the
relatively small fluctuations in the oxygen level during the combustion
tests indicated the coal and char were well-blended. Proximate and ultimate
analyses of the coal, chars, and the coal-char blends are given in Table 9.
Combustion tests were conducted by firing the chars blended with an equal
weight of Pittsburgh seam coal before the high-volatile char was fired alone.
Stable flames could be maintained in all tests with a secondary-air preheat
of 316°C (600°F), the temperature generally employed when firing coal.
Experimental results of the combustion tests are given in Table 10.
Discussion of Results
As shown in Table 9, both the nitrogen and sulfur content of the chars were
very low (0.1 percent). As expected, the NO and SO emissions were very
f X X
low (0.25 Ibm N02 and 0.18 Ibm SO /10 Btu) in the test in which the high-
volatile char was fired along. The NO emissions obtained with the 50-50
X
(by weight) coal-char blends were an average of the values obtained for the
fuels fired separately. Similarly, the SO- emissions obtained with the
blends were a weighted average of the SO emissions obtained from the fuels
73
-------
TABLE 9. TYPICAL ANALYSES OF PITTSBURGH COAL, CHARS AND CHAR/COAL BLENDS
Data
Proximate (As Received)
Moisture
Volatile Matter
Fixed Carbon
Ash
Ultimate (As Received)
Hydrogen
Carbon
Nitrogen
Sulfur
Oxygen
Ash
Higher Heating Value
Initial Deformation Temp.
Softening Temperature
Fluid Temperature
Units
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
MJ/kg
°C
°C
°C
Pittsburgh
Coal
0.020
0.377
0.533
0.070
0.053
0.749
0.015
0.019
0.095
0.070
31.26
1,166
1,193
1,332
High-
Volatile
Char
0.015
0.280
0.558
0.147
0.031
0.675
0.001
0.001
0.145
0.147
25.45
1,432
1,460
1,493
Low-
Volatile
Char
0.052
0.025
0.853
0.070
0.011
0.856
0.001
0.001
0.061
0.070
29.53
1,143
1,171
1,227
High-Volatile-
Char/Coal Blend
0.019
0.353
0.536
0.092
0.044
0.723
0.008
0.010
0.123
0.092
28.81
1,143
1,171
1,349
Low- Volatile-
Char/Coal Blend
0.017
0.226
0.687
0.070
0.032
0.816
0.010
0.009
0.063
0.070
31.07
—
—
—
-------
TABLE 10. EXPERIMENTAL RESULTS OF COMBUSTION TESTS WITH PULVERIZED COAL, CHAR AND CHAR/COAL BLENDS
Data Units
Test Number
Fineness % thru 200 Mesh
NO Emission kg NO,/ 10 kJ
X ^
S02 Emission kg S02/106 kJ
CO Emission kg CO/106 kg
Carbon Combustion
Efficiency %
Furnace Outlet
Temperature °C
Excess Air %
Secondary-Air Preheat
Temperature °C
Pittsburgh
Coal*
1
76.7
0.34
1.04
10
99.1
1,082
20
316
High-
Volatile
Char
5
67.5
0.11
0.08
10
97.3
1,038
20
316
High-
Volatile-
Char/Coal
Blend
4
75.6
0.24
0.66
10
98.6
1,054
20
316
Low-
Volatile-
Char/Coal
Blend
2
70.5
0.22
0.53
50
96.1
1,066
20
343
Low-
Volatile-
Char/Coal
Blend
3
71.1
0.23
0.57
70
98.6
1,093
20
316
*Firing rate of 227 kg/hr; char and char/coal blends fired at heat rate equivalent to 227 kg of coal per hour.
-------
separately. When fueled with the low-volatile char-coal blend in which the
•
heating value of the char was slightly lower than that of the coal, S00 emis-
sions (1.23 lbm/10 3tu) from the furnace would meet emissions regulations in
most air regions in the U. S. Since the heating value of the high-volatile
char was considerably lower than that of the coal and low-volatile char be-
cause of its higher ash and oxygen content, a greater proportion of high-
volatile char would be required in the blend to meet S02 emission regulations.
The carbon-combustion efficiency of 98.6 percent obtained with the 50-50
high-volatile-char/coal blend, and the second test with the blend of low-
volatile char and coal compared favorably with the value of 99.1 percent
obtained in the coal reference test. The lower values of 96.1 and 97.3
percent, obtained in the initial test with the blend of low-volatile char
and coal and for the high-volatile char alone, is attributed to the higher
percentage of oversized particles. Analyses of the various fuels given in
Table 11 show that the oversize (on 100 mesh) was 16.3 and 19.3 percent for
the tests in which the combustion efficiencies were somewhat low, whereas
the oversize was 9.5 and 10.4 percent in the blend tests in which the
combustion efficiencies were satisfactory. The oversize in the pulverized
coal was only 3.9 percent. The large fraction of oversize experienced is
partially attributed to the deterioration of the hammers in the impact
pulverizing mill due to the erosive nature of the char. Photographs of the
impact rotor disc and the hammers (Figures 31 and 32) show the erosion that
was experienced during 15 to 20 hours of operation. Most of the erosion was
probably due to the design of the pulverizer, in which an inventory of char
covered the impact rotor disc rotating at 3600 rpm. Erosion would probably
be nil in the large, slowly-turning roller or ball mills used in large
industrial boilers. The large amount of oversize could also be partially
attributed to carry-over of light, flake-like chips of the char, which had
3 3
an average bulk density of about 320 kg/m (20 Ibm/ft ). Despite the
3 3
relatively low bulk density, 320 kg/m (20 Ibm/ft ) compared with about 721
3 3
kg/m (45 Ibm/ft ) for coal, the char flowed freely from the bunker and in
transfer lines.
76
-------
TABLE 11. SIEVE ANALYSES OF PULVERIZED COAL, CHAR AND CHAR/COAL BLENDS
Data
Test Number
On Sieve 100
On Sieve 150
On Sieve 200
On Sieve 325
Thru Sieve 325
Thru Sieve 200
Units
—
Weight %
Weight %
Weight %
Weight %
Weight %
Weight %
Pittsburgh
Coal
1
3.9
6.7
12.5
74.7
2.0
76.7
High-
Volatile
Char
5
19.3
5.1
8.1
3.3
64.2
67.5
High-Volatile-
Char/Coal Blend
4
9.5
5.8
8.9
33.3
42.3
75.6
Low-Volatile-
Char/Coal Blend
2
16.3
5.0
8.2
16.5
54.0
70.5
Low-Volatile-
Char/Coal Blend
3
10.4
5.4
13.1
17.2
53.9
71.1
-------
1
•
Figure 31. Worn impact rotor disc and hammers after char pulverization.
-------
•
<
Figure 32. View of hammers before and after pulverization of char.
-------
CHAR-OIL SLURRY COMBUSTION/EMISSION STUDY
Description of Facility
A 60-40 blend of low-volatile char and pyrolysis oil combined with No. 6
fuel oil to produce a slurry containing 30 percent char was fired in a
981 kW (100 HP) oil-fired, package firetube boiler modified to include a
slurry preparation and feed system. A photograph of the boiler used for
the slurry combustion studies is shown in Figure 33. A different view of the
boiler and some of the auxiliary equipment is shown in Figure 34. Two No. 6
fuel-oil storage tanks appear in the background, and some of the pumps and
the char-oil slurry mixing and feed tanks are in the foreground. A simplified
flow diagram is shown in Figure 35. On the left are two steam-heated
3
storage tanks holding 26.5 m (7,000 gallons) of No. 6 fuel oil. On the
right is a mixing tank, equipped with a stirrer and recirculation pump, in
which finely pulverized char or coal is added and slurried with oil.
Following the mixing tank is the feed tank, a feed pump, and then the oil-
fired boiler. A more detailed description of the slurry combustion facility
was repored earlier [1],
Discussion of Results
The analysis of the char-oil slurry appears in Table 12, as well as analyses
of No. 6 fuel oil and a slurry of pulverized coal and No. 6 fuel oil.
Inspection reveals that nitrogen and sulfur concentrations in the char-oil
slurry are about one-half the concentration in the coal-oil slurry. The
results of a combustion test with the char-oil slurry appear in Table 13,
with results from a No. 6 fuel oil and a coal-oil combustion test shown for
comparison.
Excellent flame stability was experienced in all tests. The loss from
carbon in the stack experienced with the char-oil slurry was about the same
as that obtained with No. 6 fuel oil, and both were substantially lower than
that obtained with the coal-oil slurry. Nitrogen oxide emissions were also
lower than those obtained when firing coal-oil slurry; the SO emissions were
proportionate to the concentration of sulfur in the slurry, and about 50
percent lower than those with the coal-oil mixture. Some fouling of the small
80
-------
08
Figure 33. View of 981 kW (100 HP) Firetube boiler used
for coal-oil slurry combustion studies.
-------
:
Figure 34. View of boiler, No. 6 fuel oil storage tanks, and slurry mixing
and slurry mixing and feed tanks.
-------
00
SLURRY
TRANSFER PUMP
FUEL OIL
TRANSFER PUMP
CONDITIONER
WATER WATER
TANK
FEED WATER PUMP
Figure 35. Simplified flow diagram of the 981 kW (100 HP)
coal-oil slurry and char-oil slurry combustion
test facility.
BE-76-14 Al
-------
TABLE 12. TYPICAL ANALYSES OF OIL, COAL-OIL SLURRY AND CHAR-OIL SLURRY
Data
Ultimate Analysis
Hydrogen
Carbon
Nitrogen
Oxygen
Sulfur
Ash
Higher Heating
Value
Viscosity @ 60°C
Units
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
kg/kg
MJ/kg
N.s/m
Numb e r 6
Fuel Oil
0.122
0,855
0.002
0.001
0.008
0.000
44.12
0.082
20% Coal-
Oil Slurry
0.108
0.835
0.005
0.020
0.010
0.018
41.55
0.105
30% Char-
Oil Slurry
0.093
0.846
0.003
0.041
0.005
0.013
39.94
0.225
84
-------
Table 13. OPERATING CONDITIONS FOR SELECTED PERIODS WITH NO. 6 FUEL OIL,
COAL-OIL SLURRY AND CHAR-OIL SLURRY
Data
Solids Concentra-
tion
Fineness
Steam Pressure
Fuel Pressure @
Burner
Atomizing-Air
Pressure
Fuel Temperature
At Hold Tank
At Burner
Flue Gas Analysis
Oxygen
Carbon Dioxide
Carbon Monoxide
NO
Suffur Dioxide
Steam Flow Rate
Fuel Flow Rate
Excess Air
Dust Loadings
Carbon in Fly Ash
Loss From Carbon
in Stack
Fuel Heating Value
Units
Weight
%
% thru
200 Mesh
kN/m2
0
kN/m
2
kN/m
°C
°C
Vol. %
Vol. %
mg/kg
mg/kg
mg/kg
kg/hr
kg/hr
Vol. %
kg/hr
Weight
%
Weight
%
MJ/kg
Number 6
Fuel Oil
0
-
848
117
124 '
49
94
4.8
11.4
50
223
551
1,529
100.4
27.1
0.54
46.4
0.2
44.12
20% Coal-
Oil Slurry
20
95
855
131
131
47
94
4.3
11.5
81
273
523
1,592
104.1
23.8
1.72
49.9
0.6
41.63
30% Char-
Oil Slurry
30
92.4
827
166
172
56
98
4.4
11.8
-
235
265
1,588
114.3
23.0
1.49
29.3
0.3
39.94
85
-------
ports in the burner nozzle occurred as a result of accumulation of small
fibers that passed through the filter screen. The problem was alleviated by
a slight modification of the nozzle. Plugging would not be a problem in
larger industrial boilers in which the burner ports are proportionately
larger.
The low sulfur and nitrogen content of the char makes it an ideal fuel to
mix with either coal or oil to conserve our dwindling oil supply. When
slurried with oil and burned in an oil-fired boiler, the only pollution
abatement device required would be a baghouse for particulate control. The
low-sulfur content of the char would also permit the firing of oils higher
in sulfur content without violating SO emission standards. If a new
industrial coal-fired boiler was installed to completely eliminate the use
of oil, the requirement for a sulfur emission control system could be
eliminated by blending the coal with the low-sulfur char to meet SO emission
regulations. This is an especially attractive feature for the smaller
industrial boilers for which a flue gas desulfurization system would be
disproportionately cumbersome and expensive. The suppression of NO emissions
X
due to the low nitrogen content of the char makes it doubly attractive.
CONCLUSIONS
1. Stable combustion and satisfactory combustion efficiency could be
maintained in the 227 kg/hr (500 Ibm/hr) pulverized-coal-fired, water-wall
combustor with high-volatile char alone, or with 50-50 blends of high- or
low-volatile char and coal.
2. Excellent flame stability was experienced and carbon combustion
efficiency was equal to that obtained with No. 6 fuel oil, when pulverized
char was mixed with oil and fired as a slurry in an oil-fired boiler.
3. The low sulfur and nitrogen content of the char makes it an attractive
fuel to mix with either high-sulfur coal or oil to extend oil supplies and
to meet SO- and NO emission regulation? without emission control devices
other than a baghouse or electrostatic precipitator for particulate control.
86
-------
REFERENCES
1. Tatom, J. W., A. R. Colcord, J. A. Knight, L. W. Elston, and P. H.
Har-oz, "Utilization of Agricultural Forestry and Silvicultural Waste
for the Production of Clean Fuels," Final Report under EPA Contract
68-02-1485, Engineering Experiment Station, Georgia Tech, Atlanta,
Georgia, September 1976.
2. Tatom, J. W. , A. R. Colcord, J. A. Knight, L. W. Elston, and P. H.
Har-oz, "A Mobile Pyrolytic System - Agricultural and Forestry Wastes
into Clean Fuels," Proceedings 1975 Agricultural Waste Management
Conference published as Energy Agricultural and Waste Management,
Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan.
3. Tatom, J. W., A. R. Colcord, J. A. Knight, L. W. Elston, and P. H.
Har-oz, "Parametric Study for a Pyrolytic System for Production of
Clean Fuels from Agricultural and Forestry Wastes," Proceedings
Tenth Intersociety Energy Conversion Engineering Conference, published
by Institute of Electrical and Electronics, Inc., New York, N. Y.
4. Tatom, J. W. , A. R. Colcord, J. A. Knight, L. W. Elston, and P. H.
Har-oz, "Clean Fuels from Agricultural and Forestry Wastes - The
Mobile Pyrolysis Concept, 1974 A.S.M.E. Winter Annual Meeting,
paper no. 75 - WA/HT - 47.
5. Everett, Kevin, "Playing Ball with the Big Boys; Has Mobile Pyrolysis
Come of Age?" unpublished paper, Florida Resource Recovery Council,
December 1975.
6. Long, J. R. M., "Coal-in-Oil Fuel Study," Final Report, January 2, 1964,
Combustion Engineering, Inc.
7. Private Communication, Andrew Brown, General Motors.
8. Private Communication, J. J. Demeter of the ERDA Pittsburgh Energy
Research Center.
9. McCann, C. R., J. J. Demeter, A. A. Orning, and D. Bienstock,
"Combustion of Pulverized Char," presented at Combustion Symposium,
Div. Fuel Chemistry, ACS National Meeting, Washington, D.C.,
September 12-17, 1971.
10. Demeter, J. J., C. R. McCann, and D. Bienstock, "Further Studies of
the Combustion of Pulverized Char and Low-Volatile Coal," presented at
the Winter Annual Meeting, ASME, Detroit, Michigan, November 11-15, 1973,
87
-------
11. Demeter, J. J. , C. R. McCann, G. T. Bellas, J. M. Ekmann, and
D. Bienstock, "Combustion of Coal-Oil Slurry in a 100 HP Firetube
Boiler," PERC/RI-77/8, May 1977.
88
-------
APPENDIX A
LABORATORY PROCEDURE
The following procedures were followed in the laboratory analysis of the
input feed and the pyrolysis products:
Solid Samples
Sample Preparation—The solid samples examined consisted of the dried
peanut hulls or pine sawdust used as feed material for the waste converter,
and chars produced by the converter. The sample size received in the
laboratory ranged from one to eight liters for the peanut hulls or sawdust
feeds and from one to two liters for the char products. The samples were
thoroughly mixed and divided by quartering or by a riffle splitter to pro-
duce a representative one liter sample, which was passed through a Wiley
Model 4 mill using a six-millimeter screen. The ground sample was again mixed and
divided into approximately equal parts. One part was again passed through
the Wiley Model 4 mill using a two-millimeter screen. This material was
then mixed and reduced by quartering to approximately 100 grams. The 100-
gram sample was then passed through a Wiley intermediate mill using 40-mesh
screen, remixed, and quartered. The larger portion of the 40-mesh sample
was stored in a tightly closed glass bottle for use in laboratory analysis.
The remaining quarter of the material was again passed through the Wiley
intermediate mill using an 80-mesh screen, remixed, and stored in a tightly-
capped vial for elemental analysis.
Analytical Procedures—The following analytical procedures were used:
1. Percent Moisture in Peanut Hull or Sawdust Feeds: Duplicate 1.000-gram
samples were placed in aluminum dishes and dried for one hour at 40.5°C in
a forced air oven. The dried samples were cooled in a desiccator and weighed.
The estimated error was + 0.6 percent (absolute).
89
-------
2. Percent Moisture and Percent Volatiles in Chars: These analyses were
performed by ASTM Method D-271. The estimated error was +0.3 percent
(absolute).
3. Percent Ash and Percent Acid-Insoluble Ash in Feeds and Chars:
Duplicate 1.000-gram samples of the feed or char were weighed into tared
porcelain crucibles, ignited to constant weight in a muffle furnace at 600°C,
cooled in a desiccator, and reweighed. The ash was digested in a 1:3
mixture of hydrochloric acid and nitric acid for 30 minutes. The mixture was
then diluted to approximately 100 ml and filtered through a Whatman No. 40
paper. After thorough washing with distilled water, the filter paper and
undissolved ash were returned to the crucible used for the original ash
determinations, ignited to constant weight at 600°C, cooled in a desiccator
and weighed. The estimated error was +0.2 percent (absolute).
4. Heating Value: The heating values of the feeds and chars were determined
in a Parr Plain (Isothermal Jacket) oxygen bomb calorimeter, following the
procedures described on pages 33-38 of Oxygen Bomb Calorimeter and Combustion
Methods, Technical Manual No. 130, Parr Instrument Company, Moline, Illinois
(1960). Agreement among replicate samples was better than 2.5 percent
(absolute) for the feeds and 3.5 percent (absolute) for the chars.
5. Elemental Analysis: Carbon, hydrogen and nitrogen were determined using
a Perkin Elmer Model 240 Elemental Analyzer. (Oxygen was determined by
difference.) The manufacturer claims a precision of + 1.0 percent (relative
for pure, crystalline materials). Because of the heterogeneous nature of
the samples, loss of volatiles from the chars in the purge fraction of the
analytical cycle, and the difficulty of selecting a representative three
milligram sample, occasional variations as high as 15 percent (absolute) have
been observed in the carbon and oxygen determination on char samples.
However in most cases, the agreement was better than 6.0 percent (absolute)
for carbon and oxygen in the feeds and chars. Agreement among replicate
hydrogen or nitrogen determinations was better than 1.0 percent.
90
-------
Oil Samples
Sample Preparation—The oil samples received in the laboratory were stored in
tightly-closed glass bottles and stirred before each analysis.
Percent Moisture in Oil—The percent moisture in the oil was determined by
the method of Dean and Stark. The error is believed to be + 5.0 percent
(relative), although the oil is known to begin to decompose partially with
liberation of additional water at the temperature of the toluene-water
azeotrope, and acetone and other water soluble compounds have been detected
in the head space over stored oil samples.
Non-condensible Gas Samples
Sample Preparation—Gas samples were drawn continuously from the head space
in the waste converter or from the upstream end of the condensers. The
sample stream was passed through a series of water-cooled condensers, a
glass-wool demister, and ice-cooled trap, a chemical drying tube, and a dry
test meter to a tee in the sampling line. From the tee the major portion
of the sample was exhausted to the atmosphere through a vane-type pump. A
smaller portion of the stream was led from the tee through a tubing pump and
a wet test meter into a 96-liter "Saran" gas collection bag. The flow rate
in the gas streams was held constant throughout the sampling periods. At
the end of the test, the waters and oils from the condenser train were
measured and the gas collection bag was closed and returned to the laboratory
for analysis.
Analysis of Noncondensible Gas Samples—The gases were mixed by kneading the
sample collection bag, and their concentrations were determined by gas
chromatography. Oxygen and nitrogen were determined in a Perkin Elmer Model
990 Gas Chromatograph using helium carrier gas, a Molecular Sieve 5A column,
and a thermal conductivity detector. Hydrogen was determined in a similar
manner using argon as the carrier gas. Carbon monoxide, methane and carbon
dioxide were determined in the same instrument using helium carrier gas and
an activated carbon column. Hydrocarbons containing two or more carbon atoms
were determined in a Perkin Elmer Model 154 instrument using helium carrier
gas, a Perkin Elmer "R" column, and a flame ionization detector. The
estimated error was +5.0 percent (relative).
91
-------
APPENDIX B
LABORATORY DATA
Listed in the following pages are the results of the laboratory analysis
described in Sections 4 and 5 for the feed, char, oil and off-gases. It
should be noted that the CHNO analysis and the heating values for the oils
are for the indicated moisture content. Thus, the results for dry oil in
Tables 2 and 4 have been corrected for this moisture. The CHNO analysis
and heating values for the feed and char are on a dry basis.
92
-------
TABLE B-l. LABORATORY ANALYSIS FOR TEST 1
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
§
Oxygen
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
4.4
3.4
—
48.6
6.0
1.7
40.3
19.46
*
Char
8.3
10.9
—
75.1
2.6
2.5
8.9
25.45
+
11.9
-
—
57.0
7.6
3.5
31.9
29.12
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
VsHt^
*^2 "6
C2E,
C^Ho
36
,,.
Percent
Composition
44.37
16.88
15.78
16.17
4.60
0.52
0.72
0.13
0.24
§
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
Note that this is the volume, not the weight composition.
Oxygen computed; 02 = 1 - N2 - C - H2.
93
-------
TABLE B-2. LABORATORY ANALYSIS FOR TEST 2
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
§
Oxygen
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
4.3
2.3
-
47.0
5.8
2.0
42.9
18.40
*
Char
0.3
10.0
-
82.9
1.8
2.1
3.2
29.75
„.
33.2
-
-
55.5
7.6
3.1
33.8
22.17
Off -Gas
Non-
Condensible
Components
N2
CO
C02
H2
CM.,
C2H6
C2E,
C3H8
C3H6
<• 10
Percent
Composition
47.1
14.5
19.9
11.1
5.52
0.63
0.90
0.14
0.27
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
Note that this is the volume, not the weight composition.
§
Oxygen computed; 02
1 - N, - C - Ho.
94
-------
TABLE B-3. LABORATORY ANALYSIS FOR TEST 3
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
§
Oxygen
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
5.0
1.2
-
45.8
5.4
0.0
47.6
19.12
*
Char
4.6
6.5
-
84.4
1.7
1.0
6.4
30.73
21.1
-
-
60.6
7.7
1.3
30.4
23.97
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHi,
C2H6
C2^
C3H8
C3H6
cVho
Percent'
Composition
33.8
18.2
24.0
12.5
9.5
0.6
0.9
0.1
0.3
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
§
Oxygen computed; 02 = 1 - N2 - C - H2.
95
-------
TABLE B-4. LABORATORY ANALYSIS FOR TEST 6
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
§
Oxygen
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
4.6
2.3
—
47.3
5.7
1.2
43.5
18.99
*
Char
2.7
6.5
—
72;4
1.7
2.9
16.5
31.59
4-
Oil
17.9
-
—
60.1
8.6
2.4
28.9
No Fire
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHi^
*-*2 **6
C2Hi^
38
36
Percent*
Composition
41.1
9.8
22.4
18.7
6.7
0.6
-
0.6
-
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
* Oxygen computed; 02 = 1 - N2 - C - H2.
96
-------
TABLE B-5. LABORATORY ANALYSIS FOR TEST 7
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
R
Oxygen
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
4.6
2.3
-
47.3
5.7
1.2
43.5
18.99
*
Char
0.6
9.8
-
73.6
1.8
2.7
12.1
29.82
16.1
-
-
57.6
8.6
6.5
27.3
25.01
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHt,
C2H6
C2H,
C3H8
36
410
Percent '
Composition
41.9
24.51
8.14
15.07
8.91
0.65
-
0.78
-
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
§ Oxygen computed; 02 = 1 - N2 - C - H2•
97
-------
TABLE B-6. LABORATORY ANALYSIS FOR TEST 9
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen *
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
22.3
4.6
1.4
48.3
5.9
1.2
40.0
20.39
Char
0.6
9.8
—
73.6
1.8
2.7
12.1
28.04
20.3
-
—
56.9
8.7
1.1
33.3
27.54
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
GHtf
C2ng
^*2 **
C3H8
C3H6
-------
TABLE B-7. LABORATORY ANALYSIS FOR TEST 10
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen^
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
22.3
4.6
1.4
48.3
5.9
1.2
40.0
20.39
*
Char
1.5
13.6
4.4
74.8
1.5
0.8
10.3
27.77
Oil+
26.1
-
-
53.6
9.1
1.1
36.2
26.18
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
Cttk
C2H6
C2Hi»
C3H8
C3H6
C4H10
Percent i
Coraposit ion
53.26
17.03
11.31
12.84
4.40
0.41
0.50
0.09
0.18
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
+ The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
§ Oxygen computed; 02 = 1 - N2 - C - H2.
99
-------
TABLE B-8. LABORATORY ANALYSIS FOR TEST 11
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen §
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
22.3
4.6
1.4
48.4
5.9
1.2
39.9
20.39
*
Char
3.2
17.0
-
77.8
1.3
0.8
3.1
27.60
Oil+
28.6
-
-
51.5
8.9
1.1
38.5
24.34
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHU
C2H6
C2H4
C3H8
C3H6
C4H10
Percent*
Composition
46.98
17.91
18.18
11.13
4.63
0.41
0.53
0.09
0.16
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
+ The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
* Oxygen computed; 02 = 1 - N2 - C - H2-
100
-------
TABLE B-9. LABORATORY ANALYSIS FOR TEST 12
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
22.3
4.6
1.4
48.3
5.9
1.2
40.0
19.97
Char
1.2
20.1
—
77.3
0.9
1.1
0.6
25.22
Oil+
34.0
-
-
47.0
8.7
1.1
43.2
25.59
Off -Gas
Non-
Condensible
Components
N2
CO
C02
H2
CH(4
C2H6
C2H4
C3H8
C3H6
C<4H10
Percent *
Composition
46.88
21.86
16.36
8.72
4.84
0.43
0.63
0.09
0.19
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
+ The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
§
Oxygen computed; 02
1 - No - C - Ho.
101
-------
TABLE B-10. LABORATORY ANALYSIS FOR TEST 14
Element
Water
Ash
Acid-
Insoluble
Ash
Volatiles
Carbon
Hydrogen
Nitrogen
s
Oxygen
Heat ing
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
6.1
2.8
0.5
-
50.6
6.1
0.7
39.8
19.78
*
Char
1.2
7.1
1.0
12.2
78.5
1.8
1.1
11.5
29.12
Oil+
14.7
-
-
-
60.0
8.2
1.0
30.8
26.28
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHif
C2H6
C2H^
C3H8
C3H6
CttH10
Percent'
Composition
40.3
23.2
19.3
9.84
6.03
1.0
-
0.1
0.1
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
§
Note that this is the volume, not the weight composition.
Oxygen computed; 02 = 1 - N2 - C - H2.
102
-------
TABLE B-ll. LABORATORY ANALYSIS FOR TEST 15
Element
Water
Ash
Acid-
Insoluble
Ash
Volatiles
Carbon
Hydrogen
Nitrogen
Oxygen
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
6.1
2.8
0.5
-
50.6
6.1
0.7
39.8
19.78
Char*
0.9
10.2
3.0
11.0
78.7
1.4
0.7
9.0
23.92
Oil+
18.1
-
-
56.8
6.3
1.0
35.9
24.34
Off-Gas
Non-
Condensible
Components
N2
CO
co2
H2
CHi+
C2H6
C2Htt
C3H8
C3H6
CitH10
Percent*
Composition
47.0
11.1
26.1
0.5
3.33
0.99
-
0.20
0.13
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
' Note that this is the volume, not the weight composition.
Oxygen computed; 02 = 1 - N2 - C - H2•
103
-------
TABLE B-12. LABORATORY ANALYSIS FOR TEST 16
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen ^
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
7.2
0.9
0.6
47.3
6.0
0.1
45.7
18.48
_. *
Char
3.7
3.3
1.7
75.0
3.6
0.0
18.1
27.95
Oil*"
50.4
-
-
45.4
8.1
0.3
46.2
18.81
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHi<
C2H6
C2H»t
C3H8
C3H6
CUH10
Percent T
Composition
46.4
17.9
23.9
3.99
6.26
1.04
-
0.38
—
0.08
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
§
Note that this is the volume, not the weight composition.
Oxygen computed; 02 » 1 - N2 - C - H2.
104
-------
TABLE B-13. LABORATORY ANALYSIS FOR TEST 17
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen *
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
4.7
0.8
0.5
48.3
6.0
0.4
44.5
18.55
_, *
Char
2.0
2.7
1.1
72.7
3.7
0.0
20.9
27.43
44.9
-
—
39.7
6.6
0.6
53.1
14.41
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CnLf
C2 H£
^2 U
C3H8
36
Percent*
Composition
46.5
18.5
25.7
2.19
5.91
0.85
-
0.29
-
0.01
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
+The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
TNote that this is the volume, not the weight composition.
* Oxygen computed; 02 = 1 - N2 - C - H2.
105
-------
TABLE B-14. LABORATORY ANALYSIS FOR TEST 18
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen §
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
4.8
2.9
0.9
45.9
5.8
0.4
45.0
18.23
Char*
1.7
4.4
1.5
73.9
3.6
0.1
18.0
27.84
Oil+
25.6
-
—
45.7
7.1
0.5
46.7
20.45
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHt,
C2H6
C2H.4
C3H8
C3H6
C4H10
Percent*
Coraposit ion
48.1
19.9
21,2
4.3
5.2
0.86
-
0.26
-
0.02
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
5 Oxygen computed; 02 = 1 - N2 - C - H2 .
106
-------
TABLE B-15. LABORATORY ANALYSIS FOR TEST 19
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen *
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
3.5
2.0
0.8
48.0
5.9
0.0
44.1
18.46
Char
1.7
2.8
1.6
81.2
2.9
0.5
12.6
31.30
Oil+
12.4
-
-
53.6
6.8
0.8
38.8
22.56
Off -Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHU
C2H6
C2Hn
C3H8
C3H6
ci*H10
Percent '
Composition
37.4
20.5
31.2
2.65
6.84
1.00
-
0.34
-
0.03
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
+ The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
§ Oxygen computed; 02 = 1 - N2 - C - H2.
107
-------
TABLE B-16. LABORATORY ANALYSIS FOR TEST 20
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen*
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
6.0
2.1
1.1
47.1
5.9
0.3
44.6
18.01
Char
1.2
2.9
1.6
85.6
2.0
0.7
8.8
32.01
Oil +
21.2
-
—
48.1
7.2
0.8
43.9
21.75
Off -Gas
Non-
Condensible
Components
N2
CO
C02
H2
CH^
C2H6
C2Hi,
C3H8
C3H6
C<*H10
Percent '
Composition
36.2
21.3
21.5
10.9
8.4
1.4
-
0.32
-
0.02
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
* Note that this is the volume, not the weight composition.
§ Oxygen computed; 02 = 1 - N2 - C - H2-
108
-------
TABLE B-17. LABORATORY ANALYSIS FOR TEST 21
Element
Water
Ash
Acid-
Insoluble
Ash
Carbon
Hydrogen
Nitrogen
Oxygen §
Heating
Value
Units
Percent
Percent
Percent
Percent
Percent
Percent
Percent
MJ/kg
Feed
7.6
1.1
0.6
47.7
5.9
0.1
45.2
18.15
Char*
1.5
2.7
1.2
82.7
2.8
0.3
11.5
31.68
Oil+
5.1
-
-
56.5
6.3
1.0
36.2
25.13
Off-Gas
Non-
Condensible
Components
N2
CO
C02
H2
CHU
C2H6
C2H^
C3H8
C3H6
CUH10
Percent*
Composition
43.1
20.7
26.9
2.67
4.46
0.93
-
1.25
-
0.03
The volatile component of the char probably contains very little water
and is primarily gaseous hydrocarbons.
+ The CHNO analysis and heating values are based on oil with the indica-
ted moisture content.
T Note that this is the volume, not the weight composition.
§ Oxygen computed; Q£ = 1 - N2 - C - H2 .
109
-------
APPENDIX C
LISTING OF DATA REDUCTION COMPUTER PROGRAM
Presented in this section are listings and sample calculations illustrating
*
the use of the data analysis computer program.
To demonstrate the sample computer output, in Run Number 4 (Test 1) the
nominal laboratory CHNO and heating values for the input feed and products
(see Table B-l) are listed below:
Gas
Char
Feed
Air
N.
2
0.485
0.025
0.017
0.770
C
0.191
0.751
0.486
0
H.
2
0.021
0.026
0.061
0
°0
2
0.303
0.089
0.437
0.230
HV
6.29
25.45
19.46
0
From the testing, the char yield was 21.7 kg per 100 kg dry feed; the
measured amount of air per 100 kg of dry feed was 36.4 kg, and the amount of
the moisture in the feed was 4.6 kg per 100 kg dry feed. The energy losses
(1) were estimated at 57.0 MJ (54,000 Btu) for each 45.36 kg (100 Ibm) of
feed (or about 7 percent of the energy of the feed).
In the computation procedure, which involved an iterative approach, initial
values !
mately.
values for w and HV were chosen and equations 1-8 were solved approxi-
Then variations of + 10 percent of each of the coefficients in the eight
equations were made, and the resulting values of each of the eight unknowns
Note: All calculations within these two programs were made using the
English system of units and conversion to metric units was made during
report preparation.
110
-------
were determined. Using these results, the measured versus the computed
values of the oil composition could be compared. The results of this
procedure are presented as part of the SENSAN OUTPUT.
Comparison of the computed versus the measured oil composition shows the
following results:
Element
C
H
0
N
Measured
0.657
0.071
0.242
0.04
Computed
0.837
0.0344
0.185
- 0.056
Percent
Difference
+ 27.4
- 51.5
- 23.6
Not only was the difference between the values for C, H and 0 substantial,
but the computed value for N was physically impossible. Clearly, significant
inconsistencies between the measured and the computed results were present
using the nominal values of the coefficients.
From a study of the effect of variations in the values of the coefficients
on the deviation between the measured and computed oil composition, it was
determined that the carbon content of the char and the carbon content of the
feed have a major influence on the results. Thus, the least-squares
program made a search for that combination of w ,. and w , , within bounds of
ct ecu
+ 10 percent of the nominal values, which minimizes the square root of the
sum of the squares of the difference between the computed and measured values
of w , w , w, and w
co oo ho no
The results of this computation are presented in the ITERAT OUTPUT. Study
of the table shows that the measured versus the computed values of C, H, N,
and 0 for the oil are as follows:
Percent
Difference
+ 0.45
- 39
+ 10.7
- 15
111
Element
C
H
0
N
Measured
0.657
0.071
0.242
0.04
Computed
0.654
0.043
0.268
0.034
-------
Thus, with the slightly modified values of w f (+ 6%) and w h (+ 4%), all
the results are put into a much better overall agreement than is possible
from the direct computation of the first eight equations, and with .only minor
variations in M , M , M and HV.
112
-------
SENSAN OUTPUT
RUN N4IMOC? <•
N2 C H2 0? HV
CAS .tdS .111 .021 .303 270*.
CHAR .025 .7*1 .'J2b .089 10951
WATEP 0 0 .110 .890 U«tO
FEf.0 .017 ,i»e6 .0-M= *«0.573o HtfO= 1?<»97.7
WCO= ,85'.5'i7 WHO= .0181314 W00= 8.27837E-2 WNO= <». <»<»951E-2
-10X OF HOI /H 1 )= ,(»3e5
MG= 6<».c?' '10- ?<4.5<459 MW= 30.1971 HVO= 15199. t»
WCO= .913696 WMO= 5.55fl65E-2 WJO= .317711
NCMIHAL W( 2 )= .191
*10'^ OF NOM J( 2 >= .'101
KG= 5S.0327 10= 25.3575 :1X= 35.9093 HVO= 13599.9
MCO= .7930-JU WHO= ,&3't'»2'3 W00= .1»<»937 HNO=-1.25098E-2
-10X OF MOM M( 2 (= .1719
H&= 5i."'3?7 Mj= 25.'^75 M«= 35.=093 HVO= 1«»967.5
HCO= .dl0539 KHO= .03t.'»2'i W00= .161*997 WNO = -9.99336E-2
NCHINAL W( 3 )= .021
*10X OF N01 M( 7 •= .0231
MG= 58.0327 M0= 25.J57r' MW= 35.?09« '=
Mf,= Ofl.1393 10= 2^.093fl MW= 37.0667 HVO= l<»o4C.l
MCO= .171937 WHO= 3.0»h01€-c WdO= .i'jOSSW WKO = -&. 13207E-2
-105! OF NCM WC ? 1= 2«.33.b
HG= 57.9262 10= 26.6165 HW= 3<«.7572 HVO= 13S67.2
MCO= .79/977 MHO= 3.76i»7
-------
NOMINAL WC o )= .335
*ior OF »nn «( (. >= .0275
HC= 57.oi1o 10= 25.3716 M
WCO= .13/113 -1HO= 3
-10X OF «4CM W( 6 >=
HG= 59. 1'^f 10 = 25.
HCO = .
MVO = 1U217.7
HVO=
=-5 . 8<«l&2F.-2
NCHINAL M( 7 )= .751
+ 10'^ UF N3M ij ( 7 >= .«?<•!
MG= 51.032/ MT- 25.3575 HH= JJ.
HCO= .77252rl WHO= .OJ'*'*?^ H00=
-10X OF MOM H( 7 )= .6751
MG= 58.0327 H0= 2->.^575 Mtf= .'S
HCO=
CT^ MVO= 13301.8
lfl',9'97 WMO= fl.O<4605E-3
HVO= 15165.6
NOMINAL W( A )= .026
+ 10"? OF IIOH W( 8 ) = .03«fa
HC= 51.0337 .10= ?5.?575 !UJ= 3r>
MCO= .S36796 H'tO= .0322n<» W00=
-10X OF HOM W( 8 t= .C23<«
MG= 59.07.27 110= 25.3S.75 HW= 35.=0°d HVO= l
-------
NOMINAL rf< 13 1= .11
*10X OF NCM M( 13 )= .121
MG= D«.03?7 10= 25.3575 MH = '3.5098 4VO = 13i«05.2
KCO= . 83«>7Jb W-in = 2.01«.6lJf>2 nOO= .1«<«9C7 WNO = -t«.
-1CX OF NOr rf( 13 )= . GO*.
MG = 58.0327 10= 25.3575 MW = 35.=C9» 'IVO= 15062.2
HCO= .83f.74o WHO = «».8Jillt-2 HCO= .11i»9e7 WNO=-6.98038E-2
NOMINAL W( li» I = .IP
+ 10X OF I10M W( 1<* )=
MG= 54.0J27 M0 = ?c;.J575 MW= 35.S098 »IVO= 1U233.7
HCO= .»».3C7S6 WH3= .OJ«.«-2^ W00= 7.51Q51E-2 MNO= 5
-10*< OF UOM W( lu ) = .«01
HG= 5fl.0??7 110= 25.3075 Mt4= 35.c09» MVO= 1<»233.7
HCO= .a3d707 M0= ?5.0257 MM= 36.2136 HVO= 1«.336.5
HCO= ,a«47f.78 ^MO- 3.J5P67L-2 W00= .176306 WNO=-5.75109E-2
-10V. OF HOM M( 15 )= 1G?C
MG= 5'«323 WNQs-5.49 365E-2
NCMIHAL H( 16 I= .017
*lO*i OF NCH W( If, »= .0187
MG= 58.3869 M0= 25.3i3<« MU= o'j.5997 HVO=
HCO= .53559 WHO- 3.53!»ErJl9ft<« WNO = -6.31069E-2
-10V. OF N'OM W( It )= .0153
MG= 57.671-* MO= 25.1*016 MM= 36.22 HVO= I«»lfl3.6
WCO= .83HOOS WHO= 3.33ie>9£-2 W00= .17803'* WNO = -i«.93606E-2
NOMINAL W( 17 »= . (* 56
*10X OF N01 W( 17 )= .5T-46
MG= 58.032' M0= 25.3575 MH= 35.°09rt HVO= 17312.8
HCO= i.028!»& V»HO= .031<«,2S) W00= .lfl<.997 WHO =-. 22-) W00= 1.26Cll£-2 WNO=
115
-------
NCHINAL W( 20 )= 1*72
«10X OF NO* *( ?u 1= 9209.2
HG= *7.tt^,^ WHC; ^.1/li03C-2 H00= .322<»<«7 WMO = - 3.5831 1E-2
-10X Or N0i_ W( 2u 1= 751*.fl
HG= 58.6C15 M0= n.n2« "H= <42.070<« HVO= 17032.9
WCO= 1.13326 «MC= 9. e'
NOMINAL M( 21 >= .7?
*10X OF NCM '41 ?1 »-
MG= 63.673 M")= 2<<.6''i K'-J- '(1./96 HVO= 1509&.?
WCO= .Hfild9 MHO= 5.?3P3-'il-£ HCO= .3033"9 WNO
-1C''. OF N3M W( ?1 )= ,b'-S
MG= 52.1923 M0= 2o.01«» »M= U3.0237 MVO= 13it23.7
HCO= .S56255 WHC= 1.0606IJ-2 nOO= 7.319. 62217E-2
NCMINftL W( 23 } = 0
»10V. OF NOK M( 23 >= 0
MG= 55.0327 H0= 2S.3575 M^= 35.=jqe HVO=
WCC= .53G79& '..'10= .OJ^f-JO ;-fr!C= .l°.t.997 H'JO = -5.62217E-?
-10X OF NOM H< 23 )= 0
HG= 5«.U?27 M0= 25.357? MW= 35.=09" HVO= H23J.7
HCO= .««'!G79-j WHO= .t3M«.2^ K00= .18^997 WNO = -5.62217E-2
NCHINAL W( 2<* )= .23
«10X OP MOM M( ?U )=
HG= 55.0327 H0= 25.3575 HH= 35.909^ HVO= 1<»233.7
HCO= ,'«3r.7^6 WHO= .D3«t(.?9 W00= .21^.013 WNO=-8.
-10X OF NCK W( 2<» >= .207
HG= 58.0327 >O= 25.1575 MM= 35.CO"" HVO= 1«.233.7
WCO= .436796 MHO= .O3'.il2<) K00= ,15ic<»i KNO = -2.32059T-2
NCHIMAL H( 25 ) = 0
+ 10*X OF NCM W( 25 >= 0
MG= 5A.n^27 MO= 25.3570 MW= 25.£C9« HWO= 1<«213.7
HCO= .S3fc796 KHO= .03i|l.?9 H00= .iri.997 MNO = -5. 6221 7E-2
-10X OF MCM H( c5 )= 0
HG= 51.0327 M0= 25.757^ HW= 35.^09« HVO= 1<»233.7
MCO= .<3b796 »= .
MG= 57.H158 10= 25.'tt«.5 Mrf= 36.0997 HVO= H.203.
WCO= .837539 WHO= 3.37'."9E-2 M00= .180731 HHO = -
-102 OF NOM W( 26 )= .C3&9
HG= $4.3491 M0= ?5.33uh MW= 3i.7?03 HVO= 1<«26<».<«
HCO= .93bO?(« HHO= .035109 WOO- .149262 WNO = -. 0&0<«2 5
NCHINAL M( 27 1= 13713
116
-------
*ior OF NCM w< z? >= 159HH.:
MOs 5*.2««55 M0 = ?2. .»39« MU= 3<«.?1«.6 HVO= 1508?. •)
MCO= ,3Z72:>3 WMH= 2. 6<5?«i4E-i; «00= .11?7?7 WNO = -f>
-10V. OF NCM W( a? )= 1PH1.7
MG = 57.7671 M0= 2i.i.-iPH Md = 33.63M HVO= 13379.6
HCO= ,7<»6*3M WHO= <..lo2c6e-2 HQC= ,?57237
NCMIN4L W( ?<\ » = ?1 . /
*10V. OF MOM W( Zn )=
MG= 5^.06^5 '-10= ZS-
HCO= .(^27&^3 '-(H0= 3.61315F-2
-10X OF NG1 'J( 2» )= 19.-J3
MG= 59.Coll Mn= 27.0546 1^= 36.Ul'*'. HVO=
HCO= .SIIU76D WHO- 3.2327JE-2 HCO= .16'»2»7
NOMINAL W< 20 )= 100
*10'i OF N01 U{ ?q »= 110
MG= 57.8-355 ^0= 31.1260 HW= ^0.2779 'IVO= li»131.2 ^
HCO= .^3-^681 WHO= 3.2^011C-2 W30= .Io75«i2 MNO = -3. B5237F.-2
-IQ'< OF NOM X( ?T ) = 9C
MG= S'J.lbSI 1T= H.518^ MH= 31.5M7 HVO=
HCO= .S33901 WHO= 3.76103E-2 H00= .212733
NCMIN/VL H( 30 ) = 36.U
*10TJ OF NC* WC 30 )= 1.0.0-*
MG= 6.1.9012 M0= 34.2'J75 ^W= 3U
WCO= .'J27172 HHQ= 3.61'«91!T-2 '^00= .19711*3 WNO=-&. 0<»6 33E-2
-10X OF NO* H( 3C •= 32.76
MG= 52.1b<«l M0= 2o.Hl76 MW= 37.07e>« HVO- ii»265.6
HCO= ,1l«'56'»3 WHO= 3.2S«.62 W00= .1735^3 WNO=-5.791«»9!:-2
-10X OF NOM W{ 32 )= i^cJO
HG= 57.S-J6 M0= .'5.7916 MW= 35.5125 HVO= li«103.3
HCO= .122985 WHO= 3.v57<»?t-2 WCO= .196026 WNO = -5.i»5855E-2
117
-------
HERAT OUTPUT
RUN
GAS
CHAR
MATE"?
FEED
AIR
.025
0
.017
.770
Q
.191
.751
G
H2
.021
.026
.110
.061
0
02
.3C3
.089
.890
.1*37
.230
HV
27G<«
111*0
1372
C
OIL INITIAL VALUES:
TOTAL WEIGHTI
= 13713
FEf0= 100
HNC =
CHA3= C1.7
AIR = 36.1*
ENERGY LOSSES= 5<»000
HEIGHT FRACTIONS OF
ELEHFNTS IN OIL! CA-?80K'= .657 HYD*OGEN= .071
OXVG£N= ,2«i2 NIT*OGEN= .O**
CALCULATED VALUES A^E AS FOLLCWSl
7
17
11
11
MAS3ESI
GAS = 57.7202
OIL = 29.0537
HEIGHT FKACTIO'IS
OF E!LEM£NTS IN OIL! CAWBCK =
OXYGEN=
MOISTURf= 32.5261
HEATING VALUE IN OIL
NITROGEN=
118
-------
SENSA1I LISTING
9 FILE »i="Sr.NSAN"
10 FILE ««,="R'Jf^",«^ = "
11 FILE «13 = "'n
12 FIL£ *19 = MPUMV
20 DI*i H(32) ,A(3 ,3) .0 (
25 °='.INT 'VU'4 ff"
26 INP'JT N
30 MAT INCUT «N,K
«»0 PRINT "INITIAL -?UN"
50 GtlSUG 500
70 PRIMT "HCO = " ;K(U)' ;"'
63 °RIMT "VUN?"
90 INPUT C
100 IF C=0 THLN
102 RTSTCP.F ON
103 MAT INPUT HfJ.W
HVO=":H
105 PPINT
ti 4 ii
1
no POINT #1," PUN NUMOER";N
111 P*IHT #1
112 PRINT »1
S2
GAS
";w(11;"
":w(&) ;••
0
";w(i6>;"
.770
OIL INITIAL VALlZSl
TOTAL WEIGHT:
HATER
FfcCD
"; k {7 >:"
c
113 P*INT *1,"
115 PRINT ffl,"
116 P^IKT Hi,"
117 PRINT flt"
llfl P^INT *1,"
119 P^ir'T #1,"
120 P»INT »l,"
121 PSINT «i,"
122 PRINT ffi."
123 ^IM rfl,"
125 P^INT #1
130 PxlNT #1," HV-HtAlIKG VALUr!"," "
131 PRINT i*l," HtfO^HEATING VSLUi OF THT OIL"
132 P?INT ti," WNO=WT. F^AC. OK N2 IN OIL"
133 P-MNT ffl
13;"
ENERGY LOSSES=":H(32)
0"
MOISTU9t=":w(31l
<»50 NEXT j
-------
500 «<1,1>=H(1)
510 AC1»2)=H(?M
520 A(l,3»=W(ll>
530 At2,l)=l
500 A(2t2>=l
550 A(2,3)=l
560 A(3,1)=W(5)
570 A{3,2)=W(27>
580 A(3,3I=W<15)
600
610
620 MAT n=INV(A)
630 MAT 0=0»C
6
-------
ITERAT LISTING
9 FILE >,C(3l|g(6>«£(<»>fr<(<«)«L(<«)*MU)fHl(V THtN 215
190 \l-2tt
195 Hl(l) =H
210 Hl((.) =M(H(Hll
215 W(HC1II=M(H<1) J*.C2*M(1J
220 NEXT 0
225 W(HU> )=.g*N(l)
230 HM(2))=/MH(2M +.02*M<2)
235 N?xr N
>
265 NEXT L
270 H(H<<«» I =
271 T=H1( 1) +U1 (£) »rll(3)
272 IF T >0 THUN
121
-------
.275
zao
299
300
301
309
310
315
320
325
330
335
(«00
«»30
PPINT 11," M.GATHM Wf.IMlT FRACTION*
GO TO 25
PRINT II," lNOICfS=","NtW VALUES""
FO* 1=1 TO i.
PrflKT II,1" ' .HI III
NEXI I
1.50
<«60
500
510
520
530
5<<0
550
560
570
500
590
600
610
620
630
WfM(3l»=*l«3>
M(»HMI=H1(I*)
GOiUP HuO
G03UG OuO
GO TO 25
TOR J=l TO 10
GOSUfi 530
H{*7) =H
N£XT J
Ad, 11=^(1)
A(1,2J=W(26»
A(3,1»=W( >t
AC3,2)=W(?7)
A13,3» =«
MAT 0=I'
MUT 0=0'C
6A1
6<«2
650
660
670
660
685
666
690
600
602
601.
605
806
620
900
901
902
907
910
911
912
913
91U
915
916
917
H=C:
RETURN
PPINT ii,"
PRINT II,"
P9INT II,"
PRINT II,"
PVINT II,"
MASSES: GAS ='
OIL =";*I2) ."HHATING VALUE IN OIL
WilGHT FRACTIONS"
OF CLEMENTS IK OIL* CAR30N=" JRf. ) ,"HY OS!OG£N = "; R 15)
I,"NITROGEN=":W
="5M
920
922
923
*1, RUN
PRINT.II
PRINT II
PRINT »1,"
PRINT II
PRINT II," GA3
PRINT H," CHAR
P^UJT II," WATC«
PRINT »1," Ft£D
PRINT II," AI3
PRINT II
PRINT II," OIL INITIAL VALUESI
PHIfa II," TOTAL WCIGHTI
PRII.T I'l,"
PHI NT H,"
H2
02
H»f
o
": w < 16):'
.rra
" : H ( 3) : "
"; w t«j:"
,110 .390
"; w ("4) t"
{W
WNO =";W(26»
.230 0'
•IWO =":wi27)
AIR =MJH(JO)
EKEFGY LOSS£S=":WC32»
122
-------
925 PRIM «1," WtlGHf FRACTIONS OF"
926 "SIN! 01." CLIENTS IN ClLt C AWSON=" J t ( 1) ; " HYD^OGt N = " ;C f 2)
930
932 PRI'JT «1," CALCULflTECJ VALUuS APS AS FOLLOWSl
93«t PWINT *1
9<»0 RETURN
999ENO
123
-------
APPENDIX D
APPROXIMATE ANALYSIS OF BRAKE DE-RATING FACTOR
This appendix contains an approximate analysis of the brake de-rating factor
for an unthrottled spark-ignition (SI) engine (Otto engine). The results
are intended to be a guide for predicting the de-rating of the brake output
power of the SI engine due to a change in the fuel.
BASIC ASSUMPTIONS
For an unthrottled SI engine operating on the hypothetical Otto cycle with
a perfect gas as the working fluid, it can be shown that the thermal
efficiency depends upon the specific-heat ratio of the gas and the engine
compression ratio. Furthermore, the energy available to the engine—the
heat transferred to the gas at constant volume—is considered to be the
lower heating value of the fuel in the air-fuel mixture. (Values for this
are given in the right-hand column of Table 7, page 64, for Stoichiometric
mixtures of air and gasoline, natural gas and pyrolysis gas.) Thus, for a
given engine (i.e., a given compression ratio) and for Stoichiometric mixtures
of two different fuels and air, if the apparent specific-heat ratios for the
mixtures are practically the same,
• the thermal efficiency will be practically the same for
each air-fuel mixture; and therefore,
• the ratio of the net work (indicated work) done by each
gas (air-fuel mixture) will be essentially the same as
the ratio of the lower heating value of the fuel in a
given volume of each of the two gases (air-fuel mixtures).
BRAKE DE-RATING FACTOR
In order to develop an approximate expression for the brake de-rating factor,
it is convenient to make the following definitions:
124
-------
IPRG = indicated power produced by the reference gas
(working fluid) at a given engine speed
I^Q = indicated power produced by the test gas at
a given engine speed
FP = FPRG - FPTG = friction (includes water pump, generator, etc.)
power required to drive the engine at a given
speed—assumed independent of gas
BPRG = brake power at a given engine speed for the
reference gas
BPTG = brake power at a given engine speed for the
test gas
IDF = indicated de-rating factor at a given engine
speed IPTG/IPRG
BDF = brake de-rating factor at a given engine
Noting that
speed = BPTG/BPRG
BPRG - IPRG - FP
BPTG = IPTG - FP
the brake de-rating factor becomes
BP IP - FP
or, upon introducing the indicated de-rating factor, Eq. (D-3) becomes
IDF - FP/IP
If the "friction" power is known as a fraction of the indicated power of the
reference gas (i.e., FP/IP_,_) as a function of engine speed, the brake de-
Klr
rating factor depends only upon the indicated de-rating factor. Noting that
the indicated de-rating factor is equal to the ratio of the rate of net work
done by the test gas to that done by the reference gas which, in turn, is
essentially equal to the ratio of the lower heating value of the fuel in a
given volume of the test gas to that of the fuel in this given volume of the
125
-------
reference gas, the indicated de-rating factor may be written as
LHV
where
LHV = lower heating value of the fuel in a given
TG
volume of the test gas
"LHVRr = lower heating value of the fuel in a given
volume of the reference gas
Note that IDF is independent of engine speed.
From the data of Table 7 (page 64), using a Stoichiometric mixture of
gasoline and air as the reference gas and a Stoichiometric mixture of
simulated pyrolysis gas and air as the test gas, the indicated de-rating
factor is
IDF - 2-474 MJ/m" = 0.72
3.439 MJ/in
Since the brake de-rating factor has been obtained for simulated pyrolysis
gas (Table 6, page 63), it is possible to compute the "friction" power
fraction FP/IP_r as a function of engine speed from Equation (D-4).
Solving Equation (D-4) for FP/IP gives
FP IDF - BDF ,_. ,v
- (u—b)
Computed values of FP/IP.,.,, using the data of Table 2, are given in Table D-l.
Kd
Had data existed for FP/IP , . , it would have been possible to estimate
gasoline
the brake de-rating factor for simulated pyrolysis gas by using Equation
(D-4). For example, assuming that the factors in Table D-l are correct,
the brake de-rating factors for a test gas consisting of a Stoichiometric
mixture of natural gas and air and a reference gas consisting of a
126
-------
TABLE D-l. FRICTION POWER FACTOR
Speed
(rpm)
1,500
2,000
2,500
3,000
FP/IP
gasoline
0.20
0.30
0.30
0.26
FP
(kW)
6.4
12.6
15.4
15.7
Stoichiometric mixture of gasoline and air can be computed as follows:
• From Table 3, page 47, IDF is
LHV 3
TDTT - natural gas _ 3.101 MJ/m _ n Qn
~ LHV 3 U.:*U
gasoline 3.439 MJ/m
• From Equation (D-4), the data of Table D-l, and
DF = 0.90, compute BDF. (The results of these
computations are given in Table D-2).
TABLE D-2. BRAKE DE-RATING FACTOR—NATURAL GAS/GASOLINE
Brake
Speed De-rating Factor
(rpm) (Natural Gas/Gasoline)
1,500 0.88
2,000 0.86
2,500 0.86
3,000 0.86
Noting that the brake de-rating factor is defined by
Tjp "Dp
_ SPG _ simulated pyrolysis gas
or
SPG/G BP,, BP ..
G gasoline
BP BP
NG _ natural gas
BDGNG/G BP_ BP
G gasoline
127
-------
it is clear that
RnF
BDFSPG/NG
Thus, comparing the simulated pyrolysis gas with natural gas, the brake
de-rating factor can be obtained from the data of Table 6 (page 63) and
Table D-2. The results are given in Table D-3.
TABLE D-3. BRAKE DE-RATING FACTOR— SPG/NATURAL GAS
___ Brake
Speed De-rating Factor
(rpm) (Sim. Pyrol. Gas/Natural Gas)
1,500 0.74
2,000 0.70
2,500 0.70
3,000 0.72
128
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
. REPORT NO.
EPA-600/7-79-230
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
A Prototype Mobile System for Pyrolysis of
Agricultural and/or Silviculture! Wastes
5. REPORT DATE
October 1979 issuing date
6. PERFORMING ORGANIZATION CODE
(ooUoo)
7. AUTHOR
------- |