United States Industrial Environmental Research EPA-600/9-81-006
Environmental Protection Laboratory January 1981
Agency Research Triangle Park NC 27711
Research and Development
&ER& Symposium Proceedings:
Environmental Aspects of
Fuel Conversion
Technology, V
(September 1980,
St. Louis, MO)
Interagency
Energy/Environment
R&D Program Report
-------
EPA-600/9-81-006
January 1981
Symposium Proceedings:
Environmental Aspects of
Fuel Conversion Technology, V
(September 1980, St. Louis, MO)
F.A. Ayer and N.S. Jones, Compilers
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
Contract No. 68-02-3170
Task No. 25
Program Element No. 1NE825
EPA Project Officer: N. Dean Smith
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
These proceedings for the symposium on "Environmental Aspects of
Fuel Conversion Technology" constitute the final report submitted to
the Industrial Environmental Research Laboratory, U.S. Environmental
Protection Agency (IERL-EPA), Research Triangle Park, N.C. The sym-
posium was conducted at the Chase-Park Plaza Hotel in St. Louis,
Missouri, September 16-19, 1980.
This symposium served as a colloquium on environmental information
related to coal gasification and liquefaction. The program included ses-
sions on program approach, environmental assessment for both direct
and indirect liquefaction and for gasification, and environmental con-
trol—including the development of the EPA's pollution control guidance
documents. Process developers and users, research scientists and State
and Federal officials participated in this symposium, the fifth to be con-
ducted on this subject by IERL-RTP since 1974.
Dr. N. Dean Smith, Gasification and Indirect Liquefaction Branch, EPA-
IERL, Research Triangle Park, N.C., was the Project Officer and the
Technical Chairman. Mr. William J. Rhodes, Synfuel Technical Coordi-
nator for EPA-IERL-RTP, was General Chairman.
Mr. Franklin A. Ayer, Manager, Technology and Resource Management
Department, and Mr. N. Stuart Jones, Analyst, Technology and Re-
source Management Department, Center for Technology Applications,
Research Triangle Institute, Research Triangle Park, N.C., were sym-
posium coordinators and compilers of the proceedings.
11
-------
TABLE OF CONTENTS
Page
Opening Session 1
Keynote Address 2
Kurt W. Riegel
Session I: GENERAL APPROACH 7
Robert P. Hangebrauck, Chairman
IERL/RTP Program for Gasification and Indirect Liquefaction 8
T. Kelly Janes
EPA/IERL-RTP Program for Direct Liquefaction and Synfuel Product Use 12
Dale A. Denny
Update of EPA/IERL-RTP Environmental Assessment Methodology 17
Carrie L. Kingsbury" and N. Dean Smith
The Permitting Process for New Synfuels Facilities 40
Terry L. Thoem
The TVA Ammonia from Coal Project 64
P. C. Williamson
Environmental Control Options for Synfuels Processes .... 75
F. E. Witmer
Technical and Environmental Aspects of the Great Plains . . ... . 105
Gasification Project
Gary N. Weinreich
Session II: ENVIRONMENTAL ASSESSMENT: DIRECT LIQUEFACTION 115
D. Bruce Henschel, Chairman
Preliminary Results of the Fort Lewis SRC-II Source Test ... 116
Jung I. Kim* and David D. Woodbridge
Chemical/Biological Characterization of SRC-II Product and By-Products . 134
W. D. Felix,* D. D. Mahlum, W. C. Weimer,
R. A. Pelroy, and B. W. Wilson
Low-IMOx Combustors for Alternate Fuels Containing Significant Quantities 159
of Fuel-Bound Nitrogen
W. D. Clark, D. W. Pershing, G. C. England, and M. P Heap*
Problem-Oriented Report: Utilization of Synthetic Fuels: . 208
An Environmental Perspective
E. M. Bonn, J. E. Cotter, J. 0. Cowles,*
J. Dadiani, R. S. Iyer, J. M. Oyster
* Speaker
-------
Page
Session III: ENVIRONMENTAL ASSESSMENT:
GASIFICATION AND INDIRECT LIQUEFACTION ... 267
Charles Murray, Chairman
Environmental Test Results from Coal Gasification Pilot Plants . . ... . ... 268
N. A. Holt, J. E. McDaniel, and T. P. O'Shea*
COS-H2S Relationships in Processes Producing Low/Medium Btu Gas 289
Michael B. Faist,* Robert A. Magee, and
Maureen P. Kilpatrick
Behavior of a Semibatch Coal Gasification Unit . . • • ... 317
W. J. McMichael* and Duane G. Nichols
Carbon Conversion, Make Gas Production, and Formation . . . ... 333
of Sulfur Gas Species in a Pilot-Scale Fluidized Bed Gasifier
M. J. Purdy, J. K. Ferrell,* R. M. Felder,
S. Ganesan, and R. M. Kelly
Modderfontein Koppers-Totzek Source Test Results .... ... 359
J. F. Clausen* and C. A. Zee
An Environmentally Based Evaluation of the Multimedia ... . . 380
Discharges from the Lurgi Coal Gasification System at Kosovo
K. J. Bombaugh,* W. E. Corbett,
K. W. Lee, and W. S. Seames
Ambient Air Downwind of the Kosovo Gasification Complex: A Compendium 428
Ronald K. Patterson
Characterization of Coal Gasification Ash Leachate . . 452
Using the RCRA Extraction Procedure
Kar Y. Yu* and Guy M. Crawford
Comparison of Coal Conversion Wastewaters . 464
Robert V. Collins, * Kenneth W. Lee, and D. Scott Lewis
Session IV: ENVIRONMENTAL CONTROL 483
Forest O. Mixon, Jr., Chairman
Ranking of Potential Pollutants from Coal Gasification Processes . 484
Duane G. Nichols* and David A. Green
Effect of Sludge Age on the Biological Treatability . .... . . 504
of a Synthetic Coal Conversion Wastewater
Philip C. Singer,* James C. Lamb III,
Frederic K. Pfaender, Randall Goodman,
Brian R. Marshall, Stephen R. Shoaf,
Anne R. Mickey, and Leslie McGeorge
* Speaker
IV
-------
Page
Treatment and Reuse of Coal Conversion Wastewaters 537
Richard G. Luthy
Pilot Plant Evaluation of H2S, COS, and C02 553
Removal from Crude Coal Gas by Refrigerated Methanol
R. M. Kelly,* R. W. Rousseau, and J. K. Ferrell
Pollution Control Guidance Document for Low-Btu 595
Gasification Technology: Background Studies
W. C. Thomas,* G. C. Page, and D. A. Dalrymple
Development of a Pollution Control Guidance Document 619
for Indirect Coal Liquefaction
K. W. Crawford,* W. J. Rhodes, and W. E. Corbett
Initial Effort on a Pollution Control Guidance Document: 637
Direct Liquefaction
J. E. Cotter,* C. C. Shih, B. St. John
Appendix: ATTENDEES 653
•Speaker
-------
OPENING SESSION
-------
KEYNOTE ADDRESS
by
KURT W. RIEGEL, Ph.D.
Associate Deputy Assistant Administrator
Office of Environmental Engineering and Technology
U. S. Environmental Protection Agency
Good morning. On behalf of the Environmental Protection Agency, I
welcome you to our Fifth Symposium on the Environmental Aspects of Fuel
Conversion Technology. Since our Fourth Symposium in Hollywood last
year, much has happened, but two things in particular now inspire our
research efforts: First, the price of imported oil has continued to
skyrocket. For example, from June 1979 to June 1980, the price in-
creased from an average of $18.90 to $31.60 per barrel--not counting
spot market surcharges. Second, the President has signed into law the
Synthetic Fuels Corporation Bill authorizing up to $20 billion to en-
courage the growth of a synthetic fuels industry in the United States.
These two stimul i--among others—appear to me to insure that the synthe-
tic fuels industry will be real--establ ished and thriving—well before
the end of the century.
As environmental protection scientists and technologists, we have
had a unique opportunity to study the various synthetic fuels processes
in embryo and to lay the basis for sound environmental development of
the industry. This is in sharp contrast to the situation we have faced
with countless other industries, where after-the-fact environmental
regulations have been resented and challenged, either legally or polit-
ically. After the oil embargo in late 1973, we prepared to respond to
the environmental challenge of a rapidly growing synthetic fuels in-
dustry that, according to the Project Independence Blueprint, loomed
large on the horizon. That shadow has been looming and receding through
many cycles in the past six years. As you all know, we have suffered
on-again, off-again funding in response, but we have somehow managed to
sustain a core effort through all of these gyrations.
Perhaps it is just as well that our day of reckoning has been
delayed. We have learned a great deal more about the processes and pol-
lutants and have seen the evolution of more comprehensive Federal envi-
ronmental laws. New acronyms and areas of concern have appeared since
1974: TOSCA, RCRA, priority pollutants, hazardous solid wastes, etc.
Each new law has broadened our perception of our task to characterize
the waste streams from synthetic fuels technologies, to find appropriate
environmental control technologies, and to formulate a comprehensive
data base for the use of EPA's Program Offices, as they put together
effective, economically feasible regulations.
-------
Another important gain during this period has been the refinement
of the communications channels between DOE and EPA through interagency
programs. In response to President Carter's directive of May 23, 1977,
that EPA and DOE jointly develop procedures for establishing environ-
mental standards for all new energy technologies, a Memorandum of Under-
standing between DOE and EPA has been executed. This formalizes the
many fruitful contacts that have been developed at the various working
levels between these organizations.
Further, within the Agency the Alternate Fuels Group and the
Priority Energy Project Group have been established by Doug Costle to
consider the environmental policy issues involved in implementing the
National Energy Program and to coordinate EPA activities for appropriate
responses to these issues.
This morning I would like to briefly review the course of our
odyssey over the past six years and then discuss with you what I believe
will be done in the near future.
The EPA's Synthetic Fuels Program was initiated in the early 1970's
but received a boost in 1974, following OPEC's import embargo and i_n
para!lei with the preparation of President Nixon's Project Independence
Blueprint. The schedules that were originally laid out for our assess-
ments were based upon the apparent national schedules for synfuel com-
mercialization in the 1976 time period. However, private investors
balked at putting capital into plants to produce liquids or high BTU gas
which could not compete in price with natural fossil fuels then or in
the foreseeable future. As ERDA's (now DOE's) Synthetic Fuels
Commercialization Program had failed to gain Congressional approval,
there was no basis for expecting any major Federal support of commer-
cialization activity, and the EPA therefore targeted the completion of
the synfuels program for the 1984-86 time period, which would allow time
for application of our results to plant designs.
So, the EPA's program started rolling in needed data, ERDA/DOE's
program started rolling out development concepts, and--what nobody had
anticipated—OPEC continued rolling up crude oil prices at an ever-
increasing rate. Oil which had cost us $3.50 per barrel in mid-1973 was
over $12.00 per barrel in mid-1977. It rose to over $18.00 per barrel
in mid-1979 and was almost $32.00 per barrel in June of this year. This
escalation has had two major effects: the Federal government, seeing
the continually climbing monthly cost of supporting our crude oil de-
mands through imports and recognizing the damage being done to both our
domestic and foreign economic positions, made a decision not only to
support synfuels commercialization, but also to establish a means of
speeding permit and regulation compliance by developers. The organiza-
tion proposed to handle these tasks was the Synfuels Corporation.
Meanwhile, entirely separate from these legislative activities, a
number of commercial interests noted that the economics of operating
large-scale, coal-to-gasoline or methanol plants became favorable and
indicated a reasonable return on investment at retail unit prices of
$1.00 to $1.25 for gasoline at the pump. As a consequence, a series of
-------
completely independent, privately financed synfuel projects were an-
nounced, ranging over the major coal seams of the country, and with
schedules indicating operation in the 1984-88 time period.
I said earlier that our programs were targeted for completion in
about the same time period. It follows that there is no way that a
plant that starts operating at the time that our program is completed
could possibly utilize our input or data, and the controls on that
plant's waste streams would probably be based upon best engineering
judgement. Furthermore, neither our regional permit offices nor the
local state and county offices would have had a sound basis for evalu-
ating the permit applications submitted for that plant. Again, best
engineering judgement would have been applied in the evaluation process.
It was, therefore, very clear that the EPA needed both a means of deal-
ing with accelerated projects and a basis for rationally and objectively
evaluating forthcoming plant permit applications.
Both of these needs represented areas in which the "traditional"
EPA approaches could not be applied. Simply stated, our data acquisi-
tion and analysis program was not complete, and, therefore, we were not
in a position to write firm "traditional" regulations covering waste
discharges to all media. Furthermore, the EMB charter contained the
option of selecting and recommending certain environmental and other
regulations for executive branch set-aside, and we really didn't have
sufficient data to effectively argue all of the set-asides.
To address both of these needs, the EPA administrator created
operational arms for the use of the existing, formerly advisory, EPA
Energy Policy Committee. The first of these, the Priority Energy Pro-
ject Group focused on the development of a working relationship with the
EMB and had four major objectives:
First, the Group would draft EPA procedures and guidance for devel-
oping regulations in support of the EMB and for performing as an accel-
erator of designated priority energy projects. Second, it would be
responsible for the development of a system for tracking permit process-
ing information, from submittal through approval or rejection. Third,
it would provide information on EPA permitting procedures, thereby
influencing the development of EMB procedures and assisting both the
applicants and the permitting agencies in understanding the total pro-
cess. Finally, the Group would serve as EPA's principal liaison with the
EMB.
The second recently created working arm of the EPC is the Alternate
Fuels Group (or AFG), which has a longer listing of responsibilities in-
volving the Agency's regulatory, permitting, and research strategy for
synthetic and other alternate fuels. This group addresses all synfuels,
and its overall goal is to deal with our assessment data gap, both as a
current problem and in terms of eliminating it as a problem in the near
future. The Group's work plan logically divides into three areas:
First, defining where we are and what the Agency position on the major
issues is right now. This will be accomplished through publication of
our Agency environmental summary paper, which we plan to update period-
ical ly.
-------
Second, the group will prepare Agency guidance, in advance of our
traditional regulations, on the best available controls for application
to synfuel plant waste streams. This will lead to direct input to the
EPA regulatory offices in support of their later development of stan-
dards for the synfuels industry.
And third, the group will prepare an R, D&D plan for the overall
synthetic fuel program under the Office of Research and Development.
This plan, to cover approximately a 5-year period, will address the
options, priorities, and means of filling the data gaps and supporting
the expeditious development of regulations.
I'd like to drop back to the second element of the AFG's work plan.
Since this area--that is, the early guidance--is in current demand, I
think it's worthwhile describing where we are in more detail.
To assist in accomplishing its work assignments, the AFG has de-
fined four Working Groups, covering the major synfuel product areas.
The areas are Gasification/Indirect Liquefaction, Direct Liquefaction,
Oil Shale, and Biomass. Each of these Working Groups is drafting guid-
ance in its particular area; all are working to virtually the same
outline and format requirements; and all are treating the shared or
common technology areas in the same fashion. For example, the impact on
plant costs and operating economics is being handled in basically the
same way by all groups.
The product guidance will be Agency guidance and will cover all
media plus toxic substances and radiation. It will be approved for
release by all of the responsible EPA Program Offices as Pollution Con-
trol Guidance Documents, or PCGD's. There are three principal target of
this guidance. First are the permit reviewers, both in the EPA regional
offices and in the comparable State government agencies. Second are the
process developers or permit applicants who want to construct synfuel
plants: And third are the regulatory offices, which will utilize the
data base as an input for standards preparation.
The technical approach being taken by all Working Groups is, in
brief, to collect and analyze all available environmental and process
data in order to synthesize Agency positions on the best available
control approaches achievable at a reasonable cost. The PCGD's will
present the available process characterization and control data and the
analyses utilized in formulating guidance as an appendix. The pre-
sentation of the data base will enable the regulatory offices to eval-
uate issues (such as how to handle discharges of potentially dangerous
but presently unregulated pollutants) and aid them in deciding how and
when to develop standards. It should also serve to convince system
developers that all reasonable control options have been considered and
to show interested environmental groups that the permitting offices have
the tools needed to protect the environment through the recommendation
of specific controls. Additionally, through the implementation of a
multicycle review process, the comments and criticism of key industry
personnel are being obtained as the PCGD's evolve through several draft
stages. This direct participation will, we hope, further serve to
convince industry of the thoroughness of our approach and that it is in
-------
their best interest to use the PCGD recommendations and guidance in
their designs and permit applications.
I don't want to give the impression that we are rapidly construct-
ing some boxes and at the same time trying to convince a number of
interested groups that they'll be happy in them--not so at all.
The PCGD's will provide detailed guidance on the best control
practice (a single control) for each stream, plus provide information on
other approaches relative to cost, energy requirements and residuals. In
additional, for those streams considered to be significant environmental
problems or whose control can have major cost impacts, one or more
options for achieving greater pollutant content reductions or lesser
cost will be presented.
Options which combine controls between process segments or utilize waste
materials (both gases and liquids) as plant fuel will be included. And
for everyone's benefit, a detailed "How-to-use-the-PCGD" section, with
examples, will be provided.
So, as you can see, the boxes are designed to be comfortable for
everyone and to cover everyone's needs as best we can at this point in
time. Naturally, we'll update the PCGD's as additional data are de-
veloped and analyzed in our research program, until firm standards and
regulations are promulgated.
As you all know, the provision of the Energy Security Act which
would have set up the Energy Mobilization Board was cut out of the Act
by an overwhelming majority in the House. The Act, as signed by
President Carter, does create a Synthetic Fuels Corporation and does
provide for up to $20 billion to fund synthetic fuels projects, but the
"fast track" and environmental set-asides have been eliminated.
However, the Agency has been pleased by the responsiveness of the
Priority Energy Project Group and Alternate Fuels Group and their var-
ious affiliates. We may no longer be under pressure to "fast track,"
but we have benefited greatly from the effort to look ahead and to
coordinate research with regulatory activity and the generators of the
emerging synthetic fuels technologies. The interchanges that have
occurred over the past several months have given each participant a
keener appreciation of the pressures and, sometimes subtle, details that
must be mastered, which each of the other participants brings to the
table. Having gained this, we are loathe to let it go.
Therefore, although the pace may not be quite as frantic as it was
the first six months in 1980, we do intend to continue with the work we
have started, work which has been wel1 done.
Now that I have retraced with you the zig-zag path of legislation
and administration, I can direct your attention to the much more in-
teresting technical program that will be presented over the next four
days. Thank you for coming. I am sure that you will enjoy it.
-------
Session I: GENERAL APPROACH
Robert P. Hangebrauck, Chairman
Industrial Environmental Research Laboratory,
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina
-------
IERL-RTP PROGRAM FOR GASIFICATION AND INDIRECT LIQUEFACTION
by
T. Kelly Janes, Chief
Fuel Process Branch
Industrial Environmental Research Laboratory - RTF
U.S. Environmental Protection Agency
The synfuels program being conducted by the Fuel Process Branch of
EPA's Industrial Environmental Research Laboratory at Research Triangle
Park, North Carolina, addresses the potential environmental impacts and
control needs of coal gasification and indirect liquefaction technologies.
The purpose of this program is to support EPA's regulatory responsi-
bilities to prevent adverse health or ecological impacts when these tech-
nologies reach commercial practice. The overall goal of this effort is to
aid in the achievement of an environmentally sound and viable commercial
synfuels industry.
At the start of this program, it was recognized that certain program
objectives would have to be accomplished if this goal of an environmen-
tally sound synfuels industry was to be achieved; namely:
The characterization of the multimedia discharges from
these technologies,
The assessment of the discharges' potential health and
ecological effects,
The determination of the degree of control required to
avoid adverse impacts,
The evaluation and applicability of existing control tech-
niques ,
The identification of new control technology needs,
The development and/or support in the development of these
new needed control processes.
In 1974, the initial program effort was directed to the development
of evaluation approaches and identification of potential opportunities for
data acquisition. Due to the complexity of the technologies being ad-
dressed, the lack of ^facilities and information, and the need to undertake
-------
broad multimedia evaluations, it was decided to develop contractual "cen-
ters of expertise." These centers would provide the technical expertise
that could not be developed in-house due to limitation of personnel.
Since coal conversion technologies were only in the development stage
in the U.S. , and since the chemical breakdown of the coal structure re-
sults in the generation of aromatic organic compounds among which are
known carcinogens, the program was based on obtaining sufficient data to
identify and evaluate the total environmental effects of the discharges
rather than to focus on EPA's currently regulated pollutants only.
The program was organized into four major areas:
Environmental Assessment,
Control Technology Development,
Control Research Facilities,
Methodology Development.
Environmental Assessment involves the evaluation of technologies,
data acquisition, interpretation of results, projection of environmental
effects, and identification of control needs.
Control Technology Development involves the evaluation of the avail-
ability and applicability of existing control technologies to meet the
requirements identified by the Environmental Assessment. Additionally,
operational information, reliability, and modification capabilities are
evaluated. This effort has been dropped as a responsibility in the fed-
eral sector for control technology development, and demonstration was
shifted to the Department of Energy.
Control Research Facilities were developed to provide information
concerning the viability of control technologies and to characterize their
multimedia discharges. These facilities alsp offer capabilities to eval-
uate modification of control techniques and the testing of new approaches.
To date two such facilities have been constructed and are operating:
Gasifier with gas cleaning and acid gas removal capabili-
ties. This facility is modular and flexible in design,
allowing evaluation of different systems.
Water treatability facility to evaluate methods for treat-
ing the various wastewaters that would be generated by
synfuels plants.
Methodology Department provides uniform procedures that result in
consistent, cost-effective data gathering and interpretation. These
procedures range from sampling/analytical techniques through data inter-
pretation to report format. The procedures as originally developed by the
-------
Laboratory and other EPA organizations are continually reviewed and re-
fined .
During this initial phase of the program, considerable effort was
spent in identifying availability and viability of sites for future data
acquisition efforts. Due to lack of commercial U.S. facilities, plants in
England, Poland, Yugoslavia, Turkey, and South Africa were surveyed for
potential interest in future evaluations. These sites included the Lurgi,
Koppers-Totzek, and Winkler gasification technologies.
The second phase of this program involved the actual data acquisi-
tion, interpretation of results, and identification of projected control
needs. Domestically, various low Btu gasifiers were evaluated including
Chapman-Wilputte, Wellman-Galusha, and Stoic. Foreign sites included a
Lurgi plant in Yugoslavia and a Koppers-Totzek plant in South Africa.
Results from these evaluations will be presented during this symposium.
The Yugoslavian evaluation was by far the largest effort and was jointly
supported and conducted by U.S. and Yugoslav experts.
The third phase of this effort which we are now well into is the
compilation of data acquired to date into a data base to support EPA's
guidance and regulatory activities. The Agency is now actively developing
Pollution Control Guidance Documents (PCGDs) under the direction of EPA's
Alternate Fuels Group. The Fuel Process Branch is involved in the PCGDs
relating to low Btu gasification, medium Btu gasification, substitute
natural gas, and indirect coal liquefaction.
The PCGDs will provide guidance to protect the environment during the
periods preceding regulations promulgation and to avoid costly delays in
the commercialization of synfuels processes due to uncertainties regarding
environmental control requirements.
The primary purpose of each PCGD is to provide guidance to both
system developers and permitting authorities on control approaches which
are available at a reasonable cost for the technologies under consider-
ation. The PCGDs are also intended to provide the public with the EPA's
best current assessment of the environmental problems posed by the dif-
ferent synfuels technologies and the effectiveness and costs of available
controls. This information should (a) assist system developers at the
outset in their efforts to design facilities incorporating best available
control technologies, and (b) aid permit reviewers in their decision
making by delineating likely pollutants and their concentrations as well
as available control options. The Agency intends these PCGDs to provide
guidance only. The documents have no legal authority, contain no new
regulations of any kind, and include nothing that is mandatory.
IERL-RTP efforts to date have shown that many data gaps still exist.
Specifically, future work should address the following points:
There is a tremendous lack of information on the effective-
ness, operability, and reliability of control techniques
for coal conversion plants. Information of this type needs
10
-------
to be gathered for the whole spectrum of potential pollu-
tants from these plants, not just for those species for
which standards or criteria exist.
There is a need not only to demonstrate existing control
techniques for their applicability to coal conversion
processes, but also to initiate development of improved
methods.
There is a definite need to develop more information on the
health effects of the compounds generated by the breakdown
of the coal structure during gasification or liquefaction
and to investigate the effects of entire discharge streams
upon human health and ecological systems.
11
-------
EPA/IERL-RTP PROGRAM FOR DIRECT LIQUEFACTION AND SYNFUEL PRODUCT USE
by
Dale A. Denny
U. S. Environmental Protection Agency
Industrial Environmental Research Laboratory
Research Triangle Park, N. C.
The direct liquefaction program at EPA/IERL-RTP covers those synfuel processes
which add hydrogen to coal and form liquid hydrocarbon products directly. The
processes currently under study include SRC-II, Exxon Donor Solvent, and H-
Coal. SRC-I is also included in the program because of its similarity to SRC-
II even though the main product from that process is a solid. The synfuels
use program covers products from coal and shale synfuel processing systems.
DIRECT LIQUEFACTION OF COAL
lERL-RTP's work in direct liquefaction of coal includes both the preparation of
pollution control guidance documents, as well as involvement in support of EPA
Regional Offices.
Preparation of Pollution Control Guidance Documents
Laboratory-prepared EPA pollution control guidance documents are intended to be
used by EPA Regions as they evaluate permits, by EPA regulatory offices as
they prepare formal regulations, and by process developers as an indication of
the extent of pollution control EPA considers appropriate for the evolving
synfuel industry.
The documents contain extensive descriptions of the processes and pollutants
discharged, and detailed descriptions of control devices that might be applied
to various sources. Where appropriate, process design modifications are
proposed if they would result in an environmentally and economically more
attractive system.
The range of pollutants considered for control includes those currently
regulated, as well as those unregulated where chemical and bioassay test data
indicate control would be prudent. Synfuel products are also considered in
12
-------
the document to the extent that their on-site storing and handling impacts on
the local environment.
IERL-RTP is making every effort to ensure that the best information available
is contained in the guidance documents. A work group has been established
which has representatives from all EPA's regulatory offices. The Regions are
also represented. Representatives from DOE and the process developers in
industry participate by providing data and a critical technical review of the
accuracy of the technical components of the guidance documents. Extensive
reviews, both internal and external to EPA, are planned. Participants will
include all regulatory offices, the EPA Science Advisory Board, environmental
groups, industry, DOE, and the general public.
The schedule of activities for the next 2 years is shown in Figure 1. The
first version of the guidance document will be heavily slanted toward SRC-II.
This emphasis is the result of a paucity of data available from the H-Coal and
Exxon Donor Solvent (EDS) pilot plants. The guidance document is expected to
be updated to reflect up-to-date information on EDS and H-Coal.
Regional Support Activities
The second important use of guidance documents is as an aid to EPA Regional
Offices as they evaluate permit applications. Regions III and IV have, or
will shortly receive, Prevention of Significant Deterioration (PSD) applications
for SRC-II and SRC-I, respectively. They also have received and been asked to
comment on Environmental Impact Statements for these two processes. Since the
guidance documents are not yet available to the Regions, IERL-RTP is providing
ad-hoc assistance in the evaluation of permit applications and the review of
impact statements.
Inputs provided to date have been mainly identification of data deficiencies
in the applications or impact statements. In limited cases, where specific
control technologies have been identified by DOE, sufficient background material
has been pulled together to make an analysis of the appropriateness of the DOE
selection. Evaluation of specific control systems has generally not been the
13
-------
FIGURE 1
DIRECT LIQUEFACTION POLLUTION CONTROL GUIDANCE
DOCUMENT SCHEDULE
ACTIVITY MILESTONES
Program Kickoff
Draft chap, on Source Assessment
Draft chap, on Control Technology Options
Draft chap, on Environmental Impacts
Draft chap, on Recommended Control Practices
Vol. Ill Draft
Draft, Vol. II & Vol. Ill revised to DLWG,
OEET, DOE, IRC review
First Draft, Vol. 1 to DLWG, OEET, DOE,
IRC review
Receive review comments
Review with OEET & contractors
Review with 0 AQPS, OWPS, OSW, DOE, IRC, etc.
Review with AFG
Second Draft Vol. 1 III to DLWG, OEET
Second Draft Vol. 1 III to AFG, SAB, DOE, IRC
Receive comments on 2nd Draft
Review AFG/SAS/DOE comments with DLWG
Third Draft Vol. 1 III to DLWB/AFG/SAB
Third Draft Vol. Mil to EPC
Revise Third Draft, to DLWG/AFG
Federal Register Notice of Public Forum
Public Forum
Receive Public Comments
Review Public Comments with DLWG
Recommend Comment incorporation to AFG
EPA approval of comment incorporation
Final PCGO to OEET/DLWG/AFG/DOE/EPC
Transmit to printer with EPC approval
Date
6/12/8D
10/31/80
1/15/81
2/27/81
2/27/81
2/27/81
4/15/81
4/15/81
6/ 8/81
6/15/81
6/15-29/81
6/29/81
8/24/81
9/ 7/81
10/26/81
11/ 6/81
1/15/82
1/31/82
31 1/82
3/15/82
5/15/82
6/15/82
11 1/82
7/15/82
B/ 1/B2
9/ 1/82
9/15/82
1980
May Jun Jul Aug Sep Oct Nov Dec
A
A
1981
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
ruuur, . nimpi
A OEET - Office
A AFG • Alter.
A SAB Scien
A
A
A
A
A*
A
A
A
A
A
1982
Jan Feb Mar Apr May Jun Jul Aug Sep
Liquefaction Working Group
of Environmental Engineering & Technology
trial Review Committee
ate Fuels Group
:e Advisory Board
y Policy Committee
A
A
A
A
A
A
A
A
A
A
A
-------
prime task, however, because DOE has not progressed very far with detailed
specifications for control technology components of the SRC-II system.
West Virginia personnel are being assisted in their evaluation of a construction
permit request from DOE. The same problem occurs here: it is difficult, if
not impossible, to estimate the effectiveness of the environmental control
systan when it has not been specified in sufficient detail. These ad-hoc
support activities are expected to continue indefinitely. As a matter of
routine, all inputs to Regions and States are channeled to EPA's regulatory
offices for comment.
IERL-RTP expects to continue its direct liquefaction assessment program for
several years. Major items of concern which have been identified and will be
investigated include the nature and toxicity of emissions from heavy ends
processing, the feasibility of zero discharge water systems, the determination
of the toxic and Teachability characteristics of gasifier solid wastes, and
factors which affect stream time for sulfur cleanup systems. IERL-RTP expects
to spend about $2 million per year in this assessment and control technology
evaluation area.
SYNFUELS USE PROGRAM
EPA's Synfuels Use Program has been underway for approximately 6 months. For
the past few years much emphasis has been placed on determining the environmental
impact of synfuel production facilities. That is certainly a worthwhile
objective but it is clear that, at least in the near term, the most significant
human exposure to synfuel related materials will come from the transport,
storage, and use of the products. Very little attention has been given to
this important aspect of the evolving synfuels industry. The major objective
of the program is to estimate the human exposure associated with various uses
of synfuels and to estimate the toxicity of the materials to which people are
exposed. These estimates are of considerable importance to EPA's Office of
Pesticides and Toxic Substances as they make decisions related to the application
of the Toxic Substances Control Act to the synthetic fuels industry.
To date IERL-RTP has completed a rough-cut market penetration projection for
the various synthetic fuels. The study was limited to coal and shale oil
products because of their nearer term probability for development and uncertain
15
-------
environmental status. This market penetration projection is complemented by a
summary of all completed and on-going human effects research programs which
deal with synthetic fuels. An analysis of these two studies, planned for this
Fall, will result in a specification of the types of data still needed
to allow estimation of the risk associated with exposure resulting from
synfuels use. Priorities for completing the effects work will be established
based on the exposure estimates and estimates of the toxicity of the materials
in question: materials of higher exposure or higher toxicity will be given top
priority. These data requirements and priorities will be sent to DOE, synfuels
developers, and EPA research laboratories with recommendations for implementation.
All the effort on risk estimation has been closely coordinated with EPA's
regulatory offices. It is very important that the data generated be of the
quality and type that is directly useable for the formulation and promulgation
of regulations.
EPA's Synfuels Use Program over the next few years will continue to evaluate
the evolving synfuels industry especially from the view of risk to human
health from new uses of the products or new ways of incorporating synfuels
into the existing production system; for example, blending of synthetic and
natural crude oil in refineries. One current major deficiency is that very
little effects work is underway to evaluate the toxicity of synfuel combustion
products. As these problems become more well defined, IERL-RTP will be conducting
research to reduce the severity of the impact of the use of these products.
IERL-RTP will also begin to look at other environmental impacts such as ecological
effects, regulatory options that are available for dealing with the problems
of synfuel use, and synfuels that are preferable for development from social,
economic, and environmental points of view. IERL-RTP's budget for this
program is approximately $1 million per year for the next 5 years.
16
-------
UPDATE OF EPA/IERL-RTP ENVIRONMENTAL ASSESSMENT METHODOLOGY
Carrie L. Kingsbury
Energy and Environmental Research Division
Research Triangle Institute, Research Triangle Park, North Carolina
and
N. Dean Smith
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency, Research Triangle Park, North Carolina
Abstract
EPA's IERL-RTP has developed a systematic approach for performing each
aspect of environmental assessment to allow for consistent data gathering and
interpretation. Environmental assessment requires the determination of contam-
inant levels associated with point source discharges and comparison of those
determinations with target control levels. Procedures for conducting phased
environmental assessments involving Level 1 and Level 2 chemical analyses and
bioassays have been formalized. Multimedia Environmental Goals (MEGs) reflect-
ing potential toxicity of specific chemicals provide the target values used for
comparison. Source Analysis Models (SAMs) delineate discharge stream severi-
ties based on the components present and mass flow rates. The Level I/Level 2
chemical analysis approach has been coupled with the categorical system for
organizing chemicals addressed by MEGs.
The computerized Environmental Assessment Data System (EADS) at IERL-RTP
is used to store environmental assessment data and to provide links between
characterization and target goals. Eventually, EADS will be used to automate
large portions of the assessment data analysis.
17
-------
UPDATE OF EPA/IERL-RTP ENVIRONMENTAL ASSESSMENT METHODOLOGY
INTRODUCTION
In support of the Environmental Protection Agency's standards-setting and
regulatory functions, information is needed in response to the question, "To
what extent does a particular industrial source cause pollution damage to the
environment?" Answers to this question involve a complex mix of information
from numerous scientific and engineering disciplines. To provide a structured
and cost-effective approach to assembling and interpreting this information,
the concept of an environment assessment has been developed and procedures
established for its implementation.
An assessment of the pollution potential of an industrial source is
necessarily complex because it addresses many types of industrial discharges
into all environmental media (air, water, land). The approach to environ-
mental assessment developed by the EPA's Industrial Environmental Research
Laboratory at Research Triangle Park, N.C., is to divide the work to be accom-
plished into discrete steps with the results of each completed phase providing
guidance for succeeding efforts. Four main advantages of such a formal
approach are that:
1. Thorough screening ensures coverage of potential problems identi-
fiable on the basis of the existing effects data.
2. Attention is focused on the chemical constituents of highest con-
cern.
3. Many unnecessary samples and analyses are eliminated by virtue of
the guidance provided by the results of previous phases.
4. Results obtained from different sources by different investigators
are directly comparable.
IERL-RTP began to develop this structured approach to environmental
assessment about 5 years ago. By then, the need for a common methodology
was recognized clearly, for experiences since 1969 with Environmental Impact
Statements (required under the National Environmental Policy Act) had already
demonstrated the wide variation of outputs that could occur in assessing
possible environmental impacts. Predictably, when the first specific
18
-------
procedures and practices to be followed in environmental assessment were
spelled out in an IERL-RTP report in 1976 , the approach was met with consider-
able resistance from contractors. Some of that continues, but the advantages
of a common methodology are becoming more apparent as the volume of collected
data grows. Over the last 4 years, numerous modifications and additions have
been made in the various segments of the methodology as a result of continuous
research and in response to comments from the users. In many cases, those
applying the procedures are also the methodology developers since the develop-
ment of the methodology has proceeded concurrently with its implementation in
the preliminary environmental assessments conducted by IERL-RTP. Although the
evolution of the methodology continues, the overall approach appears to be
accomplishing its initial objectives.
Many of the conclusions that will be presented in papers at this sympo-
sium will be expressed in terms defined by the IERL-RTP environmental assess-
ment methodology. Because of the common approach, results from the different
studies are comparable, even though certain specific procedures vary to accom-
modate unique problems encountered in each assessment program. This paper
describes briefly the IERL-RTP environmental assessment methodology and its
various components at their present level of development. It is hoped that
this presentation will contribute to a better understanding of the specific
technology assessments.
APPROACH
There are five major components of the IERL-RTP environmental assessment
methodology:
Technology background development
Sampling and analysis
Environmental goals
Impact analysis
Control technology evaluation
Three levels of effort are defined for data acquisition involving sam-
pling and analysis. Level 1 was designed for initial screening or survey of
potential pollutants, and its goal is the comprehensive survey via chemical
and bioassay analyses of all discharges to the environment. Chemical analyses
at this level are primarily directed toward the identification and semiquan-
titation of categories of compounds present in the discharge streams. Level 2
19
-------
focuses on the streams and compound classes found to be of major concern in
Level 1. Analyses are aimed at identifying and quantifying the specific
chemicals present. Level 3 is presently in the conceptual planning stage, and
will involve selectively monitoring the pollutants of concern identified in
Levels 1 and 2 and determining their variation with time and process operating
conditions. Evaluation of the effectiveness of pollution control devices in
place at the test site would be a product of Level 3 data collection.
TECHNOLOGY BACKGROUND DEVELOPMENT
Much can be learned about probable pollution problems associated with a
given process or technology by reviewing existing information and applying
scientific and engineering experience. Consequently, the first step in an
environmental assessment is to obtain all the pertinent literature available.
Attention is given to the current and projected status of the commercial
development of the technology, the varieties of process units applicable, the
process chemistry, and the nature, quantities and points of discharge of waste
streams and fugitive emissions (leaks, spills, etc.). Such literature reviews
usually reveal information gaps that render difficult or impossible an ade-
quate determination of the pollution potential of the technology and associ-
ated environmental damage. Both the selection of the facilities to be tested
and the determination of the amount and types of data to be collected are
directed by the information derived from the literature review.
Once a particular facility has been selected as a test site, a detailed
engineering evaluation of existing data for that facility is made, and tenta-
tive sampling points are selected. Plant layout, temperatures, pressures,
flow rates, and other plant operation data are obtained in a pretest site
survey. The final test plan states what, how, and when required sampling and
analysis activities will be performed. It informs the sampling crew of opti-
mum sampling locations and conditions and of unusual circumstances that may be
encountered during the sampling process. Sample preservation techniques and
procedures for handling and shipment of samples are also discussed.
SAMPLING AND ANALYSIS—LEVEL 1
Sampling and analysis procedures for Level 1 environmental assessments
are set forth in the second edition of the IERL-RTP Procedures Manual. This
20
-------
manual supersedes the 1976 manual. Although the overall approach to sampling
and to organic and inorganic analysis at Level 1 remains unchanged since 1976,
incremental changes in the procedures have vastly improved their effectiveness
and reliability. In accordance with a guideline issued by IERL-RTP, all
IERL-RTP contractors and grantees performing environmental assessments are
required to use the procedures in the revised manual. The manual addresses
quality control/quality assurance as well as the specific analytical and
sampling techniques to be used. New developments in the areas of sampling,
analysis, and quality control are reported in a quarterly report called "Pro-
cess Measurements Review." This widely circulated publication of the Process
Measurements Branch of IERL-RTP announces revisions in the procedures manual
as they are adopted.
It should be emphasized that the objective of Level 1 data acquisition is
to provide a data base to allow prediction of the pollutants and streams of
concern. Once this data base is in place, as it is presently for coal-fired
power plants, it is appropriate to pursue Level 2 investigations. Thus, a
complete site-specific Level 1 study need not precede every Level 2 effort.
However, even for well-developed bases, occasional Level 1 or partial Level 1
surveys can prove informative .
Level 1 Sampling
Level 1 sampling programs are designed to permit efficient collection of
all substances in a stream, making maximum use of existing stream access
sites. Samples from each process feed stream and each process effluent stream
must be provided for the Level 1 assessment. Multimedia sampling strategies
are organized around five general types of samples: (1) gas/vapor, (2) par-
ticulates/aerosols, (3) liquids/slurries, (4) solids, and (5) fugitive emis-
sions. Particulate from gas streams is sized (four fractions recovered) in
the operation employing the Source Assessment Sampling System (SASS). The
availability of the Fugitive Ambient Sampling Train (FAST) has improved the
collection of airborne fugitive emissions. Specifics of the operation of the
SASS and the FAST are discussed in the second edition of the Procedures Manual
Sample size requirements for Level 1 are established to ensure that
analytical results will supply meaningful data. Procedures and equipment to
be used for various stream types are also specified. Table 1 indicates the
21
-------
TABLE 1. GUIDELINES FOR LEVEL 1 STREAM SAMPLING 2
STREAM
Vapors with or without
particulate
Liquid
Solids
Gas (reactive) organic
SAMPLE SIZE
30m3
20 L*
1kg
2L
LOCATION
Ducts, stacks
Lines or tanks
Open free-flowing
streams
Storage piles
Conveyors
Ducts, stacks, pipelines,
SAMPLE PROCEDURE
SASS train
Tap or valve sampling
Dipper method or
composite sampler
Coring
Full stream cut
Grab sample (glass bulb)
material with bp< 100 C;
N and S species
Gas(fixed)02, N2, C02, 10-30 L
and CO
o
Fugitive emission 2,496 m
vents
Ducts, stacks, pipelines,
vents
Ambient atmosphere
Integrated bag sample
FAST or modified hi-vol
May need additional sample volume depending on the nature of the biotesting employed.
22
-------
IERL-RTP guidelines for Level 1 stream sampling based on the detection limits
of the analytical techniques subsequently employed.
Level 1 Chemical Analysis
Samples collected from a facility are subjected to a Level 1 chemical
analysis designed to characterize both organic and inorganic constituents.
Solid samples may also receive a morphological examination. The objective of
Level 1 organic analysis is to isolate and semiquantitate (accurate to within
a factor of three) the predominant classes of organic compounds present in a
given sample. Figure 1, adapted from Reference 2, depicts the current pro-
cedure set forth for Level 1 organic analysis. Quantitative information is
provided by gas chromatography (total chromatographable organics, TCO) and by
gravimetry (GRAV). Qualitative and semi quantitative information is obtained
from conventional liquid chromatography (LC), infrared spectrometry, and low
resolution mass spectrometry (LRMS). A liquid chromatographic separation
based on polarity is employed, which results in seven fractions. Categories
of chemicals expected to elute in each fraction are recognized, and this
information is used in interpreting the LC data.
Inorganic species determined in the Level 1 program include certain
inorganic gases; the major, minor, and trace elemental constituents; and
selected anions. Inorganic gases are measured at the test site using gas
chromatographic, spectrometric, and titrimetric methods. Elemental and ion
determinations are performed on both solid and liquid samples in an off-site
laboratory. Ion chromatography or commercial test kit procedures are employed
for ion determinations. Elemental analysis is accomplished by spark source
mass spectrometry (73 elements) and atomic absorption spectrometry (for
mercury). It is recognized that analyses by spark source mass spectrometry
are better for some elements than for others, but for Level 1 screening pur-
poses the technique is sufficient. More precise determinations may be
provided at Level 2.
Level 1 Biological Analysis
While chemical characterization of a sample identifies known hazardous
chemicals, biological tests provide complementary information for mixtures
whose health/ecological effects are unknown. Biological tests conducted in a
Level 1 effort involve short-term screening tests designed to determine the
23
-------
Organic Extract
or
Neat Organic Liquid
Concentrate
Extract
Infrared Analysis
Infrared Analysis
Gravimetric
Analysis
Aliquot Containing
15-100 mg*
Solvent
Exchange
Liquid
Chromatographic
Separation
Seven Fractions:
Low Resolution
Mass Spectra
Analysis
TCO
Analysis
Repeat TCO
Analysis
if Necessary
TCO" and
Gravimetric
Analysis
*If less than 15 mg is recovered, go to LRMS.
Figure 1. Organic analysis methodology.2
24
-------
4 5
health-related and ecological effects of the samples. ' The tests to indicate
potential health-related effects include the use of both in vitro and whole
animal bioassays designed to detect evidence of any toxic or mutagenic response
in the test organisms. Ecological tests measure the response of aquatic and
terrestrial organisms to the pollutants and include the use of algae, verte-
brate and invertebrate animals, land plants, and insects. The revised Level 1
Bioassay Procedures Manual is expected to be made available this Fall from
EPA. The specific bioassay tests used in Level 1 screening are indicated in
Table 2, updated from Reference 5 to reflect the current bioassay protocol
procedures from the revised manual.
The bioassays for Level 1 screening constitute a minimum set of cost-
effective tests to evaluate the potential biological effects of a sample. The
tests were chosen after extensive evaluation and validation and reflect experi-
ence in three pilot studies and other selected applications.
INTERPRETATION OF LEVEL 1 DATA
In the phased approach to environmental assessment, Level 1 test data
need to be interpreted so that pollutant categories and waste streams can be
evaluated with respect to their potential environmental insult. Such an
interpretation of the data will lead to a decision as to what Level 2 tests,
if any, should be conducted to better characterize the problem streams. In
order to perform this evaluation, it is necessary to have a set of environ-
mental criteria against which the chemical test data can be compared. Cri-
teria which have been developed for this task are referred to as Multimedia
•7 O Q
Environmental Goals (MEGs). ' ' The procedure designed to guide the syste-
matic interpretation of Level 1 chemical analysis involves a source analysis
model called SAM/IA introduced in 1977. (A revised version of SAM/IA is
expected to be available in Spring of 1981. ) Interpretation of bioassay
data has also been systematized using rankings of responses from the various
tests performed.
Two major outputs desired from a Level 1 test effort are (1) the ranking
of pollutant classes within a stream and (2) the ranking of discharge streams.
Both rankings are based on potential adverse environmental effects.
25
-------
TABLE 2. LEVEL 1 SCREENING BIOASSAYS
HEALTH EFFECTS TESTS
TEST
EFFECT
DESCRIPTION
TEST OUTPUTS
cn
Microbial Mutagenesis
(Ames Test)
Cytotoxicity
Mutagenesis
Cellular Toxicity
Genetically sensitive strains of microorganisms
are exposed to various doses of sample with and
without metabolic activation.
Selected cells (RAM, CHO, or WI-38) are exposed
to various doses of sample, then various endpoints
are measured.
Mutagenic response is measured relative to
controls.
An index of functional impairment, toxicity,
and metabolic change is established relative
to controls.
Rodent Acute Toxicity
(RAT Test)
Whole Animal
Toxicity
Rats or other rodents are fed a quantity of sample,
then observed daily for adverse symptoms over a
14-day period. The experiment is terminated with
a necropsy exam.
Inventory of pharmacological and gross
physiological effects in a whole animal
system.
ECOLOGICAL EFFECTS TESTS
TEST
Algal Growth Response
Aquatic Animal Exposure
(Static Acute Bioassay)
Plants (Stress Ethylene
and Root Elongation)
Insect
Bioaccumulation
EFFECT
Algal
Growth Inhibition
or Promotion
Toxicity to
Fish or Daphnia
Stress or Toxicity
to Plants
Toxicity to
Drosophila
Potential
Accumulation
DESCRIPTION
Cultures of selected marine and/or freshwater algae
are used to gauge reaction to sample or dilution
thereof.
Select marine and/or freshwater fish and Daphnia are
exposed to a graded dilution series of samples.
Tests in these three areas are being evaluated.
HPLC procedure for evaluation of occurrences
in fatty tissue.
TEST OUTPUTS
Growth response measure-stimulation
or inhibition.
Gross index of toxic potential to representative
animals.
Effects on plants.
Effects on insects.
Number of components that accumulate.
Accumulation potential of each component.
-------
Multimedia Environmental Goals (MEGs)
MEGs are chemical-specific goals expressed as concentrations in air,
water, and land (or solid waste). Separate values reflect potential human
health effects and potential ecological effects. Two types of MEGs are dis-
tinguished—ambient goals (AMEGs) and discharge goals (DMEGs). AMEGs are
target concentrations of individual chemical species in the ambient environ-
ment to which receptors (i.e., human populations or ecological systems) may be
exposed on a continuous, long-term basis. DMEGs represent target concentra-
tions for contaminants in undiluted waste streams. It is assumed that recep-
tors would be exposed only for short intervals to DMEG concentrations.
Chemicals for which Federal standards or guidelines have already been
established or proposed are assigned MEG values reflecting the most stringent
standards or guidelines. Otherwise, both AMEGs and DMEGs are derived from
available toxicity data. Simple mathematical models based on worst-case
assumptions are used to transform the raw data into the needed concentration
goals for air-, water-, and land-based pollutants. The approach used to gen-
erate MEGs for chemical pollutants is illustrated in Figure 2.
Background information is compiled for each chemical and supplied with
the recommended set of MEG values. MEGs have been established for approxi-
mately 600 chemical substances, and the list is continually updated and
expanded. Chemicals addressed by MEGs are grouped in pollutant categories to
facilitate their use in Level 1 data interpretation (since Level 1 data are
expressed as chemical categories quantified in each LC fraction).
It should be emphasized that the development of MEGs is not related to
Standards setting. MEGs are established as criteria for interpretation of
environmental assessment data, which necessitates ranking a large number and
variety of chemicals, including many nonregulated pollutants.
Source Analysis Model, SAM/IA
To rank the pollution potentials of components within a single stream,
one compares the measured stream concentrations to respective DMEG values. A
difficulty is that DMEGs are species-specific, whereas Level 1 generally
reports only the concentrations of categories of compounds. To circumvent
this problem, the entire concentration of a class of compounds found to be
present is compared to the lowest DMEG for a chemical in that category. This
ratio is called the discharge severity (DS) of the component.
27
-------
Select and
classify
compounds
Determine
if regulated
Air
- Water
- Land
Assemble
Dasic toxicity
data
oo
Assemble
existing
Federal
guidelines
* Previously called EPCs
t Previously called MATEs
Transform
toxicity data
via models into
concentration
goals
Delineate
preferred models
via decision
trees
Calculate
goals—
AMEGs*
DMEGst
I
J
Present guidelines and
calculated goals in format
that allows comparison of
many pollutants
Figure 2. Approach for chemical pollutant MEGs.
-------
n_ _ (component concentration in stream)
Ubi DMEG
If good scientific evidence exists to eliminate the most hazardous species
from consideration, the next most hazardous species is selected, and so on.
In general, components or classes of compounds with discharge severities
greater than unity are considered environmentally significant. Repeating this
procedure for every category of chemicals found in the stream allows the
ranking of these categories on the basis of potential environmental damage.
Discharge severities for all components are summed to give a total discharge
severity (IDS) for the stream.
IDS = ZDS1
In comparing the potential environmental harm of different waste streams
using the DS approach, both the stream compositions and mass flow rates must
be considered. Therefore, a total weighted discharge severity (TWOS) is
defined as the product of the stream mass flow rate and the summation of the
component DS.s in the stream.
TWOS = (stream mass flow rate)(IDS)
Comparison of the TWOS for different streams that are of the same medium
allows comparison and ranking of the streams on the basis of potential environ-
mental insult. Streams with high IDS levels and those that are ranked high
using the TWOS as criteria are candidates for Level 2 sampling and analysis.
Bioassay Data Interpretation
Further indication of the potential environmental harm associated with a
waste stream is supplied by the biological tests. In Level 1 these tests are
short-term bioassays for the detection of acute biological effects. Evalua-
tion of these data is based on the maximum applicable dose for each biological
test; i.e., the maximum amount of a substance which can be administered in a
given bioassay due to experimental limitations. Test results are ranked as
high, moderate, low, or nondetectable biological responses. Table 3 (taken
from Reference 5) gives the response ranges and maximum applicable doses for
several of the Level 1 bioassays. A positive Ames test or toxic responses
from any two other tests suggest a need for Level 2 information. To aid in
the interpretation of the bioassay data , IERL-RTP released a report on data
12
formatting for Level 1 in April 1979.
29
-------
TABLE 3. RESPONSE RANGES FOR RANKING OF VARIOUS BIOTESTS'
00
o
RESPONSE RANGES
ASSAY
Health Tests
Ames
RAM,CHO,WI-38
Rodent
Ecological Tests
Algae
Fish
Invertebrate
ACTIVITY MEASURED
Mutagenesis
Lethality (LC5Q)
Lethality (LD5g)
Growth Inhibition (ECsp,)
Lethality (LC50)
Lethality (LC50)
MAO
5 nig/plate or
500 ML/plate
1,OOOMg/mL or
600ML/mL
10g/kgor
10mL/kg
1,OOOmg/Lor
100%
1,OOOmg/Lor
100%
1,OOOmg/Lor
100%
HIGH
<0.05 mgor
< 10 Mg or
<0.1
<20%or
< 200 mg
<20%or
<200mg
<20%or
<200mg
MODERATE
0.05-0.5 mg or
5-50 ML
10-1 00 MS or
6-60 ML
0.1-1.0
20-7 5% or
200-750 mg
20-75% or
200-750 mg
20-75% or
200-750 mg
LOW
0.5-5 mg or
50-500 ML
100- 1,000 Mgor
60-600 ML
1-10
75-1 00% or
750-1, 000 mg
7 5- 100% or
750-1,000 mg
75-100% or
750-1,000 mg
NOT DETECTABLE
ND at>5 mg or
NDat>500
LC50> 1,000 Mgor
LC5fJ>600ML
LD50>10
EC5g> 1,000 mg
LC5rj>100%or
LC50> 1,000 mg
LC50>100%or
LC5g> 1,000 mg
MAD = Maximum Applicable Dose (Technical Limitations)
LD5Q = Calculated Dosage Expected to Kill 50% of Population
LCso = Calculated Concentration Expected to Kill 50% of Population
ECgg = Calculated Concentration Expected to Produce Effect in 50% of Population
ND = Not Detectable
-------
Streams ranked relatively high in potential adverse health or ecological
effects on the basis of chemical composition do not always exhibit a highly
positive biological response in the Level 1 bioassay battery and vice versa.
This is because the DMEGs may be based on biological responses different from
those measured in the bioassays. Also, possible synergistic and antagonistic
effects occurring in complex mixtures of substances are often characteristic
of waste streams; these effects are not taken into account by the MEG/SAM
approach, which assumes that toxic effects of compounds are additive. There-
fore, chemical tests and biological assays complement each other and should be
run in parallel. The decision to proceed with Level 2 data acquisition should
be made on the basis of all available chemical and bioassay information.
Later this fall, lERL-RTP's Process Measurements Branch will issue a compari-
son of the sensitivities of bioassay tests and chemical analyses.
SAMPLING AND ANALYSIS—LEVEL 2
Level 2 sampling and analysis is dictated whenever Level 1 chemistry or
bioassay indicates a possible hazard. Level 2 inquiries are directed at the
confirmation of Level 1 results and at the identification and quantification
of specific compounds whose presence was inferred from the Level 1 categorical
analysis.
Level 2 generally requires a sampling and analysis scheme specifically
tailored to address questions raised by a Level 1 investigation. The appro-
priateness of a Level 1 sample or sample extract for a more detailed Level 2
study must be carefully evaluated. Was the Level 1 collection efficiency high
enough for the species in question? Is the substance to be analyzed suffi-
ciently stable so as to render still valid the original Level 1 sample? Is
the Level 1 sample truly representative of the source over a reasonable time-
frame? Would an alternative sampling procedure provide a more interference-
free sample? Upon consideration of these and similar concerns, the decision
may be made to return to the test site for a second sampling effort. While
such a Level 2 sampling effort may be expected to provide more rigorous atten-
tion to detail, it generally will not be as extensive as in Level 1 due to the
elimination of certain streams and compound classes from consideration.
31
-------
Level 2 Chemical Analysis
It is not possible or practical to formalize a single effective analyti-
cal scheme for Level 2 since each question to be answered at this stage repre-
sents a unique case. Analytical methods and/or instruments may be used which
are capable of greater selectivity and sensitivity than those employed in
Level 1. Procedures manuals addressing organic and inorganic sampling and
analysis have been issued by IERL-RTP to serve as guidelines for Level 2 data
acquisition.13'14'15
Refinement of the Level 2 chemical methodology continues. A document
prepared by A.D. Little, Inc., on Level 2 Organics Analysis Applications, soon
to be released by IERL-RTP, reports on the validation of Level 2 procedures on
actual samples. Also, IERL-RTP will soon issue a report on interpretation of
3
LRMS data, which is intended as an aid for the spectroscopist.
Level 2 Biological Analysis
In some cases, Level 2 biological tests may be as simple as those in
Level 1. Other cases may require more elaborate and classical methods. A
Level 2 biological test protocol is being developed, which will include sub-
acute and chronic effects and/or fractionation of samples for verification and
quantification of results from the Level 1 screening studies.
Interpretation of Level 2 Results
Level 2 analytical results may be interpreted by several different proto-
cols. The usual method is simply to recalculate for each stream the component
discharge severities (DS^) and the total weighted discharge severity (TWOS)
using the component-specific information now available. Such an iteration may
confirm the Level 1 results or may sufficiently alter the DS and TWOS values
to rank the components or streams of major concern differently.
Because Level 1 data are obtained for rapid screening purposes, no effort
is made to consider the dispersion of the various waste streams into the
ambient environment. At Level 2, such considerations are desirable to better
assess the environmental impact of potentially significant streams. Thus, a
second method for interpreting Level 2 data involves estimation of the ambient
concentration of a chemical, which would result from a particular source
stream, and comparison of that ambient level with the AMEG for the chemical.
32
-------
A Source Analysis Model, SAM/IA, is being developed to relate Level 2 source
test data to AMEGs. This approach represents a degree of refinement above
the comparison involving DMEGs in that AMEGs are based upon continuous recep-
tor exposures to individual chemicals in the ambient environment. DMEGs
represent goals for short-term exposures, and the use of the SAM/IA approach
assumes that human or ecological receptors will come in contact with undiluted
discharge streams.
The component-specific data acquired by Level 2 sampling and analysis and
the interpretation of that data using either of the SAM models thus provide a
reasonable basis upon which to assess the environmental impact of a source.
Discharges unsatisfactory from a health/ecological standpoint are readily
identified so that appropriate pollution control devices may be recommended.
For developing industries, such as synfuels, Level 2 data may be applied
in formulating guidance recommendations for permit writers and developers.
Level 2 data may influence standards-setting for existing industries, or the
data may trigger Level 3 investigations.
EFFECTIVENESS OF THE APPROACH
Assessments of several technologies have been completed using the Level I/
Level 2 methodology. These studies, directed toward the textile industry,
ferroalloy processes, conventional combustion, fluidized bed combustion,
low-Btu gasification, and other technologies, have been performed by different
contractors. The results of the analytical tests, however, may be compared
readily because samples were obtained by similar methods and similar labora-
tory procedures were followed. Also, the analysis data are compared to a
similar basis; i.e., the MEGs. Common formats for reporting of assessment
results have simplified the comparison of results from different sources.
The Level I/Level 2 phased approach to data acquisition has been compared
to the direct approach for environmental assessment of particulate-laden flue
gases. The Level 1 techniques were shown to be effective in narrowing the
scope of the investigation with quantitative Level 2 determinations being
directed toward the samples and components of highest environmental signifi-
cance. It was shown that the cost of the phased approach can be on the order
of 75 to 50 percent of the cost of the direct approach. The thorough
33
-------
screening provided at Level 1 ensures that problem streams or components do
not go undetected.
DATA MANAGEMENT
A data management system is imperative for storing, editing, updating,
and retrieving the vast amount of source test data generated by environmental
assessment projects. To this end, IERL-RTP has developed the Environmental
Assessment Data Systems (EADS) stored in the UNIVAC computer at EPA's Environ-
mental Research Center in North Carolina. The EADS is a comprehensive system
of computerized data bases that describe multimedia discharges from energy
systems and industrial processes. The data bases are interlinked across media
and across industries.
The EADS serves to (1) consolidate the increasing volume of environmental
data, (2) provide uniform data protocols, and (3) maintain current information
in a readily accessible mode. Four media-specific waste stream data bases are
included to address fine particle emissions, gaseous emissions, liquid efflu-
ents, and solid discharges. A fifth data base for multimedia fugitive emis-
sions will be added next year. These data bases are designed to permit entry
and retrieval of information pertinent to specific tests, sources, processes,
control devices, or specific pollutants. Coding forms for data entry are
designed to accommodate results from Level 1 and Level 2 chemical and biologi-
cal analyses.
In addition to the waste stream data bases, there are currently two
important reference data bases within the EADS. These are MEGDAT, which
stores MEG values and supporting information for MEGs pollutants, and the
Chemical Data Table which contains names, synonyms, CAS registry numbers, and
MEG ID numbers for almost 2,000 chemicals. A Quality Assurance/Quality Con-
trol reference data base for laboratory audit data is projected to be in place
in EADS in 1981. An additional reference data base called the Project Profile
System will be linked with the EADS soon. This system presently contains
profile information from conventional combustion projects but is also designed
to manage data from other technology areas.
EADS is expected to provide essential data to several EPA programs,
including:
Environmental Assessment Programs
Inhalable Particulate Standards Development
34
-------
Wastewater Treatability Manual Development
Evaluation of Control Technology Alternatives
Industrial Boiler NSPS
Identification of Hazardous Pollutant Emissions
Radionuclide Correlations with Particle Size
An IERL-RTP directive, dated May 1978, requires that all sampling and
analysis data obtained under IERL-RTP source sampling contracts awarded after
June 30, 1978, be entered in the appropriate EADS data base. User's manuals
for the existing data bases are available, and specific information requests
will be filled by the EADS Manager at IERL-RTP.17
Quality Assurance and Control
Agency policy requires participation by IERL-RTP in a centrally directed
Quality Assurance Program for monitoring and measurement efforts. The Quality
Assurance Plan developed for IERL-RTP fulfills one requirement under the
overall program managed by EPA's Quality Assurance Management Staff, Office of
Monitoring Systems and Technical Support. The plan is expected to become
effective October 1, 1980.18 Provisions in IERL-RTP1s Plan specify that all
measurement and monitoring data collected should be of known and documented
quality. Throughout the sampling and analysis segments of any environmental
assessment, a program of quality control and quality assurance must be
followed to ensure the desired accuracy and precision of results. The quality
of the data must be acceptable for its intended use. Analytical methods and
procedures should conform to EPA approach methodology when appropriate.
Customary requirements of good laboratory practice (including preservation of
samples, standardization of reagents, and calibration of equipment) must be
verified and documented. An independent group working in cooperation with the
laboratory personnel may review the laboratory's methods, engage in on-site
inspections, provide blind samples for analysis, and duplicate the sample
analyses to confirm results obtained by the test laboratory. Audited each
year will be 10 to 20 percent of the projects within IERL-RTP.
CONTINUED DEVELOPMENT TRENDS
The phased environmental assessment approach described here has been
undergoing continual development since its inception in 1976. The various
35
-------
components of the methodology have been and continue to be subjected to criti-
cal review from both inside and outside the Agency. A major peer level review
involving 15 panelists was held in January 1979.19 As a result of such reviews,
on-going research at IERL-RTP, and from user comments, refinements continue in
the sampling/analysis procedures, data reporting formats, MEGs development,
SAM models for data interpretation, nomenclature, bioassays, and mechanisms
for data management.
Areas designated for significant future development include:
1. Although the MEGs methodology makes use of most types of readily
available toxicity data, the models involve many assumptions and extrapo-
lations. Substantial refinements in the MEGs methodology are planned for
Phase II MEGs. Among the modifications will be (a) adoption of the EPA Car-
cinogen Assessment Group approach for relating concentrations of potential
carcinogens to the resulting level of risk in the exposed population; (
b) methods to address accumulation and bioconcentration; (c) category-specific
models for utilizing animal data; (d) better use of inhalation data; and
(e) improved, category-specific models to generate values for solid waste. A
review of the Phase II methodology by the EPA Science Advisory Board is being
scheduled for 1981.
2. Research is being initiated on health and ecological effects for
both individual chemical substances and complex mixtures for which inadequate
data exist to derive MEGs. As results of these tests become available, they
will be incorporated in chemical information summaries and will serve as the
basis for new MEGs values.
3. Efforts are underway to improve models for predicting risks to human
health or to the ecology as a function of exposure to hazardous chemicals.
Such models will be incorporated in MEG as data for their implementation
becomes available.
4. Development of MEGs to account for skin absorption is being con-
sidered.
5. Regional and site-specific models are needed to describe the trans-
port of pollutants from point of discharge to receptors in the ambient environ-
ment. Transformation models are also needed for use in more sophisticated
SAMs.
6. The current environmental assessment methodology does not include
evaluation of water parameters such as hardness, total dissolved solids, BOD,
36
-------
and COD. Because these parameters contribute to the environmental signifi-
cance of waste streams, MEG values are needed.
7. Level 3 sampling and analysis methodologies need to be formulated.
8. Standardization of laboratory procedures and techniques for
interpreting instrumental analysis data (especially LRMS) is essential if data
from different laboratories are to be comparable. Thus, analytical infor-
mation assimilation through IERL-RTP is being emphasized.
Assessing the potential for environmental damage from complex industrial
sources is an awesome and formidable task but one which is necessary for
providing guidance for pollution control needs, control technology development,
health and ecological research, and regulatory/standards-setting activities.
The phased approach to environmental assessment as described in this
report is indeed on the right road to fulfilling its primary purpose, namely,
to identify in a cost-effective manner the environmental problems associated
with industrial processes and fossil energy systems. This methodology is
proving especially valuable in predicting potentially adverse effects from
emerging technologies, such as coal gasification and liquefaction. In such
cases, it is vital to project the likely environmental problems while these
processes are still in the pilot or demonstration-scale stages, so that
appropriate pollution control measures will be available when the processes
are ready for full-scale commercialization.
The IERL-RTP approach to environmental assessment is an iterative and
evolutionary methodology, improving as faults are revealed and as new informa-
tion becomes available. At its present level of development, it provides a
valuable framework and focus for environmental assessments.
37
-------
REFERENCES
1. Hamersma, J. W., S. L. Reynolds, and R. F. Maddalone, IERL-RTP Procedures
Manual: Level 1 Environmental Assessment, TRW Systems Group, Redondo
Beach, CA, EPA-600/2-76-160a NTIS PB 257850 (June 1976).
2. Lentzten, D. E., D. E. Wagoner, E. D. Estes, and W. Gutknecht, IERL-RTP
Procedures Manual: Level 1 Environmental Assessment (Second Edition),
Research Triangle Institute, Research Triangle Park, NC, EPA-600/7-78-201
(NTIS PB293795) (October 1978).
3. Johnson, L. D., Process Measurements Branch, EPA/IERL-RTP, Personal
communication (September 1980).
4. Duke, K. M., M. E. Davis, and A. J. Dennis, IERL-RTP Procedures Manual:
Level 1 Environmental Assessment, Biological Tests for Pilot Studies,
Battelle Columbus Laboratories, Columbus, OH, EPA-600/7-77-043 (NTIS PB
268484) (April 1977).
5. Sexton, N. G., Biological Screening of Complex Samples from Industrial/
Energy Processes, Research Triangle Institute, Research Triangle Park,
NC, EPA-600/8-79-021 (NTIS PB 300459) (August 1979).
6. Merrill, R. G., Process Measurements Branch, EPA/IERL-RTP, Personal
Communication (August 1980).
7. Cleland, J. G., and G. L. Kingsbury, Multimedia Environmental Goals for
Environmental Assessment, Volume I, Research Triangle Institute, Research
Triangle Park, NC, EPA-600/7-77-136a (NTIS PB 276919); Volume II. MEG
Charts and Background Information, EPA-600/7-77-136b (NTIS PB 276920)
(November 1977).
8. Kingsbury. G. L., R. C. Sims, and J. B. White, Multimedia Environmental
Goals for Environmental Assessment: Volume III. MEG Charts and Back-
ground Information Summaries (Categories 1-12), Research Triangle Insti-
tute, Research Triangle Park, NC, EPA-600/7-79-176a (NTIS PB80-115108);
Volume IV. MEG Charts and Background Information Summaries (Categories
13-26), EPA-600/7-79-176b (NTIS PB80-115116) (August 1979).
9. Kingsbury, G. L., J. B. White, and J. S. Watson, Multimedia Environmental
Goals for Environmental Assessment, Volume I (Supplement A), Research
Triangle Institute, Research Triangle Park, NC, EPA-600/7-80-041 (NTIS PB
80-197619) (March 1980).
10. Schalit, L. M. , and K. J. Wolfe, SAM/IA: A Rapid Screening Method for
Environmental Assessment of Fossil Energy Process Effluents, Acurex
Corp., Mountain View, CA, EPA-600/7-78-015 (NTIS PB 276088) (February 1978)
Revision to be released in 1981.
38
-------
REFERENCES (Continued)
11. Bowen, J. S., Combustion Research Branch, EPA/IERL-RTP, Personal communi-
cation (August 1980).
12. Brusick, D. J. , Level 1 Biological Testing Assessment and Data Format-
ting, Litton Bionetics, Inc., Kensington, MD, EPA-600/7-80-079 (NTIS
PB80-184914) (April 1979)
13. Harris, J. C., M. J. Hayes, P. L. Levins, and D. B. Lindsay,
EPA/IERL-RTP Procedures for Level 2 Sampling and Analysis of Organic
Materials, Arthur D. Little, Inc., Cambridge, MA, EPA-600/7-79-033 (NTIS
PB 293800) (February 1979).
14. Beimer, R. G. , H. E. Green, and J. R. Denson, EPA/IERL-RTP Procedures
Manual: Level 2 Sampling and Analysis of Selected Reduced Inorganic
Compounds, TRW Defense and Space Systems Group, Redondo Beach, CA,
EPA-600/2-79-199 (NTIS PB80-149933) (November 1979).
15. Maddalone, R. F., L. E. Ryan, R. G. Delumyea, and J. A. Wilson, EPA/
IERL-RTP Procedures Manual: Level 2 Sampling and Analysis of Oxidized
Inorganic Compounds, TRW Defense and Space Systems Group, Redondo Beach,
CA, EPA-600/2-79-200 (NTIS PB80-200413) (November 1979).
16. Briden, F. E. , J. A. Dorsey, and L. D. Johnson, A Comprehensive Scheme
for Multimedia Environmental Assessment of Emerging Energy Technologies,
Presented at the 10th Annual Symposium of the Analytical Chemistry of
Pollutants, Dortmund, Germany (May 28, 1980).
17. Johnson, G. L., Manager EADS, Special Studies Staff, EPA/IERL-RTP, Personal
communication (September 1980).
18. Kuykendal, W. B., Quality Assurance Officer, EPA/IERL-RTP, Personal communi
cation (August 1980).
19. Environmental Assessment Methodology Workshop, Sponsored by the Environ-
mental Protection Agency, Office of Energy, Minerals and Industry, Indus-
trial Environmental Research Laboratory, Research Triangle Park, and
Industrial Environmental Laboratory, Cincinnati. Airlie House, VA,
(January 16-18, 1979).
39
-------
THE PERMITTING PROCESS FOR
NEW SYNFUELS FACILITIES
Terry L. Thoem
Director, Energy Policy Coordination Office
S. Environmental Protection Agency Region VIII
ABSTRACT
The Environmental Protection Agency and the respective State
Departments of Health are involved in a joint partnership with
shared responsibilities for protecting the environment during the
development of synthetic fuels. Legislation in the form of the
Clean Air Act, Clean Water Act, Resource Conservation and Recovery
Act, Safe Drinking Water Act, and the Toxic Substances Control
Act provide the framework for EPA's regulatory responsibilities.
The current status of implementing regulations and agency policies
vis-a-vis these Acts is provided in this paper. Also, important
aspects of State environmental regulations are provided.
Permit applications for synthetic fuels facilities are being
received by EPA Regional Offices and by State agencies. Synfuels
EISs are being reviewed. Decisions on Best Available Control
Technology are being made. These engineering judgements are also
discussed in this paper.
40
-------
THE PERMITTING PROCESS FOR
NEW SYNFUELS FACILITIES
I. INTRODUCTION
EPA has legislative mandates to protect air and water
quality, to insure a safe drinking water supply, and to provide
for an environment conducive for the enjoyment of man on this
earth. In order to accomplish these goals, EPA is involved in
a partnership with State and local agencies in the formulation
and enforcement of regulations which implement the legislative
intent. A major component of* the regulatory process is the
requirement for industrial operations such as synthetic fuels
facilities to obtain a permit for the project. This paper
discusses the EPA permit mechanism and its framework (Table 1).
II. LEGISLATION
The general process of legislation/regulations is that the
U.S. Congress establishes environmental legislation that provides
a framework for State legislation and implementation of Federal
and State regulations. State legislation and regulations can
be more (but not less) stringent than Federal requirements if
a State is delegated responsibility for administering the
program in a given media. The Federal government retains an
oversight/reviewing role for those programs that are delegated
to the States. State legislation in general parallels Federal
legislation in form and substance. The following discussion
highlights the major aspects of the legislative mandates of EPA
as it applies to a synthetic fuels industry.
Clean Air Act
Under the Clean Air Act (PL 95-95) synthetic fuel facilities
must: (a) employ Best Available Control Technology (BACT),
(b) insure that National Ambient Air Quality Standards (NAAQS)
(Table 2) are not violated, (c) not violate the prevention of
significant deterioration (PSD) ambient air quality increments
(Table 3) (40 CFR 52.21), (d) not significantly degrade visi-
bility in mandatory Class I areas (40 CFR 51), and (e) perhaps
obtain up to 1 year of baseline data before applying for a PSD
permit to construct and operate. BACT has been defined in the
form of allowable emissions limits and control device opera-
tional characteristics. Source monitoring, ambient monitoring,
record keeping and reporting requirements are also part of the
PSD permit. (40 CFR Part 60.7) Also EPA has the ability to
request monitoring data, to take enforcement actions, and to take
administrative and judicial actions if there are any emergency
episodes of pollutants that present an imminent and substantial
endangerment to public health.
41
-------
liable 1
Synfuels Permits
Permit Title Jurisdiction
1. Environmental Impact Statement (EIS) Federal
2. Resource Recovery and Conservation - Federal
definition and control
3. Toxic Substances-definition and control Federal
4. National Pollutant Discharge Elimination Federal
Systems (NPDES)
5. Prevention of Significant Air Quality Federal
Deterioration
6. Soil Prevention Control and Counter- Federal
measure (SPCC) *
7. Well Operation Permit(underground Federal
Injection)
8. Erection of Towers or Other Tall Federal
Structures
9. River and Stream Crossing Permit Federal
10. Major Fuel Burning Installation Approval Federal
11. Rights of Way Across Public Lands Federal
12. Scientific, Pre-Historic and Federal
Archeological
13. Sundry Notices and Reports on Wells Federal
14. Oil Shale Mineral Rights Lease Federal
15. Detailed Development Plan Federal
16. Collection of Environmental Data and Federal
Monitoring Plan
17. Exploration and Mining Plans Federal
18. Mine Safety and Health Federal
definition and control
19. Notice of Intent to Prospect State
20. Permits for Special Operators State
21. Permit for Limited Impact Operations State
22. Permit for Regular Mining Operations State
23. Storage of Flammable Liquids State
24. Application for Diesel Permit - State
Underground Operations
25. Operator's Notice of Activitiy State
26. Hoistman Certificate State
27. Application to Store, Transport State
and Use Explosives
28. Reservoir Construction State
29. Water Well and Pump Installation State
(requirements)
30. Air Contaminant Emission Notices State
42
-------
Table 1 (continued)
Permit Title Jurisdiction
31. Land Use Special Permit State
32. Air Contaminant Emission Permit State
33. Fugitive Dust Permit State
34. Open Burning Permit State
35. Subsurface Disposal Permit State
36. Discharge Permit State
37. Waste Disposal Plant Operator Certificate State
38. Potable Water Supply and Safety Compliance State
39. Sewage Plant Site Approval and State
Plant Approval
40. Purchase, Transportation and Storage State
of Explosives
41. Oil Facility Inspection State
42. Boiler Inspection Permit State
43. Oil Shale Leases State
44. Ground Water Well Application State
45. Application for Water Rights State
46. Mined Land Reclamation State
47- Permit for Exploration and Excavation State
48. Open Burning State
49. Fuel Burning-Sulfur Content Exemption State
50. Permit to Construct Facilities that are State
Sources of Air Pollution
51. Permit to Construct and Operate Treatment State
Works
52. Water Quality-Definition and Control State
53. Permit to Operate Solid Waste Disposal State
Site
54. Notice of Intention to Operate or State
Suspend Operations
55. Hoistman-Qualifications State
56. Escape and Evacuation Plans State
57. Boiler and Pressure Vessel- definition State
and control
58. Storage of Explosives State
59. Construction of Wastewater Ponds and State
Holding Facilities
60. Construction of Sewage Facility State
61. Subsurface Discharges State
62. Mining Permit, Mining and Reclamation Plan State
63. Notification of Mining Operations(control) State
64. Discharges-In Situ Mining State
65. Construction and Operating Permit for State
New or Modification to Existing Facility
43
-------
Table 1 (continued)
Permit Title
66. Open Burning Permit
67. Permit to Dispose of Hazardous Wastes
68. Approval for Construction and Operation
of Waste Facility
69. Construction and Operating Permit for
New or Modification to Existing Facility
70. Exploration Permit, License to Explore
71. Industrial Zone Change
72. Conditional Permit
73. Mineral Extraction
74. Rights-of-Way Approvals
75. Solid Waste Disposal
76. Rezoning Permit
77. Temporary Use Permit
78. Conditional Use Permit
79. Building Permit
80. Special Use Permit
81. Sewage Disposal
82. Solid Waste Disposal Permit
83. Conditional Use Permit
84. Sewage Disposal System
85. Installation of Utilities in Public
Right-of-Ways
86. Driveway Permit Across County Roads
87- Recreation Forest and Mining Zone
(RF&M)-definition and control
88. Mining and Grazing Zone (M&G-l)
definition and control
89. County Requirements in Addition to the
Mining and Grazing (M&G-l) and
Recreation Forest and Mining (RF&M)Zoning
Requirements
Jurisdiction
State
State
State
State
State
County
County
County
County
County
County
County
County
County
County
County
County
County
County
County
County
County
County
County
44
-------
TABLE
NATIONAL AMBIENT AIR QUALITY STANDARDS, UG/M"
***
Pollutant
so2
Particulate matter
N0x(as N02)
°3
CO
Lead
HC (non CH.)
Averaging
tine
Annual
24 hour
3 hour
Annual
24 hour
Annual
1 hour
8 hour
1 hour
Quarterly
3 hour
Primary
standard
80
365
75
260
100
240
10,000
40,000
1.5
160***
Secondary
standard
1,300
60
150
100
240
10,000
40,000
1.5
160***
* 40 CFR Part 50
** Reference conditions = 760 mm Hg and 25 C
*** Not a standard; a guide to show achievement of the 03 standard
45
-------
TABLE 3 PREVENTION OF SIGNIFICANT DETERIORATION OF
AIR QUALITY (PSD) STANDARDS*
Maximum Allowable Increase,
Pollutant
Particulate matter
so2
Averaging
time
Annual
24 hour
Annual
24 hour
3 hour
Class I
5
10
2
5
25
Class II
19
37
20
91
512
mg/m
Class III
37
75
40
182
700
* 40 CFR 52.21 and 42 USC 7401 et sec section 163.
Notes:
1. Variances to the Class I increments are allowed under certain
conditions as specified at Section 165(d)(c)(ii) and (ill) and
at 165(d)(D)(i) of the Clean Air Act of 1977.
2. EPA was to have promulgated si mi lax increments for HC, CO, O, and
NO by August 7, 1979; they are under development. Increments
A
for Pb are due to be promulgated by October 5, 1930.
46
-------
Clean Water Act
The Clean Water Act (PL 95-2]7) established goals of
(a) no discharge of pollutants into navigable streams by
1985, (b) attainment by July 1, 1983, of water quality suit-
able for protection and propagation of fish, shellfish, and
wildlife and provides for recreational use, and (c) prohibition
of discharges of toxic amounts of toxic pollutants. The Act
contains requirements in sections 402 and 404 for potential
permits for synthetic fuel facilities. A National Pollutant
Discharge Elimination System (NPDES) permit must be obtained
under requirements of Section 402 if water is discharged to a
navigable stream (defined as waters of the United States and in
fact could be a dry creek bed which flows during runoff).
Neither effluent guidelines (Section 304) nor New Source
Performance Standards (Section 306) have been promulgated for
any synthetic fuels operations. However, in their absence,
NPDES effluent limits are established on a best engineering
basis. A Section 404 permit must be issued by the Army Corps
of Engineers and concurred upon by EPA if any dredge and fill
operations take place in a navigable stream (defined for 404
purposes as stream flow greater than 3 cfs). Section 303 of
the Act provides the mechanism for establishing water quality
stream standards. Plans developed by State Water Pollution
Control Agencies must define water courses within the State
as either effluent-limited or water-quality-limited. Best
management practices (BMP's) to control nonpoint source runoff
may be defined via section 208 and 304(e) of the Act.
Safe Water Drinking Water Act
Underground injectioncontrol (UIC) regulations proposed
on April 20, 1979 (Title 40 of the Code of Federal Regulations
(CFR), Part 126)were promulgated in the May 19 and June 24, 1980
Federal Register. These regulations will govern the injection
or reinjection of any fluids. Permits (40 CFR 122.36) will be
required for in situ operations and for mine dewatering reinjec-
tion. Various States require reinjection permits under existing
regulations. The basic thrust of the UIC program is to require
containment of reinjected fluids. Monitoring (40 CFR 146.34)
and mitigation measures (40 CFR 122.42) to prevent the endanger-
ment of the groundwater system are requirements under these UIC
regulations.
Resource Coi^servajtj._on aiid_R-ec.ov_er-y__AjL£_
The Resource Conservation and Recovery Act (RCRA) will
govern the disposal of solid and hazardous wastes generated by
a synthetic fuel facility. Criteria for the identification of
hazardous wastes were proposed by EPA on December 18, 1978 at
47
-------
40 CFR, Part 250. Final regulations were promulgated in the
May 19, 1980, Federal Register at 40 CFR 261-265. It appears
that some high volume-low risk materials will not be considered
a hazardous waste. Instead, it will be subject to requirements
at 40 CFR 257 (September 13, 1979, Federal Register). A concept
of Best Engineering Judgement will govern the disposal of
hazardous wastes such as API separator sludge.
Testing of effects, record keeping, reporting, and
conditions for the manufacture and handling of toxic substances
are being defined under the auspices of the Toxic Substances
Control Act (TSCA) of 1976. An inventory of all commercially-
produced chemical compounds is now being compiled and was
published in May 1979. If a substance is placed on the
inventory, it is "grandfathered" from the TSCA pre-market
notification requirements. Ten synthetic fuels were identified
on this list of 43,000 compounds. However, these ten are
being reviewed to determine the validity of their being placed
on the list. Being on the list does not "protect" a product
from possible control requirements included in Section 8.
If a material is found to be a hazard, certain restrictions
including labeling, precautionary handling requirements or
even a ban on its production may be imposed by EPA.
The final piece of environmental legislation in which
EPA participates which is relevant to synthetic fuels is the
National Environmental Policy Act (NEPA). EPA reviews, and
in limited cases writes, the EIS when a project involves a
major Federal action. EPA's role as a reviewer is to comment
on the environmental aspects of the project.
EPA's legislation as described above normally provides a
permit process mechanism. Companies wishing to construct and
operate a synthetic fuel facility must receive a permit from
EPA or from the State permitting authority in order for the
facility to be operated. A listing of the major permits/
clearances necessary for a project appears in Table 1.
APPLICABLE FEDERAL AND STATE POLLUTION CONTROL REGULATIONS
Federal and State legislation generally prescribes the
establishment of National and State environmental standards
for a given media (i.e. air, water, solid waste, etc.).
Regulations designed to control emissions/effluents from an
individual facility are promulgated to achieve the stated
environmental standards. This section briefly describes this
concept of standards/regulations. In almost all cases, the
standards/regulations concept requires a developer to obtain
a permit to construct and operate his facility. It is the
intent of EPA to delegate the permit programs to the State.
48
-------
Air
Regulations to protect air quality exist in two forms-
ambient air quality standards and stack emission standards.
All EPA regulations are codified in Title 40 of the Code of
Federal Regulations. Applicable parts are referred to in
discussions of the various regulations below. Pursuant to
Section 109 of the Clean Air Act, EPA has established National
Ambient Air Quality Standards (NAAQS) for seven criteria
pollutants (40 CFR Oart 50). Primary standards are designed
to protect public welfare (vegetation, materials corrosion,
aesthetics, etc.). States may also establish ambient air
quality standards.
The Clean Air Act also established the concept of preven-
tion of significant deterioration (PSD) of air quality designed
to protect clean air areas (40 CFR Part 52.21). Class I areas
include national parks larger than 2,428 ha(6,000 acres),
national wilderness areas greater than 2,023 ha(5,000 acres),
and international parks, and national memorial parks that
exceed 2,023 ha (5,000 acres). Areas in the United States
that presently have lower ambient air quality than that specified
in the NAAQS are designated as nonattainment areas; the remainder
of the United States is designated Class II. Redesignation of
Class II areas to either Class I or Class III by the state is
possible. Recent court rulings have resulted in some major
changes in the PSD regulations which appear in the August 7,
1980 Federal Register.
A second ambient air quality consideration is the visi-
bility protection afforded to Federal Mandatory Class I areas
via Section 169A of the Clean Air Act (40 CFR, Part 51).
Regulations are to be promulgated by EPA (November 1980) and
the States (August 1981) that are designed to prevent visibility
impairment in the Federal Mandatory Class I areas. Since there
are many issues to be resolved, it is too early to delineate
the potential implications of the visibility regulations.
Proposed regulations appeared in the May 22, 1980, Federal
Register at 40 CFR 51.300. An EPA Report to Congress on
visibility was published in November 1979.
Limitations on the amounts of pollutants emitted from a
synthetic fuel facility are the enforceable mechanism to
assure that the NAAQS and PSD increments are not violated.
EPA establishes New Source Performance Standards (NSPS) 40
CFR Part 60), States establish emission standards, and EPA
(or the State) must define emission limits that reflect the
BACT. NSPS have not been defined for synfuels facilities, but
P>ACT has been defined for five oil shale facilities and one coal
gasification via the PSD permit process.
49
-------
Water
Water pollution control requirements exist in the form
of Water Quality Criteria, State Water Quality Standards,
Drinking Water Standards, National Pollutant NPDES limits,
and effluent guidelines. The following discussion summarizes
the major aspects of surface water and groundwater quality
standards; a complete discussion of the enforceable mechanism
to attain these standards, that is the NPDES and UIC permit
systems, may be found in other EPA references. (1)
Surface Water Quality Standards
Water quality standards are addressed in Section 303
(Water Quality Standards and Implementation Plans) of the
Clean Water Act. Excerpts and summaries of requirements
for establishment and implementation of water quality standards
of that section are presented below:
Water quality standards shall be reviewed at least every
3 years by the Governor or State Water Pollution Control
Agency and shall be made available to the Administrator.
State revised or adopted new standards shall be submitted
to the Administrator (EPA) for approval. Such revised or new
water quality standards shall consist of the designated uses
of the navigable waters involved and the water quality criteria
for such waters based upon such uses. Such standards shall
be such as to protect the public health or welfare, enhance
the quality of water, and serve the purposes of the Act(FWPCA).
Such standards shall be established, taking into consideration
their existing or intended potential use and value for public
water supplies, propagation of fish and wildlife, recreational
purposes, agricultural, industrial, and other purposes, while
also taking into consideration their use and value for navigation.
Each State shall identify those waters for which existing
or proposed effluent limitations are not stringent enough to
attain established water quality standards and establish waste
load allocations for those waters. Regulations promulgated at
40 CFR 131.11 and further discussed in the December 28, 1978
Federal Register describe the Total Maximum Daily Load concept.
Each State shall identify those waters or parts thereof
within its boundaries for which controls on thermal discharges
are not sufficiently stringent to assure protection and propa-
gation of a balanced indigenous population of shellfish, fish,
and wildlife.
(1) Environmental Perspective on the Emerging Oil Shale Industry,
November, 1980.
50
-------
The 208 Process
Section 208 of the FWPCA required States to designate
areawide waste treatment planning agencies. These 208 agencies
are to plan, promulgate, and implement a program designed to
protect surface water quality. Stream classifications and
water quality standards are to be developed.
Local input in most States on the proposed stream use
indicated a desire to assign multiple classification systems
wherever possible. Although the apparent intent of State
classification systems (1978) is simply to identify the criteria
applicable to a given stream segment, there is considerable
local concern that a single "use" classification may be used
later to restrict other uses, particularly agricultural ones.
Intermittent streams have not been classified because of
provisions made for this situation in the proposed classifica-
tion system.
As an example, the four combinations of multiple use class-
ifications that are proposed for Colorado include:
Class 1: Aquatic Life. Water Supply, Recreation, and
Agriculture
Class 2: Water Supply, Recreation, and Agriculture
Class 3: Recreation and Agriculture
Class 4: Agriculture
The proposed water quality standards allow exceptions under
certain conditions. Using the guidelines in the proposed
criteria, the water quality data base, the proposed water
quality criteria, the existing water quality problems, and a
subjective analysis of potential effectiveness of potential
control measures, three types of exceptions were identified for
Colorado:
o Permanent exception - The current criterion limit is
not valid for the drainage area because of natural
environmental conditions. It is assumed that, given
a return to prehistoric conditions, this parameter
would still violate the criterion limit. The parameter
should be monitored regularly, and any trend of increas-
ing concentration would require evaluation/investigation
of possible causes beyond natural conditions. It is
further assumed that it is uneconomical to attempt
controlling runoff.
o Temporary exception (10 Years) - This exception is
requested when a criterion violation is identified as
a possible consequence of man's activities in the basin
and management strategies are available to improve
51
-------
water quality, but it will take 19 years to evaluate
effectiveness.
o Temporary Exception (5 Years) - This exception is
requested when a limited data base indicates a problem
but more data are required to identify the cause,
extent, and correctability of the problem. The
5-year exception should allow sufficient time for
necessary additional data collection and analysis.
Ground Water Quality_Standj.r_d_s_
Federal - Federal regulations that may pertain to
groundwaters are addressed in the Safe Drinking Water Act.
This act has most recently been interpreted as applying to
well injection of waste into aquifers that do or that might
serve as sources for public drinking water. Such underground
drinking water sources, while specified to include aquifers
with less than 10,000 mg/1 total dissolved solids, must have
the potential to be sources of public water supply. Underground
injection control (UIC) regulations were promulgated at 40 CFR
126 on May 19, 1980. In situ operations will fall into the
category of "Class III wells". Drinking water standards are
listed in Tables 4 and 5. Note that pits, ponds, and lagoons
are not identified as underground injection sources at this
time. They are covered under the RCRA.
Solid and Hazardous Wastes
The RCRA requires that solid arid hazardous waste generators
and transporters receive permits and that wastes be disposed only
by safe practices. Regulations have been promulgated at 40 CFR
Part 261 for (1) the criteria to identify solid and hazardous
wastes (Section 3001); (2) disposal standards (Section 3004); and
(3) permit programs (Section 3005). If a waste is not defined
as hazardous (I.e., it is defined only as a solid waste) disposal
It 40 ™STr?e?^7 the SeCti°n 4004 regulations as promulgated
at 40 CFR Part 257 on September 18, 1979. The promulated
ar on September 18, 1979. The promulgated
regulations defined a waste as hazardous if it is ignitable
»;
-
ing are exempt ' materials ready for further process-
as
API ^ regulations probably will result in materials such as
t"n tank\ ^ 8^ ^^ C3talysts' gasifler ash, distilla-
tion tank bottoms and perhaps others being defined as a hazardous
W ct o u G •
52
-------
TABLE 4 PROMULGATED DRINKING WATEK STANDARDS (40 CFR 141)
Tha following are the maximum contaminant levels for Inorganic chemical* other than fluoridet
Contaminant Level, mg/1
Arsenic O.OS
Barium 1.
Cadmiun 0.010
Chrcnium O.OS
Lead O.OS
Mercury 0.002
Nitrate (as N) 10.
Selenium 0.01
Silver O.OS
When the average of the maximum daily air temperatures for the location in which the community water systea is
situated is the following, the maxiziun contaminant levels for fluoride aret
Temperature, °F °C Level, ng/1
53.7 and below 12.0 and below 2.4
53.8 to 53.3 12.1 to 14.6 2.2
58.4 to 63.8 14.7 to 17.6 2.0
t>3.9 to 70.6 17.7 to 21.4 1.8
70.7 to 79.2 21.5 to 26.2 1.6
79.3 to 90.5 26.3 to 32.5 1.4
The following are the maximum contaminant levels for organic chemicals. They apply only to community water syste
Compliance with maximum contaminant levels for organic chemicals is calculated pursuant to Section 141.24.
Level, mg/1
a. Chlorinated hydrocarbons]
Endrin (1,2,3,4,10, 10-hexachloro-6.7-epoxy- 0.00002
l,4,4a,5,6,7,3,8a-octahydro-l.4-endo-5,8-
dinenthano naphthalene).
Lindane (1,2,3,4.5,6-hexachlorocyclohexane, 0.004
gacmia isomer) .
Methoxychlor (l,l,l-Trichloro-2,2-bis (p-mthoxyphenyl) 0.1
ethane).
Toxaphene (ClnHlnCl -Technical chlorinate'* c.iaphenn. 0.005
67-69 percent: chlorine).
b. Chlorophenoxysi
2,4-D, (2,4-Dichlorophenoxyacetic acid). 0.1
2,4,5-TT Silvex (2,4,5-Trichlorophenoxypropionic acid). 0.01
53
-------
TABLS 5 LEVELS OF CONTROL APPLICABLE TO EXISTING SOURCES UNDER 1977 AMENDMENTS TO FWPCA
Pollutant
Naao
Abbreviation
Statutory
Deadline
301 (c) Kconnmlo
Variance
301 (ijl Environmental
Variance
"July 1, 1984, or tht.«
bjuly 1, 1984 for thoie
Conventional
Beat Conventional Pollutant
Control Technology
DCT
July 1, 19S4
110
Ho
Nonconventlonal
Boat Available Technology
Economically Achievable
OAT
July 1, 1904/aa appro-
priate. Novor later than
July 1, 1907
Ye*
Ye*
Boet Avallablo Technology
licononlcally Achievable
DAT
July 1, 19Q
-------
IV. PROPOSED PRECOMMERCIAL APPROACH TO INDUSTRY STANDARDS
The approach regulating the first synfuels facilities
must ensure compliance with existing standards, but, more
important, should emphasize characterization of residuals
from the facility. EPA Region VIII has expressed their desire
to see a synfuels industry proceed in a phased orderly manner.
Rigorous testing programs and data analyses should be performed
on the first facilities, which would be representative of
commercial size. Comprehensive monitoring of emissions, effluents,
and waste materials should be performed. Research programs
designed to define the optimum control technology for a given
pollutant for a synfuels industry should be conducted. Trade-
offs among air pollution, water, pollution, and solid waste
must be defined. The energy penalty, water consumption, and
cost of control must be defined. The comprehensive monitoring
efforts should not be limited to only the regulated pollutants,
but should characterize nonregulated pollutants.
As previously stated, emphasis should be placed on source
characterization. A moderate degree of ambient impact monitor-
ing should be performed to validate predicted impacts-and to
document trends and changes from baseline. Programs to
evaluate effects on receptors should be performed to provide
feedback on the source and ambient monitoring programs. There
are two principal bases for writing permits for synfuels
facilities. The first relies upon the transfer of pollution
control technology from related industries. The second relies
upon the development of EPA's Pollution Control Guidance
Documents.
The BACT for air pollutants must be employed for any
proposed synfuels facility with the potential for emitting 91
tonnes (100 tons) or more (controlled) per year of any regulated
air pollutant. Those facilities that have smaller potential
emissions do not need BACT but should perform comprehensive
monitoring in order to develop emissions data for potential
permit applications. Two primary mechanisms exist to define the
BACT. First, several synfuels facilities have received Preven-
tion of Significant Deterioration (PSD) permits. The BACT has
been defined on a case-by-case basis for these facilities.
Second, air pollution control technology that has been defined
as the BACT for synfuels related facilities may be considered
as transferable to the industry. It is highly likely that air
quality requirements may prove to be the governing constraing
to the size of synfuels industry in certain parts of the country.
Therefore, in order to maximize the amount of oil production
capability of oil shale country it is important to maximize the
air emissions control for each facility.
55
-------
A no-discharge-of-pollutant concept is being considered
by several developers as a means of handling their wastewater
streams. Three types of water should be considered--mine,
process, and in situ water. A no-discharge-of-process-water
concept has been written into water permits. If any water is
discharged to surface streams or reinjected into the ground-
water system, it would consist of mine inflow (but not process
or in situ water) or uncontaminated surface runoff. Treatment
may or may not be necessary. Effluent limitations will be
defined for certain pollutants including toxics for certain
process streams in the NPDES permit. Best available tech-
nology economically available (BATEA) must be provided. (See
Table 6). Major concepts to be addressed by regulatory agencies
and the developer are summarized as follows. First, because of
the semi-arid, water-short condition of potential development
areas, it may be environmentally best to encourage treatment
if necessary and discharge to a surface stream of mine water.
Second, because of salinity considerations, treatment of mine
water and/or minimization of water consumption is a desirable
policy. Third, disposal of process water onto processed
shale piles or ash piles without treatment may not be desirable.
The high organic and salt concentration of the process water may
represent too great a risk to groundwater/surface water quality
because of potential catastrophic events or unexpected
permeabilities/leaching., and they represent a deterrent to
successful revegetat ion. Fourth, maximum recycling and reuse of
process and nonprocess water will be encouraged; cost effecti-
veness must be considered. Finally, land application of
untreated mine water may be desirable only for a short period of
time because of the potential nonppint source runoff problems.
Solid and hazardous wastes should be disposed of in a
manner that avoids contact with water and subsequent toxic
concentrations. Disposal practices should also be designed that
preclude (or at least minimize) the potential for the solid
material from becoming airborne as a fugitive dust. Safe
disposal practices as defined at 40 CFR 264 apply to synfuels
facility hazardous wastes such as spent catalyst, API separator
sludge, tank bottoms, cooling tower sludge, and water treatment
plant sludge. Surface disposal for solid wastes from a synfuels
industry at a minimum should conform to those practices found in
40 CFR 257.
Pollution Control Guidance Documents
Regulating new, presently non-existent energy industries,
of course, presents different problems from regulating long-
standing segments of United States industry. The differences
are of such an extent that a unique regulatory approach is
demanded. The differences arise primarily from the facts that
the new energy industries are, for the most part, not yet
56
-------
TABLE 6. NEW SOURCE PERFORMANCE STANDARDS
FOR SYNFUELS RELATED ACTIVITIES
40 CFR 60.40 Subpart D (NSPS for Fossil Fuel Fired Steam Generators)
TS? 0.10 pound per million BTU
SO 0.30 pound per million BTU (liquid fuel)
HO 0.20 pound per million BTU (gaseous fuel)
0.30 pound per million BTU (liquid fuel)
40 CFR 60.100 Subpart (NSFS for petroleum refineries)
H S 0.10 grain/dscf
HC Floating roof or vapor recovery if true vapor pressure is >1.S psia
but < 11.1 psia reporting requirements only if true vapor pressure
is < 1.5 psia.
40 CFR 60 (NSPS for Refinery Claus Sulfur Recovery Plants)
Gaseous fuel burning 0.1 grain/dscf
Sulfur recovery
oxidation system 250 ppm SO-
reduction system 300 ppm total S
10 ppm H2S
Proposed NSPS
1. Gas Turbines >10 x 10 BTU/hour
75 ppmv NO at 15% O-
150 ppmv S0_
2. Coal Gasification (Guideline)
250 ppmv total S
99.0 percent total S removal
100 EP«W HC
3. Field gas processing units
Gaseous fuel burning 160 ppmv H.S
Sulfur recovery 250 ppmv SO. (oxidation)
300 ppmv S freduction)
-------
commercialized in the United States and have potentially
different effluents and emissions from those from existing
pollution sources.
There is , unfortunately, little or no existing source
of commercial-scale data on which to base a "conventional"
regulatory approach at this time. In some instances standards
from related industries may be borrowed. (See Table 6)
Because of these circumstances, the general approach we are
taking is to issue, as pre-regulatory guidance, a series of
Pollution Control Guidance Documents, PCGD's -- one for each
of the major energy technologies. The focal point of each
PCGD is to be a set of recommendations on available control
alternatives for each environmental discharge along with
associated performance expectations. The basis for these
recommendations will be presented. The intent is to present
guidance for plants of typical size and for each significantly
different feedstock likely to be used. PGGD's will not have
the legally binding authority of regulations but each will be
reviewed extensively both within and outside of EPA. These
documents will provide useful and realistic guidance to permit
writers within EPA and the States and to the energy industry
itself during its formative stages. As the energy industry
develops, permits for individual installations are being issued
based on best engineering judgment and, as the various PCGD's
become available, permits will be prepared in light of the
information the PCGD's contain. Then, as the energy technolo-
gies mature, EPA will invoke its normal regulatory procedures:
in the water quality area, for example, the issuance of effluent
guidelines and establishment of appropriate water quality
standards .
It is clear that for most new energy technologies,
exemplary full-scale and even pilot-scale waste treatment
installations do not yet exist. Moreover, there is a unique
chance not available to actually influence, in an environ-
mentally productive way, the choice by industry of the very
process technology to be commercialized and the overall designs
of new plants such that the most cost-effective environmental
protection methods can be incorporated into process design from
the very beginning so>that more expensive pollution control
retrofitting is minimized or eliminated. The Pollution Control
Guidance Documents, therefore, have two key purposes: (1) to
aid permit writers in preparing realistic, comprehensive permits
for the energy industry by describing and characterizing
projected waste discharges from the various energy technologies
under development and by providing the best possible information
on the expected cost and performance of the variety of control
options that appear applicable and (2) to provide guidance
to the energy industry itself with regard to the kinds of
environmental impacts with which EPA will be concerned for
their particular kind of facility, the control options which
EPA has deemed to be potentially applicable and EPA's projections
of probable cost and performance of the various options.
58
-------
Let me now elaborate on the general structure of PCGD's.
The Document will consist of three Volumes. Volume I is a
summary report including recommended pollution control tech-
nology options and related costs; Volume II is a detailed
report describing pollutants, waste streams and alternative
control options, including cost and performance; Volume III
is an appendix providing the data base for stream and pollutant
characterization and control costs and performance.
The major users of the PCGD's are expected to be the permit
writers. The Document for a particular energy technology should
help them to better understand permit applications and to
prepare a proper permit. Best available control technology will
be suggested but information on alternative control methods
will also be provided for use in considering site-specific
situations. For example, a permit writer may be faced with
having a very small allowable incremental increase in an air
pollutant, say sulfur dioxide, when conducting a Prevention
of Significant Deterioration (PSD) review. The PCGD will, hope-
fully, let him consider alternatives that achieve stringent
control but will also indicate what the cost of such a level of
protection would be.
The Documents will also serve as a beginning for future
data base developers and regulation writers. When the industry
becomes commercialized, the EPA program offices responsible
for preparing regulations will need to collect commercial-scale
data as the basis for authoritative regulations. The data base
in the PCGD's should serve as a guide to identifying needs,
organizing and carrying out these future data collection efforts.
For the developers, the PCGD's should influence the choices
they have to make on control options and even on certain
process alternatives. If industry and the other Federal and
State agencies which directly support energy development are
aware of anticipated environmental problems and available
control technologies, their development and plant design efforts
can incorporate features which will help to avoid the necessity
for future retrofitting of control technology.
It shoud be noted that providing an early indication of
EPA's concerns for various pollutants and options on pollution
limits will not just produce "passive reactions". On whatever
information EPA provided, it will receive feedback and criticism.
By precipitating this feedback process while the energy
technologies are still being developed, many issues regarding
environmental protection should be resolved prior to construction
and operation. The advance notice of EPA's thinking will permit
regulators, developers and other segments of the public to work
to a greater degree than has been possible in the past
59
-------
and should result in the development and commercialization of
an environmentally sound energy industry.
The specific energy technologies for which separate
PCGD's are now planned are the following:
o Low Btu Coal Gasification
o Indirect Coal Liquefaction
o Oil Shale (mining and milling)
o Direct Coal Liquefaction
o Geothermal (first revision of existing PCGD)
o Medium Btu Coal Gasification
o High Btu Coal Gasification
Table 8 provides the schedule for their development.
EPA has taken specific measures to assure that the devel-
opment of regulatory approaches for the energy industries
will involve a wide range of interested parties, both in the
preparation of PCGD's and in their review. These parties include
government, industry, environmentalists and the public in
general. Within EPA, we have established an Alternate Fuels
group which has the responsibility for coordinating all research
and all regulation development — on a multi-media basis — for
new energy technologies. Serving on this group are represent-
tatives from all of the major policy/program and research
offices charged with related research and regulation develop-
ment and from some of the Regional Offices which are most
concerned with synfuels commercialization. The Group's overall
responsibility is to develop the EPA regulatory approach for
the new energy technologies. Within this context the Alternate
Fuels Group is charged with producing Pollution Control Guidance
Documents, overseeing the creation of a program to insure the
development of coordinated standards taking into account cross-
media pollutional impacts and generating and updating a research
plan. Under the Alternate Fuels Group are various "work groups"
which concentrate on specific energy areas. There are separate
work groups for oil shale mining and retorting, coal gasifica-
tion, indirect coal liquefaction , direct coal liquefaction,
alcohol production and geothermal energy. The members of the
work groups are EPA employees but we have also invited partici-
pation from other involved Federal agencies, viz., the Depart-
ment of Energy (DOE), the Tennessee Valley Authority (TVA) and
the Department of the Interior (DOI).
The Pollution Control Guidancte Documents will go through
an extensive internal and external review process. Internally,
the Alternate Fuels Group and the relevant work group will be
directly involved but final sign-off will occur at the level
of the Agency's Assistand Administrators who serve on EPA's
Energy Policy Committee, the Agency's highest level energy
coordination group. Externally, the Documents will be reviewed
60
-------
TABLE 7. POLLUTION CONTROL GUIDANCE
DOCUMENT REVIEW SCHEDULE
Technology 1st Draft
(data base)
Low Btu Gasification 11/80
Indirect Liquefaction 11/80
Oil Shale 11/80
Direct Liquefaction 9/81
High Btu Gasification 4/82
Medium Btu Gasification 1/82
Public Forum
4/81
5/81
5/81
3/82
10/82
7/82
Final Publication
8/81
9/81
9/81
7/82
2/83
11/82
Table 8 Processes To Be Covered In
Pollution Control Guidance Documents Now Under Preparation
o Low Btu Gasification
(Single State, Atmospheric Fixed Bed)
Riley-Morgan
- Wilputte-Chapman
Wellman-Galusha
o Indirect Coal Liquefaction
Gasification Synthesis
Texaco
Lurgi
Koppers Totzek Fischer-Tropsch
o Oil Shale
TOSCO II
Paraho
Union
- Superior
- Occidental
Rio Blanco
o Direct Coal Liquefaction
H Coal
- SRC
- Exxon Donor Solvent
Coal-To-Methanol
Mobil "M" (Methanol for Gasoline)
61
-------
by other Federal organizations such as DOE, TVA and DOI and
by a wide variety of industrial reviewers and also public
interest groups. Associations such as the American Gas
Association, the Gas Research Institute and the National
Council of Synfuels Producers will also serve as reviewers.
A public forum providing a second opportunity for external
review will be announced in the Federal Register sixty days
prior to its occurrence. Review comments from individuals
and from technical societies such as the Federation will be
most welcome. The final Document will be revised to reflect
response to all appropriate comments. The proposed review
schedule for the six PCGD's now under preparation or planned
is shown in Table 1.
Although the major objective of a PCGD is to recommend
pollution control options, it will contain a great deal of
background information on the energy processes themselves and
on process streams and pollutant concentrations, and will,
on the basis of a series of "case studies", offer specific
technology based control guidance for various kinds of energy
processes. Processes to be included will cover those that
are expected to be built for demonstration or commercial
application first. Table 9 shows planned process coverage for
the four PCGD's currently being written). It is intended that
discussion of product (E.G., liquefied coal) uses also will be
included if use is integral with the manufacturing process.
The process descriptions will detail the key features of each
process and their pollution potential. If various process
modifications are likely to be used at different locations,
the changes in process configuration will be covered and expected
changes in pollutant releases will be indicated. Pollutant
releases that vary non-linearly with plant size or flow rates
will also be identified and quantified to the extent possible.
The environmental control alternatives to be considered
will include both end-of-pipe treatment techniques and process
changes. Candidate control alternatives will be identified
from existing United States and foreign bench-pilot-and commer-
cial-scale facilities or from different United States or foreign
processes that have similar discharges. Performance and design
will be included as will information on capital, operating and
annualized costs. Energy usage for control alternatives will
also be included. Finally, techniques for monitoring control
performance will be identified. The source of all data will be
clearly referenced to allow referral to original sources;
uncertainties in the data will be indicated.
V- CONCLUSION
Permits to construct and operate synthetic fuel facilities
must be obtained by developers. The basis for review of these
62
-------
permit applications is contained in various EPA regulations,
standards, and guidance documents. EPA and the respective
State agencies have a shared responsibility in the review,
permitting, and ensuring compliance of synfuels facilities.
G3
-------
THE TVA AMMONIA FROM COAL PROJECT
By
P. C. Williamson
Division of Chemical Development
Tennesee Valley Authority
Muscle Shoals, Alabama 35660
TVA's Ammonia from Coal Project involves retrofitting a coal gasification
process to the front end of its existing 225-ton-per-day~ammonia plant.
The purpose of the project is to develop design and operating data to assess
the technological, economic, and environmental aspects of substituting
coal for natural gas in the manufacture of ammonia. Preliminary operation
of the facility was begun in September 1980. In the absence of specific
environmental guidelines for coal gasification processes, TVA's approach
to the potential environmental problem is to meet or exceed the emission
control requirements for specific components, i.e., sulfur compounds, par-
ticulates, aqueous discharges, etc. Also, TVA's facility contract specified
limits on certain discharges based on anticipated guidelines. In addition
to a discussion of the emissions control activities, a program is described
that examines the environmental health and safety aspects of the Ammonia
from Coal Project.
-------
THE TVA AMMONIA FROM COAL PROJECT
TVA's Ammonia from Coal Project involves retrofitting a coal gasification
process to the front end,of its existing 225-ton-per-day ammonia plant.
The purpose of the project is to develop design and operating data to assess
the technological, economic, and environmental aspects of substituting coal
for natural gas in the manufacture of ammonia. Preliminary operation of
the facility began in September 1980.
The environmental considerations for this project were unique; no environ-
mental regulations presently exist specifically for coal gasification
facilities. TVA's approach to the problem was to meet or exceed the emission
control requirements for specific components, i.e., sulfur compounds, particu-
lates, aqueous discharges, etc. In addition, TVA's facility contract specified
limits on certain discharges based on anticipated guidelines.
The facility is designed to produce 60 percent of the feed gas required for
the 225-ton-per-day ammonia plant. The ammonia plant can operate at 60 percent
turndown, therefore, the ammonia plant can operate at its design rate with
60 percent of the feed gas supplied from coal and the remaining 40 percent
from natural gas; or, the plant can be operated at 60 percent of design rate
(135 tons per day of ammonia) with all the feed gas supplied from coal. The
capability of operating the ammonia plant with 100 percent natural gas feed
is retained. This arrangement will make the greatest use of the existing
ammonia plant and minimize the amount and size of new equipment required. Also,
the coal gasification facilities can be operated independently from the ammonia
plant by burning the carbon monoxide and hydrogen gas in an existing steam
boiler.
The coal gasification unit is based on the Texaco partial oxidation process.
Engineering, procurement, and erection of the coal gasification and gas puri-
fication facility was done by Brown and Root Development, Inc. The air sepa-
ration plant required to provide high purity oxygen and nitrogen for the process
was handled similarly by Air Products and Chemicals, Inc. Engineering, pro-
curement, and construction of the coal handling and preparation area, inter-
connections to the existing ammonia plant, slag disposal, and services and
utilities required for the complex were performed by TVA.
A flow scheme for the TVA Ammonia from Coal Project (ACP) is shown in Figure 1.
Coal is received by rail and is sent to open storage and later recovered by
front-end loader or it is crushed in a primary crusher to minus 1/2-inch and
conveyed directly to the coal slurry preparation area.
Coal is pulverized in disk mills as required for the gasifier operation. Water
is added to the disk mills to form a coal-water slurry. From the disk mills,
the slurry goes to one of two mix tanks where the solids content of the slurry
is adjusted to the desired level. The slurry is pumped to a feed tank and then
metered to the reactor at the process rate of about 8 tons of coal per hour.
Gaseous oxygen from the air separation plant is fed to the reactor at about 8
tons per hour through a metering system interlocked with the coal slurry feed.
65
-------
01
VENT
1
DUST
COLLECTION
COAL RECEIVING
AND
PREPARATION
OXYGEN FROM AIR
SEPARATION PLANT
CO SHIFT
CONVERSION
COS
HYDROLYSIS
SLOWDOWN TO
KASTEiATER TREATMENT
SULFUR REMOVAL
VENT
t
-»> RECOVERED SULFUR
SULFUR REMOVAL
-*C02 TO UREA MFG
-*• RECOVERED SULFUR
NITROGEN FROM AIR
SEPARATION PLANT
FINE SULFUR
REMOVAL
BOILER FEED
WATER
TO EXISTING
AMMONIA PLANT
Figure 1 Flow scheme for TVA's Ammonia from Coal Project
-------
The gasification process takes place in the reactor at a pressure of about
510 psig and at a temperature in excess of 2200 F. The carbon in the coal
is reacted with steam to produce carbon monoxide and hydrogen. Oxygen is
injected to burn part of the coal to provide heat for the endothermic re-
action. In addition to the gasification reaction, coal combustion forms
carbon dioxide (C0«), and sulfur compounds in the coal are gasified in the
reducing atmosphere to produce primarily hydrogen sulfide (H S) and some
carbonyl sulfide (COS). Small quantities of other compounds such as ammonia
and methane also are formed. According to Texaco's pilot-plant experience,
essentially no long-chain or aromatic hydrocarbons are formed.
Slag produced from the ash in the coal is removed from the reactor through
a lockhopper system. The slag is glassy in appearance and is very similar
to the bottom ash produced in a coal-fired power plant boiler. Initially,
trucks are used to transport the solids to a disposal area. A slurry pumping
system may be installed later to handle and transport the slag to the disposal
area. In such a system, the slag would be washed and screened to remove over-
size material which would be crushed to a size suitable for slurrying and
pumping.
The gas leaving the reactor is water-quenched and particulate matter (fly ash)
is removed in a scrubber. A blowdown to control dissolved solids is taken
from the water recirculating loop and pumped to a wastewater treatment facility,
which uses chemical, physical, and biological treatment processes. The waste-
water is first treated in a clarifier by addition of ferrous sulfate and hy-
drated lime. The clarifier underflow is sent to a sludge conditioning unit and
then to a filter press for solids removal.
The liquid fraction from the clarifier is steam-stripped to remove ammonia
which is recovered and routed to the coal slurry preparation area to neutralize
the acidic slurry. The stripped aqueous material containing some organic
matter, primarily as formates and cyanates, along with water from washdown
operations is sent to an equalization-cooling basin for pH control, mixing, and
cooling. After aeration, the combined waste then flows to the activated sludge
unit for biological treatment. The treated water from the unit is metered
and sampled on its way to discharge. The digested sludge flows to the filter
press where the solids are removed for disposal. Plans are to recycle the
solids to the gasifier. The filtrate is returned to the wastewater treatment
system.
The process gas from the quench scrubber flows to two carbon monoxide (CO)
shift converters. The converters are charged with a sulfur-activated catalyst
marketed by Haldor Topsoe. The design CO content of the gas entering the
converter is about 22 percent (wet basis). After full shift, the CO content
is about 2 percent which matches the CO content of the gas entering the low-
temperature shift converter in the existing ammonia plant.
The COS produced during the gasification process is not affected by the Holmes-
Stretford sulfur recovery system that is used to recover H S from the off-gas
streams from the acid gas removal system. Therefore, the quantity of COS must
be decreased to meet the sulfur emission limitations. To accomplish this, a
67
-------
COS hydrolysis unit containing a catalyst also marketed by Haldor Topsoe is
provided betweep the CO converter and the acid gas removal (AGR) system to
promote the reaction:
COS + H20 t C02 + H2S
The process gas from the COS hydrolysis unit flows to the AGR system, The
AGR system uses Allied Chemical's Selexol process (a physical absorbent
system) to remove the C0_, H S, and the remaining COS from the process gas.
This system is capable or decreasing the total sulfur in the synthesis gas
stream to less than 1 ppm.
Nitrogen from the air separation plant is added to the process gas from the
AGR system to produce an H :N ratio of 3:1. The gas then flows through a
zinc oxide bed to decrease the sulfur content to less than 0.1 ppm. Deaerated
boiler feedwater is added to bring the steam-to-dry-gas ratio to 0.44:1.
The gas is then heated to about 600 F prior to its entry into the existing
ammonia plant at a point immediately upstream of the low-temperature CO shift
converter. The pressure of the gas at the battery limits is about 385 psig.
The composition of the process gas is very nearly the same as the composition
of the gas leaving the high-temperature CO shift converter in the ammonia
plant. The approximate composition of the gas is shown in Table 1. It should
be noted that the Selexol system is capable of decreasing the CO- to a value
much lower than that shown in the table. The 10.8 percent C0_ (wet basis) is
a design requirement and is not set by Selexol process limitations.
Two reject acid gas streams are produced during regeneration of the Selexol
AGR solvent. One stream containing up to 4 percent H_S is sent to one train
in the Holmes-Stretford sulfur-recovery system. The Holmes-Stretford system,
furnished by Peabody Process Systems, Inc., uses a proprietary solution
containing an oxidized form of vanadium salts. The H?S is oxidized in the
solution to produce elemental sulfur according to the following reaction:
2H S + 0 -v 2S + 2H 0
As stated before, the COS is unaffected by the Holmes-Stretford system. The
reduced metal salt is regenerated by blowing air through the solution. This
operation also floats the elemental sulfur to the surface. The sulfur is
skimmed off and filtered to produce a wet cake. The tail gas from the Holmes-
Stretford system contains about 160 ppmv H^S, less than 30 ppmv COS, and less
than 500 ppmv CO. This stream is vented to the atmosphere under conditions
of our emissions permit.
The second stream from the AGR solution regeneration system is relatively
pure CO . This gas is sent to the second train in the Holmes-Stretford
unit and then to a vessel containing zinc oxide to decrease the total sulfur
content to less than 0.5 ppm to meet requirements for urea manufacture. This
gas will be vented to the atmosphere when the urea plant is not operating.
ENVIRONMENTAL CONSIDERATIONS
The Ammonia from Coal Project management brought TVA's environmental and
medical expertise into the project at the very beginning. They worked with
68
-------
Table 1 APPROXIMATE COMPOSITION OF GAS MANUFACTURED
FROM COAL AT
COMPONENT
HYDROGEN
NITROGEN
CARBON MONOXIDEa
CARBON DIOXIDE
METHANE
ARGON
WATER
TOTAL
THE TVA AMMONIA FROM COAL PROJECT
PERCENT BY
WET BASIS
42.0
14.1
2.33
10.8
0.1
0.1
30.6
100.0
VOLUME
DRY BASIS
60.6
20.3
3.3a
15.6
0.1
0.1
100.0
BASIS: TOTAL SULFUR =0.1 ppmv MAXIMUM
STEAM-GAS RATIO =0.44
HYDROGEN-NITROGEN RATIO =3.0
NOTE: 3THE CARBON MONOXIDE CONTENT OF THE GAS IS BASED ON
END-OF-RUN CONDITIONS FOR THE SHIFT CONVERSION
CATALYST.
69
-------
the project management team to develop the project specification covering
the environmental, health, and safety requirements. These specifications
were then included in the contract for the coal gasification project.
An environmental evaluation was made on the project and it was determined
that an environmental impact statement was not required. Also, because of
its size—180 tons-per-day coal feed rate—and because the plant is scheduled
to operate one-half of the available operating time, it was determined that
the emissions were sufficiently low so that the plant was not considered to
be a major pollution source according to EPA's Prevention of Significant
Deterioration (PSD) rules. These two facts shortened considerably the lead
time required to obtain the necessary environmental permits. Three State
of Alabama permits covering emission to the atmosphere were obtained. One
covers the coal receiving, unloading, conveying, and storage. Dust suppression
equipment is required at all transfer points as a condition of the permit.
A second permit covers the primary coal crushing operation and conveying to
the pulverizing and slurrying operation in the gasification section. This
permit requires dust suppression equipment at all transfer points and a wet
scrubber on the crusher operation. The third permit covers the coal gasifi-
cation and gas purification' unit. This permit restricts the quantity of
total sulfur compounds, CO, and NOx compounds that can be emitted to the
atmosphere. In addition, an uncontrolled vent is allowed for startup and
emergency but its use is limited to a certain number of hours per year;
combustion of the vent gases is required.
Wastewater is processed routinely as stated earlier by chemical precipitation,
stripping to remove ammonia, biological treatment, clarification, solids
separation, pH treatment and finally discharge through a flow and pH monitoring
system into an existing NPDES-permitted stream. Our efforts to meet regulations
required that we obtain a modification to the existing NPDES permit.
Solid wastes are to be disposed of in a landfill. Because we had no concrete
data proving otherwisek and as a precautionary measure considering the develop-
mental nature of the project, TVA elected to handle the slag from the gasifi-
cation operations as if it were hazardous and accordingly applied to the State
of Alabama for permission to dispose of the slag in a nearby site. We lined
the disposal pond with a minimum of 2 feet of clay having a permeability of
10~' cm/sec or less. We will accumulate the water drainage from the slag and
return it to the gasifier operation. Four monitoring wells, one upstream and
three downstream of the disposal pond, are provided for sampling to detect
any changes in the groundwater composition.
Environmental Studies
Thus far we have discussed the environmental effort in regard to meeting the
applicable regulations and emission standards. In addition to these activi-
ties, a program is planned that looks further into the environmental, health,
and safety aspects of the ACP- Table 2 lists the study areas, the sources of
the samples to be analyzed in evaluating these study areas, and the analyses
to be performed on the samples. These analyses will help to evaluate the
environmental impact of our project and also may serve as a guide in evalu-
ating the impact of future gasification projects. For instance, we fully
expect that the slag studies will show that the slag is nonhazardous and
should be handled similarly to the bottom ash from a coal-fired power plant.
70
-------
Table 2 ENVIRONMENTAL STUDY PROGRAM OUTLINE
STUDY AREA
Gaseous Emissions Monitoring
and Characterization
Liquid Effluent Monitoring
and Characterization
Solid Waste Monitoring
and Characterization
Radiological Characterization
Medical Surveillance
Basic Industrial Hygiene
SAMPLE SOURCE
Sulfur recovery tail gas
Treated effluent
Accumulator-discharge -to
wastewater treatment
Gasifier slag
Solids to landfill (from waste-
water treatment)
Background
Monitoring wells
Coal
Gasifier slag
Sulfur recovery tail gas
Accumulator discharge to waste-
water treatment
Treated effluent/Disposal pond &
monitoring wells
Solids to landfill (from waste-
water treatment)
Operating personnel (individual)
Maintenance personnel (individual)
Operating personnel (individual)
Maintenance personnel (individual)
Employee work stations
(ambient air)
ANALYSES PERFORMED
Sulfur species
Nitrogen species
Hydrocarbons
Particulates
Trace Elements
Priority pollutants (129)
Trace Elements
Other3
Trace Elements
Hazardous waste extraction
Ra-226
Ra-228
Preplacement physical examinations
Periodic physical examinations
Transfer/Termination physical
examinations
Followup physical examinations
CO
GO'S
Particulates
Aromatic Hydrocarbons
aNH , NO , and NO , organic N, TDS, TSS, VSS, BOD , alkalinity, COD, S~, anide, TOG, formate.
Also may include Ca, Mg, SO,, Si02, PO,
-------
to be performed on the samples. These analyses will help to evaluate the
environmental impact of our project and also may serve as a guide in evalu-
ating the impact of future gasification projects. For instance, we fully
expect that the slag studies will show that the slag is nonhazardous and
should be handled similarly to the bottom ash from a coal-fired power plant.
The first four items in Table 2 covering the area of gaseous emission, water
and solid discharge, and radiological characterization affect the health and
welfare of the community beyond the plant boundary limits and as such are tre-
mendously important. However, the studies listed here are routine and could
be expected tp be carried out in any program similar to the Ammonia from Coal
Project.
The last two items deserve a closer look. The purpose of the medical sur-
veillance and the industrial hygiene programs is first, to protect the
workers assigned to the TVA Ammonia from Coal Project and second, to gain
knowledge to answer the persistent questions concerning the health and safety
of workers exposed to the coal gasification environment in general.
The medical program, developed by TVA's medical staff, includes a series of
medical examinations. The first examination or preplacement examination of
the candidate workers was made to determine preexisting conditions that might
be adversely affected by work in the ACP. These people were advised of their
conditions and counseled regarding methods of protection. Particular emphasis
was placed on evaluating the condition of the skin, respiratory tract and
genitourinary tract. Also, high quality color photographs were made of the
exposed skin of the face, neck, hands, and any suspicious lesion or other skin
problem areas. Periodic examinations will be made at not more than 12-month
intervals. These will be complete physical examinations similar to the
preplacement examinations. Termination and/or transfer examinations will also
be essentially the same as the preplacement examination. In addition, followup
examinations of former ACP employees may be made on a voluntary basis as part
of an epidemological study of the employees. The epidemological study will
involve pairing the ACP workers as a group with two other similar groups
(comparable sex, age). One, a similar group of workers with histories of work
in chemical plants except for this group's lack of exposure to the gasification
environment. The second comparative group will have "clean" histories with no
exposure in chemical plants. Statistical analysis will include a comparison
between the two control groups and the ACP workers to determine the contri-
bution, if any, of the gasifier environment to adverse health effects of ex-
posed workers.
The primary objective of the ACP industrial hygiene program is to protect ACP
employees from developing occupational diseases during the operation of the
projects and at any time in the future. But, because of the demonstration
nature of the ACP, another goal is to determine as completely as possible any
health and safety hazards associated with the process. This overall assess-
ment is expected to supply data for future coal conversion projects.
The possible hazardous agents that are of interest from an industrial hygiene
standpoint which might be found in the environment and their maximum limits
for unprotected workers are listed in Table 3.
72
-------
Table 3 POSSIBLE HAZARDOUS AGENTS AND THEIR STANDARDS
AGENT
CARBON MONOXIDE
HYDROGEN SULFIDE
CARBONYL SULFIDE
COAL DUST
AROMATICS
COAL TARS
NOISE
HEAT
STANDARD (8 hr. TWA)
ft
50 ppm
10 ppm
(no standard)
ob
2 mg/nr
10 ppm as benzene
0.2 mg/m^ as benzene soluble
fraction**
90dBAb
30 C WBGT (Wet bulb ,
globe temperature)
Source: American Conference of Government Industrial Hygienists
Source: Department of Labor, Occupational Health and Safety
Administration
73
-------
As a result of review of the plans and specifications for the gasification
facilities by industrial hygiene personnel, control measures such as area
monitors with audible alarms for carbon monoxide and hydrogen sulfide have
been or will be built into the physical plant. Other control measures
identified so far through the review process are: personnel protective
equipment such as protective clothing, hearing protection, and safety glasses;
positive pressure ventilation in control and analysis rooms; and provision
of deluge showers and eye baths.
Before the initial startup of the AGP facilities, a walk-through inspection
and evaluation of the plant was conducted. Area monitors and alarm systems
were tested; control systems were evaluated; and procedures for the personal
hygiene, protective clothing, and protective equipment were reviewed. The
plant operational procedures will be reviewed periodically to evaluate their
health and safety impacts.
A concentrated effort was begun during startup and will continue through pre-
liminary operation of the ACP facilities to identify and measure hazardous
agents produced by the operation of the facilities and equipment. Individual
worker environment is being sampled by portable devices attached to the
individual. Area samples are taken by fixed, automatic sampling stations
located at strategic points throughout the plant. Samples from these sources
are being analyzed in an attempt to identify unexpected as well as expected
agents that could be generated. A statistically valid number of samples will
be taken for each agent so that the confidence level will be maintained. This
means that the individual worker environment probably will have to be sampled
several times during the startup phase. If during the initial survey an un-
expected hazardous situation is discovered, additional sampling will be
scheduled.
Results from the initial survey will be evaluated and will serve as the basis
for developing a secondary workplan that will cover all future industrial
hygiene activities for ACP. The secondary workplan will cover at least the
following items: the hazardous agents that will be periodically measured;
the employees' exposure history; and the decision points concerning protective
clothing usage. The workplan will be a dynamic guideline that will be subject
to continuous change depending on the requirements of the ACP program.
The list of activities discussed above for the medical and industrial hygiene
studies on the ACP is by no means complete. However, it does cover the major
items of interest and indicates the degree of health protection and surveillance
that is built into the ACP program. We anticipate that hindsight will show
that we have considerable overprotection and overcaution in this area, but
at this stage we are taking no chances.
74
-------
ENVIRONMENTAL CONTROL OPTIONS FOR SYNFUEL PROCESSES
F. E. Witmer
Environmental and Safety Engineering Division
U.S. Department of Energy
Washington, D.C. 20545
Ultimately, the large scale production of synfuels from U.S. coal and oil shale
will become a reality. The U.S. Department of Energy (DOE) has a charge to
foster the commercialization of energy conversion technology that is environ-
mentally acceptable. "Environmental acceptability" is perceived to extend
beyond meeting environmental compliance standards at a given plant and to include
the "acceptability" of subtle, longterm health and ecological effects and the
composite of low level environmental effects associated with an aggregate of
synfuel installations. DOE has a hierarchy of site-specific environmental
assessments integral to DOE development and demonstration activity. The
objective of these assessments is to provide a data base for a determination
of environmental readiness by the Assistant Secretary for Environment. An
evaluation of the adequacy of the environmental control technology is a key
component of these determinations.
In assessment of control adequacy, many alternative approaches present them-
selves. Some of these control options result from a natural synergism of
combining process needs; for example, an auxiliary power plant that recovers
flue gas S02 in a concentrated stream can be advantageously coupled to t^S
recovery from the conversion process to produce by-product sulfur via Glaus,
or an entrained type gasifier can be included with a series of Lurgi gasifi-
cation units to handle rejected coal fines and oxidize highly contaminated
condensate wastewaters. Other control options follow from making controls more
cost-effective and/or environmentally superior. Wastewater reuse to extinction
(zero discharge) and the catalytic incineration of process tail gases are
examples of improvements over conventional technology. In the case of small,
site oriented industrial gasifiers, process simplicity and reliability are a
driving force for improved controls or the absence thereof; for example, in-
gasifier sulfur scavenging to eliminate subsequent t^S cleanup or "dry-quenching"
of product gas to eliminate the difficulty of wastewater treatment.
This presentation will overview a number of select environmental control options
whose technical and economic feasibility has been recently established. The
direction that future resultant control technology is expected to take will be
outlined.
75
-------
ENVIRONMENTAL CONTROL OPTIONS FOR SYNFUEL PROCESSES
INTRODUCTION
There has been considerable activity within the Department of Energy recently with
regard to synfuels related initiatives. Some of this proliferation results from
synfuel process development activity, which has been a long time in being and is
now reaching the critical pilot plant or demonstration phase (Figure 1). However,
much of this activity stems from industrial response to DOE's alternative fuels
initiative (Figure 2). Most of these synfuel projects are in various stages of
engineering and design. The alternative fuels efforts include both feasibility
studies (preliminary design efforts) and cooperative agreements to share precon-
struction and construction costs.
To one who has been "exposed" to these designs, several premises become clear:
o the energy conversion process design is tailored to the feedstock,
end-product mix, and specific site;
o the environmental control technology is integrated with the process
(end-of-the-pipe philosphy does not generally prevail); and
o a large number of environmental control options exist.
The innovative integration of environmental controls with the conversion processes
is a relatively new area of process design. This innovation has resulted in new
and different controls required as a result of recent and evolving environmental
standards (especially in the synfuels area). The evolution of controls with the
technology facilitates a beneficial synergism that can be missed if considered
mutually independent. The development of such control synergisms can involve
different sections of the plant and be based on the integration of both multimedia
and multipollutant interactions. It has long been the contention of the Assistant
Secretary for Environment that environment control development should be handled
integral to the technologies.
I.n this symposium Pollution Control Guidance Documents (PCGD) will be discussed.
These documents attempt to develop an environmental data base for synfuels process
configurations. A number of representative plant configurations have been selected
and preferred control options concomitantly delineated. These generalized studies
reinforce the fact that a large number of control options exist for a given synfuel
process. Because of these many options and their different effect on overall process
characteristics, it is indeed a challenging and difficult task to specify a "Best
Available Control Technology" (BACT) for these emerging technologies. Perhaps it
is best to return to the BACT concept after a brief discussion of control options.
In this presentation I would like to develop an appreciation for the complexity of
the control systems and their high variability as reflected in recent designs, to
stress the potential benefits resulting from integrating multimedia controls to
the conversion process, and to outline some control options that possess an economic
incentive for further development. The intent is to provide an overview of the
numerous control options that are emerging and the direction future controls may
take. The discussion will be confined to coal based synfuel processes and the
conversion process per se, however, it may be considered representative of other
areas such as oil shale and biomass conversion.
76
-------
FIGURE 1. MAJOR DOE FOSSIL ENERGY DEMONSTRATION ACTIVITY - COAL SYNFUEL
PROCESSES (SEPT 1980)
Gasification <
PROJECT
'Gasifiers-in-lndustry
Memphis
Grace
Conoco
I^ICGG
/SRC I
\ SRC II
Liquefaction^ H.Coa,
I^EDS
LOCATION
Duluth, Minn.
Memphis, Tenn.
Baskett, Ky.
Noble County, Ohio
Willisville, III.
Newman, Ky.
Morgantown, W.Va.
Catlettsburg, Ky.
Baytown, Tex.
COAL
DEMAND
Ton/ Day
75
3200
2300
1080
2300
6000
6000
200-600
250
MAJOR
PRODUCTS
Heating Gas, Fuel Oil
Medium Btu Fuel Gas,
SNG
Ammonia
SNG
SNG. Fuel Oil
Solid Boiler Fuel
Fuel Oil
Fuel Oil, Syncrude
Fuel Oil. Syncrude
STATUS
Operative
In Detailed Design
In Preliminary Design
(Reoriented toward
Methanol and Mobil-M
Gasoline)
In Detailed Design
In Detailed Design
In Detailed Design
In Detailed Design
Pilot-Plant in
Shakedown
Pilot Plant in
Shakedown
-------
FIGURE 2. SELECT COAL SYNFUELS ALTERNATIVE FUELS SOLICITATION - FEASIBILITY
STUDIES AND COOPERATIVE AGREEMENTS (JULY 1980)
oo
Feasibility
Studies
CONTRACTOR
Cooperative
Agreements
W.R. Grace
Clark Oil & Refining
General Refractories
Houston Natural Gas
Central Me. Power
EG&G
Crow Tribe
Nakota Co.
Phil. Gas Works
Celanese Corp.
Transco Energy
Union Carbide
Hamphire Energy
Texas Eastern Synfuels
Great Plains Gasification
Wycoal
POTENTIAL
SITE
Moffat Co., Colo.
S. III.
Florence, Ky.
Covent, La.
Waldo Co., Me.
Fall River, Mass.
East Billing, Mont.
Dunn, N. Dak.
Phil., Penn.
Bishop, Tex.
Calvert, Tex.
Houston, Tex.
Gillette, Wyo.
Henderson, Ky.
Beulah, N. Dak.
Douglas, Wyo.
FUNDING
REQUEST*
$ 786.477
$4,000.000
$ 922,555
$3,260,000
$3.624,558
$4,000,000
$2.729,393
$4,000,000
$1,168,108
No Cost
$1.874.005
$3.945.676
$4,000,000
$24.3M
$22M
$13.1M
MAJOR PRODUCT
Methanol
Gasoline
Low Btu Industrial Fuel Gas
Fuel Grade Methanol
Medium Btu Gas for Combined Cycle
Combined Cycle Power and Methanol
SNG
Methanol
Medium Btu Gas
Syngas
Medium Btu Gas
Low/Medium Btu Gas
Gasoline
SNG-44%, Transportation Fuel-30%
SNG
SNG
•To be Negotiated
-------
CONTROL OPTIONS
In considering the environmental impact of coal conversion, the total process train
should be taken into consideration (coal mining, beneficiation, transporation,
preparation, synfuels production, and product upgrading, distribution and end-use).
The conversion process is typically supported by an auxiliary boiler/power plant.
At the synfuel plant site, the auxiliary boiler plant is normally the major source
of emission of criteria pollutants.
The major synfuel-conversion processes, gasification and liquefaction (direct and
in-direct), are environmentally similar relative to inorganic pollutants, i.e.,
sulfur, NOX precursors, particulates, solid wastes, trace elements, etc. With
regard to the production of heavy organics, there is a wide variation between
processes, not so much as to "type" of organics, but to degree, since a wide range
of aromatic based tars and oils are typically produced. However, there can be a
marked difference in the bioactivity of the liquid fractions; as a disproportionate
portion of mutagenicity (which is indicative of carcinogenicity) has been found to
reside in high boiling primary aromatic amines which can vary widely between processes,
Entrained gasification, being a high temperature process, cracks most of the organics
thereby producing a product gas and quench water which is nearly devoid of heavy
organics. This is in contrast to the heavily organic laden condensate/quench waters
associated with direct, low temperature gasification processes and/or liquefaction.
For catalytic processes, the effect of spent catalyst on solid and aqueous wastes
varies process to process.
Environmental control options are conventionally segregated into types which deal
specifically with gaseous, liquid and solid pollutants. This follows in part from
the environmental legislation which is primarily concerned with impact on the
accpetor media, e.g., air, water, and land. However, in evaluating a control option,
effects on other media must be taken into consideration. Ideally, the pollution
control process is fully integrated with the conversion process to take advantage
of economics of energy consumption, reduced pollutant production, water reuse
potential and by-product production.
Complexity and Variability of Environmental Controls
Major potential pollutant sources which require the use of control processes are:
1. flue gas from auxiliary power plant/boilers
2. sulfur containing tail gases from acid gas separation
3. wastewater from multiple sources (product gas quench, coal pile
runoff, sanitary sewer, etc.)
4. auxiliary power plant/boiler solids (bottom ash, fly ash, scrubber
sludge)
5. conversion process solids (ash/slag, wastewater sludges, spent catalyst,
etc.)
power plant/boiler flue gas -
EPA, DOE, and industry continue to develop a large inventory of control options to
79
-------
reduce the emissions of sulfur oxides, nitrogen oxides and particulates from
the combustion of coal. For sulfur control, coal beneficiation and lime/limestone
flue gas desulfurization (FGD) have received primary emphasis and are considered
commercial processes. A number of other alternatives are at various stages of
development and demonstration, e.g., double alkali, dry-FGD, fluidized bed combus-
tion (FBC), and co-generation. In the area of NOX control, combustion modification
including low excess air, staged combustion, and burner modifications appears capable
of meeting the emission requirements specified by current New Source Performance
Standards (NSPS). NSPS particulate release standards (0.03 Ib/MBtu) can be met by
deploying enhanced electrostatic precipitators or fabric filters. It is emphasized
that these NSPS apply to compliance criteria and are current. Future changes can
be expected in the regulations concomitant with major synfuels activities over
the next 10-20 years.
tail gases -
The gaseous sulfur compounds generated during the coal conversion process (primarily
H2S, some COS, CS2, mercaptans, and thiophenes) are generally removed along with
C02 by the acid gas treatment train. The acid gases may be non-selectively
absorbed and partitioned into a t^S enriched stream (40-60%) and a I^S lean
stream (2-10%); the enriched and lean streams are typically routed to a Glaus
unit and a selective absorption unit, respectively, for sulfur recovery (Figure
3). The nominal C02 tail gases from these systems generally contain trace residual
sulfur—the Glaus system removes all but a few percent of the H^S, while the
absorption system can produce a tail gas with about 100 ppm H~2S. Incineration
represents the preferred treatment for the E^S-depleted streams which also may
contain some low level hydrocarbons. Stringent sulfur emission standards could
necessitate additional IL^S absorption prior to incineration or scrubbing of the
incineration flue gas with a conventional FGD system. In any event, it is
apparent that high H2S removal efficiency (>97%) can be confidently achieved
with existing commercial equipment.
wastewater -
Coal gasification and liquefaction typically produce a highly contaminated
"condensate" water which represents a by-product of the conversion reaction, extra
steam for cooling, a quench for direct cooling and scrubbing product gases, etc.
A wide range of organic loading is experienced; however, compositions tend to be
similar with phenolic compounds usually predominating. Condensate waters originat-
ing from a high temperature process (non-tar producer) can be essentially devoid
of organic material} Most plants tend to design for "zero" discharge of conden-
sate waters, that is, no condensate water is discharged to a surface acceptor;
however, such water may be rejected to the atmosphere through evaporation and
concentrated aqueous wastes, or may be disposed of via land-fill, ash surface
wet-down, deepwell injection (in accordance with applicable underground injection
control regulations), etc. Some process schemes consume the contaminated water
as recycle to gasification. In addition to condensate waters, various blowdowns
produced from feedwater treatment, boiler and cooling tower operation, coal pile
runoff, and sanitary wastes are generally integrated into the overall wastewater
treatment train. For example, if one examines the design of the wastewater treat-
ment trains for the major gasification projects DOE is involved with, one finds a
wide variation of process trains (Figures 4-8). The wastewater treatment options
may involve the combination of streams to enhance treatability and evaporation of
salt laden blow-downs. The variability between these wastewater treatment schemes
is stressed.
80
-------
FIGURE 3. COMMERCIAL ADSORPTION PROCESSES FOR CONTROL OF HYDROGEN SULFIDE
oo
Efficiency of
S Removal Absorbent Characteristics
Process
Chemical Solvent
Type:
1. ME A
2. DEA
3. TEA
4 Alkazid
5. Benfield
6. Catacarb
Physical Solvent
Type:
7. Sulfinol
B. Selexot
9. Rectisol
Direct
Conversion:
10. Stretford
11. Townsend
Drybed Type
12. Iron Sponge
Absorbent
Monoethanolamine
Diethanolamine
Triethanolamina
Potassium Di
methylamino
Acetate
Activated Potas-
sium Carbonate
Solution
Activated Potas
sium Carbonate
Solution
Sulfolane + Diiso
propanamine
Polyethylene Glycol
Ether
Methanol
Na-CO- •+ Anthra
qumone Sulfonic
Acid
Triethylene Glycol
Hydrated Fe^O.,
Type of
Absorbent
Aqueous Solution
Aqueous Solution
Aqueous Solution
Aqueous Solution
Aqueous Solution
Aqueous Solution
Organic Solvent
Organic Solvent
Organic Solvent
Alkaline Solution
Aqueous Solution
Fixed Bed
Temp.
op
80 to 120
. 100 to 130
100 to 150
70 to 120
150 to 250
150 to 250
80 to 120
20 to 80
<0
150 to 250
70 to 100
% H2S
Pressure Influent
Insensitive to 99
Variation in
Pressure
Insensitive to 99
Variation in
Pressure
Insensitive to 99
Variation in
Pressure
Insentive to 99
Variation in
Pressure 1 80 Atm
99
Insensitive to 99
Variation in
Pressure gen
erally >300 psi
High Pressure 99
Preferred
99
99
99.9
99.9
99
Effluent Regen-
HjS ppm Life eretion
MOO Thermal
MOO Thermal
MOO Thermal
MOO With
steam
H2S 4 COS Unlimited. With
MOO No deOfa steam
dation
H2S + COS With
steam
H2S + COS Low pres
MOO sura heat-
ing or
with
steam
H2S + COS
MOO
MOO
MO
MO
H2S + COS
MOO
Selectivity
Toward
Forms nonre-
flan. comp. with
cos, cs2
Absorbs CO«
does not absorb
cos, cs2
H2S
H2S
H2S is high
H~S partial
afso absorbs
cos. cs2
H-S and also
absorbs COS,
CS2 and mer-
captans
H2S, also ab-
sorbs COS
H2S
H2S
H2S
•HjS and also
towards COS.
CS~ and mer
captans
Form of
Makeup Sulfur
Rat* Recovery
50 to As H-S gas
100%
% As H2gas
<5% As H2S gas
As H2S gas
As H2S gas
-------
IU-QAS FLUID BED OASIFIER)
RECOVERED OIL TO
OFF-SITE DISPOSAL
UNTREATED HASTE
TO MUNICIPAL SEWII
STOW HATER
SPENT SERVICE
HATER
00
no
CREDIT GENERATION
CONDENSATE
CLARIFIED WATER
FROM GAS
CODLING AND
SCRUBBING
OXYGEN GRANULAR
FROM CARBON
AIR (MAKEUP)
SEPARATION
TREATED
MATER TO
MISSISSIPPI
DIVER
DEUATERED
SLUDGE TO
DISPOSAL
COOLING
TOHER
BLOHDMI
FIGURE 4. WASTEWATER TREATMENT SYSTEM - MEMPHIS INDUSTRIAL FUEL GAS PLANT
-------
(TEXACO ENTRAINED GASIFIER)
CONDENSATE WATER
SITE DRAINAGE
00
co
RETURN TO GASIFIER
DISPOSAL
OILY
WATER'
API
SEPARATOR
EQUALIZATION
(ASH POND)
t
I
I
I
LIME
CLARIFICATION
\
^ POLYMER
p^ ADD'N
SLUDGE
DEWATERING
ACTIVATED
CLARIFIER
TO REUSE
(COOLING TOWER
FILTER -
• BACKWASH. ETC.)
• EFFLUENT
FIGURE 5. WASTEWATER TREATMENT SYSTEM - GRACE AMMONIA DEMO PLANT
(CURRENTLY IN REDESIGN TO PRODUCE METHANOL AND M-GASOLINE)
-------
(BRITISH GAS/LURGI SLAGGING FIXED BED GASIFIER]
STORM WATER
DRAINAGE FROM LANDFILL
CONDENSATE ft
OILY WATERS
AMMONIA RECOVERY EFFLUENT
SANITARY SEWER
BRINE TO
• SULFUR
RECOVERY
SLOWDOWN
TO SLAG QUENCH
FIGURE 6. WASTEWATER TREATMENT SYSTEM - CONOCO SNG DEMO PLANT
-------
(COGAS STAGED PYROLYSIS GASIFIED)
oo
en
SITE DRAINAGE'
CONDENSATE &
OILY WATERS "
SEPARATOR
SEPARATOR
HOLD
BASIN
-^•SLOP OIL
»
1
EQUALIi
TO RAW WATER FEED
(INTERMITTENT)
r
rATIOM ._*. LIME/SODA
ZATION •— *» SOFTENING
SLUDGE
1
L RECAHBONATION
— (INTERMITTENT) "~"
SANITARY SEWER-
\
SLUDGE
TO REUSE
SOLIDS
FIGURE 7. WASTEWATER TREATMENT SYSTEM - ICGG SNG DEMO PLANT
-------
ILUROI FIXED BED GASIFIER1
CO
cr>
ALL FLOWS ARE FOR PHASE I ONLY — DOUBLE FLOWS FOR FULL PLANT
FIGURE 8. SCHEMATIC OF WATER SYSTEM FLOW ANG COAL GASIFICATION COMPANY
(REVISED OCTOBER 31, 1979)
-------
One common characteristic of the wastewater systems that must handle an organically
charged condensate water (Conoco, ICGG, and ANG) is that there is "zero discharge"
for this stream. The rationale for the selection of the "zero discharge" alterna-
tive with respect to condensate waters is that while activated sludge tends to be
a universal process for adequately treating condensate waters to effluent qualities
reflective of current regulations, the nature of these wastewaters, i.e., high
organic loading, toxicity of certain compounds, presence of refractory organics,
heavy metals and trace elements, causes uncertainty with respect to the evolving
Federal regulations resulting from the Toxic Substance Control Act (TSCA) and the
Resource Conservation Recovery Act (RCRA). While the technical feasibility of
additional steps to the conventional activated sludge train for controlling
effluents to more stringent standards has been demonstrated, the treatment processes
become more complicated and costly.
solid wastes -
The major solids produced by coal conversion facilities obviously result from the
mineral content of the coal feedstock. The characteristics (state) of the slag
or ash associated with the conversion process are dependent on the nature of the
process per se, since high temperature entrained gasification produces a relative
inert glassy material while non-slagging fixed bed gasifiers produce an ash.
Preliminary leaching tests indicate that both forms have weathering properties
similar to power plant bottom ash. Depending on the method of controlling SC^
emissions, there may be considerable scrubber sludge from the auxiliary power
plant which typically gets disposed of along with wastes from the conversion
process. Wastewater sludges, salts from evaporator ponds and/or concentration
equipment, spent catalysts and absorbents are representative of relatively low
volume secondary wastes that are likely to require special treatment in order
to be disposed in a manner consistent with RCRA requirements. The individual
treatment and/or disposal methods must be tailored to the specific waste and
site.
Integration of Multimedia Controls within Coal Synfuel Processes
In incorporating the afore discussed controls into a plant design, a number of
trade-offs exist, e.g., situations where by-products, contaminated water, spent
solids, waste heat, etc. can advantageously be used within the process and/or
environmental control area (Figure 9). A number of these "options" have appeared
in process designs and the literature. Others have been "conjured up" to give
some indication where innovative engineering might lead to improve the efficacy
of the process. In my judgement, this is an area that deserves further analysis
to determine the more promising options and their respective incentives.
One might ask "What are the economic incentives for some of the synergisms which
have been projected?" That is, are they really worth the undertaking of the
development and associated risk in the application? The answer to this question
is best satisfied by a detailed trade-off analysis. However, one can develop a
"feel" for potential savings. A very approximate breakdown of costs of environ-
mental controls for a major coal synfuels facility is given in Figure 10? Product
costs are estimated to be in the neighborhood of $5-8/MBtu for SNG, thus environ-
mental controls should typically account for 10-20% of the total product cost.
Reducing overall environmental control costs by say 50% (which is highly unlikely)
would result in a saving of merely 5-10% in product costs, not a large incentive
87
-------
FIGURE 9. CANDIDATE SYNERGISMS FOR COAL CONVERSION PROCESS AND
ENVIRONMENTAL CONTROLS
ENVIRO
CONTROL
SECONDARY
CONTROL/
UNIT OP
SYNERGISM
POTENTIAL BENEFIT
Wastewater
Concentration
Wastewater
Treatment
Oil/Tar
Disposal
Tail Gas
Control
Vent Gas
Control
Wastewater
Incineration
Wastewater
Concentration
H^S Recovery
Ash/Slag
Disposal
Wastewater
Treatment
Wastewater
Treatment
Wastewater
Disposal
Wastewater
Disposal
Wastewater
Concentration
Cooling Tower Wastewater Cone-Heat Rejection
Water Reuse
Aux Heat/
Power
Aux Power
Combustor/FGD
Aux Power
Combustor
Entrained
Gasifier
Aux Power FGD
Regenerative
FGD, i.e. Dual
Alkali
FGD Sludge
Disposal
FGD Sludge/
Slurry
Disposal
Oxygen
Production
Ash Cool Down
Wetdown of
Ash Piles
and Mine
Tailings
Heat Rejection
High Quality Effluent from Treat-
ment Train-Boiler Water Makeup
and Process Water Requirements
Combustion of Organics-Heat
Recovery
Existing Boiler and Flue Gas
Clean-up Train Used to Control
Tail Gas HC and Sulfur Releases
Existing Boiler Used to Control
Vent Gas HC Release in Lieu of
Flare
Destruction of Organics, Cone of
Solids-Provide Steam Req'mts
Wasteweter Cone-Makeup to
Flue Gas Scrubber
H2S and SO2 Control Combined
in Claus Unit
Mutual Disposal
Flocculation/Clarifi cation-Com-
bined Wastewater/FGD Sludge
Disposal
Relatively Cheap Oxygen Used to
Abet Bioxidation and/or Ozone
Production
Wastewater Further Concentrates
While Quenching Hot Slag
Wastewater Disposal-Control of
Fugitive Emissions
Wastewater Cone by Envapora-
tion and/or Freezing Adsorption
System-Low Quality Steam Uti-
lization
Wastewater Wastewater Addition of Lime to the Waste-
Treatment Stripping water Abets NH3 Stripping and
Flocculation/Clarification
Precludes or Reduces Effluent Release,
Reduces Raw Water Requirements
Reduces Raw Water Requirements
Maintains Potentially Hazardous Material
Within Plant Boundary
Avoids Special Controls and Insures High
Quality Emission
Potentially Better Control Especially if
Stack Gas Clean-up Practiced
Avoids Elaborate Treatment Train to Pro-
duce High Quality Effluent
Reduces Effluent Release and Raw Water
Requirements
No Scrubber Sludge, By-product Ele-
mental Sulfur
Alkaline Sludge will Discourage Trace
Metal Leaching from Ash/Slag
Reduction of Wastewater Lime Req'mts
Improved, Cost-Effective Treatment
Facilitates Disposal of Wastewater Con-
centrate
Facilitates Disposal of Wastewater Con-
centrate, Dust Control and Mine Res-
toration
Improved, Cost-Effective Wasteweter
Desalination and Reduction of Organics
More Complete NHo Stripping and Cost-
Effective Use of Lime
88
-------
FIGURE 10. ESTIMATE OF ENVIRONMENTAL CONTROL COSTS COAL SYNFUEL FACILITY
Overall process efficency assumed to be 65%
Auxiliary power plant assumed to use 20% coal input
Coal: 10,000 Btu/lb, 10% ash, 3.5% S
synfuel output
00
in
Auxiliary power plant
SO2 scrubbing
NOX burner control
Particulates - bag house
Solid disposal (ash and sludge)
Conversion Process1
Sulfur2
Tail gas incineration
Wastewater treatment
Slag disposal
Cost
basis
5-10 mills/kwhr
nil
1-2 mills/kwhr
$10/ton
10-20 $/MBtu
5-10 $/MBtu
$10-20/1000 gal
$3-10/ton
low
15
high
30
10
5
10
2
47
20
10
40
6
114
MBtu = 106 Btu
Excludes mining - environmental aspects included in cost of coal.
2in some instances high level removal required to preserve catalysis activity
-------
from the perspective of the producer and potential risks incurred, if the control
processes encounter difficulties and disrupt operations. However, if one looks
at the incentive in absolute terms, for a single major facility, a 10c/MBtu
saving translates into $7.5 M/yr. or $200 M over the life of the facility.
Savings of 10c/MBtu in the environmental control area are not unrealistic. It
is this driving force that has encouraged the study of the feasibility of
improved environmental control options in DOE's Environmental and Safety
Engineering Division (ESED).
Control Options Studied
As a result of a continuing assessment of environmental control adequacy within
DOE/ESED, a number of candidate control options have become worthy of a
determination of technical-economic feasiblity:
sulfur -
Sulfur absorption technology is well established and based on experience in the
petroleum industry. There has been some minor concern for possible contamination
of the absorption media with complex hydrocarbons, trace elements and dust;
however, operating experience on coal gases indicate such effects can be
accommodated.
With the intent of simplifying the clean-up technology for an on-site industrial
fuel gas producer, the control of sulfur within the gasifier proper using a
calcium treated coal has been studied (Figures 11 and 12)-? An important advan-
tage of the use of a treated coal feedstock to small users is that it eliminates
the environmental problems associated with the treatment and disposal of sludges
and waste water generated from flue gas clean-up and fuel gas desulfurization.
Another significant advantage to consider is the^improved process reliability
expected from this approach relative to product (fuel) gas cleanup and FGD options
The user simply needs a supply/inventory of treated coal to keep running or make
a fuel switch. For those applications where intermittent operations are contem-
plated due to prime fuel curtailment, the use of treated coal would eliminate
the need to operate and maintain a chemical scrubbing system.
Laboratory screening studies have demonstrated that a coal treated with CaO at
ambient conditions can effectively remove sulfur and produce a low-sulfur fuel
gas in a moving-bed, a fluidized-bed, or an entrained bed gasification system.
The sulfur captured in the gasification ash is converted to essentially inert
calcium sulfate for environmentally safe disposal. Sulfur removal efficiencies
of calcium treated coal relative to untreated coal are shown in Figure 13.
A preliminary economic evaluation of "conversion to coal" (oil/gas backout) by
typical industrial users has shown the treated coal to be competitive with the
direct combustion of coal and with the gasification of untreated coal that
require flue gas desulfurization and fuel (product) gas desulfurization respec-
tively, for controlling sulfur emissions. Results of a preliminary cost evalua-
tion of industrial steam generating systems with a peak load of 100,000 Ib/hr
steam and an average load of 60,000 Ib/hr steam are presented in Figure 14 to
compare various fuel-replacement/retrofit options.
90
-------
?lue Gas
Coal
Air
Steam
Gasifier
Cyclone
To Ash Disposal
Water
Quench
Steam
Hot Gas
Condensed
Water Sulfur
Boiler
Feed Water
Air
Waste Water
Water Recycle
GASIFICATION OF UNTREATED COAL WITH H2S REMOVAL
Calcium
Treated
Coal*
Steam
Air
Gasifier
'Cyclone
w
To Ash Disposal
• Supplied by off-tit*, central treatment facility.
Flue Gas
Fuel Gas
Boiler
Feed Water
Air
GASIFICATION OF CALCIUM TREATED COAL
FIGURE 11. PROCESS VARIATIONS FOR FUEL GAS RETROFIT APPLICATIONS
91
-------
Slaker
Water
Slurry
Mixing
Tank
ked Lime
Lime
Tre
Slu
I
ated
rry
.
-•••Grit to Disposal
Water
Recycled Water
Blnde
I
l^J Centrifuge
Ilrlquetting'
Machine
Makeup Water
FIGURE 12. PRODUCTION FACILITY FOR CALCIUM TREATED COAL
Product to Storage
Contractor - Battalia
-------
FIGURE 13. NOMINAL SULFUR CONTROL LEVELS CALCIUM TREATED COAL
(LABORATORY SCREENING STUDIES)
SULFUR REMOVAL, PERCENT
UNTREATED TREATED
Moving-Bed Gasification
Fluidized-Bed Gasification
COAL
(bl
COAL
80
85
Entrained Gasification
CONCENTRATION. PPM
UNTREATED TREATED
COAL COAL
PRODUCT GAS
H2S
HCN
SCRUBBER WATER FLASH GAS
H2S
HCN
SO2
4500
33
5300
180
8000
370
10
25
25
200
(a) Agglomeration occurred but gn flow through pellets allowed ten to be completed.
(b) Test uneuccauful due to eevere •gglomeretion of untreated coal in fluidiied-bed gasification.
Contractor — Battelle
FIGURE 14. PROJECTED ECONOMICS FOR CONVERSION OF INDUSTRIAL GAS-FIRED
BOILERS TO COAL
SYSTEMS
Coal-Fired Boiler with FGD
(Boiler and Scrubber New)
Gasification with FGD
(Boiler Retrofit, New Scrubber)
Gasification with H2S Removal
(Boiler Retrofit)
Gasification of Calcium Treated Coal
(Boiler Retrofit)
CAPITALCOST.
$106
9.1
9.8
10.4
7.7
OPERATING COST,
$108/YR
2.8
3.1
3.1
3.0
STEAM COST,
$/1000 LB STEAM
10.7
11.6
11.9
10.3
93
-------
tail gases -
The reference control technology for the tail gases associated with acid gas
stripping operations is direct incineration at approximately 1,600°F with a
clean fuel gas. Alternative control methods which showed promise in a preliminary
assessment study were incineration in a coal fired boiler at 4c/MBtu (product gas
basis) and catalytic incineration at 5c/MBtu, while tail gas incineration with
clean fuel gas is projected to cost in the neighborhood of 10-12£/MBtu7 Commercial
catalyst have been screened to determine the effect of temperature, space velocity,
and the presence of H2S and COS on hydrocarbon and carbon monoxide conversion
(Figure 15)^ These bench scale studies indicate the most effective catalysts
are precious metal catalyst on a monolith substrate and a non-precious metal oxide
deposited as micro spheres on a solid substrate (Figure 16). The more promising
catalysts H, G, and A are currently undergoing life tests. A detailed analysis
of the coal-fired incineration option is to be made by a commercial incinerator/
boiler manufacturer.
wastewater -
The control options for treating condensate wastewaters in a conventional mode
have been demonstrated at bench scale. It appears that activated sludge is
sufficient for coal wastewaters to meet existing discharge standards. Prior to
biotreatment, gross ammonia and organic removal is required to render the feed
non-toxic.
Coal condensate waters contain dissolved ammonia, up to 2%. This NH3 is usually
neutralized by dissolved CC>2 that is produced in driving the conversion process;
thus the condensate waters are strongly buffered and to change the pH via the
addition of chemical reagents is normally quite expensive. Some coals contain
high chloride which enters the condensate water and provides a strongly acidic
anion to retain the NH3 as NH^Cl. In such instances, it is necessary to add a
strong base (CaO) to enhance NHo strippability. Normally such coals occur in
the East and the additional salt loading due to reagent addition presents no
critical problem with effluent discharges.
Phenolic compounds contribute to the bulk of BOD (5,000-10,000 ppm) and along with
other organics, pose a severe stress on sludge microorganisms. One typically
resorts to solvent stripping and/or dilution to bring the levels down to 1000-
2000 ppm, at which level acclimated organisms can do a reasonable job. An on-going
study is determining the trade-offs between NH3 and organic stripping options
attempting to conserve reagents and at the same time, reduce steam requirements'
Coal wastewaters contain some ring structures, polynuclear aromatics (PNA's) and
heterocyclics (1-10 ppm), some of which are biorefractory. The more refractory
compounds are adsorbed on the sludge, with effluent concentrations running in the
range of 10-50 ppb. Laboratory bench testing has indicated that a significant
reduction of PNA type materials can be achieved if the effluent is subjected to
partial ozonation followed by activated carbon adsorption. It appears important
that the ozonation precede the sorption step, lest the large ring-structure
compounds be too large for the pores of the carbon. Current efforts are focused
at determining the efficacy of regeneration techniques for the spent carbon.
Another study is attempting to demonstrate the viability of powder activated
carbon (PAC) to help stabilize the biooxidation of solvent gtripped condensate
waters and improve the efficacy of activated sludge systems'! Biological screening
tests are being performed on the various intermediate process waters to help
ascertain the completeness of the treatment with regard to mitigating any low
level adverse biological impact that may result by the release or use of partially
94
-------
FIGURE 15. SUMMARY OF INCINERATION CATALYSTS TESTED
CODE COMPOSITION
COMMENTS
10
en
A.F
C
E
H
'Spherical and Extrudate Forms, Non-
Precious Metal Oxide on Support Material
0.1% Pt, 5% Ni
0.1% Pt, 3% Ni
Pd on Metal Lessing Rings
Pt on AI2O3 Monolith Support
Precious Metal on Ceramic Honeycomb
Mn and Cu Oxides
Inexpensive, ~640 $/m
Not Poisoned by Pb, Zn, Halides.
<10 ppm SO3 in Effluent. Used for CO.
H/C and Other Organic Removal.
"Some" SO3 in Effluent.
No Experience with Similar Streams.
An NOX Removal Catalyst Via NH3 Reduction.
Same as D, a Hydrogenation Catalyst
<10 ppm SO3 in Effluent. Expensive, ~1.4 x 10
$/m . No Comment on Poisons. Can be
Recycled 3 Times. Primarily Used for H2
Removal.
Favors SO3 Production. Expensive,
~5.3 x 10 $/m . No Comment on Poisons. For
Industrial Tailgas Cleanup.
~100 ppm SO3 in Effluent.
Expensive, ~6.4 x 104 $/m3.
For Industrial Tailgas Cleanup.
Poisoned by S and Heavy Metals. Inexpensive
~710 $/m3. Designed Removal of H/C's and
CO from Breathing Air.
Contractor - ORNL
-------
IO
O
5
IT
O
u.
O
ui
cr
O
tu
DC
0.
5
1200
(1700°F)
1000
(1330°F)
800
(980°F)
600
(620°F)
400
(260°F)
200
METHANE
CARBON MONOXIDE
CATALYST
Note: ethane oxidation found comparable to methane, while catalysts tended to
oxidize ethylene at lower temperatures, 600-800°K
FIGURE 16. METHANE AND CARBON MONOXIDE REMOVAL AS A FUNCTION OF CATALYST
Contractor — ORNL
-------
treated effluents (Figure 17)? Note that the toxicity after biotreatment
is suspected to result solely from inorganic species, i.e., the conversion of
thiocyanates to ammonia during biotreatment (laboratory unit not as fully aerated
as a commercial operation) and conversion of trace, residual cyanates to cyanide
on ozonation. In some instances, a color problem has been associated with the
aging of trace polyhydric phenols which may be overcome with a carbon polishing
step or the addition of PAC to the activated sludge system. Unit operations can
be arranged in a condensate treatment train that would produce almost drinking
quality water. Relatively high treatment costs are likely to bar such intensive
treatment (Figure 18); however, it should be noted that the cost impact under
current standards is considerably less, expecially since only 10-20 gallons of
condensate water may be produced per MBtu. Costs also can be reduced if it is
practical to resort to PAC in lieu of ozonation and activated carbon.
As indicated in the plant designs, the trend for wastewater control is to perform
some partial treatment on the wastewater stream (solvent extraction, activated
sludge) and use cooling towers to concentrate the stream to a point where a reason-
ably sized blowdown stream can be fed to evaporation ponds or multiple effect
evaporators. Ideally it is economically desirable to use as poor a quality of
water as the reuse application will permit. An on-going study is evaluating
water quality requirements for a number of reuse applications, such as cooling
towers, many of these applications have been previously outlined.
Special attention has been given to reducing the quantity of wastewater associated
with the quench operation by instituting a two stage quench - the initial stage is
a low volume recycled highly contaminated water while the second stage consists of
a much larger volume of relatively clean water, the strong acid gases condensing
out in the first stage. The incentive for such a system has been shown to reside
with coals having a halide content greater than 0.15% Cl, i.e., generally Eastern
coals (Figure 19). It is likely that future plant designs will adopt water
conservation measures and desalting technology to preserve the water balance
within the plant so that a concentrated, highly contaminated, low volume waste
stream will be produced. Thermal oxidation techniques, e.g., gasification (recycle
to the conversion process), wet-air oxidation, and even incineration, are expected
to become viable treatment practice for the concentrate.
solid wastes -
As indicated, it is desirable to dispose of solid wastes in a manner tailored to
the specific properties of the individual waste. Studies have been supported
to classify major gasification and liquefaction slags/ashes as hazardous or
non-hazardous under EPA/RCRA protocols (Figure 20). It appears that such material
may be disposed in a conventional manner, which can mean landfilling during mine
restoration for strip mining operations near to the conversion facility. With
the intent of better defining the true environmental acceptability of waste
disposal practice for such materials, a series of laboratory column leaching
and lysimetric tests are being performed to develop an understanding of leaching/
mobilization phenomena and identify viable control procedures. Preliminary
studies have shown high initial sulfur releases from gasifier slags and their
auto-oxidation to sulfuric acid, may preclude the natural capacity of geologic
material to adsorb migrating trace heavy metals. Incorporating an alkaline
material (limestone, spent scrubber sludge, etc.) with the slag ash would tend
to discourage acid formation during these critical, early leach cycles (Figure
21).
97
-------
FIGURE 17. ACUTE TOXICITY TO DAPHNIA MAGNA OF HYDROCARBONIZATION
WASTEWATER BEFORE AND AFTER VARIOUS TYPES OF WASTEWATER TREATMENT
SAMPLE APPROXIMATE 48-HR LCgp (%)
Raw Scrubber Water 0.65
Biofeed Water 2.3
^ Biotreated Water -70
CO
Water After Ozonation ~18
Water After Ozonation and Charcoal Adsorption =0.1
Water After Charcoal Adsorption and Ozonation =4.5
Contractor — ORNL
-------
RAW WASTE WATER
STEAM-
SOLVENT-
AIR-
NUTRIENT-
OZONE-
NH,
APPROXIMATE COST
$/1000 GAL
NIL
2-5
PHENOLS
3-7
SLUDGE 2-8
SUBTOTAL 7-20 (PRESENTSTANDARDS)
0.1-0.2
2-5
TOTAL
10-15
19-40 (FUTURE STANDARDS)
EFFLUENTS
FIGURE 18. REPRESENTATIVE WASTE WATER TREATMENT PLAN
FOR COAL CONVERSION EFFLUENTS
99
-------
FIGURE 19. TWO STAGE QUENCH OPTION
*. DESULFURIZATION
O
o
O
o
§2
0.1
TWO STAGE
J
DIFFERENCE BETWEEN APPARENT COAL
COSTS FOR SINGLE STAGE QUENCH AND
TWO-STAGE QUENCH, I/TON
CONTRACTOR - ARTHUR G. McKEE
-------
FIGURE 20. EPA-EP LEACHING RESULTS FOR SIX GASIFICATION/LIQUEFACTION SOLID WASTES
1
ELEMENT
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Mercury
Selenium
Silver
Nickel
Zinc
WASTE C
0.27
<200
0.054
1.6
2.7
<0.3
0.64
<5
<0.03
281
63
WASTE E WASTE G WASTE H
(All Concentrations in ppb)
0.06
<500
0.97
0.44
3.7
0.26
0.03
2
<0.03
219
10
<1
20
<1
<5
10
<10
<1
<1
<2
30
13
-------
As
FIGURE 21. INFLUENCE OF pH AND REDOX POTENTIAL ON METAL CONCENTRATIONS IN
WATER (GASIFICATION WASTE, SOLID:SOLUTION RATIO, 1:50)
Contractor - ORNL
102
-------
CONCLUSION
Hopefully what has been conveyed by this broad-brush presentation is that a large
number of environmental control options exist, that many of these control options
are integrated into the process to improve the efficacy of the overall conversion
process and lessen the concomitant environmental insults of the conversion process
The inventory of viable control options are rapidly evolving: under such
a dynamic situation where actual performance data on full-scale, environmentally
acceptable facilities is lacking, it appears premature to develop firm BACT
criteria. What would appear to be of greater service to the nascent industry
would be a set of reasonable technology based emissions regulations or guidelines
that would provide industry with the requisite freedom and flexibility and the
incentive for innovation to operate within such bounds. In a nut shell, let's be
prudent.
References
1. Lee, M. L., et al, "Study of By-Products and Potential Pollutants from High
Temperature Entrained Flow Gasifiers," Brigham Young University, DOE Contract
EE-77-S-02-4377, April 16, 1980.
2. Witmer, F. E., "Environmental Concerns for Coal Synfuel Commercialization,"
International Journal of Energy Research, John Wiley & Sons, Volume 4, No. 2,
April-June 1980, Pages 185-195.
3. Kim, B. C.; Feldman, H. F.; el al, "Control of Emissions from Gasifiers
Using Coal with a Chemically Bound Sulfur Scavenger," Battelle Columbus
Laboratories, DOE Contract W-7405-Eng-92, April 18, 1980.
4. Fisher, J. F.; Peterson,G. R., "Control of Hydrocarbon and Carbon Monoxide
Emissions in the Tail Gases From Coal Gasification Facilities," ORNL, DOE
Contract W-7405-eng-26, ORNL/TM-6229, August 1978.
5. Brown C. H., Klein, J. A., "Control of Hydrocarbons and Carbon Monoxide
via Catalytic Incinerators," ORNL, DOE Contract W-7405-eng-26, m publication.
6. Workshop Report: "Processing, Needs and Methodology for Wastewater from the
Conversion of Coal, Oil Shale and Biomass to Synfuels," University of Cali-
fornia, DOE Grant DE-AT03-79EV10227, May 1980.
7- "Environmental Control Options for Gross Treatment of Condensate Waters,"
LBL Contractor, J. King, Principal Investigator, DOE RPIS No. 800407.
8. "Assessment of Environmental Control Technology for Coal Conversion Wastewater
Systems," ORNL Contractor, J. Klein, Principal Investigator, DOE RPIS No. 80060.
9. "Use of Powdered Activated Carbon to Improve the Efficacy of Activated Sludge
Systems on Coal Conversion Wastewaters," ANL/Carnegie Mellon, W. Harrison/
R. Luthy, Principal Investigators, DOE RPIS No. 800516.
103
-------
10. Witmer, F. E., "Status of Synfuel Wastewater Treatabillty Options," Energy
Optimization of Water and Wastewater Management for Municipal and Industrial
Applications, DOE/ANL New Orleans, La., December 10-13, 1979.
11. "A Study of the Control of Environmental Impact of Condensate Wastewaters
for Coal Conversion Plants," Water Purification Associates, D. Goldstein,
Principal Investigator, DOE Contract DE-AC02-80EV10367.
12. "Improved Water Management of Coal Conversion Processes By Preliminary
Absorption of Halides," Arthur G. McKee and Company, DOE Contract No. EE-
77-C-02-4375, January 1979.
13. "Hazard Evaluation of Solid Waste," ORNL Contractor, C. Gehrs and C. Francis,
Principal Investigators, DOE RPIS No. 800387
104
-------
TECHNICAL AND ENVIRONMENTAL ASPECTS
OF THE GREAT PLAINS GASIFICATION PROJECT
Remarks of Gary N. Weinreich
Manager, Environmental and Community Affairs
American Natural Service Company
Ladies and gentlemen, it's a pleasure to have this opportunity
to speak before you today about the Great Plains Coal Gasification
Project. Unlike our presentations during the last seven years,
today we can talk about a synthetic fuels facility that is under
construction, a facility that will be the first commercial-sized
substitute natural gas (SNG) plant in the United States, and a
facility that represents a signal to the world that this country is
serious in its efforts to reduce its dependency on foreign countries
for its crucial energy supply- While this plant is by no means a
panacea, it most definitely represents a major and difficult first
step on the part of industry and government that will eventually
lead to a successful new synthetic fuels industry in this country.
Synthetic fuels, coupled with energy conservation and successful
developmental efforts in the areas of solar power, non-conventional
and renewable energy sources, will enable the United States to enter
the twenty-first century in a much better energy supply and national
security posture than is maintained today.
We must give a great deal of credit to the US Department of
Energy for their assistance in the form of a federal loan guarantee
for the project. With DOE's pledge of assistance, Great Plains was
able to maintain the 1980 construction start date and avoid further
105
-------
delays in this long overdue venture. As you may be aware, the
Federal Energy Regulatory Commission approved the Great Plains
Project in November, 1979, but General Motors Corporation and three
state commissions opposed the consumer-backed financing arrangements
approved by the FERC. The federal loan guarantee alleviates this
situation and has permitted the project to proceed. Ground was
broken in August and construction of the facility will continue
through to the completion date in 1984.
I was asked to speak on the technical and environmental
considerations involved in a coal gasification facility such as the
Great Plains Project. As you can imagine, this is a very broad
subject to cover in 25 minutes. I will try to address the highlights
and the bases for some of the environmental decisions involved in
our project.
A brief organizational description of the Great Plains Project
might be appropriate for those of you who are unfamiliar with the
project. Great Plains Gasification Associates is a consortium made
up of subsidiaries of five major natural gas pipeline companies. The
project was originally proposed by ANG Coal Gasification Company,
a subsidiary of American Natural Resources Company of Detroit, Michigan.
ANG is now an equal partner in the project as well as the project
administrator responsible for the design, construction and operation
of the facility for the consortium. The other members of the consor-
tium are subsidiaries of the Peoples Energy Company, Transcontinental Gas
Pipe Line Corporation, Tenneco, Inc. and Columbia Gas Transmission Co.
The project consists of a 275-million cubic-foot per day high-BTU
coal gasification plant which is being built in two half-size phases.
106
-------
The project is located in Mercer County, North Dakota, six miles
northwest of the town of Beulah (population approximately 3,000)
and seven miles south of the plant's water supply, Lake Sakakawea.
The plant is located immediately adjacent to an 880-megawatt steam
electric generating plant currently being constructed by Basin
Electric Power Cooperative of Bismarck, North Dakota. Together, the
two plants will share common facilities such as water supply, rail-
road, plant access and coal mining. The power plant will supply
electricity to the Great Plains facility while using the lignite
fines which are unusable in the Lurgi gasifier. Together, the two
plants complement each other and provide economic advantages while
reducing the adverse environmental impacts of two separate plant sites
The air pollution control systems included in the design of the
Great Plains facility represent the largest single pollution control
cost. The air emissions control system can be divided into four
broad categories: 1) coal gasification, 2) steam generation, 3) coal
handling, and 4) incinerators, flares and miscellaneous sources.
Each category is unique and merits a brief explanation of the control
alternatives.
The Great Plains' gasification system, like that of many other
proposed SNG plants in the United States, will employ the Lurgi
Rectisol process to remove acid gases from the synthesis gas stream.
The Rectisol process uses a cold methanol wash to absorb C02/ H2S
and other sulfur compounds from the product gas, and the methanol is
107
-------
then regenerated. Our engineers considered several options for
treating the sulfur-containing off-gas streams from Rectisol. At
first a Glaus unit with tail-gas clean-up and a Wellman-Lord stack
gas scrubber was considered. Detailed investigation, however,
raised a number of questions about the operating reliability of the
Glaus system on a feed stream containing variable concentrations of
H2S. For this reason as well as high cost, a system utilizing the
Stretford sulfur recovery process was selected for the Great Plains
plant. The Stretford process is known to effectively reduce H2S to
less than lOppmv; however, the Stretford process has not been proven
on streams with as high a CO2 content as that of the Rectisol off-gas.
For this reason, our plant includes a Stretford system designed
to remove H2S to a level less than lOppmv, but our permit takes credit
only for the vendor-guaranteed removal efficiency or lOOppmv. Of
course, we are hopeful that the higher removal efficiency will be
achieved and the plant-wide sulfur emission will be much lower.
The tail-gas from the Stretford unit will contain residual H2S
and virtually all the organic sulfur and hydrocarbons present in the
feed from Rectisol. For this reason, incineration of the Stretford
tail-gas is required. In the case of the Great Plains plant, this
tail-gas will be incinerated in the plant boiler system, recovering
the BTU value of the gas while converting the H2S, organic sulfur and
hydrocarbons to compounds acceptable for emission to the atmosphere.
Although the Stretford tail-gas contains a very small BTU value on a
cubic foot basis, it constitutes a major fuel source by virtue of
108
-------
its large volume. We, therefore, have found that combusting the
Stretford tail-gas is preferable to flaring from an energy utilization,
conservation and environmental standpoint. The environmental
benefit results from increased energy efficiency which reduces the
need to burn additional sulfur-containing fuel. In addition, with
this boiler design, the gasification section of the Great Plains plant
will comply fully with EPA's guidelines for the Control of Emissions
from Lurgi Coal Gasification Plants (EPA-450/2-78-012).
This brings us to our second air emission source, the plant
steam generation system. Several sources of steam generation are
available to the designer of a modern SNG facility, including genera-
tion from coal fines or liquid by-products, recovery from exothermic
processes (such as methane production), and recovery from gasifier
steam jackets. The Great Plains plant will utilize plant byproduct
tar, tar oil, naphtha, and phenols plus the Stretford tail-gas to
generate the steam required above and beyond that recovered in an
extensive in-plant steam recovery, reuse and conservation system.
EPA's new source performance standards for steam generation apply to
this section of the plant. However, the EPA emission standards are
not suited to direct application in the case of Great Plains due to
the innovative energy conservation approaches utilized. First, EPA
has no sulfur emission standard for a sulfur-containing gaseous fuel
such as the Stretford tail-gas. Further, EPA's NOX emission standard
does not consider NOX emission from a liquid fuel (e.g. tar and
tar oil) with a higher entrained nitrogen value than conventional
liquid fuels. Fortunately, the North Dakota State Department of
Health, from the time of our first project announcement, has been willing to
109
-------
evaluate our proposals in detail, carefully considering and balancing
environmental, economic, energy conservation and safety criteria.
After a thorough review with an invitation for public comments, the
Health Department made determinations of 1) best available control
technology for the project, 2) compliance with the federal guidelines
for the Control of Emissions from Lurgi Coal Gasification Plants,
3) compliance with ambient air quality standards and 4) compliance
with the Prevention of Significant Deterioration regulations at the
Class I area 100 kilometers west of the plant site. The North Dakota
State Department of Health, in their 167-page analysis of the Great
Plains Project, found that the facility as proposed would comply with
all federal, state and local air quality regulations. The EPA,
Region VIII, then reviewed the state's analysis and congratulated the
Health Department, stating that their technical effort "may well become
the standard to which new source reviews of this office and the other
Region VIII States are compared".
It is evident that in this case a very thorough evaluation of
a new synthetic fuels facility was completed by means of a "case-by-
case" review. The existence of new source performance standards,
pollution control guidance documents or the like could very possibly
have made permitting of the facility more difficult due to the inherent
inflexibility of the regulations and the restrictions they impose when
considering special situations and innovative techniques. A case in
point is EPA's 1979 Environmental Assessment Report on Lurgi Coal
Gasification Systems for SNG (EPA-600/7-79-120). This report contains
an excellent overview of the environmental aspects of a Lurgi SNG
facility. However, when applying the EPA guidelines and new source
no
-------
performance standards, the report incorrectly states that the Great
Plains Project (refered to in the report as ANG) exceeds federal
standards for S02 emission from the gasification section, exceeds the
federal standards for SC>2 emissions from the steam and power genera-
tion section, and exceeds the federal standard for TSP emission from
the steam and power generation section. This is after the Health
Department and Region VIII certified that the facility is in 100%
compliance with all regulations. The lesson to be learned is that
hard-and-fast standards are not appropriate for complex emerging
technologies such as those found in the synthetic fuels industry. A
very thorough case-by-case review is highly preferable until such
time as sufficient operating data on modern facilities have been
compiled and verified and valid standards can be developed.
The other two sources of air emissions are 1) coal handling and
2) incinerators, flares and miscellaneous sources. Particulate emissions
from the coal handling facilities will be controlled through the use
of covered conveyors and baghouse collectors at all transfer points.
EPA new source performance standards for Coal Preparation Plants
applies to this section of the plant. The low-volume intermittent
gaseous streams in the plant will be incinerated where such treatment
is appropriate and does not represent a safety hazard. Start-up gases
and expansion gases from gas-liquor separation will be routed to a
start-up incinerator for controlled combustion. The majority of the
coal lock-gas will be recovered, desulfurized and reused, resulting
in a very small vent, less than 2% of the total lock gas volume. The
111
-------
flare system is the primary plant safety system and is capable of combus-
ting the entire gas flow from either train of the plant in the event of an
emergency shut-down of a gas processing unit.
The water pollution control systems included in the Great Plains
Project are designed to eliminate the discharge of process wastewaters
to surface streams. A complex recycle and reuse system will be
employed within the plant followed by utilization of the plant cooling
tower, multiple effect evaporators and a liquid incinerator to
concentrate, then destroy all organic components of the plant waste-
water. A brine solution from the regeneration of demineralizers and
softeners will be disposed of via a deep well into an aquifer where
the natural water quality is six times more brackish than the waste
stream. Stormwater runoff will be collected in sedimentation ponds
prior to discharge and the coal pile has been covered to minimize
suspended particulate loading from that potential source. Sanitary
wastewater will be treated in a package plant and the effluent will
be discharged to the runoff pond which will provide tertiary treat-
ment in the form of a polishing pond prior to discharge.
This system for handling liquid effluents was selected over other
alternatives such as solar evaporation ponds, activated carbon
adsorption and biological treatment after detailed engineering,
economic and environmental review revealed that the present system
is the best suited for our particular plant design and location.
Solid waste from the gasification plant consists primarily of
coal ash from the gasifiers and from the liquid incinerator. Approxi-
mately 2200 tons per day of ash will be generated by the full plant.
112
-------
This waste does not qualify as hazardous under the EPA's extraction
procedure toxicity test and is further exempted as a coal combustion
waste. Nonetheless, care will be taken in selecting and developing
disposal areas within the mine. Disposal will be limited to dry
locations where natural or emplaced clay barriers will prevent the
formation and migration of ash leachates. In west-central North
Dakota, the natural soil and groundwater conditions exhibit a rela-
tively high pH and acid formed by the oxidation of pyrites is quickly
buffered. Acid conditions and the resulting leachate problems
evidenced in other parts of the country are not encountered in the
Northern Great Plains region.
The in-mine disposal technique proposed to be used at the Great
Plains Project represents a considerable improvement over the primary
alternative which is ash sluice ponds. In-mine disposal eliminates
four problem areas that occur with sluice ponds: 1) the commitment
of large acreages for ponds, 2) the need to dispose of decanted water,
3) the need to reclaim the filled pond to a useful end-use and
4) the need to protect the groundwater from infiltration of sluice
water. For these reasons, it is felt that proper in-mine disposal
represents state-of-the-art in solid waste disposal.
In the area of employee health and safety, the Great Plains
Project is designed to protect the worker from the potentially
hazardous substances that are present in all synthetic fuels facili-
ties. Containment of these substances and a good work practices
control program coupled with a thorough medical surveillance program,
are the essential elements of the occupational health and safety
113
-------
program. Our consulting agreements with the South African Coal Oil
and Gas Corporation, Ltd. of South Africa enabled our engineers to
discuss possible solutions to various air, water and process emissions
and to select the most efficient means of control based on years of
operating experience. As you may know, the Sasol plant was visited by
an investigative team from the National Institute for Occupational
Safety and Health (NIOSH) in 1977. The plant was given a clean bill
of health by that group, a remarkable achievement for a facility that
has been in operation for over 25 years.
In summary, we are confident that the Great Plains Coal Gasifi-
cation Project can be built and operated in compliance with all
requirements for environmental, health and safety control. In addition,
our monitoring and surveillance programs will go beyond that required
by regulation and will include data gathering programs necessary to
develop a data base for future synthetic fuels projects. As always,
we pledge our cooperation and assistance to the EPA and the other
federal and state agencies wherever possible and we look forward to
sharing the non-proprietary portions of our operating data so that
sound substantiable regulations may be developed.
On behalf of the partners of the Great Plains Gasification
Associates, I appreciate the opportunity to speak before you today
and wish to extend an invitation to each of you to come to Beulah,
North Dakota in 1984 and visit the first operating commercial-sized
synthetic fuels plant in the United States.
Thank you.
114
-------
Session II: ENVIRONMENTAL ASSESSMENT;
DIRECT LIQUEFACTION
D. Bruce Henschel, Chairman
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina
115
-------
Preliminary Results of the
Fort Lewis SRC-II Source Test
Jung I. Kim, Ph.D.
David D. Woodbridge, Ph.D.
Hittman Associates, Inc.
9190 Red Branch Road
Columbia, Maryland 21045
116
-------
Introduction
The SRC pilot plant was designed to convert coal into a
low sulfur and ash product in either solid or liquid form.
The process that yields the solid product is called SRC-I,
while the liquid product mode is referred to as SRC-II.
This paper deals with the SRC-II operation.
The primary objective of this study is to evaluate
environmental implications of the SRC-II technology on the
basis of data obtained from the Fort Lewis SRC-II pilot
plant. Efforts were made to sample and analyze non-site-
specific streams that could be scalable to a full-size
commercial plant. Although the characteristics of some of
the streams collected may differ somewhat from their commer-
cial counterparts, they may provide general qualitative
information on pollutants expected from a commercial facility.
Data obtained from this pilot plant must be carefully evalu-
ated in order to determine their applicability and scalability
to a commercial-size facility.
This paper first establishes basic similarities and
differences in process and operation between the Fort Lewis
SRC-II pilot plant and an expected commercial SRC-II facility.
It then discusses an SRC-II sampling and analytical program
being conducted by Hittman Associates, Inc. (HAI), and
provides the data obtained thus far.
SRC-II Process Description
The SRC-II process involves non-catalytical treatment
of coal with hydrogen at an elevated temperature (45A°C) and
pressure (13.8 MPa). In this process, a dried, pulverized
coal is mixed with a process-produced recycle slurry to form
a coal slurry. The coal slurry is then mixed with hydrogen
and pumped through a preheater to a reactor where coal is
dissolved and hydrocracked, liberating gases such as H^S,
J^O, NHo, C02, and hydrocarbons. The reactor effluent
enters a series of pressure let-down vessels where process
gases and liquid are separated. The gases are sent to an
acid-gas absorber unit for the removal of H^S and CC^ • The
HjS is further processed into a salable sulrur product.
Light hydrocarbons and unconverted excess hydrogen leaving
the absorber are cryogenically separated; the hydrogen gas
is recycled to the process and the light hydrocarbons are
processed into salable product gases. The light liquid
stream is fractionated into naphtha and fuel oil. The
product slurry is split into two streams. One of the streams
117
-------
is sent to the front end as recycle slurry to be mixed with
feed coal, while the other stream passes to vacuum distilla-
tion where fuel oil is further recovered. The high-ash and
low-sulfur residue (referred to as vacuum bottoms) from the
vacuum distillation tower is sent to a gasifier for the
production of make-up hydrogen or synthetic gas.
The Fort Lewis SRC-II pilot plant (Figure 1) does not
have some of the process features described above. Many of
the processes it employs are unique to the pilot plant and
therefore would differ from those of an anticipated commer-
cial facility. These differences are given in Table 1.
Only if and when these differences are fully understood, can
the data obtained be successfully extrapolated to the
commercial operation to provide pollutant characterization
and control technology information.
Sampling and Analysis Program
Background
HAI, under contract to the U.S. Environmental Protection
Agency, began an SRC-II sampling and analysis effort in
March 1978. The purpose of this effort was to evaluate the
SRC wastewater treatment system and characterize the SRC-II
products. Because of the important role of coal liquefaction
to our nation's energy self-sufficiency and the environmental
implications of this technology, this initial effort soon
evolved into a comprehensive environmental assessment program
to measure pollutants associated with the SRC-II operation.
This program uses the EPA phased sampling and analytical
approach to characterize emission and effluent streams from
various processes and control units.
The first phase (Level 1) environmental assessment be-
gan in February 1979, and is now completed. Environmentally
significant streams and their chemical components were
identified, screened, and prioritized for more detailed
second phase (Level 2) analysis. However, the SRC-II pilot
plant underwent major system modifications and since then
experienced start-up problems, which delayed the planned
phase 2 sampling program. Meanwhile, the original SRC-II
operation schedule was altered and the feedstock used (Pow-
hatan No. 5) during the Level 1 sampling period was replaced
with Powhatan No. 6. As a result of the process modifica-
tions and coal type change, the original Level 2 test plan
was revised to include Level 1 and Level 2 sampling to be
performed simultaneously to obtain the required sequential
data. This combined Level I/Level 2 sampling and analytical
effort began in March 1980. Analyses of these samples are
118
-------
RECYCLE GAS
303
FLARE
COAL-
DISSOLVER AND
HIGH PRESSURE
FLASH DRUM TO
QUENCH
100's = LIQUID STREAMS
200'5 = SOLID STREAMS
300'5 = GASEOUS STREAMS
US
MIDDLE DISTILLATE
116 HEAVY DISTILLATE
Figure 1. Overall process flow diagram of the Fort Lewis
SRC-II pilot plant.
-------
TABLE 1. THE FORT LEWIS SRC-II PILOT PLANT
Vs. COMMERCIAL SRC-II FACILITIES
Fort Lewis Facility
Commercial Facility
Affected Stream
Characteristics
No gasification of
Vacuum Bottoms.
Vacuum Bottoms
currently stored
for outside dis-
posal.
A portion of Sour
Water is being re-
cycled to provide
a quenching stream.
Middle and Heavy Dis-
tillates produced
separately.
Sour Water is not
treated but diluted
with non-process
water prior to
treatment.
Fuel gases and
purged hydrogen are
being flared.
No hydrotreating
of product fuels
including Naphtha.
Vacuum Bottoms will
be gasified, and re-
sultant slag will be
landfilled.
Oil quenching is
currently under
consideration.
Blended to yield
fuel oil.
Sour water will be
pretreated to recover
NH_, H_S, and phenols.
Fuel gases will be
recovered. Cryogenic
hydrogen separation
obsoletes hydrogen
purge.
Products may have
to be upgraded.
No emissions and waste
discharge associated
with Vacuum Bottoms
solidification. However,
in commercial practice
slag and quenching water
from gasification may
pose disposal problem.
Alteration in process
sour water character-
istics expected.
Will not affect overall
pollutant balance. How-
ever, chemical con-
stituents in the fuel
oil may vary depen-
ding on the blend ratio.
The pretreatment of sour
water will affect the
stream entering the waste-
water treatment system.
Consequently, different
treatment process may have
to be considered.
Flare input stream is not
representative of that
of commercial facilities.
Lower heteroatomic
compounds in the hydro-
treated products.
120
-------
still in progress. Preliminary data obtained from selected
sampling streams are presented in this paper.
With the exception of analyses which called for non-
composite sampling, such as volatile organic analysis, each
aqueous or solid stream was sampled three times per day, 8
hours apart, for six sampling days, and was composited to
constitute a single representative sample for a given
stream. All aqueous samples were preserved according to EPA
procedures, by organic extraction, or by refrigeration.
Product streams were sampled once a day for six sampling
days. In addition, a total of 36 samples were collected
from four streams - wastewater treatment plant influent and
effluent, and middle and heavy distillates - in order to
perform a comprehensive statistical evaluation of process
variability, sampling and analytical variability-
Gaseous streams were sampled once or twice per stream
during the entire sampling period. Inorganic and organic
species were collected in evacuated glass flasks, teflon
bags, and Tenax GC and XAD-2 sorbent columns. Impinger
bottles were used for species such as ammonia, cyanide, and
volatile elements which could be collected and analyzed more
effectively by wet-chemical or other methods. Collected
volatile species such as H^S, CO, COS, S0~ , and mercaptans
were analyzed immediately using onsite GC columns equipped
with species-specific detectors. Tenax GC columns were
thermally desorbed and analyzed on a GC/MS system for the
volatile species lost during extraction. Higher boiling
organic compounds were extracted with methylene chloride in
a Soxhlet extraction apparatus and subjected to GC/MS analy-
sis. Table 2 presents the environmental source tests being
performed on the collected SRC-II stream samples.
Table 3 shows metals present in dried coal ( 2 percent
moisture) with their distribution among various products/
by-products and their recycle process water (process sour
water). As expected, most of the non-volatile metals pre-
sent in the feed coal find their way into the vacuum bot-
toms. Use of the vacuum bottoms for a commercial gasifier
will generate slag material which consists primarily of
inorganic elements. Leaching characteristics of this material
must be thoroughly investigated for the development of a
safe method of disposal. This slag contains high levels of
metals such as aluminum, iron, and titanium (see Table 3).
The recovery of these elements may provide a potential
disposal alternative. High levels of vanadium, sodium,
iron, and other elements present in the elemental sulfur do
not originate in the feed coal, but rather in the Stretford
solution. Currently, the Fort Lewis plant produces unwashed
sulfur which is transported for outside disposal. The
121
-------
TABLE 2. SUMMARY OF TESTS TO BE PERFORMED
ON THE SRC SAMPLES
101 Condensed water from
coal dryer
T
102 Lean DEA Solution
103 Recycle Process water
104 Flare Knockout Condensate
105 Solvent Fractionation Area Runoff
106 Dissolving and Separation Area Runoff
107 Stretford Pad Runoff
108 Feed Cooling Water
109 Cooling Water
110 Wastewater Treatment Plant Influent
111 Bio-unit Influent
112 Bio-unit Effluent
113 Sand Filter Effluent
114 Naphtha
115 Middle Distillate
116 Heavy Distillate
201 Pulverized i Dried Coal
202 Recycle Slurry
203 Vacuum Bottoms
204 Elemental Sulfur
205 Flottazur Skimmings
206 Clarifier Sediment
207 Digester Contents
301 Slurry Blend Tank Vent
302 Purge Hydrogen to Flare
303 Light Hycrocarbons from Naptha
Flash Drum to Flare
304 Off Gas from Stretford Unit
305 Stretford Oxidizer Tank Vent
306 Hot Well Tank Vent
307 Input Stream to Flare System
122
-------
TABLE 3. METALS PRESENT IN FEED COAL, PRODUCT/BY-PRODUCTS, AND PROCESS SOUR WATER
IN3
CO
Feed Coal
(Pittsburgh Seam,) Vacuum Elemental
Powhatan #6 Bottoms Sulfur
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Phosphorus
Potassium
Selenium
Silicon
Silver
Sodium
Strontium
Tin
Titanium
Tungsten
Uranium
Vanadium
Zinc
*Determined by
1.3%
<15
11
44
- 0.3
_
_
2.5
0.267.
18
3.1
12
2.3%
•=0.1
460
34
-
5.8
12
310
0.13
-
-
_
410
69
< 3
630
-
-
34
13
flameless
**Expected concentrations
Concentrations
in wg/g,
2.8%
(12%)**
23
96
<0.3
_
-
<2.5
0.53%
52
7.6
45
4.9%
(21%)**
<0 1
o ; 1%
74
-
17
35
660
0.27
-
_
_
730
150
<3
0 . 147.
(0.6%)**
-
-
77
39
1.4
0.3
^0 .6
o!i 2
<0.006
< 1 . Q
<0.02
<0.05
14.5
0.3
<0.04
2.0
110
2.1
0.7
0.8
-
0.8
0.4
1.2
20
-
2.5
<0.06
0 . 1%
0.1
<0.06
0.1
-
-
34
4.0
AAS . All other elements were
present in commercial
gasifier
Heavy Middle
Distillate Distillate
7
<0
0
0
<0
< \
0
<0
2
11
<0
0
51
<0
0
0
-
0
4
<0
0
0
3
<0
1
0
<0
0
-
-
0
0
determined
slag.
7
3
006*
04
006
0
34
05
1
04
17
16
35
86
77
1
8
9
003
7
06
1
05
06
35
07
1
by ICP.
<0
0
0
0
<0
1
0
<0
0
0
<0
<0
0
<0
0
0
-
<0
<0
<0
0
0
1
<0
5
C0
<0
<0
-
-
<0
0
.4
.003*
. 004*
.02
.006
.0
.1
.05
.35
.39
.04
.03
.69
.16
.35
.04
.08
.05
.8
.19
.002
.75
.06
.1
.002
.06
.012
.02
.03
Naphtha
<0
2 x
0
<0
<0
< 1
<0
<0
0
<0
<0
0
0
<0
0
<0
-
<0
<0
<0
<0
6 x
2
<0
0
<0
<0
<0
-
-
<0
0
4
ID'4*
006*
002
006
0
04
05
18
06
04
18
44
16
08
006
08
05
8
02 ,.
10°
3
06
84
002
06
012
02
35
Process Sour
Water
<0
5 x
0
0
<0
.15
ID'4*
.007*
.008
.003
'0. i
230
<0
1
<0
<0
0
2
<0
0
0
-
<0
<0
<0
0
3.4
3
'0
1
0
<0
<0
-
-
<0
0
.025
.1
.03
.02
.015
.1
.08
.08
,03
.04
.025
.13
•4 -4
x 10
.4
.03
.0
.004
.03
.006
.01
.03
unless otherwise designated.
-------
levels of metals found in the products are generally related
to product volatility. Generally, levels of trace elements
present in the heavy distillates are high when compared with
either the middle distillate and naphtha. Heavy distillates
are least volatile, middle distillates are next, and naphtha
is most volatile. Process sour water contains low levels of
metals, with the exception of boron. High pH and sulfide
appear to be responsible for low metal concentrations in
this stream.
Table 4 shows the reductions in various water quality
parameters and trace elements from the wastewater treatment
system. The wastewater treatment system is depicted in
Figure 2. On the average, a 20 to 93 percent reduction in
metals was accomplished by the treatment process. The table
also shows trace elements found in the clarifier sediment
and flottazur skimmings. Trace element analyses on RCRA
extracts of these streams are currently being performed.
Table 4 reveals that a high level of phosphorus is entering
the treatment plant. The high level of phosphorus is attri-
buted primarily to the blowdown from the cooling tower and
boiler systems.
Figure 3 shows the effectiveness of this treatment in
reducing organic class compounds. This figure, which was
derived from the previous Level 1 data from the SRC-II
operation with Powhatan No. 5 coal, indicates that the
treatment system appears to be effective in lowering levels
of organics such as aliphatic hydrocarbons, benzene and
substituted benzenes, and fused polycyclic hydrocarbons.
The effectiveness of the treatment system in reducing biologi
cal toxicity is shown in Figure 4. This figure was also
derived from the previous Level 1 data. Neither the influent
nor effluent demonstrated toxicity on the Ames or the rodent
tests.
Analytical results of the SRC-II gaseous streams are
shown in Table 5. While the slurry blend tank vent, the
oxidizer tank vent, and the hotwell tank vent are emission
streams discharged directly into the atmosphere, the Stret-
ford offgas stream is sent to the flare system. Although
the existing flare system receives emissions from the various
pressure relief vessels, major input sources are the purged
hydrogen, offgas from the Stretford unit, and light hydro-
carbons from the naphtha scrubbing unit. Since fuel gases
were not recovered but were being flared at this pilot
plant, the characteristics of these flared gases would be
quite different from those of a commercial facility. From
an operational standpoint, the pilot plant flare unit is
very similar to a commercial flare system operating under
plant upset conditions.
124
-------
TABLE 4. CHARACTERISTICS OF WASTE STREAMS FLOWING
THROUGH THE WASTEWATER TREATMENT SYSTEM
cn
Ammonia
Sulfide
Cyanide
COD
Aluminum
Antimony*
Arsenic*
Barium
Boron
Calcium
Chromium
Copper
Iron
Magnesium
Manganese
Nickel
Phosphorus
Potassium
Selenium*
Silicon
Sodium
Strontium
Titanium
Vanadium
Zinc
Influent
61
5.1
0.12
950
22
0.002
0.03
0.09
1.9
19
0.03
0.2
45
5.4
0.06
0.025
9.1
4 .
3 x 10"4
23
140
0.11
0.06
1.2
0.9
Sand
Filter
Effluent
46
0.4
0.1
300
1.6 ,
2 x 10'4
0.006
0.04
0.6
15
<0.03
0.04
8.5
4
0.04
0.025
0.9
2-5 4
2 x 10 4
12
100
0.07
<0.006
0.12
0.1
Treatment
Efficiency
25
92
17
(68)
93
90
80
56
68
21
-
80
81
26
33
-
90
38
33
48
29
36
>90
90
89
Primary
Clarifier
Sediment
mg/g
1770
51
0.04
-
5.7
0.12
0.4
72
1.1
0.07
0.06
0.4
4.6
0.06
0.2
2.1
1.8
Flottazur
Skimmings
dry base
1860
29
0.02
-
3.3
0.07
0.2
44
0.7
0.04
0.04
0.26
3.1
0.04
0.12
1.2
1.1
^Determined by flameless AAS. All other elements were analyzed by ICP.
-------
CLARIFIER AND 205
INLET
WATER *
SURGE
RESERVOIR
110
^" i-LUIHI IUIN "^
SKIMMING
CLARIFIER
y-
DIS
FLC
(FL
SOLVED AIR
ITATION UNIT
OTTAZURTM)
206 1
SETTLED CLARIFIER
^HYDROCARBONS SEDIMENT
PRODUCT
SOLIDIFICATION
WATER
1 NH ADDITION
HYDROCARBON
SAND
FILTER
SANDJ
FILTER
F
113
112
BIOLOGICAL
UNIT
STEAM"1
ADDITION
n
HOLDING
TANK
CHARCOAL
FILTER
BACKWASH WATER
U- ,
2°7
L
NOTE: 100':
200':
LIQUID SAMPLES
SOLID SAMPLES
FILTER
BACK-
WASH
TANK
DISCHARGE
]WHEN WASTEWATER IS LOW IN NUTRIENTS DURING THE SRC PROCESS PLANT
SHUTDOWN
2WHEN THE TEMPERATURE OF WASTEWATER IS LOW FOR NORMAL BACTERIA
ACTIVITY
3THE SPENT BACKWASH WATER IS ROUTED TO THE SURGE RESERVOIR
4CHARCOAL FILTER WAS NOT IN USE DURING THIS SAMPLING
Figure 2. Overall flow schematic of the SRC pilot plant
wastewater treatment system showing sampling points.
126
-------
10
5 -
10
k _
a>
<
f£.
O
o
10
3 _
10
2 _
10
S INFLUENT
.
~ (] EFFLUENT
-
S
s
s
s
s
s
s
s
s
s
s
V
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
^
V
s
y
s
s
s
V
s
s
s
V
V
s
s
s
s
s
s
s
s
V
s
V
\
^
s
\
s
s
S
s
s
V
s
s
s
s
s
s
s
x
,\
\
V
V
s
s
s
s
V
s
s
s
s
s
s
s
PM
H
^
1
M
ro
s
s
^
V
V
1
•i
rm
V
s
s
s
V
s
S
s
s
V
V
s
s
V
s
N,
s
s
V
s
s
s
s
s
V
s
s
s
V
s
s
s
s
s
s
V
s
s
s
s
••
R
s
s
s
s
s
s
s
s
s
s
s
s
s
s,
s
s
s
N
s
s,
s
s
V
s
s
s
s
s
s
s
s
s
s
s
s,
s
V
s
s
s
s
s
s
s
s
N,
M
T
s
s
s
s
V
s
s
s
s
s
s
s
s
s
s
s
V
s
s
s
s
s
s
s
V
s
s
V
s
s
s
M
^
1
s
$
s
1
s
^
s
M
^•1
1
M
1 _
22
23
2k 25
1 3 7 8 15 18 21
ORGANIC CATEGORY
1 - ALIPHATIC HYDROCARBONS
3 - ETHERS
7 - ALDEHYDES, KETONES
8 - CARBOXYLIC ACIDS AND DERIVATIVES
15 - BENZENE AND SUBSTITUTED BENZENE HYDROCARBONS
18 - PHENOLS
21 - FUSED POLYCYCLIC HYDROCARBONS
22 - FUSED NON-ALTERNANT POLYCYCLIC HYDROCARBONS
23 - HETEROCYCLIC NITROGEN COMPOUNDS
2k - HETEROCYCLIC OXYGEN COMPOUNDS
25 - HETEROCYCLIC SULFUR COMPOUNDS
(Based on the average concentrations of three independently taken grab
samples on February 11, 12, and 16, 1979, for the influent, and 2 in-
dependent grab samples taken on February 12 and 16, 1979, for the
effluent).
Figure 3. Levels of organics present in the
treatment plant influent and effluent.
127
-------
HIGH h
ro
oo
g MODERATE
>-
o
LOWk
-4 1
£3
0
-1
o
LA
O
_l
\
IS
O
_J
-2
-3
Y////////////////////////////////A
«j
_i
UJ
0
0
3Z
O
INFLUENT
EFFLUENT
rq
//////////////////////////////////A
a:
UJ
1—
^
X
CO Z
LU CO
(£. —
U. U.
s
s
C£.
UJ
< <
—
I
—
a:
LU
^
§
X Z X Ul
co u: co <
ui a.
(X. <
LU U
DC _l
U. Q Ll- <
O
LA
O
-H
H HIGH
x
o
-2
MODERATE 3
LOW §
Figure 4. Reduction in biological toxicity by wastewater treatment.
-------
TABLE 5. COMPOSITION OF THE SELECTED SRC-II GASEOUS STREAMS
IN3
Slurry Blend
Tank Vent
Parameter (2-Day Average)
C^s 360
C2's 280
C3's 230
C4 ' s 280
C5's 1,400
C6's 1,400
H2S 1,020
COS 3
Methyl Mercaptan 23
Ethyl Mercaptan ND
Nos. of unidentified
Sulfur Species ND
CO ND
NH3 11
HCN ND
Species identified
by GC/MS phenol
xylenes (0, M, & P-)
Benzenes (C-, C, & C.-)
23 4
benzofurans (methyl-)
naphthalenes (C,, C2 & C -)
phenanthrene/anthracene
pyrene/f luoranthene
Stretford Offeas Oxidizer Tank Vent
1.4 x 104 ND
6,200 ND
5,000 ND
1 , 300 ND
ND ND
ND ND
5,900 ND
40 ND
400 ND
40 ND
3 ND
2.5 x 104 ND
120 8
0.1 ND
xylene methyl benzofuran
benzenes (C., & ^,-) naphthalene (C, , C- , &
naphthalene (C , C -) C -)
tetralin fluorenes (C & C--)
phenanthrene /anthracenes phenanthrene/anthracene
(methyl-) (methyl-)
pyrene/f luoranthenes pyrene/f luoranthene
(methyl-)
Hotwell
Tank Vent
2
290
50
50
29
12
ND
ND
ND
ND
ND
ND
ND
ND
benzene (C3, C,-)
naphthalene
tetra/methyl benzo-
furan
methyl teralin/
C,j -benzofuran
Inout to Flare
1.8 x 104
2.9 x 104
3.6 x 104
1.3 x 104
5,000
3,000
4.2 x 104
40
40
220
3
5 x 104
88
0.04
cyclopentene
cyclohexanes
phenols
cresols
xylenols
xylenols
benzenes (C. , C_ & C.-)
1. J 4
toluene
furan
xylenes (0, M, S, P-)
benzofuran
naphthalenes
(clf c2, c3. & c,->
fluorenes (methyl-)
phenanthrene/anthracene
pyrene/f luoranthene
tetralin
-------
The Stretford offgas and the oxidizer tank vent are the
Stretford process-related streams. The slurry-blend tank
vent was designed to remove various fumes and vapors gener-
ated during the slurry/coal mixing. These pollutants are
cooled and further condensed by a steam ejector prior to
atmospheric release. Because sampling occurred at a point
before the steam ejector, the information on pollutant
characteristics shown in Table 5 is of limited value. For
the hot well tank vent, the sampling probe was not placed in
the vent duct, but rather, over the open end of the vent.
Furthermore, the vent cycle could not be determined; thus,
the concentration data shown in Table 5 provide only compara-
tive quantitative information on the identified pollutant
species. Table 5 shows the organic species identified by
GC/MS. Compounds present in the streams did not vary greatly
Quantitative information on the identified species is not
yet available, but is expected to be in the yg/m range. It
should be noted here that accurate sampling of high molecular
weight compounds was difficult because samples could only be
taken from existing sampling valves which were connected
through a long, unheated sampling line to the main process
streams. As a result, many high boiling organic compounds
probably condensed out, and therefore, were not collected at
the outlet.
For the selected liquid stream samples, volatile organic
compounds were identified by GC/MS using the purge and trap
technique (Table 6). Although the treatment plant influent
contained volatile compounds which were collected from
various sources, no detectable amounts of these compounds
were present in the effluent. This probably resulted from
atmospheric loss in the aeration unit rather than actual
biological degradation of these substances.
Table 7 shows several important water quality para-
meters of the recycle process water. This stream was char-
acterized by extremely high alkalinity with very low hard-
ness and low levels of alkali metals. Actual COD for this
stream should be somewhat higher than the value shown in the
table. Volatile organic substances, including some phenolic
compounds, were believed to be lost by purge gases (mostly
HpS) formed during acidification for sample preservation.
Tfie phenol level shown in the table was somewhat higher than
expected (normally about 0.7 percent). Since a portion of
this stream is recycled to the process, the phenol level at
a given time is dependent on the recycle ratio, assuming
that all other process conditions are constant.
130
-------
TABLE 6. VOLATILE ORGANIC COMPOUNDS PRESENT IN THE
SELECTED FORT LEWIS SRC-II STREAMS
Recycle Process
Water
Pyrroles
Furans
Pyridines
C, Hydrocarbons
C Hydrocarbons
C, Hydrocarbons
Benzene
Ethyl Benzene
Toluene
Xylene
Unidentified - CN
Chloroform
8.6
0.3
0.21
1.8
0.98
1.1
ND
ND
(4.2-16)
(0-0.8)
(0.05-0.3)
(0.4-4.2)
(0.2-1.4)
(0.5-2.0)
11 (5-17.3)
0.64
8 (4
(0-1.5)
.6-11.3)
Solvent Frac- Wastewater
Condensed Water tionation Area Treatment Sand Filter
From Coal Dryer (Fugitive Effluent) Plant Effluent Effluent
0.005 (0-0.03) 0.
0.007 (0-35) 2.
0.05 (0-0.08) 0.
0.06 (0-0.1) 0.
0.
0.21 (0.04-0.5) 0.21 (0.08-0.4) 0.
0.3 (0-0.7) 0.26 (0.07-0.2) 0.
0.33 (0.1-0.5) 1.3 (0.34-3.7) 0.
0.13 (0-0.5)
0.
02
1
08
05
03
06
06
24
01
(0-0.1)
(0.04-3.7)
(0.02-0.2) None
Detected
(0.02-0.1)
(0-0.06)
(0-0.1)
(0-0.1)
(0.15-0.3)
(0-0.06)
co
NOTE: Concentrations in mg/L.
The numbers in parentheses represent the ranges of concentration variation over a 6-day sampling period.
ND = Not Detected
-------
TABLE 7. CHARACTERISTICS OF RECYCLE
PROCESS WATER (SIX-DAY AVERAGE)
pH 9.0
Alkalinity (as CaC03) 97,000 mg/L
Hardness (as CaCOo) 10 mg/L
Ammonia (as N) 36,000 mg/L
Sulfide (as S) 30,000 mg/L
Cyanide (as CN) 1.3 mg/L
Chemical Oxygen Demand 26,000 mg/L
(as 02)
Phenol 7,600 mg/L
Cresols 2,850 mg/L
Xylenols & C^ phenols 1,250 mg/L
C Phenols 2,200 mg/L
132
-------
Conclusions
More detailed analytical data and plant process infor-
mation are still forthcoming. The results discussed herein
are preliminary in nature, and require further confirmation
and expansion as more data become available.
Most of the metals present in feed coal were almost
entirely recovered in the vacuum bottoms. Use of this
material for a commercial gasifier will generate slag, con-
sisting almost entirely of inorganic elements. Detailed
leaching characteristics must therefore be investigated in
order to develop a safe method of disposal. The recycle
process water contained mostly ammonia, sulfide, and phenols,
and was essentially free of metals, except for boron. The
boron level in this stream was over 200 mg/L. Because at
levels exceeding 1 mg/L, boron has deleterious effects on
the human body and the ecosystem, it may be necessary to
remove it, along with ammonia, sulfide, and phenols, from
this stream. In the coal drying process at the Fort Lewis
pilot plant, moist air from the coal dryer is cooled with a
dehumidifier and the condensed water is sent to wastewater
treatment. This stream contains a number of pollutants of
environmental significance. Although their levels are
relatively low, these pollutants may have to be controlled
since, in commercial facilities, the moist air resulting
from coal drying is expected to be discharged as vapor into
the air.
Due to several process upsets, the wastewater treatment
samples may not fully reflect normal operating conditions.
ACKNOWLEDGMENTS
The authors would like to express their appreciation to
the U.S. Environmental Protection Agency Energy Assessment
and Control Division, Industrial Environmental Research
Laboratory, for supporting this study. The authors would
also like to acknowledge the Hittman Associates Laboratory,
who performed the analytical work for this study.
133
-------
CHEMICAL/BIOLOGICAL CHARACTERIZATION OF SRC-II PRODUCT
AND BY-PRODUCTS
W. D. Felix, D. D. Mahlum, W. C. Weimer
R. A. Pelroy and B. W. Wilson
Pacific Northwest Laboratory
Richland, WA 99352
ABSTRACT
Biological and chemical tests in concert with engineering analyses of
plant operations have been used to provide data for the assessment of health
and environmental effects of a mature coal liquefaction industry. In this
report, we describe the methodology whereby biological testing is used to
guide the chemist in the analysis of fractions of selected pilot plant mate-
rials. The principal components of an unmodified distillate blend from the
SRC-II process are two-and three-ringed aromatic and heteroatomic species.
Phenolic and pclynuclear aromatic components are generally present at higher
levels than expected in petroleum crudes. Biotesting, with the Ames test as
the primary first tier method, revealed mutagenic activity. Chemical frac-
tionation in conjunction with Ames testing implicates the primary aromatic
amines as the compound class of primary concern. Chemical biotesting of a
hydrotreated distillate blend showed a significant reduction of the primary
aromatic amines as well as polynuclear aromatic hydrocarbons. Hydrotreating
also can result in the reduction of sulfur- and oxygen-containing compounds,
e.g., thiophenes and phenols.
Prepared for the U.S. Department of Energy under Contract DE-AC06-76RLO 1830
134
-------
CHEMICAL/BIOLOGICAL CHARACTERIZATION OF SRC-II PRODUCT
AND BY-PRODUCTS
Dependency of the United States upon foreign oil has led to the rapid
implementation of programs oriented toward the development of new energy
technologies. Simultaneously with the development of these synfuel processes,
it is necessary to perform studies which will determine the potential health
and environmental effects associated with the given technology. The purpose
of this paper is to discuss the method and approaches used at the Pacific
Northwest Laboratory in providing chemical and biological data dealing with
SRC (Solvent Refined Coal) materials. The approach we have taken is designed
to provide meaningful health effects data to the technology developers within
the time frame which permits technology changes to be made optimally to ame-
liorate potential problem areas.
In evaluating the health effects associated with a coal conversion in-
dustry, it is essential that the chemist and biologist coordinate their re-
search efforts toward a common goal. The usual scenario, however, results in
the biologist asking the chemist to give him the compounds or materials with
which he should be performing his assays. The chemist, on the other hand, asks
the biologist which compounds are biologically active in order to orient his
analyses toward these selected materials. The end result is usually one of
utter frustration and mutual distrust leading to the confirmation as far as
the chemist is concerned that the biologist doesn't really know what he is
doing. The biologist, of course, already knew that about the chemist.
The problem is that the chemist is oriented toward the precise measure-
ment of specific elements or compounds. Given a defined compound, a chemist,
in many cases, can measure to femtogram levels. However, in the early stages
of a developing technology such as coal liquefaction, the given compounds of
concern have not yet been identified by the chemist nor has the biologist de-
fined those materials which are biologically active. The chemist is thus
faced with a horrendous task. He has in front of him what amounts to
Beilstein's bucket of compounds and the effects with which the biologist is
concerned may involve compounds whose toxicity or biological effects are so
potent that miniscule quantities in this milieu of compounds may indeed be
135
-------
important. On the other hand, the engineer, who is concerned about the de-
velopment of the process, usually doesn't give serious consideration to the
problems of controlling his processes at micro levels. Yet, as we'll see in
this paper, changes in the process will significantly affect the biological
and chemical response of end products present at extremely low concentrations.
The evaluation, therefore, of the biological impact of a given process re-
quires effective coordination among the activities of the biologist, the
chemist, and the engineer. In this paper, we will describe how this inter-
action has led to the definition of specific compounds of probable concern
within the SRC process. Interaction with engineering personnel has led to
the logical investigation of process parameters which may directly impact
biological activity in coal liquefaction materials. One of the results of
such interaction at Pacific Northwest Laboratory has been the identification
of primary aromatic amines as compounds of principal concern. Hydrotreating,
as will be seen, leads to a reduction of the biological activity of the SRC
materials.
Chemical and biological characterization studies at the Pacific North-
west Laboratory have included GC, GC/MS, LC/MS analyses, specialized separa-
tions procedures for providing biological testing materials, microbial muta-
genesis, in vitro mammalian cell toxicity and transformation assays, epider-
mal carcinogenesis (skin painting), acute and subchronic oral toxicity,
developmental toxicity, dominant lethal assays, inhalation toxicity, and
dosimetry and metabolism studies.
The approach to the study of SRC materials proceeds in basically three
steps: in the first step, an engineering analysis defines the process and
effluent streams in the pilot plant which are expected to be important in
the final developed technology or to which there are expected to be high
levels of occupational or populace exposure; in the second phase, materials
selected in Phase 1 are subjected to biological screening tests and chemical
characterization. Biological activity is usually detected using microbial
assay systems. On evidence of activity, the material is chemically fraction-
ated and the fractions subjected to bioassay. On the basis of the results
of the microbial assay and the chemical characterization studies, materials
are then selected for further study using mammalian cell cultures. The
136
-------
combination of results from cellular and microbial systems along with chemi-
cal characterization are then used to select materials which will be exten-
sively analyzed by animal assays in the third phase. In this phase, mate-
rials are entered into animal systems for study of acute, subchronic,
mutagenic and developmental effects. Certain long-term effect studies are
also designed. Obviously, at each level of testing, other materials are
employed including shale oil, petroleum crudes, other fossil-derived mate-
rials and pure known chemical mutagens and carcinogens for comparative
purposes.
Material used in the studies described were obtained from the SRC pilot
plant at Ft. Lewis, Washington. This pilot plant is operated by the Pitts-
burg and Midway Coal Mining Company for the Department of Energy. Mate-
rials from the pilot plant were selected on the basis of engineering design
data for the projected demonstration plants of both the SRC-I and SRC-II
processes. The selection of materials was based upon one or all of the
following criteria:
a) The material is produced in significant quantity;
b) The material has potential for occupational and/or ecological enviorn-
mental exposure;
c) The material can be obtained in a form which is considered by the best
engineering estimates to be representative of demonstration or commer-
cial level plant operations;
d) The material contains components which are already of known biological
concern.
Consequently, the following process streams in the SRC pilot plant have been
investigated: light oil, wash solvent and process solvent from the SRC-I
process; and light, middle and heavy distillates from the SRC-II process.
The boiling point ranges and specific gravity ranges for these materials are
given in Table 1. The materials in all cases were obtained during equilib-
rium run conditions when the process was being operated for material balance
determination. Given the conditions of pilot plant operations and pilot
plant design objectives, these materials are probably not fully representa-
tive of materials expected from a commercial or demonstration plant.
137
-------
However, the materials do provide information that may be of use in evaluat-
ing areas of toxocological concern whthin a given proposed process slate of
products and effluents.
TABLE 1. Boiling Point Ranges of SRC Materials
Used in Biological Experiments
Process Material Boiling Range (°F) Density
SRC-I Light oil ambient to 380 0.72
Hash solvent 380 to 480 0.96
Process solvent 480 to 850 1.04
SRC-II Light distillate 134 to 353 0.82
Middle distillate 366 to 541 0.99
Heavy distillate 570 to 850 1.10
CHEMICAL AND BIOLOGICAL STUDIES
The Ames mutagenesis assay provides a low cost method for the analysis
of large numbers of samples in preliminary screening activities. In our lab-
oratory, tests are carried out by mixing the test material with the Salmo-
nella TA98 strain in the presence of mammalian liver microsomal enzymes (S9).
By counting the number of revertants (from dependency on histidine in the
media to nondependency on histidine) an index of mutagenicity induction is
obtained for various test materials. As seen in Table 2, the heavy distil-
late and process solvent streams exhibit substantial mutagenic activity
whereas the light oil, wash solvent, light distillate and middle distillate
show no detectable activity. (0 By comparison, raw shale oil showed limited
activity, and a crude petroleum (Prudhoe Bay) does not show activity in the
Ames system.
To further define the response from the heavy distillate and process
solvent materials, two fractionation procedures were employed: an acid-base
scheme and a method based on LH20-Sephadex coupled with HP/LC. These schemes
are diagrammed in Figures 1 and 2. While the acid-base sequence produces
larger quantities of materials in a short period of time, the LH20-Sephadex
method, when coupled with HP/LC, ultimately produces more refined cuts of
138
-------
N-TAR
co
vo
SAMPLE
ISOOCTANE
ISOOCTANE-SOLUBLE
INHCI/ISOOCTANE
ISOOCTANE
AQUEOUS
Iso-o
pH9
ISOOCTANE
AQUEOUS
PPt
B-TAR
I NNaOH/1 SOOCTANE
I
ISOOCTANE
DMSO/ISOOCTANE
AQUEOUS
pH3
ISOOCTANE
is
D-O
NEUTRAL
DMS
PAH
Iso-o
AQUEOUS
ppt.
A-TAR
FIGURE 1. Acid-Base Fractionation Scheme
-------
SAMPLE
a
GEL SWOLLEN WITH
MeOH:H?0 (85:15 v
HEXANE FRACTION
TOL/HEX FRACTION
(10:90)
MeOH FRACTION
FIGURE 2. Sephadex LH-20 Fractionation Scheme
-------
material with less crossover among fractions. Fractions for biological test-
ing are collected from HP/LC separations made on reverse phase NH2 columns.
Where minimal amounts of materials are required for biological testing, thin-
layer chromatography has been effectively used to provide both separation
and material for analysis. Acid and neutral fractions derived from HD by
using the acid-base separation scheme showed relatively little response to
the Ames test whereas the basic, basic tar and neutral tar fractions were
mutagenically active.^2) The data for the basic and tar fractions yielded
essentially linear dose-response data as seen in Table 3. While the spe-
cific activity was about one-half that of the basic fraction, the total muta-
genic activity in the basic tar and neutral tar fractions was greater than
that in the basic fraction because of the substantially greater mass of the
tars. It is interesting that the neutral (non-tar) fraction which contains
*
most of the PNAs exhibited little activity. This is probably due to the
large number of compounds in this material which potentially prevent meta-
bolic activation of the PNA components.
TABLE 2.(i) Comparison of the Mutagenicity of Solvent Refined
Coal Materials, Shale Oils, and Crude Petroleums
in Salmonella Typhimurium TA98
Materials Revertants/yg of Material
SRC-I
Process solvent 12.3 ± 1.9
Wash solvent <0.01
Light oil <0.01
SRC-II
Heavy distillate 40.0 ± 23
Middle distillate <0.01
Light distillate <0.01
Shale Oil
Paraho-16 0.60±0.19
Paraho-504 0.59 ±0.13
Livermore L01 0.65± 0.22
Crude Petroleum
Prudhoe Bay <0.01
Wilmington <0.01
Pure Carcinogens
Benzo(a)pyrene 114 ± 5
2-Aminoanthracene 5430 ± 394
141
-------
TABLE 3.(2) Mutagenicity of Basic and Tar Fractions
from SRC-II Heavy Distillate (HD)
Sampl e
Basic fraction
Basic tar fraction
Neutral tar fraction
198
88
78
7
4
10
1.00
0.9
0.89
Controls
2-Aminoanthracene 14,000 rev/yg/y£ DMSO
benzo(a)pyrene 406 rev/5 yg/5 y£ DMSO
DMSO only 41 ± 15 rev/5 y£
Data for HD is in form Y = ax + b in rev/yg where a is
the slope, b is the interrupt, and $ is correlation
coefficient, x is the amount of material in yg.
Analysis by TLC using a solvent system designed to preferentially sepa-
rate the polar compounds from less polar constituents is presented in Figure
3 for the heavy distillate (HD) basic fraction. The TLC chromatograms were
cut into strips, extracted with hexane/acetone mixtures and the extractant
subjected to Ames assay using an S9 enzyme system. The activity associated
with each of the separated fractions is shown in the section of Figure 3
designated S9. The chromatographic behavior of the materials shown here cor-
responds very closely to that expected for polar compounds such as aromatic
amines. Similar results were obtained with the basic and neutral tar frac-
tions of heavy distillate. High resolution mass spectrometry and GC/MS
studies on the materials also indicated the presence of nitrogen containing
compounds and, specifically, aromatic amines including aminonaphthalenes,
aminoanthracenes, aminophenanthrene, aminopyrenes and aminochrysenes.
High resolution MS data also allowed a tentative identification based
on elemental compositions for aminofluorenes and aminocarbazoles; confirma-
tion of these assignments will require further work with adequate stan-
dards. (2) Isomers of the various amines were separable by capillary GC as
142
-------
BASIC I
SOLVENT
CO
11
10
9
1 8
S 5
4
3
2
1
0
A- MFAO
i
D
i
i
B- S9
D
i
i
(
(
f
£
(
c
(
1
c
(
TLC
REGION
11
10
9
8
7
6
5
4
3
2
1
n
FLUORESCENT
COLOR
YELLOW
FAINT PURPLE
YELLOW
ORANGE
FA INT PURPLE
PURPLE
YELLOW AND BROWN
LIGHT BLUE
•
VIOLET
TAN
noir.iw
10 20 100
REVERTANTS TA98 x 103
200
FIGURE 3. Ames Mutagenicity Analysis of Materials Eluted from Thin Layer Chromatogram
of the Basic Fraction of SRC-II Heavy Distillate.
-------
shown in Figure 4. Assignments specifically indicated in the figure were
made on the basis of retention times of authentic standards.'2)
The correlation of the aromatic amine content with the biologically
active regions from TLC of the heavy distillate basic fraction is shown in
Figure 5. The relative concentrations of aminoanthracenes, aminophenanthrenes,
aminopyrene and aminochrysene are seen to be highest in the regions with the
strongest mutagenic activity. With the exception of aminqnaphthalene, pri-
mary aromatic amines were not found in regions that lacked mutagenic activity.
Aminofluorenes and aminocarbazoles have also been tentatively identified in
the active regions. Analyses of these materials indicate that three and four
ring primary aromatic amines are important mutagens, but that two ring amino-
naphthalenes contribute little to mutagenic activity.
Since both GC/MS analyses and Ames results from the TLC fractions impli-
cated the aromatic amines as the mutagenically active agents in the basic,
basic tar, and neutral tar fractions of HD, a series of experiments were per-
formed to further support this conclusion. One approach used the unique
catalytic properties of mixed-function amine oxidase (MFAO), a purified liver
enzyme system. This enzyme is specific for the matabolic transformation of
primary aromatic amines to a mutagenically active state but is inactive with
BaP and other polycyclic aromatic hydrocarbons. The 2-aminonaphthalenes are
also not activated probably due to instability of the enzyme product. Muta-
genic activity after activation of the HD basic fraction with S9 appears pri-
marily in TLC regions with rf's of approximately 0.08 to 0.20. When activa-
tion was performed using MFAO, the same distribution of mutagenic activity
among the TLC regions was found as with S9 as is seen by referring again to
Figure 3 and comparing the MFAO with the S9. These results thus provide fur-
ther evidence that aromatic amines are both present and capable of expressing
their mutagenic activity in the basic fraction of HD.^1'3'
The above data were considered as presumptive for the involvement of the
primary aromatic amines as causative agents in the mutagenic activity of the
basic fraction and of the heavy distillate. Another more direct approach is
also available to support this premise. Treating HD and its basic fraction
with nitrous acid diazotizes aromatic amines and renders them nonmutagenic
144
-------
m/e 243
m/e 217
m/e 193
m/e 143
10
12 14 16 18
RETENTION TIME IN MINUTES
22
24
FIGURE 4.
Single-ion chromatograms for the m/e 143 (M+ for AN), the m/e 193 (M+ for AA and APH), the
m/e 217 (M+ for AP) and m/e 243 (M+ for AC) shown above the total ion current chromatogram
of a mutagenic neutral tar subfraction of HD. Groups of peaks preceding the ami no com-
pounds arise from the methyl homologs of the corresponding nitrogen heterocyclic (e.g.,
methylacridine precedes aminoanthracene).
-------
CTl
REVERTANTS TA 98 RELATIVE CONCENTRATION
^ M (ARBITRARY UNITS)
§ i 9
o o o -» -» o
SRC II HE,
BASIC
- t| II
1 -TB? m
—
Im
1
(0.08)
&VY D
: FRAC
*!$
&'•':
:•::•:
ft:"
i
77 jll*
*s.' "Xv"'"
* /y '•'.•".••
7/W;
'yM
/yi&
'/''$%:
ys,t&
'/, W:?
ISTILLATE ^^) AMINOCHRYSENE
'TION AMINOPYRENE
1
!
"v7
;-'iv.,
^
."•"'• -
"'•^i-
'•. 'i^:
?•;
AMINOPHENANTHRENE
[i"-S^j AMINOANTHRACENE
IS?S$ftft?l AHAIMOCI IIODCIVIC
PR IXXxX?sl AlvlllNUrLUUntlMt
« AMINONAPHTHALENE
1 j
ll ^1
ii ^ii^li , 11 ll
G3 f1"!
2345
(0.10) (0.11) (0.17) (0.23)
FIGURE 5. Identification and Relative Concentrations of Primary Aromatic Amines in Thin-
Layer Chromatography Regions from SRC-II Heavy Distillate Cut, Basic Fraction
-------
in the Ames system. Thus, disappearance of mutagenic activity in the basic
fraction or in the heavy distillate after nitrous acid treatment would pro-
vide direct evidence for the mutagenic importance of this class of compounds.
In Figure 6, it can be seen that the mutagenic activity of a pure aromatic
amine 2-aminoanthracene is almost completely lost while the activity of
benzo(a)pyrene or benzacridine is not affected by the nitrous acid treatment.
Activity seen in heavy distillate, process solvent and their basic fractions
is also mostly eliminated by nitrous acid treatment. As shown in Figure 6,
activity of these materials after treatment with nitrous acid, is reduced to
less than 10% of the original activity. It thus appears that much of the
mutagenic activity is probably due to the presence of primary aromatic amines
in both the crude material and in the basic fractions.(3'
HYDROTREATING
Since materials from coal liquefaction processes may at some point be
used for chemical feedstocks or for further refining, it is possible that
hydrotreating processes may eventually be employed in commercial SRC based
plants. Hydrotreating, however, may also be expected to significantly impact
nitrogen-containing compounds, particularly on deamination of the primary
aromatic amines. Carbon-carbon bond cleavage will also occur which will also
result in destruction of larger ring systems to form lighter weight alkylated
and/or hydrogenated species. Loss of sulfur, nitrogen and oxygen in the form
of H2S, NH3, and H20 is also expected in heterocyclic compounds. Materials
from the Ft. Lewis pilot plant which had been subjected to hydrotreatment
were therefore examined.
While the hydrotreated samples were generated under process conditions
which represent current commercial practice, final demonstration scale de-
signs are not yet available. Thus the results of the hydrotreatment proc-
essing can be evaluated only in general terms.
Material obtained from tankage accumulated over a series of pilot plant
runs extending from October 1978 into the early part of 1979 was subjected
to hydrotreating by Universal Oil Products. A middle distillate to heavy
distillate blend ratio 2.9 to 1.0 was determined from the average yield
ratios of runs during this period. Obviously because of the long-term
147
-------
100
o
g
CO
o
cd
80
co
60
o
S 40
UJ
o
C£
UJ
Q-
20
0
-x-
Q
UJ
2:
CM
^3351
i
5
<
C£
u.
O
to
CO
2
EZ
o
0£
ff
O
CO
CO
CO
Q_
O —
i—
UJ
>;
o
CO
CO
CO
s
u_
OS
ff
_j
|
UJ
a
DC
O
§
O.
RACTION
CO
<
CO
to
o_
1
CONTROL CHEMICALS
SRC MATERIALS
o z
zr o
o
CO
<
CO
o;
I
UJ
CO
o_
VS/W/A
FIGURE 6. Effect of Nitrosation on Mutagenicity of SRC Materials
-------
accumulation period, there are some difficulties in assessing sample repre-
sentativeness and processing history due to the numerous modes ranging from
steady state to upset conditions of operation and to the unavoidable product
variability from one run to another. Materials were hydrotreated in standard
research fixed bed reactors using a commercial UPO catalyst. Analysis of the
materials of the distillate plant before and after hydrotreatment showed dra-
matic differences in gross chemical composition. GC/MS runs were made with
SE2250 or SE52 coated capillary columns. Examples are given in Figure 7. The
reconstructed total ion chromatograms of the materials show that there is a
dramatic reduction of multiring compounds and phenols with subsequent conver-
sion into hydroaromatic materials, specifically tetralins and their alkylated
homologs. Table 4 summarizes the GC and GC/MS data and gives the concentra-
tions in ppm for various compound classes before and after hydrotreatment.
Severe hydrotreatment resulted in the reduction of total phenols from 130 ppm
to 17 ppm in the total distillate blend. Aromatics and N-heterocyclic com-
pounds show significant reduction. Introduction of hydrogen to the rings is
obviously demonstrated by the appearance of compounds such as tetrahydro-
quinoline, tetrahydrocarbazole and tetrahydrozapyrene, tetralins and other
hydrogenated multiring compounds. Primary aromatic amines, initially present
at a total concentration of 1.9 ppm, are below the detectable range of GC and
GC/MS following hydrogenation under the conditions employed. Figure 8 gives
a graphic summary of the results for the compound classes affected.
Biological activity associated with the basic, base-induced tar, acid-
induced tar and isooctane-induced tar fractions of the distillate blend fol-
lowed the trend shown by chemical characterization in loss of the primary
aromatic amines (Figure 9). Moderate hydrotreatment, for example, reduced
the mutagenic activity of the basic fraction from 16.2 to 2.2 revertants per
microgram (Table 5). This is a reduction in the weighted contribution to
total mutagenicity from .86 to .03 revertants per microgram feedstock. The
tar fractions were reduced in potency to levels below the limits of detec-
tion. While the specific effects of hydrotreatment upon chemical composition
and biological activity of a given coal-derived fuel product will depend up-
on reaction conditions, catalysts, and starting material composition, it
nonetheless appears that hydrotreatment will, in general, result in products
with reduced mutagenic activity. This is probably due to the reduction of
149
-------
ioo%-
•z. t2
O o;
Q. ce
on ID
UJ (_)
>- 50%-
OH
10
15 20
TIME, MIN
25
30
FIGURE 7a. Reconstructed Total Ion Chromatogram of SRC-II
Distillate Blend^)
100%
26
10
15 20
TIME, MIN
25
30
FIGURE 7b. Reconstructed Total Ion Chromatogram of Severely
Hydrotreated SRC-II Distillate Blendl1*)
(See legend on page 158)
150
-------
LEGEND TO FIGURE 7
Reconstructed total ion chromatograms comparing unfractionated SRC-II
feedstock, Figure 7a, with the severely hydrotreated material, Figure 7b.
Principal peaks are identified in both chromatograms: (a) 1: phenol,
2: Cj. phenol, 3: tetralin, 4: naphthalene, 5: indole, 6: C3 phenol,
7: C, naphthalene, 8: biphenyl, 9: ^2 naphthalene, 10: phenylether,
11: dibenzofuran, 12: acenaphthene, 13: fluorene, 14: GI fluorene,
15: dibenzothiophene, 16: phenanthrene. (b) 1: methyldecalin, 2: methylindan,
3: methyltetralin, 4: tetralin, 5: dimethylindan, 6: dimethylindan,
7: methyltetralin, 8: dimethylindan, 9: methyltetralin, 10: ethyltetralin +
dimethylbenzofuran, 11: ethyltetralin, 12: ethyletralin, 13: biphenyl +
hexahydroacenaphthene, 14: phenylether, 15: C.-indene, 16: C^-tetralin,
17: C3-dihydronaphthalene, 18: tetradecahydroanthracene,
19: tetradecahydrophenanthrene, 20: C^-dihydronaphthalene,
21: C.-tetralin, 22: Cg-indan or C^-tetralin, 23: Cg-indan or
C4-tetralin, 24: C4-dihydronaphthalene, 25: hexadecahydropyrene,
26: octahydroanthracene.
151
-------
en
ro
TABLE 4. Alteration in Chemical Composition of SRC-II Distil|at
Due to Hydrotreatment for Five Compound Classesvd.em
Aromatlcs and Polynuclear
Material Phenols13' N-heterocyclesl ' Primary Aromatic Amines' ' Hvdroaromaticsta) Aromaticslc)
Feedstock
Moderately
Hydrotreated
Severely
Hvdrotreated
£phenols 130
C, phenols 41
(L phenols 35
phenol 27
C, phenols 16
o-cresol 9
£]phenols 30
C, phenols 5.5
C? phenols 2.6
phenol 1.6
Cj phenols 1.2
^phenols 17
C, phenols 4.3
C2 phenols 2.1
phenol 1.2
Cj phenol 0.8
£N-heterocycles 28
quinoline 3.6
C, quinoline 1.4
carbazole 1.3
Co quinoline 0.8
ac r 1 d i ne 0.1
£N-heterocycles 1.2
tetrahydroquinoline 0.08
tetrahydrocarbazole 0.06
carbazole 0.04
Cj quinoline 0.03
tetrahydroazapyrene 0.02
£>-heterocycles i.o
tetrahydrocarbazole 0.07
tetrahydroquinoline 0.05
carbazole trace
C, quinoline trace
^primary aromatic amines 1.9
aminonaphthalenes 0.09
aminoanthracene/ 0.07
aminophenanthrene
aminoblphenyls 0.03
aminopyrene/ 0.03
aminof luoranthene
aminochrysene 0.02
amlnocarbazoles trace
^primary aromatic £0.005
amines
none detected £0.005
^primary aromatic £0.005
amines
1
none detec.ted <0.005
J^aromatics + 450
hydroaromat ics
naphthalene 97
Cj naphthalenes 82
Co naphthalenes 65
tetralin 57
Cj tetralin 26
biphenyl 24
^aromatics + 660
hydroaromat ics
tetralin 71
C2 tetralins 51
Cj tetralins 48
Cj naphthalenes 41
X]aromatics + 780
hydroaromat ics
Cj tetralins 120
Co tetralins 41
tetralin 34
C, tetralins 27
Xipolynuclear 110
aromatics
C14H10 38
C16H10 9'2
C18H12 3'5
benzo(a)pyrene 0.041
benzol e)pyrene 0.077
X^polynuclear 18
aromatics
C14H1Q 2.5
C16H10 0.8
C18H12 0.4
benzo(a)pyrene £0.010
benzo(e)pyrene £0.010
X!polynuclear 7.5
aromatics
C14H10 U^
C16H10 I0'5
benzo(a)pyrene 4o.010
benzo(e)pyrene £0.005
(a) Estimated directly in the unfractionated material by GC and GCMS.
(b) Estimated in the basic fraction by GCMS. Concentrations given have been calculated for the unfractionated material.
(c) Estimated in the unfractlonated material and In the PAH fraction by GC and GCMS. Concentrations given have been
calculated for the unfractlonated material. ,
(d) The contributions listed do not total 100X due to the presence o.f compound classes not listed (e.g., aliphatics) and
losses during extraction. Specific compounds listed under each heading are those found in the highest concentrations within that class.
(e) Concentrations are given in parts per thousand.
-------
en
co
40%
JL
20%
10%
FEEDSTOCK
100
60%
V.
MODERATELY
HYDROTREATED
2.5
Q
UJ
H
O
UJ
UJ
Q
ill
Z
o
Z
ll 1
80%
SEVERELY
HYDROTREATED
AROMATICS
| [ HYDROAROMATICS
PHENOLS
PNA
[NITROGEN
1 HETEROCYCLICS
I I PRIMARY AROMATIC
I 1 AMINES
O
UJ
O
UJ
LJJ
Q
UJ
O
Z
0.8
RELATIVE MUTAGENIC ACTIVITY
FIGURE 8. Gross Chemical Composition Related to Severity of Hydrotreatment^)
-------
MATERIAL
20.0
C71
-=» >
;~
o
LU
O
QX LU
10.0
1.0
8£ .5
o o
fc! CD
FEEDSTOCK
^~
1
''.
1
\
il
0
I • 1. 20
MODERATELY
HYDROTREATED
I - 0. 03
SEVERELY
HYDROTREATED
I - a 01
f^^^
B AT BT NT B AT BT NT B AT BT NT
FRACTION
FIGURE 9. Specific and Weighted Activity Test Results for the Basic (B), Base Induced tar (AT)
acid induced tar (BT) and isooctane induced tar (NT) from Raw and Hydrotreated SRC-II
Distillate Blend, (a) Strain TA98 with S9 enzyme activation, (b) Specific activity
weighted by the gravimetric yield of the fraction.(*0
-------
TABLE 5. The Effect of Catalytic Hydrogenation on the Mutagenldty of SRC-II Coal Liquid
01
O-:
Chemical
Material Fract1on
-------
those compound classes in coal liquids which are primarily responsible for
induction of mutagenic activity, namely, the nitrogen containing aromatics
and especially the primary aromatic amines as well as reduction of the con-
centrations of polynuclear aromatic hydrocarbons. Other biological assays
including mammalian cell culture and skin painting studies are also under
way but are not reported in detail here. Generally, there has been rela-
tively good agreement among the assays used. Table 6, a comparison of data
from three biological assays, demonstrates this agreement. Differences do
show up, however, in the results from 2-aminoanthracene and for heavy dis-
tillate. The mutagenic activity of 2-aminoanthracene is very high whereas
tumorigenic activity is only moderate. The reverse is true for heavy dis-
tillate; tumorigenicity is high whereas mutagenicity is moderate relative to
standard control compounds.'3'
Information such as reported here will obviously have some impact upon
the development of a liquefaction industry. Samples used were selected with
engineering guidance. Criteria included suitability and relevance to future
demonstration or commercial design and operation. However, since one can,
in practice, only anticipate or scale up to a limited number of the condi-
tions in a final design configuration, caution must be applied in the appli-
cation of pilot plant derived data. Certainly further data is required.
But more important, interaction between chemists, biologists, ecologists and
process engineers must be on a continuous basis such that pertinent and
meaningful data is prepared within a time frame commensurate with the process
development.
156
-------
TABLE 6. Comparison of Mutagenic and Carcinogenic Activity
for Several Crude Fossil-Derived Materials
Material Ames Assay Mammalian Cell Culture Skin Tumorigenesis
Light distillate — — —
Heavy distillate ++ ++ ++++
Shale oil + + ++
Crude petroleum -- slight +
Benzo(a)pyrene ++ ++ +++
2-Aminoanthracene ++++ +++ ++
157
-------
REFERENCES
1. Pacific Northwest Laboratories, Biomedical Studies on Solvent Refined
Coal (SRC-II) Liquefaction Materials: A Status Report, PNL-3189,
October 1979.
2. B. W. Wilson, et al., Identification of Primary Aromatic Amines in
Mutagenically Active Subfractions from Coal Liquefaction Materials.
Mutation Research, in press, 1980.
3. D. Mahlum, et al., Toxicologic Studies of SRC Materials, 2nd DOE En-
vironmental Control Symposium, Proceedings, in press, March 1980.
4. W. C. Weimer, et al., Initial Chemical and Biological Characterization
of Hydrotreated Solvent Refined Coal (SRC-II) Liquids: A Status Report,
PNL-3464, July 1980.
158
-------
LOW-NO COMBUSTORS FOR ALTERNATE FUELS CONTAINING
X
SIGNIFICANT,QUANTITIES OF FUEL-BOUND NITROGEN
W. D. Clark
D. W. Pershing
G. C. England
M. P. Heap
Energy and Environmental Research Corporation
8001 Irvine Boulevard
Santa Ana, California 92705
ABSTRACT
This paper summarizes data generated on two EPA-sponsored programs
concerned with the development of low-NO combustors for high nitrogen
containing fuels. EPA Contract 68-02-3125 is concerned with NOX produc-
tion .and control from liquid fuels containing significant quantities of
bound nitrogen. It was found that fuel nitrogen content is the primary
composition variable affecting fuel NO formation and that emissions from
both petroleum and alternative liquid fuels correlate with total fuel
nitrogen content. Conditions were identified which allow high-nitrogen
fuels to be burned satisfactorily with minimal NO emissions. Certain
coal-derived fuel gases may contain ammonia. Data is presented from a
series of bench-scale reactors designed to minimize the conversion of
this ammonia to NO . Lowest NO emissions were produced in a.-rich/lean
combustor utilizing either a diffusion flame or a catalyst in the fuel-
rich primary stage.
159
-------
SECTION 1
INTRODUCTION
Combustion of liquid fuels derived from petroleum sources accounts for
a significant fraction of fossil fuel consumption in stationary combustors.
As petroleum reserves grow smaller, the United States is projected to place
heavy reliance on coal, the most abundant fossil fuel available, in the
search for new energy supplies. Coal can be burned directly or converted
into either a liquid or a gaseous fuel. The potential for low sulfur emis-
sions makes combustion of gasified coal an environmentally attractive alter-
native to direct-fired coal combustion. However, low-Btu coal gases can con-
tain ammonia concentrations as high as 0.38 percent (1). In a conventional
combustor, much of this ammonia may be converted to nitrogen oxides resulting
in significant pollutant emission: up to 1370 ng/J (3.2 lbm/10 Btu) for
full conversion of NH, to N0«.
A balanced fuel economy necessitates that in the future many industrial
users will burn petroleum and coal- or shale-derived liquid fuels. Since
these liquid fuels have relatively high nitrogen content and low hydrogen-to-
carbon ratios, there will be the potential for adverse environmental impact
due to the increased emission of combustion-generated pollutants unless pre-
ventative measures are taken (1-2). The pollutant of major concern in this
paper is nitrogen oxides (NO ). The paper addresses the impact of switching
X
from conventional fuels to alternative gaseous or liquid fuels and of the
mechanisms of combustion modification techniques used to control NOX emissions.
Alternative liquid fuels can be broadly classified as those synthesized
from the products of coal gasification, and those derived directly as liquids.
The fuels in the first category tend to be clean, low-boiling-point fuels
such as alcohols, and are essentially free from nitrogen and sulfur; thus,
their impact upon pollutant emissions is minimal. The liquids in the second
category may be compared to crude petroleum oils containing a wide range of
hydrocarbon compounds with boiling points from 300°K to greater than 900 K.
The bound nitrogen content of crude synfuels is generally higher than petro-
leum crudes, and for many applications it might be necessary to upgrade the
fuel by removing the nitrogen. Recognizing that alternative liquid fuels
160
-------
contain more bound nitrogen than the petroleum fuels that they would be
replacing, one key factor in their production is to what extent combustion
modification will allow control of NO emissions and reduce the necessity
for substantial denitrification, thereby reducing the cost of synfuels.
Nitrogen oxides produced during combustion emanate from two sources.
Thermal NO is formed by the fixation of molecular nitrogen and its forma-
tion rate is strongly dependent upon temperature (3). Fuel NO is formed by
the oxidation of chemically-bound nitrogen in the fuel by reactions with a
weak temperature dependence, but a strong dependence upon oxygen avail-
ability (4-5-6-7). Thus, those emission control techniques which minimize
peak flame temperature by the addition of inert diluents (e.g., cooled recycled
combustion products or water addition) minimize thermal NO formation, but
have a minor impact upon fuel NO production. Staged heat release (staged com-
bustion) provides the most effective NO control technique for nitrogen-
ux
containing fuels because fuel NO formation is mainly dependent upon local
stoichiometry- It can be accomplished either by separating the combustion
chamber into two zones and dividing the total combustion air into two streams,
or by appropriate burner design which promotes localized fuel-rich conditions.
Minimizing fuel NO formation requires the existence of a fuel-rich
X
primary combustion zone to maximize the conversion of fuel nitrogen to
molecular nitrogen since the fate of fuel-bound nitrogen is strongly con-
trolled by the reactant stoichiometry. Many studies (8-12) have shown that
under fuel-rich conditions the efficiency of conversion to N increases
significantly. Thus, there are two fuel nitrogen reaction paths leading to
the production of N? or NO, namely:
Path A. Fuel-lean
XN + Oxidant -> NO +
Path B. Fuel-rich
XN + -> N2 +
The objective of staged combustion emission control techniques is the pro-
vision of conditions which maximize N« production via Path B. Two factors of
practical importance are the residence time and the stoichiometry required to
maximize N? production in the fuel-rich primary zone.
161
-------
If the residence time is insufficient, then the original fuel nitrogen
species will exist in the gaseous state as some XN compound which can be
converted to NO in the second-stage heat release zone. The stoichiometry
required to achieve minimum XN concentrations at the exit of the primary
stage will be determined by (1) the rate of evolution of nitrogen species from
the fuel; (2) the inevitable distribution of stoichiometries from fuel-rich
to fuel-lean which occurs because the primary zone is supplied by a diffusion
flame; and (3) the overall temperature of the primary zone. From equilib-
rium considerations the total fixed nitrogen (TFN given by NO + HCN + NH )
is a minimum at approximately 65 percent theoretical air with levels less
than 10 ppm depending upon temperature and fuel C/H ratio. Exhaust NO emis-
X
sions are considerably greater than levels predicted by equilibrium, suggest-
ing the existence of kinetic limitations in the fuel-rich primary stage.
NO formation during combustion of alternate fuels is not well-understood;
X
however, recent test results have indicated that replacing a petroleum oil
with a coal-, or shale-derived liquid may result in a major increase in NO
X
emissions. Bench-scale experiments (13) have shown that the smoke and com-
bustion characteristics of the SRC-II coal liquids are equivalent to light oil,
but uncontrolled NO emissions are high due to the 0.8 to 1.2 percent N in the
X
fuel. Pilot-scale SRC-II studies (14-16) have demonstrated that both fuel
blending and staged combustion are effective in reducing NO emissions and
X
that improved atomization, increased preheat, and increased excess 0_ increase
NO . Full-scale testing (17) has confirmed the need for optimized combustion
X
modifications. Similar results have also been achieved during bench-scale
(18) and field tests (19) with shale-derived liquids.
162
-------
SECTION 2
EXPERIMENTAL SYSTEMS
The experimental systems used to investigate NO formation from gaseous
X
and liquid fuels have been described in detail elsewhere and only a brief
summary will be presented in this paper.
LEG GAS STUDIES
The apparatus for the bench-scale experiments can be divided into four
subsystems: LEG supply, modular combustors, sample train and control systems.
A simplified schematic of the facility can be seen in Figure 1.
Synthetic LEG was produced from hot air premixed with vaporized water
and heptane passed through a catalytic reformer. The reformer was operated
at pressures between 6.4 and 11.9 atmospheres at a stoichiometry of 45 per-
cent theoretical air, the richest stoichiometry attainable without excessive
sooting. The water acts as a diluent to maintain the maximum catalyst bed
temperature at around 1370°K. The reformer product gas passed through a
variable heat exchanger, cooling it to the desired preheat temperature.
Ammonia and methane are added to trim the gas to the desired fuel nitrogen
and hydrocarbon content. The LEG passed through a soot filter and into a
valve system, controlling the fraction of the LEG which goes to the combus-
tors and the fraction which is bypassed. If none of the gas was bypassed,
maximum combustor capacity was 60,000 J/s (200,000 Btu/hr).
The combustors consisted of a series of modules with 5 cm (2 in) ID reac-
tion/flow chambers enclosed in 15 cm (6 in) OD low-density insulation and
housed in flanged steel pipe. Primary ignition modules include the catalyst
and the diffusion flame. Secondary burnout was achieved in the jet-stirred
secondary air injector. Plug flow modules of various lengths allowed con-
trol of primary and secondary residence times. The primary ignition modules
are shown in Figure 2. In the catalyst module, premixed LEG and primary air
passed through a stainless steel flow straightener/flame arrestor and into
the graded cell catalyst. The catalyst, supplied by Acurex, consisted of
three zirconia honeycomb monoliths of decreasing cell size, coated with nickel
oxide. Platinum had been added to the coating of the upstream monolith to
promote ignition. In the concentric diffusion flame module, LEG is introduced
163
-------
en
AIR
WATER
REFORMER AIR
PRIMARY AIR
SECONDARY AIR
EMISSIONS
CONSOLE
HEPTANE
FLOW
RESTRICTOR
SAMPLE
TRAIN
CATALYTIC
REFORMER
3 L
MODULAR
COMBUSTORS
TRIM
GASES
BYPASS
Figure 1. Bench-Scale Pressurized Test Facility.
-------
BENCH-SCALE CATALYST
BENCH-SCALE DIFFUSION FLAME
To Second Stage
cn
en
Catalyst Wall
Thermocouples Q
Flash Rack
Thermocouple
To Secondary
11
XXXXXXX'XXXXXI
'XXXXXXXXXXXXXl
xxxxxxxxxxxxxl
Plug Flow
Module
Catalyst Bed
Thermocouples
Graded Catalyst
Cells
Low Density
Insulation
Injector/Flow Straightener
Primary
Air
IRG
Primary
Air
Low Density
Insulation
Premixed LBfi ' Primary Air
Figure 2. Primary Ignition Modules.
-------
through a removable fuel tube of variable diameter. Straightened primary
air passes annularly around the fuel tube in the direction of the fuel flow.
Samples are taken in the secondary region, through a water-cooled stain-
less steel probe situated on the centerline of the flow chamber. The cooling
water is preheated and the stainless steel sample lines are wrapped with heat
tape to maintain the sample system above the dewpoint of the exhaust gases.
The sample stream is throttled to nearly^atmospheric pressure.
LIQUID-FIRED TUNNEL FURNACE
The downfired tunnel furnace illustrated in Figure 3 was designed to
allow utilization of commercially-available spray nozzles, and yet be capable
of testing with artificial atmospheres. This combustor, which has been de-
scribed in detail elsewhere (6), was 2.1m long and 20 cm in inside diameter.
The walls consisted of insulating and high-temperature castable refractories
and the full-load firing rate was 0.53 cc/sec, which corresponds to a nominal
heat release of 20 kW. All airstreams were metered with precision rotameters.
The main combustion air was preheated with an electric circulation heater; the
atomization air was not preheated. In certain tests the "air" was enriched
or replaced with varying amounts of carbon dioxide, argon, and oxygen, all of
which were supplied from high-pressure cylinders.
ANALYTICAL SYSTEMS
Exhaust concentrations were monitored continuously using a chemilumi-
nescent analyzer for NO and NO , a NDIR analyzer for CO and C0«, and a para-
i *C l ^
magnetic analyzer for 0 . The flue gas was withdrawn from the stack through
a water-cooled, stainless steel probe using a stainless steel/Teflon sampling
pump. Sample conditioning prior to the instrumentation consisted of an ice
bath water condenser and glass wool and Teflon fiber filters. All sample
lines were 6.3 mm Teflon and all fittings 316 stainless steel.
In-flame temperature measurements Were made with a standard suction
pyrometer containing a platinum/rhodium thermocouple. In-flame gas samples
were withdrawn with a long, stainless steel water-quench probe. HCN and NH.,
were absorbed in a series of wet impingers and concentrations determined
using specific ion electrodes. Sulfide ion interference was minimized by
the addition of lead carbonate (20)- Hydrocarbons were measured using a water-
cooled probe, heated sample line, and an FID analyzer.
166
-------
Ultrasonic Twin-fluid Atomizer
... . _ ^ Burner
Viewing Port Section
j
o
1 »
I »
1 \
0
o
o
o
0
Tunnel Furnace
Thermocouple
Connection ~
Oil Heater
Connection
Combustion'
Air X
Atomizing Air
Oil Pressure
Tap
Oil Inlet
Viewing. Port
Ultrasonic
Nozzle
Burner Detail
Insulating
Block
Insulating
Refractory
High-Temperature
Refractory
Flue
Furnace Cross-Section
Figure 3. Details of the Tunnel Furnace System.
167
-------
LIQUID FUELS
Figure 4 illustrates the wide spectrum of composition for the distillate
oils (half-filled symbols), heavy petroleum liquids (open symbols), and alter-
native liquid fuels (solid symbols) investigated to date. The petroleum-
derived fuels had sulfur contents ranging from 0.2 to 2.22 percent with a
maximum nitrogen content of 0.86. The nitrogen content of the alternative
fuels range from 0.24 to 2.5 percent. Table 1 lists the complete chemical
analysis and physical properties of each fuel as determined by an independent
laboratory. The shale liquids included crude shale from the Paraho process
(A3) and four refined products: diesel fuel marine (DFM, Al) residual fuel
oil (A5), a 520-to-850°F distillation cut (A7), and a 5.75/1 medium/heavy
SRC-II blend (A6), a heavy SRC-II distillate (A9), and an SRC-II blended with
the donor solvent (A4).
168
-------
2.4
2.Z
2.0
1.8
c 1.6
cu
O)
|M
•z.
.p 1-2
O)
£ i.o
__ Q.
S .B
.6
.4
.2
•
*
fc
•
-
- .
°0 £>
V
Ik °
^O
^ "^ Q
O ^
* 1 Al 1
1.0 2.0 3.0
Wt. Percent Sulfur
e
1 1—" 1
*
t
-
-
•* "
o<> °
V
^ D
L O
^V
0 ^
°' 0 ' A
i.O '-0 «-0 9.0
Carbon Hydrogen Ratio
i
•
-
B "
-
°o<>
•• ••
k D
D li o
" O k
1 1 1 1 1 1 1 1 1 1 1 p 1 1 1 1 1
2 46 8 10 12 14 16 \a
Conradson Carbon Residue (%)
Figure 4. Properties of Fuels Tested.
-------
Table 1. Detailed Fuel Analyses
Symbol
Ultimate Analysis:
Carbon, X
Hydrogen, X
Nitrogen, X
Sulfur, I
Ash, t
Oxygen, X
Conradson Carbon Residue, X
Asphaltene, X
Flash Point, "F
— ' Pour Point, °f
O API Gravity at 60°F
Viscosity. SSU, at 140*F
at ZIO'F
Heat of Combustion:
Gross Btu/lb
Net Btu/lb
Calcium, ppm
Iron, ppm
Maganese, ppm
Magnesium, ppm
Nickel, ppm
Sodiun. ppm
Vanadium, ppm
01
Alaskan
Diesel
0
86.99
12.07
0.02
0.31
<.001
0.62
33.1
33.0
29.5
02
W. Texas
Diesel
A
88.09
9.76
0.026
1.88
•c.OOl
0.24
18.3
32.0
28.8
03
California
No. 2 OH
3
86.8
12.52
0.053
0.27
<.001
0.36
32.6
30.8
29.5
19,330
Rl
East
Coast
0
86.54
12.31
0.16
0.36
0.023
0.61
2.1
0.34
205
50
24.9
131.2
45
19.260
18,140
7.1
16
0.09
3.7
6.7
37
14
R2
Middle
East
k
86.78
11.95
0.18
0.67
0.012
0.41
6.0
3.24
350
48
19.8
490
131.8
19,070
17,980
1.2
2.6
0.02
0.08
13
0.98
25
R3
Low Sulfur
No. 6 Oil
0
86.57
12.52
0.22
0.21
0.02
0.46
4.4
0.94
325
105
25.1
222.4
69.6
19,110
17.970
9.52
123.6
0.46
2.23
14.10
3.74
3.11
R4
Indo/
Malaysian
A
86.53
11.93
0.24
0.22
0.036
1.04
3.98
0.74
210
61
21.8
199
65
19,070
17.980
14
16
0.13
3.6
19
15
101
R5
Venezuelan
Desulphurized
Q
85.92
12.05
0.24
0.93
0.033
0.83
5.1
2.59
176
48
23.3
113.2
50.5
18,400
17.300
8.7
6.5
0.09
3.6
R6
Pennsylvania
(Amarada Hess)
t\
84.82
11.21
0.34
2.26
0.067
1.3
12.4
4.04
275
66
15.4
1049
240
18.520
17,500
9.2
13. Z
0.10
3.3
32.7
64.5
81.5
R7
Gulf
Coast
<0
84.62
10.77
0.36
2.44
0.027
1.78
14.8
7.02
155
40
13.2
835
181
18.240
17,260
4.4
19
0.13
0.4
29
3.6
45
R8
Venezuelan
D
85.24
10.96
0.40
2.22
0.081
1.10
6.8
8.4
210
58
14.1
742
196.7
18,240
17,400
9.1
11
0.09
3.8
52
32
226
R9
Alaskan
D
86.04
11.18
0.51
1.63
0.034
0.61
12.9
5.6
215
38
15.6
1,071
194
18.470
17,580
6.9
24
0.06
1.4
50
37
67
RIO
California
V
85.75
11.83
0.62
1.05
0.038
0.71
19.5
246.1
70.00
-------
Table 1. Detailed Fuel Analyses (Continued)
Rll
R12
R13
R14
Al
A2
A3
A4
AS
A6
A7
A8
A9
California
California. California California (Kern County)
Ultimate Analysis:
Carbon, X
Hydrogen, X
Nitrogen, X
Sulfur, X
Ash, X
Oxygen, X
Conradson Carbon Residue, X
Asphaltene, X
Flash Point, °F
Pour Point, *f
API Gravity at 60'F
Viscosity, SSU, at 140*F
at 210°F
Heat of Combustion:
Gross Btu/lb
Net Btu/lb
Calcium, ppm
Iron, ppm
Manganese, ppm
Magnesium, ppm
Nickel, ppm
Sodium, ppm
Vanadium, ppm
O
85.4
11.44
0.77
1.63
0.043
0.71
8.72
5.18
38
15.4
854
129
18,470
17,430
21
73
0.8
5.1
65
21
44
0
85.33
11.23
0.79
1.60
0.032
1.02
9.22
5.18
150
30
15.1
748.0
131.6
18,460
17,440
14
53
0.1
3.8
82
2.6
53
a
86.66
10.44
0.86
0.99
0.20
0.85
15.2
8.62
180
42
12.6
720
200
18,230
17.280
90.6
77.2
0.87
31.4
88.0
22.3
66.2
0
86.61
10.93
0.83
1.16
0.030
0.44
8.3
3.92
255
65
12.3
4630
352
18.430
17,430
4.4
15
0.15
1.1
68
3.4
39
Shale-
Derived
DFM Syntholl
^
86.18
13.00
0.24
0.51
0.003
1.07
4.1
0.036
182
40
33.1
36.1
30.7
19,430
18.240
0.13
6.3
0.06
0.43
0.09
<.l
B
86.30
7.44
1.36
0.80
1.56
2.54
23.9
16.55
210
80
S-1.14
10,880
575
16,480
15.800
1,670
109
6.2
170
2.6
148
6.5
Crude
Shale
*
84.6
11.3
2.08
0.63
.026
1.36
2.9
1.33
250
80
20.3
97
44.1
18,290
17,260
1.5
47.9
0.17
5.40
5.00
11.71
•O
SRC II
Blend
A
89.91
9.27
0.45
0.065
0.004
0.30
6.18
4.10
70
<-72
10.0
40.6
32.5
17.980
17,130
0.33
3.9
<0.5
0.17
<0.5
0.31
<1.0
Shale
Residual SRC II
k. +
86.71 85.91
12.76 8.74
0.46 (0.96)
0.28 0.30
0.009 0.04
0.03 4.08
0.19 ' 0.51
0.083
235 1.73
90 -55
29.0 ( )
54.3 39
37.3 ( )
19,350 17,100
18.190
4.20
<0.5
<0.5
0.15
<0.5
2.51
<1.0
Shale
Fraction
( 520-850' F)
Ik
85.39
11.53
1.92
0.72
0.002
0.44
0.07
0.12
255
70
22.3
62.9
41.8
18,520
17,470
<.05
2.9
0.033
0.021
<0.5
<0.1
<0.2
Shale
Fraction
(+850-F)
•
85.92
10.61
2.49
0.63
0.24
0.11
9.3
4.24
370
95
12.0
3050
490
18,000
17,030
238
86
1.3
51
7.4
11
1.1
SRC II
Heavy
Distillate
A
88.98
7.64
1.03
0.39
0.058
1.90
265
8
1.3
67.2
41.3
17,120
16,240
-------
SECTION 3
RESULTS - LEG GAS
Encouragingly low NOX levels have been achieved on the bench scale
utilizing a catalytic reactor and a diffusion flame reactor. An effective
fuel nitrogen-reducing catalyst was identified in laboratory-scale experi-
ments and the effects of scale and stoichiometry were examined in the bench-
scale experiments. A fuel-lean diffusion flame was identified as an attrac-
tive low-NOx combustor concept in laboratory-scale experiments and effects of
scale, stoichiometry, hydrocarbon content of the fuel, fuel tube size, pressure,
and primary residence time were examined in the bench-scale experiments.
Effects of catalyst type on fuel nitrogen processing in LEG combustion
were examined on the laboratory scale in an unstaged catalytic reactor operated
at a constant adiabatic flame temperature of 1473°K. Figure 5 shows the variable
stoichiometry results for two catalysts. The alumina supported platinum cata-
lyst converted almost all fuel nitrogen to NC) in fuel-lean combustion and had
X
a minimum conversion of 40 percent in fuel-rich combustion. At stoichiometries
richer than 60 percent theoretical air, decreasing NO concentrations were over-
whelmed by increasing NH-j, and HCN concentrations, causing a sharp rise in EXN.
The zirconia supported platinum/nickel oxide catalyst converted 80 percent of
the fuel nitrogen to NOX in lean combustion, but had very low conversions in
rich combustion. For a 500ppmNH3 in LEG dopant level, less than 10 ppm £XN
were measured at stbichiometries as rich as 40 percent theoretical air. Tests
of the platinum/nickel oxide catalyst over a range of adiabatic flame tempera-
tures (1273-1673 K) and with CH4 as the fuel yielded similar results.
A rich/lean series staged platinum/nickel oxide primary catalytic reactor
was selected as a potential low NOX concept for bench scale testing. The
scale-up results were in general agreement with the laboratory-scale results.
Figure 6 compares the results of staged combustion of a 500 ppm NHg doped LEG
at two scales: 1200 and 20,000 J/sec (4000 and 70,000 Btu/hr). Each had high
conversions of NH^ to NOX in fuel-lean combustion. Minimum conversions occurred
172
-------
OJ
300
I
O
D
200
100
NO
HH3
HCN
O £XN or NOX
NH3 in LBG = 500 ppm
Pt/NiO Catalyst
Approximate
1 Full Conversion
I
100 120
Ptrcmt Theoretical Air - Primary
200
100
40
O
D
A
O
NO
NH3
HCN
or NO
NH3 in LBG = 500 ppm
Pt Catalyst
Approximate Full
Conversion
R Q
_L
100 200
PcrcMt Theoretical Air - Prlaary
Figure 5. Laboratory Scale Catalyst Comparison.
-------
100
X
0
z
1
,80
g
•H
E
" 60
o
3
4-*
g
u
01
PM
20
—
mH^
—
BENCH SCALE
Q - NOX, NORMAL OPERATION
• - NOX, BREAKTHROUGH
NH3 In LBG =553 PPM
VARIABLE ADIABATIC FLAME
TEMPERATURE
D 0 °0|
1 1 1 1
LABORATORY SCALE
D-NOX
NH3 In LBG = 500 PPM
ADIABATIC FLAME TEMPERATURE »
1473°K
}
1 1 1
^^
—
^_
|
400
•1
300 §
-------
in rich/lean staged combustion when the primary was operated close to
stoichiometric. At a primary stoichiometry of 90 percent theoretical air,
the laboratory-scale catalytic reactor converted 8 percent of the input NH_
to NO while the bench-scale combustor had an overage conversion of 14 per-
cent. Conversions in the bench-scale combustor remained low (less than 18
percent) over all rich primary stoichiometries under normal operation; but
breakthrough occurred if the primary was operated richer than 75 percent
theoretical air: the temperatures on the walls of the catalyst monoliths
dropped and the conversion rose sharply. Breakthrough was not observed in
i
the laboratory-scale experiment where the adiabatic flame temperature was
maintained at a constant 1473 K by varying the amount of nitrogen diluent
in the reactants. The undiluted flame reactor LEG had a higher heating
value (HHV) of 6.7 x 106 J/m3" (180 Btu/ft3) while the HHV of the bench-
/: " o o
scale LBG was only 3.0 x 10 J/m (80 Btu/ft ). This indicates that rais-
ing the hea-ting, value of the gas could extend the operating range, of the
Pt/NiO catalyst, and that catalyst effectiveness is limited by a threshold
flame temperature below which breakthrough occurs.
It is difficult to compare the laboratory and the bench-scale diffusion
flame combustors. Figure 7 shows laboratory- and bench-scale results for
diffusion flame; combustion of LBG containing about 500 ppm NH-j and varying
amounts of methane. In the unstaged laboratory-scale experiment, performed
at atmospheric pressure in a cold-wall reactor under attached laminar-flow
conditions, the hydrocarbon''content of the LBG had the most significant
effect on XN conversion. Conversions as low as 10 percent were observed
for combustion of hydrocarbon-free LBG under nearly stoichiometric condi-
tions. Under richer conditions, conversion increased due primarily to
increasing ammonia concentrations. However, under leaner conditions conver-
sions remained quite low. Similar trends were observed in combustion of
LBG containing 5 percent methane, but XN conversion was much higher. In
the staged bench-scale experiment, performed at 8 atmospheres in a nearly
adiabatic combustor under turbulent-flow conditions, effects of hydrocarbon
content and stoichiometry were not So pronounced.
The bench-scale flame was not visible and there was no reliable indi-
cator as to whether the flame was attached or lifted. However, throughput
and tube size ranged from conditions where the flame should definitely be
175
-------
•--1
CTl
80
60
LABORATORY SCALE
UNSTAGED
NH3 in LBG •= 471 PPM
O EXN, CH4 = 0
Q IXN, CH4 = 5%
BENCH SCALE
STAGED
NH3 in LBG = 553 PPM
£ NOX, CH4 = .6%
B NOX> CH4 = 2.1%
a
ODD
B
„
2
g 40
u
u
g
o
8 20
B.
O
o
(
1 1
20 40
^
0
o
1 1
60 HU
D
•
^
O
•
D p nji D
*
o
o
1 1 1 1 1 1
100 120 140 160 180 200
Percent Theoretical Air - Prlamry
Figure 7. Diffusion Flame Scale-up.
-------
attached to conditions where the flame should definitely be lifted. No sharp
changes were observed in NOX emissions or in other measured parameters, indi-
cating that the attached/lifted transition was not an important factor. This
agreed with previous variable-throughput laboratory-scale tests of a hydro-
carbon-containing diffusion flame, where a smooth NCL. transition was observed
2i
as the flame became detached (3) .
Figure 8 shows the effect of fuel tube size on XN conversion in the
bench-scale diffusion flame operated at 8 atmospheres. In constant-pressure
operation at a fixed stoichiometry, fuel flow and primary residence time
were independent of fuel tube size, while Reynolds number was inversely pro-
portional to the fuel tube I.D., fuel tube size had little effect on XN con-
version in fuel-rich combustion. However, in lean combustion, increasing
tube size (decreasing Reynolds number) decreased NH3 conversion to NOX.
Increased tube size also decreased the NOX noise level (high frequency con-
centration fluctuation shown by the error bars in the figure) , perhaps an
indication of flame stability.
Figure 9 shows the effect of pressure on rich/lean and lean diffusion
flames. In the bench-scale system, pressure is maintained by passing the
exhaust gases through a critical-flow orifice. For a fixed stoichiometry,
fuel flow and Reynolds number are proportional to pressure while primary
residence time is independent of pressure. The staged tests were performed
at a primary stoichiometry of about 95 percent theoretical air. For low-
hydrocarbon LEG, NHo conversion to NO remained constant at 33 percent over
pressures ranging from 4 to 8 atmospheres . For LEG containing 2 . 1 percent
CH^, conversions remained constant around 40 percent with changing pressure.
The lean tests were performed at a stoichiometry of about 150 percent theo-
retical air. Noise levels were higher than in the staged case. Conversions
increased slightly with increasing pressure in low hydrocarbon combustion.
Little change in conversion was seen with changing pressure for the 2.1 per-
cent CH LEG.
Primary residence time appeared to have the most pronounced affect
on XN conversion in a staged diffusion flame. Residence time was varied at
constant pressure by changing the pressure control orifice size. In constant-
pressure operation at a fixed stoichiometry, fuel flow and Reynolds number
were inversely proportional to primary residence time. Figure 10 shows XN
177
-------
co
x 80
i
1 60
X
c
o
C 40
I
1
§ 20
01
o.
• NOX, 3/8 FUEL TUBE, Re = 90,000
O NOX, 3/4 FUEL TUBE, Re = 40,000
NH3 in LBG = 553 PPM
100% CONVERSION = 430 PPM
-
—
|
1 0
$ f ft
.
1 1 1 1 1 1 1 1
20 40 60 80 100 120 140 160
400
300 -r
*
200 i
I
1
100 .1
Percent Theoretical A1r - Prlnary
Figure 8. Bench-Scale Diffusion Flame: Effect of Reynolds Number.
-------
REYNOLDS NUMBER
100
80
X
i
1 60
g
| 40
c
o
o
§ 9n
Jf 20
•i
o.
1 1 1
16000 24000 40000
NH3 in LBG = 553 PPM
3/4 O.D. FUEL TUBE
T Primary - 120 msec
95% T.A. PRIMARY
Q NOX , CH 4 = . 6%
H NOX, CH4 = 2.1%
| RICH/LEAN STAGED | —
^_
—
—
' 0^ 0 ol
_
^^
1 1 1 1
400
TJ
91
1
"l^
c
3003
5
200 ji
X
§
1
100 1
a.
REYNOLDS NUMBER
40 60 80 100
Coflfcustor Pressure (pslg)
1 rtp
1 UU
80
X
§
80
r>
i
c
o
-------
CD
o
IUU
80
X
1,60
3c
Percent Conversion 1^
ro *>
0 0
1 1 1 1 1 1 1 1
NH3 in LBG = 553 PPM Q ^ ? pr^fy = 12Q msec
PRESSURE = 100 PSIG
3/4 O.D. FUEL TUBE ^ m^ T pH|Mry s 25Q msec
—
Its t? "
If: : -
1 1 1 1 1 1 1 1
400
300
I
— ' ro
o o
0 0
Approxlmte NOX (ppnv
20 40 60 80 100 120 140 160
Percent Theoretical A1r - Primary
Figure 10. Bench-Scale Diffusion Flame: Effect of Residence Time.
-------
conversion with primary stoichiometry for two different primary residence
times. Using a large pressure-control orifice, a pressure of 8 atmospheres
was achieved at fuel-tube Reynolds numbers around 40,000 and primary resi-
dence times around 120 msec. A minimum XN conversion of 34 percent was
observed at a primary stoichiometry around 90 percent theoretical air. Using
a smaller pressure control orifice, a pressure of 8 atmospheres was achieved
at fuel-tube Reynolds numbers around 20,000 and primary residence times
around 250 msec. A minimum XN conversion of about 22 percent was observed
at a primary stoichiometry around 90 percent theoretical air. For a 553-ppm-
doped LBG burned out to 150 percent theoretical air, this XN minimum corres-
ponded to a NOX concentration of 100 ppm.
Figure 11 shows NOX concentration as a function of NI^ in the LBG for
rich/lean staged combustion in a diffusion flame and in a platinum/nickel
oxide catalytic combustor. For both the catalytic and the diffusion-flame
combustors, NOX emissions increased with increasing fuel nitrogen content,
but the increase in N0__ was much less than proportional to the increase in
X
fuel nitrogen content. The 3/8 OD tube diffusion flame, operated at a pri-
mary stoichiometry of 76 percent theoretical air and a pressure of 4.4 atmo-
spheres, converted 40 percent of its fuel nitrogen to NO., at 553-ppm NHo in
A -J
the LBG and had conversions of only 11 percent at a 3800-ppm doping level.
,The catalyst, operated at a primary stoichiometry of 80 percent theoretical
air and a pressure of 2.4 atmospheres, had XN conversions of 16 percent at
the low NHo doping level and 6 percent at the high doping level. Similar
trends, but higher NOX concentrations, were observed for both combustors in
fuel-lean combustion.
181
-------
600 _
oo
ro
O
1000
RICH/LEAN STAGED
80% T.A. PRIMARY
O NOX, Pt/NiO Catalyst
{) NOX, 3/8 0.0. Tube Diffusion Flame
— ——Full Conversion - Catalyst
————— Full Conversion - Diffusion Flame
o
I
2000
NH3 in LBG (ppmv wet)
3000
4000
Figure 11. Bench-Scale Reactor Comparison: Effect of Dopant Level.
-------
SECTION 4
LIQUID FUEL - EXCESS AIR RESULTS
PETROLEUM LIQUIDS
To define the influence of fuel composition on total and fuel NO emis-
sions, each oil was tested under similar conditions in the tunnel furnace.
Fuel NO formation was determined by substitution of the combustion air with
X
a mixture of argon, oxygen, and carbon dioxide. The argon replaced the nitro-
gen, thereby eliminating thermal NO formation and the C09 provided the proper
X ^
heat capacity so that flame temperatures were matched. Total and fuel NO
X
emissions were measured with an air preheat level of 405 +5 K and an atomiza-
tion pressure of 15 psig. Figure 12 presents a composite plot for total and
fuel NO (defined by argon substitution) as a function of weight percent
X
nitrogen in the fuel for a wide range of petroleurii and blended distillate
fuels. In Figure 12 the various symbols represent different base fuels (see
Table 1 for symbol key). Those symbols shown with a line refer to distillate
or residual fuels doped with pyridine or thiophene. It can be seen that both
total and fuel NO increase with increasing fuel nitrogen content, and that
total fuel nitrogen level is the dominant factor controlling fuel NO forma-
X
tion in this system. The form of the nitrogen does not appear to signifi-
cantly influence fuel NO formation under excess air conditions, as doping
with a volatile nitrogen compound (pyridine) resulted in NO emission similar
X
to that from a less volatile residual oil of the same nitrogen content. Since
the data is for a system where very fine oil droplets (approximately 25 micron
mean diameter) are well-dispersed in the oxidizer under hot fuel-lean condi-
tions, it is not surprising that fuel NO emissions are somewhat higher than
X
those achieved in practical systems.
183
-------
1000
S-
-a
CVJ
o
800
/ Indicates addition of pyridine
and/or thiophene to base fuel.
0.2 0.4 0.6
Fuel Nitrogen (weight, percent)
0.8
Figure 12.
Total and Fuel NOX Emissions From Pure and Doped (pyridine
and thiophene) Petroleum Fuels Tunnel Furnace.
184
-------
ALTERNATIVE FUELS
Figure 13 presents a composite plot of total and fuel NO for the range
X
of petroleum fuels together with alternative fuels and mixtures. The Paraho
shale was mixed with the same low sulfur oil used by Mansour (19). Synthoil
could not be pumped without blending and the results presented in Figure 4
refer to 80 and 90 percent Synthoil blends with distillate oil. The SRC-II
blend refers to a mixture of SRC-II and the donor solvent. Under the con-
ditions tested, fuel NO emissions increase approximately linearly with
X
increasing fuel nitrogen and it can be seen that the fate of fuel nitrogen
in alternative fuels is similar to that in petroleum-derived fuels. Figure 14
presents the fuel NO data plotted as a percentage of the fuel nitrogen con-
X
verted to fuel NO . For low fuel nitrogen contents, the conversion decreases
X.
rapidly (from greater than 90 percent) as fuel N increases. Eventually,
however, the conversion becomes almost independent of fuel nitrogen content;
hence, the linear dependence shown in Figure 13.
The absolute level of the fuel N conversion can be influenced by alter-
ing the fuel/air contacting and/or the fuel atomization (2), but the results
obtained in this study suggest that fuel nitrogen is the only first-order
fuel composition parameter controlling NO formation in fuel-lean flames. This
X
conclusion applies to petroleum-, coal-, and shale-derived liquid fuels. How-
ever, there appear to be second-order effects where the volatility of the fuel
nitrogen compound does have an influence upon fuel NO formation. Comparison
jC
of the data for the fuels with fuel nitrogen content of approximately 0.24 per-
cent indicates that the highest conversion is achieved with a shale-derived
distillate fuel with a large volatile nitrogen fraction.
185
-------
T T I
2000 _
X
x" FUEL no^
O PETROLEUM DERIVED
k II BLEND
k SHALE RES I DUAL
>SRC II
[SYNTHOIL/BLENDS
PAR/\HO SHALE/BLENDS
) .4 .8 1.2 1.6 2.0
Fuel Nitrogen (weight, percent)
Figure 13. The Effect of Fuel Nitrogen Content on Total
and Fuel HQX (5 percent excess oxygen).
186
-------
oo
-I 1 1 r—
O Petroleum Derived
Coal Derived
Shale Derived
0 ,2 .4 .6 ,8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Fuel Nitrogen (weight, percent)
Figure 14. Fuel Nitrogen Conversion - Comparison of Alternate and Petroleum-Derived Fuels,
-------
SECTION 5
LIQUID FUELS STAGED COMBUSTION RESULTS
POTENTIAL FOR NOV CONTROL
A
Staged combustion, i.e., the operation of a combustion system in which
the fuel originally burns under oxygen-deficient conditions, provides the
most cost-effective control techniques established to date for reducing fuel
NO . Figure 15 shows the influence of primary zone stoichiometric ratio on
X
total NO emissions for two coal-derived and two shale-derived liquids under
X
staged combustion conditions and 3 percent overall excess 0^. All the data
in Figure 15 were obtained in the tunnel furnace with ultrasonic atomization
and with a first-stage residence time of approximately 800 ms. As the primary
zone becomes more fuel-rich, NO emissions decrease dramatically to a minimum
and then increase again. This trend is in agreement with previously-reported
data on petroleum fuels (21).
FUEL CHEMISTRY
First Stage Stoichiometry
In an effort to better understand the mechanisms of NO formation under
staged combustion conditions, the original furnace was modified to allow in-
flame sampling of the XN (NO, HCN, NH~) species and cooling of the first-stage
and/or second-stage combustion products, as illustrated in Figure 16. A "radia-
tion shield" (choke) was installed near the top of the furnace to minimize
the effects of downstream changes on the fuel vaporization zone. A secondary
air injection ring and cast refractory choke were installed at 41 in. to insure
isolation of the first stage. Variable cooling was achieved by insertion of
multiple stainless steel water-cooling coils.
Figures 17, 18, and 19 show typical results of the detailed in-flame measure-
ments made at the exit of the first stage for a distillate oil (Dl-Alaskan
188
-------
I
I
800
600
CM
O
0*400
Q.
Q-
200
A Coal N - 0.44%
k Shale N - 0.46%
Coal N - 0.99
Shale N - 2.08
0.60
0.70
0.80
0.90
SR
1
Figure 15. Minimum NOX Levels Achieved With Alternate Fuels
(tunnel furnace primary zone residence time .83 sec)
189
-------
Nozzle
Position
g Air oil
II ^ Combustion Air
W «
Ti
^
^
&
1 1
§
O
0
I
.6
.12
.18
.24
.30
.36
.42
.48
.54
.60
.66
.72
.78
.84
.90
Secondary Air
Injector
Cooling Coils
Cast Refractory Choke Section
Figure 16. Modified Tunnel Furnace.
-------
240
ON01
D HCN
ONH
Figure 17. Detailed In-Flame Species Measurements With
Alaskan Diesel Oil (Dl).
191
-------
1200_
Figure 18. Detailed In-Flame Measurements
With Kern County Crude Oil (R14).
192
-------
5000 —
4600 -
0.5 0.6 0.7
*°*°2 cp.
— =U— — — fc
0.8 0.9
1
Figure 19. Detailed In-Flame Measurements
With +850°F Shale Fraction (A8)
193
-------
diesel), a high nitrogen residual oil (R-14-Kern County, California) and an alter-
native liquid fuel (A8-+850 F shale fraction). These measurements were made
on the centerline of the furnace at a distance of 104 cm (approximately
630 msec) from the oil nozzle. Detailed radial measurements indicated that
the concentration profile was essentially uniform at this location. All of
the in-flame data are reported on a dry, as-measured basis. After each
in-flame measurement, second-stage air was added at 107 cm and exhaust NO
X
measurements were also made (shown on a dry, 0% 0» basis). In general,
decreasing the first-stage stoichiometric ratio reduced the NO concentration
leaving the first stage. However, below a stoichiometric ratio of approxi-
mately 0.8 significant amounts of NH_ and HCN were measured. Thus, there
exists a minimum in exhaust NO concentrations because of a competition
•3£
between decreased first-stage NO and increased oxidizable nitrogen species
such as HCN. Figures 18 and 19 indicate that the petroleum-derived oil (0.83
percent N) and the heavy shale liquid (2.49 percent N) produce large amounts
of HCN. In addition, both fuels exhibited a minimum in TFN at a first-stage
stoichiometry of approximately 0.8.
Data for the Alaskan diesel oil (Figure 17) also show the presence of
much smaller but significant concentrations of HCN and NH-, although this
fuel is essentially nitrogen-free. Total conversion of the fuel nitrogen
would produce 21 ppm TFN at SR^O.7. This confirms previous work (10-12)
which demonstrated that reactions involving hydrocarbon fragments and N~ or
NO can produce HCN.
Hydrocarbons
The rapid increase in HCN concentration below SR =0.8 was accompanied by
an increase in hydrocarbon content of the partially oxidized combustion prod-
ucts. Figure 20 summarizes the in-flame hydrocarbon measurements for the
Alaskan Diesel (Dl), three petroleum-derived residual oils (Indonesian-R4,
Alaskan-R9, Kern County-R14), three alternative liquids (SRC-II heavy dis-
tillate-A9, crude shale-A3, heavy shale fraction-A8) and methane containing
0.75 weight percent nitrogen as NH. (JZ? ). Hydrocarbon concentrations correlate
well on the basis of first-stage stoichiometry. At very low stoichiometric
ratios the distillate oil «^) and CH,/NH produced slightly higher hydrocarbon
concentrations than the heavier liquid fuels.
194
-------
10,000
T3
HI
M
CO
CO
I
CO
cd
-------
XN Distribution
Figure 21 shows typical results on the percentage of the original fuel
nitrogen existing as either NO, NH~ or HCN at various stoichiometric ratios
for four fuels. Above SR =0.8, NO was the dominant TFN species; at lower
stoichiometries HCN dominated with all fuels tested except the CH,/NH_.
Axial profiles with the liquid fuels indicate that near SR^O.8, signifi-
cant amounts of NH, may be formed early in the rich zone but they decay
rapidly. These data are in strong contrast to similar results obtained with
pulverized coal (20) which indicate that the preferred TFN species is a
strong function of coal composition.
In general, both the alternate and petroleum-derived liquid fuels
behaved very similarly with the exception of the Kern County, California
crude (R14). It produced less HCN under rich conditions, and this tendency
cannot be readily associated with common fuel properties. Hydrocarbon and
nitrogen distillation data indicated that in terms of equilibrium volatile
evolution the Kern County fuel is intermediate among the liquids tested.
The Indonesian oil was the lightest of the liquid fuels and it produced the
highest TFN concentration at the minimum (SR =0.8).
SECOND-STAGE NOV FORMATION
A
Exhaust NO emissions in a staged combustor result from conversion of
x
TFN exiting the first stage and any thermal NO production during burnout.
X
Thermal NO production was not considered to be significant in this study
X
because changes in heat extraction in the burnout region had almost no effect
on final emissions. Figure 22 shows exhaust NO emissions as a function of
total fixed nitrogen in the first stage at stoichiometries between 0.5 and
0.8 for all fuels. The form of this correlation can be compared with that
presented in Figure 12 for excess air conditions since the second stage burnout
can be considered an excess air flame. Exhaust emissions increase with
increasing oxidizable nitrogen content, but the conversion efficiency
decreases as the TFN concentration increases. There are three possible
explanations for the data scatter shown in Figure 22: (1) TFN is not indi-
cative of the oxidizable nitrogen compounds that are leaving the first stage;
(2) TFN conversion in the burnout zone is dependent upon the form of the TFN;
196
-------
100
sol—
0)
o
•3 100
I
O
JJ
g
u
VI
0)
(X,
I I I
INDO/MALAYS IAN (-R4)
CH.-HIH
SRC-II HEAVY DISTILLATE (A9)
KERN COUNTY (R14)
0.6
SR,
Figure 21. Distribution of First-Stage XN Species for Alternative
and Petroleum-Derived Fuels.
-------
400
10
oo
300
fr
T3
*~s
W
"~><
O
55
200
£
100
j i
I I II
200
400
600
PPM
800
1000
1200
1'400
(dry, 0%02>
Figure 22. Exhaust NO Versus TFN at the Exit of the First Stage.
-------
and (3) TFN conversion is also dependent upon the oxidation of the partial
products of combustion at the exit of the fuel-rich zone.
IMPACT OF THERMAL ENVIRONMENT
The TFN concentrations shown in Figure 21 are in excess of equilibrium
levels and Sarofim and co-workers (25) have suggested that increasing the
temperature of the primary zone would prove beneficial. The results presented
in Figure 23 were obtained with the shale crude (A3) to demonstrate the impact
of first- and second-stage heat removal on the fate of fuel nitrogen. Fig-
ure 23a indicates that adding the radiation shield with cooling coils in both
the first- and second-stage (hence, increasing the temperature of the vapori-
zation zone) reduced the minimum NO emissions and shifted the optimum stoi-
chiometry more fuel-rich. Figure 23b shows that removing the water cooling
coils from the first stage reduced the exhaust emissions. Removing the second
stage coils did not alter the minimum level; however, it did shift the minimum
SR more fuel-rich. Thus, the optimum thermal environment has a high tempera-
ture vaporization zone, a hot, rich hold-up zone, and a cooled second stage
(Figure 23c).
The axial profiles (22) provide an explanation for this shift in the
minimum emission levels. Heat extraction in the first stage impacts the
rate of decay of TFN. Under cold conditions, both NO and HCN essentially
freeze, whereas without heat extraction the initial rate of decay for all
three species is much faster leading to low TFN concentrations at the exit
of the fuel-rich first stage. It should be noted that heat extraction also
affects the rate of CO oxidation.
199
-------
ro
o
o
300
00
o
200
a.
100
^ Without Radiation Shield
r-\ With Radiation Shield -
LJ First Stage
Primary + Secondary Cooling
NO Cooling
Secondary Cooling
Primary + Secondary Cooling
Radiation Shield +
Secondary Cooling
007
0.9
Q.7
,0.8
0.9
0,7
0.8
0.9
SR,
(a)
Figure 23.
(b)
(c)
Influence of Heat Extraction Profile in the First and
Second Stage Upon Exhaust NOV Emissions.
J\
-------
SECTION 6
CONCLUSIONS
A rich/lean series staged combustor with a platinum/nickel oxide primary
was the most promising low-NOx combustor investigated with LEG. It had low conver-
sions of fuel nitrogen to NO over a wide range of fuel-rich primary stoichiometries.
X.
Thus, it could be operated rich enough to maintain the adiabatic flame tempera-
ture relatively cool, prolonging the life of the catalyst. However, catalyst
coated ceramics are often short-lived due to loss of activity of the coating
and structural problems of the support caused by thermal shock. During the
course of the bench-scale experiments there was a great change in the appearance
of the Pt/NiO catalyst. A green coating formed on the surface. Also, the zirconia
honeycombs became quite fragile after repeated thermal cycling, especially the
fine-cell downstream monolith which was almost completely destroyed in the final
experiments. Further investigation is necessary of catalyst aging and of pressure
and throughput effects under optimized combustor conditions before a catalytic
combustor could be considered a serious candidate for a gas turbine combustor.
A rich/lean series staged combustor with a diffusion flame primary also
had low conversions of fuel nitrogen to NOX. Primary stoichiometry and residence
time had the most significant effects on fuel nitrogen conversion. Minimum NOX
emissions were achieved at primary stoichiometries around 90 percent theoretical air
for long primary residence times (250 msec or longer). Pressure and Reynolds
number had little effect on NOX in a staged diffusion flame, while an increase
in the hydrocarbon content of the LEG caused a slight increase in NOX emissions.
Combustion of a hydrocarbon-free LEG was not tested on the bench scale, but
laboratory-scale tests indicated that the absence of hydrocarbons in the fuel
could cause a significant reduction in NOX emissions.
A lean unstaged diffusion flame produced higher NOX emissions than the
rich/lean staged diffusion flame. However, because of its simplicity, it
remains an attractive low NOX combustor concept. The influence of Reynolds
number on NOX levels in the lean flame suggests that NOX emissions could be
lowered by utilizing larger fuel tubes, perhaps approaching the levels achieved
by the staged diffusion flame.
201
-------
It is planned to investigate other combustor configurations including a
premixed backmixed simulated stirred reactor and a combination diffusion flame/
catalyst hybrid combustor. The zero dimensional stirred reactor is easy to
model. It will provide experimental feedback for the fuel nitrogen processing
kinetics code to be used in future prototype combustor design. The hybrid
system will input low ZXN containing fuel-rich diffusion-flame exhaust into a
Pt/NiO cleanup catalyst prior to secondary burnout.
The results of the bench-scale studies on the influence of liquid fuel
properties and thermal environment on NOX formation indicate that:
• With liquid fuels, fuel nitrogen content is the primary composi-
tion variable affecting fuel NO formation. NO emissions increase
X
with increasing fuel nitrogen. Alternative liquid fuels correlate
with the high-nitrogen petroleum oils.
• Staged combustion dramatically reduces both fuel and thermal
NO formation. Minimum emissions occur at a primary zone
stoichiometric ratio between 0.75 and 0.85 depending on the
combustion conditions.
• First-stage stoichiometry determines the dominant TFN species.
Below SR =0.8 HCN is the dominant species, and above SR =0.8, NO
is the dominant species. NH_ concentrations at the first-stage
exit generally accounted for less than 20 percent of the fuel nitrogen,
• Exhaust NO emissions are directly related to the TFN concentra-
tion at the first-stage exit. NO control for high-nitrogen fuels
X
is most effective when a rich primary zone is held at an optimum
stoichiometry to minimize the TFN concentration. This concentration
is further minimized by increasing the temperature of the fuel-rich
zone.
202
-------
Figure 24 summarizes the impact of fuel-bound nitrogen content on mini-
mum emissions observed under staged combustion conditions and the associated
TFN. Under optimum staged conditions NO emissions (and TFN) correlate well
X
with total fuel nitrogen content. Only the SRC-II heavy distillate ( A )
exhibited unusually high emissions and this was the direct result of a high
TFN yield. These results suggest that NO emissions resulting from the com-
j£
bustion of coal- or shale-derived liquid fuels can be controlled in a cost-
effective manner by modification to the combustion process. Low-NO combus-
3£
tors can be designed which are tolerant to wide ranges in fuel-bound nitrogen
content. Thus, the production of alternate fuels should be optimized without
regard for the reduction of fuel nitrogen content as a method of controlling
NO emissions from stationary sources.
203
-------
400-
I I I I I I
III I I I
MINIMUM (NOY)E
A
AT
MINIMUM (NOX)E
100 —
0.4 0.8 1.2 1.6 2.0
Fuel Nitrogen (weight - percent)
Figure 24. The Effect of Fuel-Bound Nitrogen Content on
Exhaust NOX and TFN in the Primary Zone
(SR = 0.78, 3% overall excess 0^.
204
-------
REFERENCES
1. Mansour, M. N. and M. Gerstein. Correlation of Fuel Nitrogen Conversion
to NOX During Combustion of Shale Oil Blends in a Utility Boiler. In:
Proceedings of Symposium on Combustion of Coal and Synthetic Fuels,
American Chemical Society, March 1978.
2. Heap, M. P. et al. Control of Pollutant Emissions from Oil-Fired Package
Boilers. In: Proceedings of the Stationary Source Combustion Symposium,
EPA-600/2-76-1526, NTIS, Springfield, Virginia, 1976.
3. Bowman, C. T. Kinetics of Nitric Oxide Formation in Combustion Processes.
In: Proceedings of Fourteenth Symposium (International) on Combustion,
The Combustion Institute, Pittsburgh, Pennsylvania 1973.
4. Martin, G. B. and E. E. Berkau. An Investigation of the Conversion of
Various Fuel Nitrogen Compounds to NO in Oil Combustion. In: Proceed-
ings of AIChE Symposium Series No. 126, 68, 1972.
5. Turner, D. W., R. L. Andrews, and C. W. Siegmund. Influence of Combus-
tion Modification and Fuel Nitrogen Content on Nitrogen Oxide Emissions
from Fuel Oil Combustion. In: Proceedings of AIChE Symposium Series
No. 126, 68, 1972.
6. Pershing, D. W., J. E. Cichanowicz, G. C. England, M. P. Heap, and
G. B. Martin. The Influence of Fuel Composition arid Flame Temperature
on the Formation of Thermal and Fuel NOX in Residual Oil FLames. In:
Proceedings of Seventeenth Symposium (International) on Combustion, The
Combustion Institute, Pittsburgh, Pennsylvania, 1979.
7- Heap, M. P. NOX Emissions from Heavy Oil Combustion. International
Flame Research Foundation Report for Contract 68-02-0202, IJmuiden,
Holland, 1977.
8. Malte,. P, C. The Behavior of NH and CN in Nitrogen-Doped High Intensity
Recirculative Combustion. Paper presented at the WSS Combustion Institute,
Berkeley, California, October 1979.
9. Takagi, T., T. Tatsumi, and M. Ogasawara. Nitric Oxide Formation from
Fuel Nitrogen in Staged Combustion: Roles of HCN and NHi. Combustion and
Flame, 35, 17, 1979.
205
-------
10. Fenimore, C. P. Studies of Fuel Nitrogen species in Rich Flame Cases.
In: Proceedings of Seventeenth Symposium (International) on Combustion,
The Combustion Institute, Pittsburgh, Pennsylvania, 1979.
11. Haynes, B. S. Reactions of NH3 and NO n'n the Burnt Gases of Fuel-Rich
Hydrocarbon-Air Flames. Combustion and Flame, 28, 81, 1977.
12. Gerhold, B. W., C. P- Fenimore, and P. K. Dederick. Two Stage Combustion
of Plain and N Doped Oil. In: Proceedings of Seventeenth Symposium
(International) on Combustion, The Combustion Institute, Pittsburgh,
Pennsylvania, 1979.
13. Haebig, J. E., B. D. Davis, E. R. Dzuna. Preliminary Small-Scale Com-
bustion Tests of Coal Liquids. Environmental Science Technology, 10:3,
243, 1976.
14. Muzio, L. J., and J. K. Arand. Small-Scale Evaluation of the Combustion
and Emissions Characteristics of SRC Oil. Paper presented at the ACS
Fuel Chemistry Symposium, Anaheim, California, March 1978.
15. Downs, W., and A. J. Kubasco. Characterization and Combustion of SRC-II
Fuel Oil. EPRI Report No. FP-1028, Palo Alto, California, 1979.
16. Mansour, M. N. Factors Influencing.NOX Production During the Combustion
of SRC-II Fuel Oil. Paper presented at the WSS Combustion Institute,
Berkeley, California, October 1979.
17. Hersch, S., B. F. Piper, D. J. Mormile, G. Stegman, E. G. Alfonsin, and
W. C. Rovesti. Combustion Demonstration of SRC-II Fuel Oil in a Utility
Boiler. Paper presented at the ASME Winter Annual Meeting, New York,
New York, December 1979.
18. Dzuna, E. R. Combustion Test of Shale Oils. Paper presented at the CSS
Combustion Institute, Columbus, Ohio, April 1976.
19. Mansour, M. N., and M. Gerstein. Correlation of Fuel Nitrogen Conversion
to NOX During Combustion of Shale Oil Blends in a Utility Boiler. Paper
presented at the ACS Symposium on Combustion, Anaheim, California,
March 1978.
20. Chen, S. L., M. P. Heap, R. K. Nihart, D. W. Pershing, and D. P. Rees.
The Influence of Fuel Composition on the Formation and Control of NOX in
Pulverized Coal Flames. Paper presented at the WSS Combustion Institute,
Irvine, California, 1980.
21. England, G. C., D. W. Pershing, J. H. Tomlinson, and M. P. Heap. Emis-
sion Characteristics of Petroleum and Alternate Liquid Fuels. Paper
presented at the AFRC NOX Symposium, Houston, Texas, October 1979.
22. England, G. C., M. P. Heap, D. W. Pershing, R. K. Nihart, and G. B. Martin.
Mechanisms of NOX Formation and Control: Alternative and Petroleum-Derived
Fuels. Paper presented at the Eighteenth Symposium (International) on Com-
bustion, The Combustion Institute, Waterloo, Canada, August 1980.
206
-------
23. Corley, T. L. Development of a Kinetic Mechanism to Describe the Fate
of Fuel Nitrogen in Gaseous Systems. Paper presented at the Fifth
E.P.A. Fundamental Combustion Research Workshop, Newport Beach, California,
1980.
24. Folsom, B. A., C. W. Cpurtney, and M. P. Heap. The Effects of LEG Com-
position and Combustor Characteristics on Fuel NOX Formation. Paper
presented at the ASME Gas Turbine Conference and Exhibit and Solar
Energy Conference, San Diego, California, 1979.
25. Sarofim, A.F., J.. H. Pob.1 and B. R. Taylor. Mechanisms and Kinetics of
NOX Formation: Recent Developments. Paper presented at the 69th Annual
Meeting AICHE, Chicago, Illinois, 1976.
207
-------
PROBLEM-ORIENTED REPORT:
UTILIZATION OF SYNTHETIC FUELS:
AN ENVIRONMENTAL PERSPECTIVE
E.M. Bohn, J.E. Cotter, J.O. Cowles,
J. Dadiani, R.S. Iyer, J.M. Oyster
TRW
Energy Systems Planning Division
8301 Greensboro Drive
McLean, VA. 22102
ABSTRACT
This paper discusses the potential environmental problems arising
from the refining, transportation, storage and utilization of fuels
produced by a synthetic fuels industry. Scenarios defining possible
build-up rates for synfuel products from oil shale and coal conversion
are developed to scope the magnitude of potential exposures. The
market infrastructure for the use of these products is examined and
the potential public health risks during the handling, transportation
and utilization of these synfuel products is evaluated. Significant
issues regarding environmental impacts and the need for regulatory
attention are discussed.
208
-------
SUMMARY
PLANNING FOR SYNFUELS UTILIZATION MUST BEGIN NOW
This document is a preliminary overview intended to broadly sketch out
the essential facts of interest to EPA about the utilization of synfuels
and their potential environmental impacts. It is also intended to present
an overall environmental perspective. A Final Environmental Market
Analysis Report will be developed with the purpose of analyzing specific
areas of relevance to EPA in greater depth and noting possible EPA
activities for mitigating potential environmental impacts of synfuels.
EPA is currently sponsoring projects focussed on the environmental
aspects of coal and shale conversion processes. This document deals more
with the fate of synthetic fuel products after they leave the plant gate.
Future work will be concerned in more detail with the estimated national
flow rates and paths of such products and byproducts, their hazards to
human health, and the risks of public exposure to these synthetic fuels.
In carrying out its mission of preserving the quality of our natural
environment, EPA has the responsibility to keep fully abreast of synthetic
fuel developments because a reasoned approach to dealing with the
environmental impacts of a synfuels industry requires accurate knowledge
about current synfuels processes and commercial applications.
Current trends in the international energy situation are rapidly
increasing the probability that a domestic synthetic fuels industry will
emerge in the 1980s. Because government incentives and private ventures in
the synfuel arena are burgeoning in response to soaring world oil prices
and decreasing reliability of oil imports, forecasters are now projecting
earlier start dates, faster growth rates, and larger ultimate sizes for
such an industry.
Several synfuel technologies are under consideration for commercial
production. A wide range of synfuel products are expected to be produced
and they will be utilized in a broad category of end uses (reference Table
1). Synfuels products will most likely be used largely as transportation
fuel, including gasoline and diesel fuel from refined shale oil and coal
conversion processes and jet fuel from refined shale oil. Utility and
209
-------
Table 1. Synfuels Market Overview
HHAT TECHNOLOGIES PRODUCE SYNFUELS?
WHAT MAJOR PRODUCTS/
BYPRODUCTS WILL THEY
MAKE?
WERE WILL THE PRODUCTS/
BYPRODUCTS BE USED?
WAT ARE THE RELATIVE POTENTIAL
EXPOSURE LEVELS TO THE PRODUCTS?
OIL SHALE: NUMEROUS RETORTING Syncrude upgraded and
PROCESSES. INCLUDING refined to yield:
TOSCO. PARAHO. UNION, LPG
OCCIDENTAL Gasoline
Jet Fuel
Diesel Fuel
Residuals
Lubricants
taxes'
DIRECT COAL LIQUEFACTION: SRC-II LPG
Naphtha
Fuel 011
SNG6 ,
Tar Oils'
EXXON DONOR SOLVENT Propane
Butane
Naphtha
Fuel 011
Solvent*
H-COAL Naphtha
Fuel 011
• Commercial and military
transportation. Including
highway vehicles, aircraft,
ships
• Utility and Industrial
boilers
• Commercial and residential
heating
• Industrial lubricants
• Utility and Industrial
boilers
• Commercial and residential
heating
• Chemical feedstocks
• Utility and Industrial
boilers
• Commercial and residential
heating
e Paint thinner s
e Utility and Industrial
boilers
• Comnerclal and residential
heating
Low for transport of crude shile
to refinery; moderate during re-
fining and end use is boiler fu»);
Increased exposure level Nhen used
In transportation sources ind
space heating.
Low for LPG, SNG, Napnthi, Butint;
Moderate t«posure for fuel oils it
Industrial sites *ith exposure In-
creasing when used in spict
heating.
INDIRECT COAL LIQUEFACTION: FISCHER-TROPSCH Gasoline
LPG
Diesel Fuel
Heavy Fuel Oil
Medium Btu Gas
SNG
Tar Oils'
Phenols1
Chemical Feedstocks'
Pesticides
Fertilizers'
M-GASOLINE Gasoline
LPG
"ETHANOL Methyl Fuel
Methanol
e Commercial and military
transportation
e Utility and industrial
boilers
e Commercial and residential
heating
e Chemical feedstocks
e Agricultural uses
• Commercial and military
transportation
e Commercial and military
transportation
e Chemical feedstocks
Low for LPG. SN6. IP* NeaMunritu
Gas. Moderate) exMture 'When fuels
used in transportation sourcei ind
boilers. Low 10 imerajte iiposurt
is als6 estimated xneo products
used as cha*(carfeed»tocks.
HIGH BTU COAL GASIFICATION:
MEDIUM/LOW BTU COAL GASIFICATION:
NUMEROUS PROCESSES.
INCLUDING LURGI,
COED-COGAS, TEXACO.
Shm-KOPPERS
NUMEROUS PROCESSES
SHGb
Medium Btu Gas
Low Btu Gas
e Commercial and residential
heating
• Captive fuel use for
Industrial heating and
chemical feedstocks
Very low - similar to current
distribution of naturil gis.
Very low since it Is prUnrlly
captive use.
S0nly representative byproducts ire Indicated.
bSubst1tutt Ntturil Gas
210
-------
industrial boilers will utilize the fuel oils produced from coal liquids.
High-, medium-, and low-Btu gases from coal will find use in commercial,
residential, and industrial heating applications. The products from most
synfuel processes will be used as chemical feedstocks in a large variety of
industries.
The national environmental impacts of a large-scale synfuels industry
could be significant. The environmental concerns of end use, including
handling and transport, will have to be investigated in detail. Since
there is limited information concerning the end-use exposure effects of
synthetic fuel products and by-products, the nature of these future impacts
is largely speculative. In fact, since synthetic fuel technology is highly
evolutionary, even the composition and amounts of future industrial
synthetic fuel products and by-products are not well known.
In this report the term synfuel product refers to primary products of
the synfuel industry such as gasoline, high-, medium-, and low-Btu gas,
whereas the term by-product has been used to identify secondary useful
products that are likely to be produced from synfuels such as plastics,
solvents, varnishes, and fertilizers.
211
-------
A SYNFUEL INDUSTRY IS EMERGING
INCENTIVES FOR SYNFUEL DEVELOPMENT ARE HERE
The primary incentive for synfuels development is the imbalance
between domestic supply and demand for petroleum liquids and natural gas.
The long-term decline in domestic oil production coupled with increased
demand has resulted in a level of oil imports of 9 million barrels per day
(MMBPD) of oil or about 50 percent of U.S. consumption. The proven domes-
tic reserves of natural gas are also declining and demand is now being met
with increasingly higher priced supplies.
A substantial market for liquid synthetic fuel products and chemical
feedstocks is expected by 1990. A recent analysis concludes that about 2.9
Quads of energy or about 1.5 MMBPD will have to be supplied from synthetic
liquid fuels (reference Table 2). As indicated in this analysis, use of
synfuels is expected to be heavily directed toward transportation. Industry
concern over potential interruptions in gas supplies has provided the
incentive to develop coal gasification processes to supplement current gas
supplies and for use as chemical feedstocks.
It is these considerations, along with the uncertainty inherent in the
import supplies and the increasing problem of balance of payments, that now
provide the impetus for Federal support for synfuels development. Recent
Federal action creating the Synthetic Fuels Corporation (SFC) is aimed at
alleviating some of the factors that to-date have discouraged development.
The goal of the SFC, with authorized funds for loan guarantees, cooperative
agreements, and price supports, is to reduce and share the investment risk
of establishing a commercial synfuels industry.
Now, as the U.S. synfuels industry is a developing reality, the EPA
will need to initiate close coordination with the SFC. As EPA takes the
lead in regulatory approvals, other regulatory agencies will be encouraged
to participate. A well organized, coordinated approach on the part of all
Federal agencies will be viewed as an added incentive by the developing
synfuels industry.
212
-------
Table 2. Anticipated Liquid Fuel Products Demand in 1990a/
Gasoline
Jet Fuel
Kerosene
Heating Oils
Residual
Asphalt
Misc. Product
LPG
Supplied to
Consumers
Quads
14.2
2.1
0.3
6.4
3.2
1.0
2.4
0.7
30.3
Petroleum
Supplies
Quads
12.7
1.8
0.3
5.7
3.2
1.0
2.0
0.7
27.4
Syn fuels
Quads
1.5
0.3
-
0.7
-
-
0.4
-
2.9
a Coal Technology Market Analysis, ESCOE, January 1980. Assumes
U.S. refineries will operate with the same mix as 1978.
b 1 Quad/yr » 0.5 MMBPD
SYNFUELS UNDER CONSIDERATION
The term "synfuels" has become synonymous with any combustible
nonpetroleum fuel source which may include coal- and shale-derived fuels
and feedstocks as well as those derived from agricultural products such as
grain, wood, and cellulose. However, industry has become increasingly more
interested in synfuel technologies with products that are easily substi-
tuted, in a marketing and utilization sense, for petroleum and natural gas.
These synfuel technologies are those relating to coal- and shale-derived
products. The following discussion is limited only to these products.
213
-------
OIL SHALE RETORTING TECHNOLOGY IN HIGH GEAR
SHALE OIL MAY BE FIRST SYNFUEL TO ENTER THE MARKET AS A PETROLEUM
REPLACEMENT
As a-direct substitute for large volumes of liquid fuels* oil shale
technology is perhaps closest to commercialization in the U.S. Several;
consortia and companies with established shale oil projects have been
engaged in the development of shale oil technology for some time.
(reference Table 3). These projects are all located in prime shale areas
of Colorado and Utah.
Many technologies are being developed and tested which are aimed at
extracting kerogen, a waxy organic material, from shale. Most involve
heating shale to about 480°C and pyrolyzing the kerogen into a viscous.
liquid called shale oil. They differ in the manner in which this heating
process is accomplished; surface retorting, in situ retorting, or modified
in situ retorting.
In surface retorting, oil shale is mined, crushed to the proper size
and then fed to a large kiln for heating. Several surface retorting
processes are under development and they differ primarily in the heating
method employed. Internal-combustion retorting heats shale by the circu-
lation of hot gases that are produced inside the retort by the combustion
of residual carbons in the shale. Gas-cycle retorts used by Union Oil heat
the shale by circulating externally heated fluids; No combustion occurs
inside the retort. In solid-heat-carrier retorting, shale is mixed with
hot solids that are heated outside the retqrt. and..cycled through the.shale.
TOSCO II is an example of this method, using ceramic balls as the heat
carrier.
In situ retorting pyrolyzes "oil shale while it is still in the ground.
1 ',„''""' * " • •
The shale bed is ignited and sustained by injection wells, the shale is
pyrolyzed, and the oil produced is pumped out of the retort volume through
a production well. The spent shale remains in place. Tor successful in
situ retorting, the shale bed must be made permeable to the flow of heat
' "' - - •••- : • :••.-.,' - r v '•' ••<•-, ., .- .-•-•• ,; .
and product oil; various techniques of bed leaching or fracturing are
:."••". ' * " f '..•-•";•;,.'•-• • -••-.!.
employed. The difficulty of creating a permeable shale bed has led"'to the
development of modified in situ processes. Vertical modified in situ
214
-------
Table 3. U.S. 011 Shale Projects
Project
Location1
Technology
Production
Capacity. Goal
(bbl/day)
Status
Chevron
Colony (TOSCO. EXX.ON C)
Equity Oil
Geoklnetlcs, Inc.
Getty Oil
Mobil
Occidental Oil
Occidental Oil - Tenneco
Paraho (Development Engineering.
Inc.)
Rio Blanco (Gulf, Amoco)
Superior OH
TOSCO-Sand Wash
Union Oil
White River tSohlo, Sunoco,
Phi 111 pi)
Plceance Basin
Parachute Creek
Plheance Creek
Ulnta County
Plceance Basin
Plceance Basin
Logan Wash
Tract C-b,
Piceance Basin
Anvil Points;
Tract C-a,
Plceance Basin
Piceance Creek
Uinta Basin
Parachute Creek
Tracts U-a and U-b,
Uinta Basin
Undecided 50.000
Surface retorting 47,000
Solution injection. Modified
in-sltu
Horizontal Modified In-sltu 7-13
: 2,000-5,000
Surface thermal extraction
Undecided 50,000
Vertical modified In-sltu 70,000
Vertical modified In-sltu 57.000
Surface retorting 150-200
Vertical modifted in-sltu, 50,000
surface retorting
HultiMineral recovery, sur- 13,000
face retorting
Modified in-situ, surface 50.000
retorting
Surface retorting 50,000
Modified In-sltu, surface 100.000
retorting
Technical assessment phase.
Construction of commercial nod-
ules scheduled for 1980.
Steam-injection feasibility:
Several snail retorts successfully
burned; work on larger retorts in
progress.
Getty RID proposal being con-
sidered by DOE
May start module in 1987
Six retorts burned; 48>000 bbl
produced. Retorts 7 and 8
scheduled for-cluster bum.
Shaft sinking in progress; con-
struction of initial retorts sched
uled for 1982.
Shut down due to lack of funding;
88,225 bbl produced over about
one year period.
Modular program consisting of 5
retorts scheduled for completion.
by 1982.
Company seeking land exchange
with Federal Government which
was denied in febuary 1980.
Feasibility studies in progress
Construction of experimental
mine and plant scheduled for 1982.
Operations suspended due to
legal proceedings on ownership
of lands.
State Locations: Piceance Basin
Anvil Points - Colorado.
- Colorado; Parachute Creek - Colorado; Uinta County - Utah; togan Wash - Colorado;
-------
(VMIS) retorting removes a portion of the shale from the bottom of the
deposit and fractures the remaining shale to create a chimney of shale
rubble. The shale is retorted in this chimney from top to bottom.
Occidental Oil Company has been testing VMIS retorting on shales at Logan
Wash and Piceance Creek Basin in Colorado. Horizontal modified in situ
retorting lifts the overburden in some cases, and fractures the shale seam
to retort the shale from side to side. Geokinetics, Inc. is developing
this technique in Utah.
The technology for surface retorting is more advanced than in situ
retorting. Process variables are easier to monitor and control in above
ground retorts than in underground retorts. However, large-scale
commercial surface retorting requires large-scale oil shale mining,
hauling, and crushing; and large-scale disposal of spent shale. It is also
limited to that portion of the shale resources that is mineable. In situ
retorting without mining is applicable to a greater variety of shale beds,
and eliminates the requirements for handling, crushing, and spent shale
disposal. Attempts to demonstrate this technology have identified many
development problems. Modified in situ processes present a compromise,
requiring some mining and handling, but offering more process control and
easier development.
The crude shale oil produced by retorting will be upgraded by further
processing. This upgraded shale oil, or syncrude, will be used as a
refinery feedstock or boiler fuel. It is well suited for refining into
middle distillate fuels. If hydrocracking is chosen for the refining
process, the yield and range of products is particularly desirable: motor
gasoline - 17 percent; jet fuel - 20 percent; diesel fuel - 54 percent;
and residuals - 9 percent.
Several oil shale projects, with identified participants, plan to
begin operation during the 1980s. The technologies, which are proprietary
in many cases, appear to be sufficiently mature to move ahead to
commercialization. Several retorts have been successfully operated by
Geokinetics, Inc., Occidental Oil, Paraho, Union, and TOSCO. Colony, Union
Oil, and Occidental Oil have announced plans to begin commercial
development in 1980. All technologies have been demonstrated at pilot
scale or larger.
216
-------
BOTH DIRECT AND INDIRECT ROUTES TO COAL LIQUIDS ARE AVAILABLE
DEMONSTRATION AND FULL-SCALE UNITS ARE BEING ENGINEERED
Coal, hydrogen, and a coal-derived oil are mixed at high temperature
and pressure to accomplish direct liquefaction. Under these conditions,
the coal decomposes, and the fragments react with hydrogen to form addi-
tional derived oil, which is separated from the unreacted solids and
further refined to produce usable liquid fuels. Indirect liquefaction
processes react the coal with oxygen and steam in a gasifier to produce a
synthesis gas composed mainly of carbon monoxide, carbon dioxide, and
hydrogen. After the carbon dioxide and other impurities are removed from
the gas, the carbon monoxide and hydrogen are chemically combined in a
catalytic reactor to produce liquid products for use as chemical feedstocks
or liquid fuels.
There are three major direct coal liquefaction processes currently
undergoing development: SRC II, Exxon Donor Solvent (EDS), and H-Coal
(reference Table 4). These processes differ mainly in the manner in which
the hydrogen is made to react with coal fragments to produce the unrefined
coal liquids. In the SRC II process, the coal feed and hydrogen are mixed
with a process recycle stream that contains unreacted coal ash as well as
coal-derived oil. The iron pyrite in the unreacted ash catalyzes the
reaction between the coal fragments and hydrogen. In the EDS process, the
coal feed and hydrogen are mixed with a specially hydrogenated coal oil
called the donor solvent. The hydrogen added to the coal fragments is
provided by the solvent and the hydrogen gas mixed in the reactor. The
donor solvent is made by catalytically hydrogenating coal-derived oil using
conventional petroleum refinery hydrotreating technology. In the H-Coal
process, the unreacted coal and hydrogen are mixed with coal-derived oil
and an added solid catalyst in a special reactor referred to as an
ebuHated bed.
Once the gases and distil Table liquid products have been separated
from the reactor effluent, the remaining "bottoms" material is processed.
This material contains significant quantities of heavy hydrocarbons which
must be efficiently utilized to enhance process economics. The principal
bottoms processing step under consideration for the EDS process is
217
-------
Table 4. Major Coal Liquefaction Processes
PROCESS
PROCESS TYPE
PRODUCTS
STATUS
Solvent Refined Coal,
SRC II (Gulf Oil)
ro
co
H-Coal
(Hydrocarbon Research,
Inc.)
Exxon Donor Solvent, EDS
(Exxon Research and
Engineering Company)''
Flscher-Tropsch
(M.M. Kellogg/Lurgi)
Mobil M
Direct liquefaction by sol-
vent extraction: coal dis-
solved in solvent, slurry
recycled, catalytic hydro-
genatlon
Direct liquefaction by
catalytic hydrogenation,
ebullated catalyst bed
Direct liquefaction by
extraction and catalytic
hydrogenation of recycled
donor solvent
Indirect liquefaction,
liquefaction of synthesis
gas in an fluid bed
catalytic converter
Indirect liquefaction,
liquefaction of synthesis
gas in fixed bed using
molecular size-specific
zeolite catalyst
LPG
Naphtha
Fuel Oil
SNG
Naphtha
Fuel 011
Propane Butane
Butane
Naphtha
Fuel Oil
Gasoline
LPG
Diesel Fuel
Heavy Fuel Oil
Medium Btu Gas
SNG
Gasoline
LPG
Pilot Plant under operation.
6700 ton/day of coal (20,000
barrels/day of oil equivalent
demonstration module under
design and schedule for oper-
ation In 1984-1985
600 ton/day (1400 barrels/
day of oil equivalent)
pilot plant under construc-
tion, testing will begin
in 1980. Plant is located
at Catlettsburg, Kentucky
250 ton/day (500 barrels/
day of oil equivalent)
pilot plant under construc-
tion, testing will begin in
1980. Plant is located at
Baytown, Texas
SASOL I. 800 tons/day, pro-
ducing over 10,000 bbl day of
liquids in commercial produc-
tion since 1956. SASOL II,
40,000 tons/day, producing
over 50,000 bbl day of liq-
uids has been completed and
will begin start-up in 1980.
SASOL III with approximately
the same capacity as SASOL II
is currently being plan-
ned.
Commercial scale plant to
produce 12,500 barrels of
gasoline using reformed
natural gas is planned for
New Zealand in 1984-1985
-------
FLEXICOKING, which consists of thermal cracking of the bottoms to produce
additional liquids and coke. The coke is subsequently gasified to produce
plant fuel gas or hydrogen for the liquefaction step. Bottoms processing
for the SRC II and H-Coal processes probably will be partial oxidation
(i.e., gasification) to produce hydrogen for the liquefaction step.
There are two major indirect coal liquefaction processes: Fischer-
Tropsch which is commercial now in South Africa, and Mobil-M which is
expected to be commercial in 1983-84. In the Fischer-Tropsch process, the
purified synthesis gas from the gasifier is reacted over an iron catalyst
to produce a broad range of products extending from lightweight gases to
heavy fuel oil. The broad product distribution from this process is
generally considered as a disadvantage where large yields of gasoline are
desired. Improved catalysts are currently being developed at the bench
scale to maximize the yield of gasoline-range hydrocarbons. In the Mobil-M
process, the synthesis gas is first converted to methanol using commer-
cially available technology. The methanol is then catalytically converted
to high-octane gasoline over a molecular-size-specific zeolite catalyst.
Indirect coal liquefaction is successfully operating on a commercial
scale at the SASOL I plant in South Africa using the Fischer-Tropsch
technology. The SASOL I plant produces gasoline, jet fuel, diesel oil,
middle distillates, and heavy oil. SASOL II, producing 50,000 barrels per
day of coal-derived liquids, has been completed and will begin operation
later in 1980. Active interest in this technology has developed and plans
to license and construct similar plants in the U.S. are progressing. There
is strong interest in the Mobil-M gasoline indirect process because of its
attractive high-octane gasoline yield. A commercial-scale plant producing
12,500 barrels per day of gasoline is planned for operation in New Zealand
by 1985.
Direct coal liquefaction technologies are in various stages of
development. SRC I and II processes have been tested at the pilot plant
level and are entering into the demonstration plant stage.
Large pilot plants are currently under construction for testing of the
H-Coal and EDS processes. These plants are located at Catlettsburg,
Kentucky, and Baytown, Texas, respectively.
219
-------
SRC I process produces primarily a solid product with a small amount
of useful liquid product. However, SRC II process produces primarily
liquid products.
In addition to these major coal liquefaction technologies, several
other processes have received attention, including the Dow process, Riser
Cracking, Synthoil, and the Zinc Halide process. All have been tested in
small-scale units.
220
-------
GASEOUS FUELS AND CHEMICAL FEEDSTOCKS FROM COAL
A WIDE VARIETY OF COAL MAY BE USED IN THE SYNFUELS INDUSTRY
Most ccal gasifiers react coal, steam, and oxygen to produce a gas
containing carbon monoxide, carbon dioxide, and hydrogen. When air is used
as the oxygen source, the product gas contains up to 50 percent nitrogen
and is referred to as low Btu gas since its heat of combustion is only 80
to 150 Btu/standard cubic feet (scf). Synthesis gas or medium-Btu gas
ranges from 300 to 500 Btu/scf.
Low-Btu gas is used as a fuel gas near its point of generation since
its low heating value makes it uneconomical to distribute over long
distances. Medium-Btu gas can be used as a fuel gas and transported
economically over distances of up to 200 miles. It can also be used as a
chemical feedstock for the production of methanol or gasoline. Finally, it
can be converted catalytically to substitute natural gas (SNG), having a
heating value of about 1,000 Btu/scf. Additionally, medium-Btu gasifica-
tion is an integral part of all indirect liquefaction technologies.
There are many coal gasification technologies that differ in design
and operation, depending upon the type of coal used and the product
desired. High- and medium-Btu gasification technologies using noncaking
coals characteristic of U.S. western coals are relatively well developed.
Severe operational problems are encountered with commercially available
gasifiers in processing caking coal such as those found in the eastern U.S.
Several gasification technologies for high- and medium-Btu gases are under
active development (reference Table 5). Many additional processes are
being tested, but at less advanced stages of development (reference Table
6).
A fixed-bed gasifier, such as the Lurgi, feeds coal to the top of the
gasifier. The descending coal is successively dried, devolatilized, and
gasified in contact with gases rising from the bottom. Steam and oxygen
are introduced at the bottom of the gasifier, and solid ash is removed
through an ash lock. In some gasifiers, such as British Gas Company (BGC)
Lurgi, the temperature at the bottom of the bed is sufficient to melt the
221
-------
Table 5. Coal Gasifiers for High, Medium and Low Btu Gas
Process
Lurgi Dry Ash
British Gas
Company (BGC)
Lurgi
Texaco
U-Gas Institute
of Gas Tech-
nology (IGT)
Westlnghouse
Shell (Coppers
Koppers-Totzek
Process Type
Pressurized fixed
bed, dry bottom
Pressurized Fixed
bed slagging bottom
Pressurized single
stage entrained,
slurry feed
Pressurized fluid
bed, ash
agglomerating
Pressurized single
stage fluid bed,
ash agglomerating
Pressurized entrained,
dry feed
Atmospheric entrained,
dry feed
Potential
Products
Substitute Natural Gas
(SNG, also known as High
Btu Gas), Medium Btu
Fuel9 Gas. Low Btu Fuel
Gas
SNG, Medium Btu Fuel
Gas, Low Btu Fuel Gas
SNG, Medium Btu
Synthesis Gas, Low
Btu Fuel Gas
SNG, Medium Btu Fuel
Gas, low Btu Fuel
Gas
SNG. Medium Btu Fuel
Gas, Low Btu Fuel
Gas
Medium Btu
Synthesis Gas,
Low Btu Fuel Gas
Medium Btu
Synthesis Gas,
Low Btu Fuel Gas
Most Suitable
Products
SNG. Medium Btu Fuel
Gas, Low Btu Fuel Gas
SNG. Medium Btu Fuel
Gas, Low Btu Fuel
Gas
Medium Btu Synthesis6
Gas
Medium Btu Fuel Gas
SNG, Medium Btu Fuel
Gas
Medium Btu
Synthesis Gas
Medium Btu
Synthesis Gas
Status
40 years of commercial development and 14
commercial plants located in Australia,
Germany, UK, India, Pakistan, South Africa.
Korea. Average module size 800 tons/day
{2000 BOE)C
790 tons/day (of coal) (2000 BOE) pilot
plant tested In Westfleld, Scotland
160 ton/day (400 BOE) plant operating In
West Germany
14000 tons/day of coal plant (35000 BOE)
producing Medium Btu Fuel Gas, under
design for construction in Tennessee
IS ton/day (40 BOE) process development
unit, under testing at Waltz Mill. Pa.
150 ton/day (400 BOE) pilot plant In oper-
ation in W. Germany. 1,000 ton/day
scheduled in 1983/1984.
1,000 ton/day (2500 BOE) plant In opera-
tion In South Africa for the production
of ammonia
ro
CO
a Medium Btu Gas with significant concentration of methane is more suitable for use as fuel, and therefore identified as Medium Btu Fuel Gas.
b Medium Btu Gas with low concentration of methane is more suitable for chemical synthesis,ard therefore Identified as Medium Btu Synthesis
Gas
c BOE - Barrels per day of oil equivalent
-------
Table 6. Status of Other Coal Gasification Processes
DEMONSTRATION PLANTS
HYGAS
COED-COGAS
U-GAS
SCALE
(tons/day
coal feed)
7340
2210
3160
STATUS
Conceptual Design
Detailed Design
Detailed Design
PILOT PLANTS/PDUs
BELL HIGH MASS FLUX 6
BIGAS 120
COMBUSTION ENGINEERING 5
DOW 24
EXXON CATALYTIC 100
GEGAS 24
HYDRANE 4
MOLTEN SALT 24
MOUNTAIN FUEL 12
SYNTHANE 72
TRI-GAS 1
Operational
Operational
Operational
Under Construction
Proposed
Operational
Proposed
Operational
Proposed
Mothballed
Operational
Conceptual design incorporates all important details of major unit
areas in the plant. Material balances are provided around all major
unit areas. (Unit area is a section of the plant consisting of
several components integrated to perform a single transformation on
the product stream. Examples are gasification, raw gas cooling, gas
cleanup, or methanation.)
All equipment and detailed pipeline diagrams are prepared as part of
detailed design. In addition, detailed material balances are prepared
for each piece of equipment.
GThe plant is either operating or has operated successfully in the
past.
223
-------
ash, allowing its removal as molten slag. The slagging feature provides a
distinct advantage in contending with the caking characteristics of eastern
U.S. coals.
Lurgi high-pressure operation, in conjunction with relatively low
gasification temperatures, favors the formation of significant quantities
of methane in the gasifier, enhancing the heating value of the product.
These conditions also favor production of by-products such as tars and
impurities like phenols, organic nitrogen compounds, and sulfur compounds.
In fluid-bed gasifiers currently under development, high-velocity
gases pass up through the bed to fluidize the coal, providing excellent
mixing and temperature uniformity throughout the reactor- Operability
with caking coals (eastern U.S.), as well as low tar production and
tolerance to upsets in fuel rates, has been demonstrated at the pilot scale
for both the Westinghouse and U-Gas gasifiers.
The Texaco and Koppers-Totzek gasifiers are representative of
entrained-bed technology in which the solid particles are concurrently
entrained in the gaseous flow. Flame temperatures at the burner discharges
are in the range of 1370 to 1925°C, resulting in melting of the coal ash
with minimum production of impurities. Entrained-flow gasifiers may be
favored for the production of synthesis gas for indirect liquefaction.
They can operate with caking coals. However, compared to fluid-bed
gasifiers, they have very low carbon holdup capability in the reactor and,
therefore, have limited safeguard against possible formation of explosive
mixture in the reactor in case of coal feed interruption.
There has been extensive commercial experience in the U.S. with
low-Btu coal gasification technologies operating near atmospheric pressure.
However, these applications have been limited to small-scale captive
applications for providing industrial process heat and space heating. For
example, the Wellman-Galusha gasifier designed for atmospheric pressure
operation was used extensively by industry years before pipeline-supplied
natural gas was readily available at comparatively lower cost. Pressurized
gasification processes capable of yielding high-Btu gas for pipeline use
and medium-Btu gas for chemical feedstocks are less developed, with the
exception of the Lurgi fixed-bed process. The Lurgi process is based on 40
years of commercial development at 14 commercial plants that are located in
224
-------
Australia, Germany, U.K., Korea, India, Pakistan, and South Africa. A
great deal of interest in the Lurgi technology is emerging in the U.S. with
several announced plans for SNG production by pipeline and gas utility
companies. Several projects utilizing the Texaco process for captive
applications (chemical feedstocks and on-site power generation) are in the
planning and design stage with at least one project (Tennessee-Eastman)
scheduled for construction in 1980.
225
-------
DEVELOPMENT OF THE SYNFUELS INDUSTRY OVER THE NEXT 20 YEARS
Three scenarios or projections of synfuel industry buildup rates to
the year 2000 have been developed to illustrate the potential range of
synfuel product utilization:
• A "National Goal" scenario driven by Federal incentives
• A "nominal production" or most likely scenario
f An "accelerated production" scenario representing an upper bound
for industry buildup.
ACHIEVING THE NATIONAL GOAL - SCENARIO I
In July 1979, President Carter announced new energy initiatives for
the U.S. aimed at reducing our dependence on imported oil. One of the key
elements of this policy is the provision of Federal funds to stimulate
production of synthetic fuels at the rate of 2.2 million barrels per day
(MMBPD) by 1992. Specifically, the national synfuel goals are:
Coal Liquids. To stimulate and accelerate the construction and
operation of the first few plants to provide sufficient data on the
competing commercial coal liquefaction processes so that industry, with its
own investment, stimulated by Government incentives if required, will build
plants with sufficient capacity to provide upwards to 1 MMBPD liquid
fuels by the year 1992.
Shale oil. To stimulate shale oil production at the rate of 0.4 MMBPD
by 1990.
High-Btu Gas. To develop and implement a program that enables the
U.S., by 1992, to produce significant quantities of pipeline quality gas
(0.5 MMBPD - oil equivalent ) from commercial HBG plants in an
environmentally acceptable manner. This is facilitated by the short-range
goal of having two or three commercial HBG plants in operation by the
mid-1980s.
For easy comparison with petroleum supply/demand figures, synfuel
production rates are expressed in barrels of oil equivalent in this
document. This does not imply that high-, medium- and low-Btu gases from
coal that are substituted for domestic natural gas will have any direct
effect on the reduction of imported oil.
226
-------
Low-/Medium-Btu Gas. To stimulate an initial near-term commercial
capability for several medium-Btu commercial plants in key industries as
well as utilities, for energy and feedstock applications for both single
and multiplant use, and for multiple applications of low-Btu gas in each of
the prime industry markets. Commercial-scale development will depend on
the long-term economics of this technology, vis-a-vis the price of domestic
oil and natural gas. Once a capability has been established, capacity will
be accelerated to achieve at least 0.29 MMBPD oil equivalent by 1992. Of
this total, up to 0.04 MMBPD oil equivalent will be provided from 40 to 50
low-Btu facilities and up to 0.25 MMBPD oil equivalent from 25 to 30
medium-Btu plants. Again, it must be mentioned, that if this low- and
medium-Btu coal-gas is substituted for natural gas, there will not be a
direct effect on the reduction of imported oil.
The key assumptions allowing achievement of these goals are: (1)
Federal funds provided are sufficient to reduce investment risk by the
synfuel industry through 1992, and (2) other requirements for industry
development are satisfied, i.e., environmental permits, material, equip-
ment, and labor. A likely buildup rate profile for the synfuel industries
under this scenario is shown in Figure 1.
LOW/ MEDIUM BTU GAS PLANTS
x<\°X'X°°-"*'-*-"'V«'-vX-X'Xv
Figure 1. Synfuels Industry Buildup for the National Goal Scenario
227
-------
For shale oil, several of the most advanced projects were selected as
a basis. The planned operation startup schedules and capacity buildup
rates for these projects were used to generate the industry production
buildup profile to about 0.4 MMBPD by 1992. The period beyond 1992 is
viewed as one of technology consolidation: gaining a firm footing with
regard to environmental and economic performance and technology improve-
ments. This type of industry production profile is not without precedent;
for example, the Federal support of the synthetic rubber industry during
World War II.
The goal of 1 MMBPD of coal liquids will be met predominantly by
indirect coal liquefaction. At present, the only commercially demonstrated
coal liquefaction process is the Fischer-Tropsch embodied in the SASOL
plants in South Africa. The Mobil-M process should be commercially
demonstrated within the next five years. Considering construction and
permitting lead times, plants of this type could begin operation around
1985. To meet the production goal, 10 to 15 plants of a nominal 0.05 MMBPD
capacity must be in operation by 1992. A potential drawback to the commer-
cialization of SASOL technology in the U.S. is the broad product distribu-
tion, ranging from light hydrocarbon gases to heavy fuel oil. The Mobil-M
technology, on the other hand, produces an all-gasoline product which would
be particularly well suited to the U.S. market demands. Given this
apparent advantage of Mobil-M technology over SASOL, it is believed that
industry should favor commercialization of both Mobil-M and SASOL tech-
nology during the next few years, with the breakdown being roughly 50/50.
Approximately 75 percent of the coal liquids production will be due to
these indirect liquefaction processes.
For the direct liquefaction processes, there will not be sufficient
experience and information to attract any more than developmental interest
over the next few years, under this scenario, By 1985 there should be
sufficient information available from the operation of the EDS, H-Coal and
SRC II plants to support a commercialization decision concerning these
processes. Federal incentives will likely be distributed such that by
1992, three or four pioneer commercial-scale plants employing direct lique-
faction will begin to appear. Of the total production goal of 1 MMBPD of
coal liquids it is estimated that 25 percent will be produced by these
228
-------
first commercial direct liquefaction plants embodying the basic SRC II, EDS
and H-Coal technologies, or improvements and modifications to these. It is
projected that for the next few years after 1992, production will remain at
1 MMBPD while technological evaluations are performed. These direct lique-
faction plants will be located near the major eastern U.S. coal areas.
The Lurgi fixed-bed process is the lead high-Btu coal gasification
technology and has been commercially demonstrated outside the U.S. It is
expected to be utilized in all commercial plants constructed over the next
10 years. As the process requires noncaking coals, these plants will most
likely be located in the western U.S. Interest will continue in other
high-Btu gasification technologies such as the Slagging Lurgi which is
capable of using eastern caking coals. At least one of these alternate or
advanced processes probably will be supported under Federal incentives but
it is unlikely that a commercial plant will appear until the early 1990s,
and this would probably be located near a midwestern coal resource.
The Lurgi fixed-bed medium-Btu process is the lead technology for
medium-Btu gas. Texaco partial oxidation gasification or similar pressur-
ized entrained-bed gasifiers such as pressurized Koppers-Totzek, will be
under development and demonstration during the early 1980s and will likely
serve as the prime medium-Btu gasification process for eastern coals. To
1992, however, the major buildup in medium-Btu gasification will come from
Lurgi plants located in the western U.S.
For low-Btu gasification, the several technologies that are currently
available and providing commercial service are assumed to be easily
applied, under the incentives existing to 1992, to generate the 0.04 MMBPD
production rate goal. Low-Btu gas will generally be captively employed as
fuel gas or used on-site for combined-cycle power generation.
The production buildup profile for major synfuel products resulting
from of the synfuels industry buildup in Scenario I is shown in Figure 2.
These product quantities are projected to enter commercial use and are to
be considered in assessments of potential environmental impacts from
synfuels. Naturally, these major products are presented for the sake of
clarity, but there are many other products and byproducts that will be
produced and distributed into the market place. These products and
byproducts will also vary in greater or lesser quantities in Scenarios II
and III which follow. 229
-------
LOW/MEDIUM BTU GAS PLANTS
l?55^*5!£3?f??f*=s?»»sii
^-^--====5--:^=c--2^r.-^3
GASOLINE, NAPHTHA. AND LPG PLANTS.*
MIDDLE DISTILLATES HANTS:
Figure 2. Major Synfuel Product Buildup for
the National Goal Scenario
PRODUCTION AT A NOMINAL RATE - SCENARIO II
Recent studies of the technical capability of the U.S. to meet the
synfuel national goal point out that there are significant concerns
regarding achieving this goal. They include:
t Availability of skilled manpower: it is expected that the supply
of engineers and construction labor will be severely taxed to meet
the synfuel production goal set forth in Scenario I.
t Availability of critical equipment: certain critical equipment for
the synfuel industry such as compressors, heat exchangers, and
pressure vessels are expected to be in short supply unless
corrective measures are taken now, thus slowing the synfuel
industry buildup rate indicated in Scenario I.
• Diversion of investment to competing technologies: demand on the
limited capital available in the economy by competing energy supply
technologies, such as coal liquefaction, coal gasification, oil
shale, geothermal, and solar technologies, could result in the
slowing of buildup rates for some technologies.
• Environmental data: lack of environmental data needed for
regulatory approvals could slow down the buildup rate.
230
-------
• Licensing: time and construction schedule constraints imposed by
State and Federal licensing and permitting requirements could
hinder synfuel industry buildup rate.
Taking these concerns into consideration, a nominal synfuels
production buildup - Scenario II - has been developed, as indicated in
Figure 3. A production rate of about 2.1 MMBPD is estimated by the year
2000, instead of 1990 as indicated in Scenario I. The technologies
expected to contribute to both Scenarios I and II are the same; the major
difference is in the rate of buildup: it is slower and delayed in time.
Figure 3. Synfuels Industry Buildup for the
Nominal Production Scenario
For shale oil, a nominal production rate of 0.4 MMBPD should be
achieved by the year 2000. The buildup rate is estimated to lag about 4
years behind that of Scenario I and is based on the following observations:
t Some technologies are still considered developmental, such as the
modified in situ process.
• Land problems, including availability of off-tract disposal sites,
may take longer to resolve.
Under this scenario, no large-scale commercial coal liquefaction
plants are projected to be on line until 1992 with a growth rate beyond
yielding 1 MMBPD by the year 2000. It is believed that Federal incentives
will be applied to support construction of one each of the indirect
231
-------
liquefaction plants and a direct liquefaction plant only after sufficient
assessment has been made of the operations of the EDS and H-Coal pilot
plants and the SRC II demonstration plant. Rather than commit sizable
resources to the commercialization of indirect liquefaction, a decision
probably will be delayed resulting in no operating commercial liquefaction
plants before 1992 under this scenario. During the 1980s it is believed
that improvements will be made in both the operating indirect liquefaction
plants and the designs of the direct liquefaction processes. These
"advanced" technologies with product slates yielding primarily trans-
portation fuels, will be sufficiently attractive to encourage development
of 1 MMBPD of coal-derived liquid production by the year 2000.
Currently there is a great deal of interest in SNG technology.
Several gas utility and pipeline companies have expressed plans to con-
struct high-Btu plants. With incentives, several of these plants will be
constructed and in operation by 1985. However, as a result of the pro-
jected improved outlook for gas supplies, including potential from uncon-
ventional sources, the availability of "imported" conventional natural gas
(Alaskan, Canadian and Mexican) and the current unfavorable rate-structure
pricing policy, the complete commercialization of HBG will be hampered.
Its production rate is not likely to expand beyond the 0.25 MMBPD-level
attained around 1992 under this scenario.
The buildup of medium-Btu gas plants will also be impeded by the
availability of natural gas; however, for certain industrial applications
requiring large volumes of uninterrupted supplies (e.g., chemical feed-
stocks, cogeneration) low-/medium-Btu plants will remain attractive. It is
estimated that production of low-/medium-Btu gas will reach a level of 0.45
MMBPD by 1992.
ACCELERATED PRODUCTION - SCENARIO III
The accelerated production scenario is based on the assumption that
Federal incentives are sufficient to synfuels production to meet the
national goals in 1992, that operation of synfuels plants up to 1992 is
successful to the extent confidence in processes is gained, and all
resource requirements are satisfied. Licensing and permitting procedures
must also be streamlined. It is assumed that demand for coal-derived
synfuels remains at a level such that new plant capacity continues to be
232
-------
added to the year 2000 at about the same rate as the buildup to 1992. For
shale oil, the production of 0.9 MMBPD by the year 2000 is based on a
survey and analysis of the desired goals of each industrial developer. As
indicated in Figure 4, a total synfuels production rate of 5 MMBPD may be
reached by the year 2000. This includes 2.6 MMBPD of coal liquids, 1.5
MMBPD of gas and 0.9 MMBPD of shale oil.
I
VIA! I9«4
If 14
I9M
1990
199]
1994
199*
I99« 9000
Figure 4. Synfuels Industry Buildup for the
Accelerated Production Scenario
233
-------
However, in view of the limitations facing the synfuel industry, some
of which were discussed earlier, the accelerated production scenario is
highly unlikely. The synfuels industry buildup rate (Figure 4) for this
scenario can be considered an upper bound to synfuels utilization over the
next 20 years.
The three scenarios describing possible synfuel industry buildup
profiles provide a basis for projecting the market penetration of synfuel
products in the near future. As these products enter the market, potential
environmental impacts related to synfuels utilization must be considered.
234
-------
THERE IS A LARGE POTENTIAL MARKET FOR SYNFUEL PRODUCTS AND BY-PRODUCTS
The major synfuel products could be broadly classified into five
groups:
• Gaseous Products
- High-Btu gas
- Medium-Btu gas
- Low-Btu gas
- Liquified Petroleum Gas (LPG)
• Light Distillates
- Gasoline
- Naphtha
t Middle Distillates
- Jet fuel
- Kerosenes
- Diesel oil
• Residue
- Heavy fuel oil
- Lubricants
• Petrochemicals.
GASEOUS PRODUCTS
The high- and medium-Btu gases are suitable for essentially all
industrial fuel applications that can be serviced by coal, oil or natural
gas. In some cases equipment modifications or special controls will have
to be implemented to retrofit existing plants for medium-Btu gas, whereas
this problem may not exist for high-Btu gas installation. However, there
should be no difficulty in employing either high- or medium-Btu gas in new
industrial installations. These products will be utilized by major energy
consuming industries such as food, textile, pulp and paper, chemicals, and
steel. It appears that only chemical, petroleum, and steel industries
will require sufficient fuel gas at a single location to economically
justify the dedication of a single gasification plant. Other industrial
plants will have to share the output distributed by pipeline from a central
gasifier, or tap into the existing natural gas pipeline system for their
need. Preliminary economic studies indicate that it is not economical to
235
-------
transport medium Btu gas through pipelines for more than 200 miles.
Medium-Btu gas can also be utilized by the petrochemical industries as
chemical feedstock for the production of ammonia, methanol, and
formaldehyde. Currently most of this requirement is met by reforming
natural gas. The use of medium-Btu gasification appears especially
attractive when integrated with new combined plants for utility
applications.
The major characteristics of low-Btu gas are its high nitrogen
content, low carbonmonoxide and hydrogen content, and resulting heating
value typically below 150 Btu/SCF. Its flame temperature is also about 13
percent lower than that of natural gas. Because of these characteristics
low-Btu gas is limited to on-site use, industrial processes requiring
temperature below 2800°-3000°F, and is generally unsuitable for use as a
chemical feedstock. Further, because of its low energy density it requires
significant equipment modifications for retrofit applications. Today there
are operating and planned low-Btu gasifiers in the U.S. for:
• Kiln firing of bricks
• Iron ore pelletizing
• Chemical furnace
• Small boilers
Liquified petroleum gas (LPG) has applications for industrial,
domestic, and transportation uses. In domestic applications LPG is used
mainly as a fuel for cooking and for water and space heating. In industry,
LPG finds a large number of diverse outlets. Apart from use as a fuel in
processes which require careful temperature control (glass and ceramics,
electronics) or clean combustion gases (drying of milk, coffee, etc.), LPG
is also used in the metallurgical industry to produce protective
atmospheres for metal cutting and other uses. The chemical industry,
particularly on the U.S. Gulf Coast, uses petroleum gases for cracking to
ethylene and propylene as well as for the manufacture of synthesis gas.
Small portions of LPG are also used to fuel automotive vehicles. Another
use of LPG is to enrich lean gas made from other raw materials to establish
proper heating value levels. On a volume basis, production of LPG in the
U.S. exceeds that of kerosene and approaches that of diesel fuel. About 40
percent of LPG production is used by the chemical industry, another 40
percent is for domestic use, 10 percent for automotive use, and the
236
-------
remaining distributed among other industrial and agricultural fuel uses.
Currently LPG is supplied primarily from refineries handling petroleum
crudes. With the anticipated shortfall in the supply of these crudes, the
resulting shortage of LPG will be met to some extent by LPG from synfuel
plants.
LIGHT DISTILLATES
Gasoline, which is a major light distillate, is generally defined as a
fuel designed for use in reciprocating, spark ignition internal-combustion
engines. Other uses for gasoline are of small volume. Primarily it is
used as fuel for automotive ground vehicles of all types, reciprocating
aircraft engines, marine engines, tractors and lawn mowers. Other small-
scale uses include fuel in appliances such as field stoves, heating and
lighting units, and blow torches. By far the primary use of gasoline
produced from coal will be for transportation applications. Currently we
consume nearly 6.8 MMBPD of petroleum-derived gasoline and this corresponds
to about 40 percent of the total petroleum consumption.
Naphthas have a wide variety of properties and serve many industrial
and domestic uses. Their primary market is the petrochemical industry
where they can be used for the manufacture of solvents, varnish, turpen-
tines, rust-proofing compounds, Pharmaceuticals, pesticides, herbicides,
and fungicides. However, preliminary analysis indicates that there will be
a relatively small amount of coal-derived naphthas entering the market.
MIDDLE DISTILLATES
The market for middle distillates, which essentially are jet fuel,
kerosene, diesel oil and light fuel oil, are jet aircraft, gas turbines,
and diesel engines used for transportation and stationary applications, and
residential and commercial heating.
RESIDUES
The market for residues, consisting mainly of fuel oil, is primarily
for industrial, utility and marine fuel use. Other applications for
residues include preparation of industrial and automotive lubricants,
metallurgical oils, roof coatings, and wood preservative oils. Coke is
another likely useful product from residue.
237
-------
PETROCHEMICALS
Many synfuel products, in addition to their primary use as fuel, are
likely to be used by the petrochemical industry for the production of
several other by-products. Currently over 3000 petrochemical by-products
are derived from petroleum and natural gas sources. These include items
like synthetic rubbers, plastics, synthetic fibers, detergents, solvents,
sulfur, ammonia and ammonia fertilizers and carbon black.
Petrochemicals from synfuels will generally fall under three broad
groups based on their chemical composition and structure: aliphatic,
aromatic, and inorganic. An aliphatic petrochemical is an organic compound
which has an open chain of carbon atoms. Important petrochemicals in this
group include acetic acid, acetic anhydride, acetone, ethyl alcohol, and
methyl alcohol. Most aliphatic petrochemicals are currently made from
methane, ethane, propane or butane. Aliphatics currently represent over 60
percent of all petrochemicals and are the most important group
economically.
An aromatic petrochemical is also an organic compound but one that
contains or is derived from a basic benzene ring. Important in this group
are benzene, toluene, and xylene, commonly known as the B-T-X group.
Benzene is widely used in reactions with other petrochemicals. With
ethylene it gives ethyl benzene which is converted to styrene, an important
synthetic-rubber component. As a raw material it can be used to make
phenol. Another use is in the manufacture of adipic acid for nylon.
Toluene is largely used as a solvent in the manufacture of trinitrotoluene
for explosives. Xylene is used as a source of material for polyester
fibers, isophthalic acid, among other petrochemicals.
An inorganic petrochemical is one which does not contain carbon atoms.
Typical here are sulfur, ammonia and its derivatives such as nitric acid,
ammonium nitrate, ammonium sulfate.
The different end-use applications of major synfuels products are
summarized in Table 7. We see from this discussion that coal-derived
synfuel products are likely to be used not only as a fuel, but also in the
manufacture of a number of other by-products which will be used in
multitudes of other applications.
238
-------
Table 7. Major End-Use Applications of Synfuel Products
Major Synfuel Products
Likely Major
End Use Applications
High and medium Btu gas
Low Btu gas
LPG
Gasoline
Naphtha
Middle distillates
(kerosene, diesel,
light fuel oil)
Residues
Food, textile, pulp and paper,
chemicals, iron and steel
industries; residential/
commercial heating
Small boilers, kilns, pelletizing
Glass, electronics, chemical
industries; domestic cooking and
heating; automotive
Transportation
Petrochemical industry; solvents;
varnish; turpentines
Transportation, gas turbines,
residential and commercial
heating
Industrial, utility and marine
fuel; matallurgical oils; roof
coatings; wood preservatives,
lubricants
239
-------
ANTICIPATED SYNFUELS MARKET PENETRATION IN THE VARIOUS
SECTORS OF THE U.S. ECONOMY WILL EXPAND OVER TIME
As an indication of the time frame over which the EPA must consider
issues regarding the use of various synfuel products, market development
and penetration of these products must be anticipated. For example, the
synfuels market may develop as illustrated in Figures 5, 6 and 7, over
1985-1987, 1988-1990 and 1991-2000 time frames.
SYNFUEL PRODUCT UTILIZATION EMPHASIS, 1985-1987
Oil shale-derived synfuels will be introduced into the petroleum
product markets about 1985, and based on Scenario I as much as 0.2 MMBPD of
shale oil can enter the market by 1987. The first stage of synfuels market
infrastructure development will be oriented towards transportation fuels
(reference Figure 5) because oil shale that is hydrotreated can be refined
in existing refineries to such products as gasoline, jet fuel, diesel and
marine fuels. The bulk of this supply will be in the form of middle
distillates comprised of jet fuel and diesel oil. The demand for
transportation during the late 1980s is expected to be around 10 MMBPD.
Of this, about 5 percent is likely to be consumed by the military sector.
It is conceivable, therefore, that the bulk of the shale oil products could
be utilized by the military, possibly with a Government synfuel purchase
guarantee program.
It is anticipated that the oil shale industry will continue to grow
producing as much as 0.45 MMBPD by year 2000 as per Scenario I and II and
as much as 0.9 MMBPD as per Scenario III. The Bulk of this production is
anticipated for the transportation sector-
SYNFUEL PRODUCT UTILIZATION ADDITIONS, 1988-1990
Subsequent buildup of the synfuels industry during the 1988-1990 time
period (reference Figure 6) is expected to come from commercial-size, high-
Btu gasifiers. As per Scenario 1, the output from these high-Btu gasifiers
may be as high as 0.4 MMBPD of oil equivalent by 1990; however, the
conservative estimate based on Scenario II is that only around 0.17 MMBPD
of oil equivalent is likely to be produced by that time. The high-Btu
gasification will serve some of the energy needs of both the industrial and
residential/commercial sectors as direct gas sales or through electric
240
-------
SYNFUEL UTILIZATION DURING 1985-1987
ro
-P.
MAJOR SYNTUIL PRODUCTS/
•VPROMICTS
CHI ShaU Product*
• Gasoline
• Mlddlo Oiitlllatos
l, K«rei«iM,
Ught rw«l Oil)
• ••sldwal*
(•4., Mario* fuels,
Lubricants)
COMMERCIAL
RESIDENTIAL
SECTOR
Figure 5
-------
SYNFUEL UTILIZATION DURING 1985-1990
ro
MAJOt SVNHia MOOUCTS/
•vptooucrs
Oil ShoU
Gasoline
MJtldl* M»HMat«s
U0h» ' «•»• Oil)
Gasification
Product*
• SNO
• Low and Medium
Gas
• MUthanol
• Petrochemical*
Marin* Pu*U,
lubricanH)
Figure 6
-------
-p=>
oo
.•". .COAL .* •/..
'GASIFICATIONS
MAJOR SVNFUEL PRODUCTS/
SYPRODUCTS
Oil Shot*
(Products
Coal Gasification
Products
• Gasoline
• Middle
Distillatos
(o.g.. Jot
Fuel.
Diosol,
Korosono,
Light Fu*l Oil)
• Residuals
(*.g., Marino
SNG
Low and
Medium
Gas
Mothanol
Petro-
chemicals
Lubricants)
Coal liquefaction
Products
• Gasoline
• Middle Distillates
(0.9., Jet Fuel,
Diesel, Kerosene,
Light fuel Oil)
• Residuals
(e.g.. Marine Fuels,
Lubricants)
• SNG
• Low and Medium
Gas
• Methanol
• Petrochemicals
Figure 7
-------
power generation by utilities. Some of the major industrial users of
high-Btu gas are likely to be textile, food, steel, and chemical
industries. Initially, following the current use pattern, it will be used
not only as an industrial fuel but also as a chemical feedstock. It is
expected that the existing natural gas pipeline network, with the exception
of a few connecting pipelines, will be utilized for the distribution of
high-Btu gas and, therefore, introduction of high-Btu gas is not likely to
cause major problems concerning distribution for end-use applications.
During this time period it is also likely that low- and medium-Btu
gasification plants will be used by industries in a captive mode to supply
some of their fuel and chemical feedstock needs. This may amount to as
much as 0.3 to 0.4 MMBPD of oil equivalent based on the first two
scenarios. The medium-Btu gas could be used as a synthesis gas for the
production of different chemical products such as ammonia which in turn
could be used for the manufacture of such products as fertilizers, fiber
and plastic intermediates, and explosives. Currently the petrochemical
industry derives its synthesis gas by reforming natural gas or naphtha.
During this time period, it is likely that one to three small plants
possibly producing methanol from medium-Btu gas may come on line. These
are likely to be owned by industries primarily to supply internal needs.
This could be for the production of formaldehyde, a product with a number
of end-use applications. It is unlikely that products from these plants
will be entering the open market directly, on a large scale, for public
consumption. During this time period the use of low-Btu gas will be
limited to an industrial fuel in such applications as kilns, chemical
furnances and small boilers. However, the use of low-Btu gasification by
utilities in one or two demonstration units for combined-cycle applications
cannot be ruled out.
During this time frame the shale oil output will continue to grow
reaching as much as 0.4 MMBPD in accordance with the National Goal
Scenario. As a result it is anticipated that increasing amounts of shale
oil products will be entering the transportation sector, with limited entry
into the industrial sector for use as fuel.
244
-------
SYNFUEL PRODUCT UTILIZATION ADDITIONS, 1991-2000
During the 1991-2000 time frame (reference Figure 7), central coal
liquefaction plants will introduce into the market a spectrum of products
and by-products that will be consumed by the transportation, industrial,
and residential/commercial sectors. Based on the nominal and accelerated
scenarios, by the year 2000 1.5 to 2.5 MMBPD of coal liquid products will
be entering the market. Under these conditions, a significant segment of
the transportation fleet could be running on synthetic fuel. Coal-derived
liquids will be utilized not only by industry as a fuel source and chemical
feedstock, but also by the residential and commercial sectors for space
heating, hot water supply and other domestic uses. Furthermore, many of
the oil-fired utility plants given exemption from converting to coal in the
interim will be burning coal-derived fuel oil. SRC II plants will be the
likely candidate which will be supplying the bulk of this fuel. It is also
expected that methanol from indirect coal liquefaction could be entering
the market for use as turbine fuel for the production of electricity.
during this time period. In addition, SNG produced from the liquefaction
processes will be also entering the market, supplementing the output from
high-Btu gasification plants. The SNG output from liquefaction plants
could be as high as 20 percent of the total useful output from these plants
in terms of heating value. LPG and naphtha produced from direct and
indirect coal liquefaction processes and oil shale are likely to be used
primarily by the petrochemical industries. For example, LPG may be used by
the petrochemical industry as a raw material for the production of
alcohols, organic acids, detergents, plastics, and synthetic rubber
components. Naphthas may be utilized for the manufacture of such items as
solvents, adhesives, pesticides, and chemical intermediates. Currently the
petrochemical industry uses about 11 percent of our crude oil supply for
the production of various petrochemicals. During the 1990-2000 time frame,
it is possible that the same percentage of available synfuels will be
utilized by the petrochemical industry for the production of hundreds of
petrochemical products. A major use of residuals from coal liquefaction
processes and oil shale is likely to be the manufacture of different types
of lubricants. These could be for such applications as lubrication of
engines and general machinery, steam turbine bearings and reduction gears,
compressors, insulating oils, metal working and cutting oils.
245
-------
So we see in the above discussion that the synfuels products and
by-products are likely to enter all the end-use sectors, in course of time.
The potential for exposure and for environmental impacts must be carefully
considered. Early planning by the EPA will require that synfuel
products/by-products be assessed with regard to their environmental
acceptability.
246
-------
POTENTIAL ENVIRONMENTAL EXPOSURES DUE TO SYNFUELS UTILIZATION
A major concern of the emerging synfuels industry is the potential
environmental, health and safety impacts associated with the use of
synfuels. The potential exposure of the public to synfuels will depend on
the rate of development of the synfuels industry's specific end-use
markets. Since the market may cover a wide range of products and end uses,
a significant portion of the population may be exposed. The products will
enter the markets in varying quantities over the coming years. To illus-
trate the important environmental concerns, synfuels product production
rates based on the National Goals Scenario (Scenario I) are considered and
three time periods are examined for potential environmental exposure,
1985-1987, 1988-1990, and 1991-2000.
POTENTIAL EXPOSURES: 1985-1987
During this period, synfuels entering the market will be mostly
limited to shale oil products. Approximately 0.2 MMBPD of products by 1987
is projected by Scenario I. Crude shale oil will most likely be
transported to refineries in either the Gulf Coast or Midwest and is
expected to be distributed by existing pipelines. Product quantities will
be limited. The hazards of transporting and storing crude shale oil and
shale oil products are expected to be minimal. Shale oil products will be
used primarily as transportation fuels such as gasoline, diesel oil, and
jet fuel and will be distributed by railroads, tankers, trucks and barges.
During this period the quantities handled are estimated to amount to 0.04
MMBPD of gasoline and a combined total of 0.16 MMBPD for the middle
distillates. The major exposure to these products occur at storage
terminal unloading operations and service station storage tank loading
operations, both of which have high spill potential. The end user (a
passenger car, truck, or other vehicle) also poses a potential spill
problem due to the rapid expansion of self service stations. Combustion of
the fuels may expose a large segment of the population since most
automobile traffic is generated in central business districts and their
suburbs. By-products from shale oil refining such as lubricating oils and
greases will be shipped from refinery bulk packing plants in secure
containers, minimizing the likelihood of exposure.
247
-------
Products from shale oil production could reach approximately 0.3 to
0.4 MMBPD during this period, as suggested by the accelerated rate
scenario, with the potential exposure reaching twice the level suggested by
the National Goal Scenario.
POTENTIAL EXPOSURE: 1988-1990
During this period, in addition to increased shale oil production,
SNG low and medium Btu gas, and some indirect liquefaction products will
also be entering the market, which increases the complexity of the synfuels
distribution network and increases the potential for public exposure to the
products. It is a time period by which the EPA must have identified
potential problems and have developed a plan for meeting the synfuels
challenge.
Shale oil production during this period is projected to be 0.3 to 0.4
MMBPD under the National Goal Scenario, but could range from 0.2 MMBPD
(nominal production rate scenario) to 0.8 MMBPD (accelerated production
rate scenario) in 1990. The exposure potential to the products will
increase proportionally during this time period compared to the previous
period.
The SNG entering the market is projected to amount to an oil
equivalent of 0.4 MMBPD by 1990, and will be transported by existing
pipeline to the various markets. Although pipelines transporting SNG or
crude shale oil present a low accident potential, pipelines either transect
or terminate in densely populated areas, providing some degree of exposure
potential to these products. First generation coal gasification technology
(Lurgi) buildup will occur near western U.S. coal deposits, the Northern
Great Plains/Rocky Mountains area. The SNG from this area will enter the
northern tier pipeline network and will be distributed across the upper
Midwest. Medium- and low-Btu gases will also be in the market during this
period, although they will probably be used for internal plant needs. This
will minimize the exposure potential since these gaseous products will not
require any transportation.
Some synthetic gases have different compositions than natural gas, and
may cause internal corrosion and stress-corrosion cracking in pipelines.
Effects of impurities on the long-term degradation of some pipeline
248
-------
materials are unknown. Synthetic gaseous fuels also have different
flamnability and explosion limits that may require new techniques in the
management of pipeline leaks. Gases with a high CO concentration are toxic
and could present significant exposure problems.
In addition to their use in transportation and boiler applications,
synfuels products will be used as feedstocks for industrial processes.
These applications, although limited during this time period, present
another avenue of exposure for which EPA must be prepared. The population
exposed could include industrial plant personnel as well as the end users
of the industrial products. During this period, medium-Btu gas could be
used as a synthesis gas for the production of methanol and ammonia, each of
which can be utilized as a finished product.
Although this period will be characterized by the emergence of many
synfuels products, the main population exposure potential will occur frcm
crude shale oil transport by pipelines, product storage and the combustion
of these products.
A basic environmental concern with the transportation of liquid
synfuels is the possibility of an accidental spill. A recent (1979)
Department of Transportation analysis shows that of all the accidents
resulting from pipelines carrying liquid petroleum products, the largest
spillage occurs from LPG (58.6 percent) followed by crude oil (25.3
percent), with fuel oil (6.1 percent) and gasoline (4.5 percent) being the
other major contributors.
As an example to illustrate the relative exposure of transporting
petroleum products by pipeline in order to provide an awareness of the
potential exposure in transporting shale oil, Table 8 presents a listing of
oil pipeline accidents. Since existing pipelines will be used during this
time period for transporting crude shale oil, these potential exposures and
risks in each component of the carrier system must be considered by EPA.
249
-------
Table 8. Number of Oil Pipeline Carrier System Accidents
YEAR 79 78 77 76 75 74 73 72 71
Line Pipe 207 194 177 169 185 203 215 238 264
Pumping Station 20 30 32 11 24 13 23 31 14
Delivery Point 644455532
Tank Farm 5 15 12 14 30 22 21 24 11
Other 11 13 12 14 10 13 9 10 19
Total Accidents 249 256 237 212 254 256 273 306 310
Source: Department of Transportation
POTENTIAL EXPOSURE: 1991-2000
This period is characterized by the large-scale entry into the market
of direct and indirect liquefaction products and by-products for use
primarily by the transportation, industrial and utility sectors. Based on
the National Goal Scenario, 1.0 MMBPD of coal liquids will be in the market
by 2000, but may range up to 2.5 MMBPD. Utility and industrial boiler
fuels produced by coal liquefaction processes will be most in demand in the
Gulf Coast, Northeast, and Southern California regions, as shown in Figure
8. These regions contain a significant portion of the U.S. population.
The use of these fuels will also have some beneficial effect in areas that
are sensitive to particulate and sulfur dioxide since these fuels have
lower ash and sulfur contents. As more liquefaction capacity develops in
the Appalachian and interior regions, liquid fuels will more readily be
used in the industrial areas of Indiana, Illinois, Ohio, and the upper
Northwest. Shale oil products during the period may reach a level of 0.4
MMBPD under the National Goal Scenario and could reach as high as 0.9 MMBPD
under the accelerated rate scenario. High-Btu gas under these two
scenarios is estimated at 0.5 MMBPD and 1.0 MMBPD respectively by 2000. As
coal gasification technology develops, it is likely that a key area for
gasification will eventually be Appalachia, with SNG entering the existing
pipelines and being distributed along the east coast to both industrial and
residential users.
This period is also characterized by increased use of coal liquid
products for chemical feedstocks and in the housing and commercial sectors
250
-------
Figure 8. Potential Regions for Synfuel Demand for Industrial
Fuel Application
en
n
^.
High Demand Regions
-------
for space heating and hot water supply. Naphthas produced by liquefaction
processes are likely to be used by the petrochemical industries for
manufacturing solvents, pesticides and chemical intermediates. Residues
from coal liquefaction processes may be used to manufacture several types
of lubricants with a wide variety of applications. This market penetration
significantly increases exposure potential as there is virtually no segment
of the population that would be excluded from the use of synfuel products
and by-products.
In addition to synfuels utilization, EPA must also consider the
transportation and handling aspects of the synfuels products and
by-products. As the synfuels develop during this period, transportation
modes other than pipelines will be utilized. Although there are associated
risks, pipelines are considered to present less risk than other modes such
as railroads, trucks, and tankers. As these modes are currently used for a
wide variety of petrochemical products, it is expected that they will also
be used as synfuels penetrate the market, thereby presenting another
concern that EPA must address.
Table 10 presents an estimate of the range of synfuel products to be
shipped by the various transportation modes beginning in the 1990s. Nearer
to the year 2000, the relative amounts of products transported between the
modes may vary. The majority of the synfuel products as well as crude
shale oil will be transported by pipelines, which presents the least amount
of exposure potential. On the other hand, railroads which have a high
accident potential will transport the least amount of products. In order
to supply the high demand regions (reference Figure 8) the transport
distribution networks may develop as illustrated in Figure 9. The
distribution system indicates that the crude shale oil, refined products,
SNG, and coal liquids will each be transported across areas of high
population density and industrial concentration, mostly in the eastern U.S.
A market for 2.2 MMBPD of synfuels products and by-products by 1992 under
the National Goals Scenario indicates the magnitude of the problem for
which EPA must prepare.
The transportation modes that will be utilized by the synfuels
industry and which pose a greater accident potential than pipeline
transport are railroads, trucks and tankers. Railroads will be used
252
-------
Table 10. Range of Synfuels Distributed by Node of Transportation In the 1990's
SYNFUEL PRODUCT
HI Btu GAS (MMSCFD)(2)
MEDIUM Btu GAS (MMSCFD)
LIQUEFACTION PRODUCTS (MMBPD)
HEAVY FUEL OILS AND
MIDDLE DISTILLATES
GASOLINE
NAPHTHA
LPG
CRUDE SHALE OIL (MMBPD)
REFINED* SHALE (MMBPD)
GASOLINE
JET FUEL
DIESEL OIL
RESIDUAL OIL
ACCIDENT RISK
PIPELINE
1900 - 4800
8300 - 5600
0.025 - 0.080
0
0
0
0.389 - 0.750
0.040 - 0.076
0.047 - 0.090
0.084 - 0.162
0.025 - 0.047
LOW
RAIL
0
0
0
0
0.042 - 0.067
0
0
0
0
0
0
HIGH
TRUCK
0
0
0.014 - 0.044
0.072 - 0.520
0.005 - 0.008
0.0 - .018
0
0
0
0.084 - 0.162
0.004 - 0.007
HIGH
TANKER OR BARGF
0
0
0.007 - 0.022
0.018 - 0.130
0.005 - 0.008
0.0 - 0.005
0
0.026 - 0.051
0.031 - 0.061
0.042 - 0.162
0.007 - 0.014
MODERATE
en
OJ
(1) MMSCFD = Million Standard Cubic Feet per Day
(2) MMBPD = Million Barrels per Day
-------
Figure 9. Synfuel Resources and Distribution System in the 1990's
rv>
en
n
. x—ir—*——J
Coal Regions
Oil Shale Regions
A Gasification Plants
-T Existing Crude Oil Trunkline to Refinery
-New Trunkline to Refinery
===== Existing Natural Gas Trunkline
ssi=.«New SNG Trunkline
• Liquefaction Plants
• Shale Oil Plants
-------
primarily for the transportation of naphthas which in 1992 are estimated to
range from 0.04 to 0.07 MMBPD. This mode of transportation presents a high
degree of accident risk due to the poor condition of the Nation's rail
system. Derailments, grade crossing accidents, and collisions between
trains pose potential risks to the transportation of any hazardous or toxic
substances. Tank car accidents with hazardous materials are shown in Table
11, providing another example of potential risks associated with
transporting synfuels products.
Table 11. Railroad Tank Car Accidents with Hazardous Materials
1979 1978
Total Accidents 937 1014
Accidents Involving 165 228
Atmospheric Release
Source: Federal Railway Administration
The use of tanker trucks will be extensive in transporting coal
liquids and refined shale oil products. In 1992 under the National Goal
Scenario, approximately 0.4 MMBPD of gasoline from coal liquefaction may be
transported by truck, and up to 0.5 MMBPD under the accelerated rate
scenario. Other products using this mode are middle distillates and
naphthas. Potential exposure to the general population is high with tanker
trucks since much of these products will be delivered to urban areas where
trucks will face the normal amount of traffic accidents in congested areas.
In addition, exposures to the products will occur in loading and unloading
of trucks at storage terminals and service stations. Due to vapor recovery
requirements mandated by state implementation plans, evaporative emissions
of volatile compounds are gradually being controlled, but pollution control
systems must be improved to further reduce emissions. Accidents or
defective emission control systems provide the chief potential for release
of synfuels products by truck transport.
Tankers and barges will also be used for the transportation of the
refined shale oil and coal liquids products and could be used extensively
if the markets are accessible to the gulf coast. Under the National Goal
255
-------
Scenario, 0.1 MMBPD each of gasoline and diesel oil may be transported by
these modes in 1992. Other products to a lesser extent are naphtha, LPG,
jet fuel, and residual oil. A significant amount of petrochemical products
currently move along the Mississippi River to northern markets. The major
emission source for this operation involves loading and unloading; however,
the accident rate is less than that of surface transportation mode. As
with truck loading, increased emission controls are being initiated for
ship and barge loading which will significantly decrease evaporative
emissions by the time the synfuels industry is developed. Improvements are
also being made to reduce spills of petrochemical products into waterways.
Reduction of accidental spills and prevention of intentional releases are
currently under regulation by the Coast Guard and EPA.
In addition to transportation and handling, the storage of synfuel
products and by-products may pose potential environmental problems. These
problems may occur primarily with refined shale oil and coal liquids. As
with other petroleum products they will be stored at bulk storage terminals
until used. By 1992, a total of 1.4 MMBPD of synthetic liquids will be
produced under the National Goal Scenario, and ranging up to 1.7 MMBPD
under the accelerated rate scenario. Exposures to these products at the
terminals may occur during the loading and unloading operations, as well as
breathing losses from the tanks during product storage. The potential for
exposure depends upon the volatility of the products and the frequency of
loading operations. Since storage facilities are located at refineries,
utility and industrial plants, airports and numerous other facilities,
exposure potential is significant. Concern over the uncertainties of the
constituents of synfuels may lead to storage procedures for these products
that are more rigid, and new storage vessels or containers for liquids may
be required under stringent specifications. Some emissions may also occur
from low-level leakage.
As with other major control requirements for loading and unloading
petroleum products, vapor recovery techniques for bulk storage facilities
are being improved, primarily by the use of floating roof tanks. Synfuels
such as SRC II liquids have vapor pressures similar to No. 6 fuel oil which
has very low evaporative emissions and working losses compared to gasoline.
Fugitive emissions of synfuels will always be present as they are with
256
-------
other petroleum products. However, there will be new control systems
developed and emissions will be reduced over the next several years through
improvements in emissions control procedures in transportation, handling
and storage operations. Only after thorough toxicity testing of synfuel
products and by-products can an assessment be made of whether synfuels
transportation, handling and storage will pose environmental, health, and
safety problems greater than those experienced in the petroleum refining
and chemical manufacturing industries.
257
-------
ENVIRONMENTAL CONTROL TECHNOLOGY NEEDED FOR SYNFUELS UTILIZATION
The utilization of synfuels products and by-products will require
improvements in existing environmental control technology and the
development of new technologies. In order to assess the control technology
requirements, it is necessary to first understand the hazards associated
with synfuel utilization. This may be accomplished by determining, the
constituents of synfuels products and by-products, their transformation
upon use, and ultimate fate in the environment. These data in turn must be
tied closely to the product buildup rate described in each of the scenarios
since these impact the types of products produced and their rate of
penetration into the market. Once these factors are understood, then
control technology options may be evaluated. This cycle must be completed
within the next 10 years in order for EPA to meet the synfuels challenge.
EXISTING DATA REGARDING HAZARDS OF SYNFUEL PRODUCTS IS SPARSE
At the current time there is a lack of sufficient data available to
properly assess the potential risks associated with the utilization of
synfuel products and by-products. The development of these data will
require significant efforts on the part of the government, industry and the
academic community to generate sound, reliable information to assure
minimum risks to the health and welfare of the nation as synfuels are
introduced into the market. This synfuels data base must contain not only
accurate and representative information about the physical properties,
chemical composition and biological activities of synfuels, but must also
contain equally comprehensive data on the end uses of the products and by-
products.
The DOE and EPA are presently conducting significant research efforts
on synfuels product characterization. The results of some of the shale oil
and coal liquid products are becoming available. An example of the prelim-
inary analysis of these two products compared with petroleum crude is
presented in Table 12. There are some similarities in the diaromatic
content between shale oil and petroleum crude, with coal liquids having the
highest content. This factor may be significant if a spill of these
products occurred, as impacts on water pollution would be less than from
coal liquids. A comparison between coal and petroleum derived gasolines is
258
-------
Table 12. Diaromatic Content of Synthetic Crudes and
Crude Oils
Concentration
Constituent
Naphthalene
2-Methyl naphthal ene
1 -Methyl naphthal ene
Biphenyl
2, 6-Dimethyl naphthal ene
1, 3/1, 6-Dimethyl naphthal ene
2, 3-Dimethyl naphthal ene
1 , 5-Dimethyl naphthal ene
1, 2-Dimethyl naphthal ene
Acenaphthalene
Acenaphthene
TOTAL
T = trace, ND » not detected
Typical
Shale
Oil
1.39
0.91
0.68
0.06
0.10
1.63
0.28
0.03
0.19
0.26
T
5.23
Coal
Syncrude
1.68
3.47
1.11
0.44
0.81
3.01
1.53
0.67
0.23
2.19
0.30
15.4
, mg/g
Petroleum
Crude
0.87
1.04
0.75
T
0.08
1.48
0.51
0.08
0.31
0.30
ND
5.42
Source: Oak Ridge National Laboratory
259
-------
presented in Table 13, indicating significant variations in aromatics and
unidentified compounds. Due to the high aromatic content of the coal-
derived gasoline, potential adverse health effects may occur from
widespread use of this fuel in automotive applications. Some of the
synfuels products and by-products may be classified as toxic chemicals
under the Toxic Substances Control Act (TSCA).
Preliminary health effects studies have indicated that coal liquids
have industrial toxicity ratings similar to those of benzoic acid,
phosphoric acid, sodium tartrate, and polychlorinated biphenyls (PCB).
Coal liquids have also been found to be less toxic than pesticides such as
dieldrin and chlordane, and more toxic than crude petroleum and shale oil.
Historical epidemic!ogical and animal studies have established that coal
tars and pitches from coal coking, gasification, and combustion possess a
carcinogenic nature. Although these studies are not all directly
comparable, it would appear that some high-boiling point products from
direct liquefaction processes or from coal pyrolytic processes may possess
a high degree of carcinogenicity.
It is apparent that although work has started in the right direction
to assess synfuels hazards, much work still needs to be conducted. As the
physical, chemical and biological results are analyzed, and potential risks
evaluated, decisions can start to be made as to the various pollution
control technologies that can be most effectively applied in the
utilization of synfuels.
260
-------
Table 13. Major Chemical Component Classes of
Petroleum and Coal-Derived Gasoline
CHEMICAL GROUP
Total
Total
Total
Total
Saturates
Alkenes
Aromatics
Unidentified
GASOLINE
Petroleum-Derived
56.
5.
24.
38
00
32
0
- 68.
- 7.
- 32.
- 3.
68
69
91
02
Coal
20.1
-Derived
- 68.5
0
2
34.20 - 75.63
0 - 12.8
1
Data are from Sanders and Maynard (1968) and Runion (1975).
The range of numbers are for different grades of gasoline
of low, medium, or high octane.
"Data are from EPRI (1978). The range of numbers correspond
to different amount of hydroprocessing. Increased hydro-
processing results in fuel with a lower aromatic content.
261
-------
PRODUCT BUILDUP RATES WILL DETERMINE MAGNITUDE OF ENVIRONMENTAL IMPACTS
Once the hazards of synfuels products and byproducts are known, their
relative impacts on the environment will depend upon the product buildup
rate and market penetration as described for each of the scenarios. All
media, air, water and land, must be considered.
Air pollution impacts will occur primarily from the combustion of
synfuels in stationary and mobile sources, with some impacts from fugitive
emissions occurring during transportation, storage, and handling
operations. Under the National Goal Scenario, coal liquids and shale oil
products will contribute the greatest percentage of products. A level of
1.4 MMBPD of these fuels will be produced in 1992 and continue through
2000. Coal liquids will most likely be used in all sectors of the market
including utilities, transportation, industrial, and commercial. As most
of the products will be used in stationary sources, the air pollution
impacts are expected to be less than from shale oil products, all of which
will be used by transportation sources. The individual mobile sources do
not lend themselves to as effective emission controls as centralized
stationary sources. Due to the moderate amount of petroleum product use
that is expected to be replaced by synthetic liquids, the air pollution
impacts are expected to be moderate.
Under the nominal rate scenario (Scenario 2), only 0.5 MMBPD of liquid
fuels will be produced in 1992, and the 1.4 MMBPD level will not be reached
until 1998. This will provide relatively lower air pollution impacts from
liquids combustion than the National Goal Scenario. The use of low- and
medium-Btu gas is projected to be higher under scenario 2 than scenario 1,
although air pollution impacts are not considered to be significant since
these products will most likely be used for in-pi ant and feedstock
applications.
The greatest relative impact would occur under the accelerated rate
scenario, as the quantities of each product are higher than for each of the
other two scenarios. By 1992, shale oil and coal liquids production reach
a level of 1.8 MMBPD and as much as 3.5 MMBPD by 2000. Shale oil in this
period is in excess of 0.9 MMBPD, all of which is used in transportation
sources. As shale oil products can be used virtually anywhere in the U.S.,
262
-------
there is very little of the population that may not be exposed to the
combustion products* If the majority of the products are used by the
military sector, the geographic area of use may be better defined.
Significant market penetration under this scenario will also be made by SNG
which may be used in all sectors with the exception of transportation. As
this product has widespread application, its composition must be accurately
defined to determine if combustion will produce air pollution impacts
different from use of natural gas.
Water pollution will occur primarily from spills associated with the
transportation of synthetic liquids. As the production of these is
greatest under the accelerated rate scenario, it provides the greatest
potential for these impacts. The crude and refined shale oils, as well as
coal liquids will be transported over long distances by pipelines, and then
to the markets by various modes of transportation. The loading of tankers
and barges, and transportation of the products by waterways provides a
moderate degree of spill potential.
Solid wastes will be generated primarily by the pollution control
systems used during synfuels utilization. These systems will be limited to
stationary source applications where the coal liquids and gases are used in
utilities and for industrial processes. As the quantities of solid wastes
produced will be dependent on the amount of these fuels used, it will have
the greatest impact under the accelerated rate scenario. Oil shale
products will not contribute to these impacts since they will be used in
transportation sources. By 1992 under this scenario, coal-derived fuels
will be produced at a level of 1.8 MMBPD and 4.1 MMBPD by 2000. The
majority of these products will be used in stationary sources with emission
control systems producing solid wastes. Under scenario 2, only 1.7 MMBPD
of coal-derived fuels will be produced by 2000, and 1.8 MMBPD under the
National Goal Scenario. As another example of the need to determine
synfuel composition, the solid wastes generated by control systems may
contain toxic or hazardous components which upon disposal may leach into
groundwaters at waste disposal sites.
On the basis of the information presented, significant data need to be
developed to assess control technology options. The optimal method of
control, if achievable, would be to upgrade the products to remove as much
263
-------
of the pollutant source content as possible rather than rely on downstream
pollution controls. This would have significant benefits on pollution
impacts that may occur prior to product utilization. As an example,
fugitive emissions into the atmosphere, or spills into waterways would not
be expected to be severe if the majority of pollutants were removed during
the manufacture of the product.
Once the products are ready for combustion, emission controls will be
necessary if product upgrading is unsuccessful. Recent small-scale tests
of synfuels combustion have provided encouraging results from an
environmental perspective. Several combustion tests of SRC liquids and
solids, EDS and H-Coal liquids, shale oil, and coal derived gases have been
conducted. For test purposes, some of the combustion devices were not
equipped with high-efficiency pollution control devices. Once the products
are used in commerce, Best Available Control Technology (BACT) will be
required.
EPA is currently proceeding to develop Pollution Control Guidance
Documents for all of the synfuel technologies that are being considered
under the three scenarios. The purpose of these documents is to foster the
development of acceptable synfuels technologies with a minimum of
regulatory delays. A similar series of documents may be prepared for the
utilization of the products from these technologies.
264
-------
SIGNIFICANT ENVIRONMENTAL ISSUES FOR SYNFUELS UTILIZATION
• Although a few synfuels products have been included in the toxic
substances inventory, most synfuels may be designated as new
products under TSCA. EPA will have to identify potential risks
associated with the transport and use of synfuels products and
by-products, as well as their end uses. Risk and exposure concerns
depend on the market infrastructure and likely end use of the
variety of products that will result. More diverse end uses and
methods of handling, storage, and distribution will increase the
exposure potential.
• In addition to TSCA, stipulations of the Clean Air Act will also
impact the synfuels market. Atmospheric emissions from fugitive
sources are potentially an environmental concern, as well as end-
use combustion emissions. These emissions must be characterized so
that BACT determinations can be made. Similarities and differences
with related petroleum products need to be evaluated.
• Potential atmospheric emissions are much more diverse than the
limited set of criteria pollutants which constitute the majority of
air pollution concerns today. A critical issue is not so much that
hydrocarbons may be an emission, but rather an assessment is needed
of the kinds of other organic emissions and the associated risks.
• The potential of accidental spills in the transport and storage of
synfuels products and by-products is one of the most critical
concerns for protection of groundwater quality and dependent
drinking water sources, as stipulated by the Clean Water Act.
Additional contamination of receiving waters could be caused by
area washdown and stormwater runoff at facilities where minor
leakage occurs.
• RCRA requirements will include an integrated solid and hazardous
waste management program. Waste oils, storage tank sludges,
disposable materials (seals, packing, etc.), and ash residues can
all be anticipated from synfuels usage, in addition to waste
by-products.
• There is a high probability that synfuels will be blended with
petroleum products, either as refinery and petrochemical feeds or
as products at end-use locations. EPA will have to judge the
applicability of existing regulations covering petroleum product
transport and use when the product characterizations are related to
blend ratios. Furthermore, synfuels materials that will be used as
chemical feedstocks will require environmental assessments
regarding their physical, chemical, and biological acceptability.
• The eventual complexity and diversity of the synfuels market
infrastructure will represent a challenge to traditional
environmental monitoring and inspection procedures, as well as
control technology assessment.
265
-------
• Some of the control approaches will be equipment and operations
oriented. This characteristic will require a close EPA interface
with other regulatory agencies (such as DOT, ICC, and Coast Guard)
regarding transportation operations which are both safe and
environmentally acceptable.
• The feasibility of segregating the handling and end-use of
potentially hazardous synfuels will certainly have to be evaluated.
Proper assessment of environmental risks from synfuels product
end-use will be needed to establish exposure estimates.
PERMITTING AND PROGRESS
EPA's regulatory role in an emerging synfuels market will involve
permitting for the production, storage, transportation, and end use of the
products. Permitting procedures will have to be streamlined to eliminate
unnecessary delays in the long-range national goal of reducing petroleum
imports. TSCA requirements will be particularly critical in this emerging
industry. Plans have been announced by some industries to begin
construction of plants to supply SNG and chemical feedstocks. Synfuels
projects scheduled for the mid-1980s include shale oil development in
Colorado and Utah, and the SRC II demonstration plant in West Virginia.
With typical engineering and design efforts requiring 2 years, and
construction another 2 to 3 years, it is essential that all permitting be
complete within 1 year to keep these critical developments on schedule.
266
-------
Session III: ENVIRONMENTAL ASSESSMENT:
GASIFICATION AND INDIRECT LIQUEFACTION
Charles F. Murray, Chairman
TRW
Redondo Beach, California
267
-------
ENVIRONMENTAL TEST RESULTS
FROM
COAL GASIFICATION PILOT PLANTS
N. A. Holt, J. E. McDaniel, T. P. O'Shea
Electric Power Research Institute
Palo Alto, California
Environmental awareness and the world oil situation are having a pro-
found impact on the U.S. Electric Power Industry. "Environmental accepta-
bility" has been redefined and it is emerging as one of the major criteria
for selection of a power generation process to satisfy increasing load de-
mand or to replace retired units. Furthermore, the fact that the cost of
fuel has risen in real terms dictates that more fuel efficient plant config-
urations will be deployed. Fuel efficiency and environmental tolerability
come only at the expense of increased monetary cost.
These fundamental changes certainly are creating problems for the power
industry but they are also creating opportunities for new and more appropriate
power generation processes.
EPRI has high expectations that combined cycle power systems fueled by
gas from coal will be cleaner and more efficient than the competing processes
for equivalent capital cost. Advantages accrue to these Gasification-Combined
Cycle (GCC) systems primarily from the relative ease of cleaning fuel gas,
the benign nature of the waste products, and the inherent and proven high
thermodynamic efficiency of the combined cycle configuration.
These and other advantages will be discussed. Coal gasification pro-
cesses will be identified which most effectively capitalize on these advan-
tages. Environmental test results on these processes will be summarized.
Finally, the plans for commercial scale demonstration of a GCC system will be
reviewed. This demonstration will be a critical milestone since no technol-
ogy can be considered to be a real option until it has been operated at an
appropriate scale.
268
-------
ENVIRONMENTAL TEST RESULTS
FROM
COAL GASIFICATION PILOT PLANTS
INTRODUCTION
The combined circumstances of rapidly escalating oil prices, reduced
availability of oil and natural gas, strict plant emission standards and the
prospect of continued delays in nuclear implementation plans, provide the
electric power industry with urgent incentives to develop economically com-
petitive and environmentally acceptable new methods of power generation based
on our most plentiful fossil fuel resource - coal.
Of all these motivations, it is probably the environmental aspects which
constitute the major incentive for coal gasification based power systems,
since without the requirement for post-combustion clean up of the flue gases
it would clearly be less costly to simply burn coal directly.
Coal gasification based systems offer distinct environmental advantages
over conventional direct coal fired plants with flue gas clean up, since
emission forming constituents are removed prior to the combustion process.
When coupled with combined cycle power generation the resultant Integrated
Gasification Combined Cycle (IGCC) plants will be more efficient and use less
water than direct coal fired units. Studies show that such IGCC plants when
designed to current emission standards and using currently commercial combus-
tion turbines are economically competitive with direct coal firing. If emis-
sion standards become more restrictive the competitive position of IGCC tech-
nology will be further enhanced. There are also considerable prospects for
future improvements in both coal gasification and combustion turbine tech-
nology, which will enable the industry to resume its historic learning curve
for more efficient less costly systems.
EPRI CLEAN GASEOUS FUELS PROGRAM
The overall goal of the EPRI Clean Gaseous Fuels Program is to develop
economically competitive and environmentally acceptable coal gasification-
based generating systems.
The principal technical objective of the EPRI program is to design and
operate an integrated Texaco entrained gasification-rcombined -cycle demonstra-
tion plant of about 100 MW by 1985. A second demonstration plant based on
another gasifier is also planned. The program also includes work to improve
gasifiers, gas clean-up technology, heat recovery boilers, fuel gas combus-
tors and other components of gasification-based generating systems.
Coal gasifiers react coal, steam and air or oxygen to produce a gaseous
fuel, primarily carbon monoxide and hydrogen. The sulfur in the coal is con-
verted to hydrogen sulfide (H2S1, which can be removed from the gas and con-
verted to elemental sulfur by processes currently used widely in the natural
gas, chemical and petroleum industries. The mineral matter is withdrawn
primarily as ash or slag from the gasifier or from the gas stream as part of
the gas cleaning process. The coal nitrogen is converted either to ammonia,
which can readily be scrubbed from the gas, or to nitrogen itself.
269
-------
Gasifiers are also important components of other coal conversion tech-
nologies of potential benefit to utilities. The CO-H2 product gas can be
catalytically converted to methanol for use in peaking or intermediate service.
Gasifiers can be used to provide hydrogen for use in Exxon, H-Coal or SRC
coal liquefaction plants by gasifying the liquefaction residues.
ECONOMIC ATTRACTIVENESS OF GCC PLANTS
EPRI studies show that integrated gasification combined cycles using
commercially available combustion turbines (2000°F inlet temperature) and
based on Texaco or BGC/Lurgi slagging gasifiers are competitive with conven-
tional coal-fired power plants with stack gas cleanup. Table 1 shows a per-
formance comparison between conventional coal firing and gasification-based
power systems. The data presented in this table reflect 1978 environmental
control regulations. Cost estimates are included for cycles with advanced
high temperature turbines to illustrate the further performance improvement
potential of this technology. As environmental control regulations become
more stringent, the economic advantages of gasification combined cycle (GCC)
power plants will increase markedly. Table 2 shows estimated costs for more
stringent projected mid-1980s standards. GCC systems offer better efficiency,
lower emissions, reduced water consumption and land requirements, less fuel
and chemicals consumption, and reduced solid waste volume. The solid waste
from the Texaco, BGC/Lurgi slagger, and Combustion Engineering gasifiers is
in the form of extremely inert slag which should be readily disposable at
lower cost than solid waste from a coal-fired plant.
Gasification may also offer fuel for retrofit to existing gas and oil-
fired boilers, combined cycles and combustion turbines. Gasifiers might be
installed in an existing plant or in some cases remotely, with fuel distrib-
uted fay pipeline. Gasification may allow repowering existing boilers with
combustion turbines to reduce the heat rate and provide increased generation
capacity in convenient increments at an existing site with probably reduced
permitting periods.
ENVIRONMENTAL ADVANTAGES OF GASIFICATION«-BASED POWER PLANTS
The potential environmental advantages of gasification-combined cycle
power plants over direct coal fired plants with flue gas cleanup are sum-
marized in Table 3. GCC plants offer better resource utilization - more
kilowatts per ton of coal mined, less water usage per kilowatt, and less land
since sludge disposal is not required. They are also capable of achieving
markedly reduced emissions compared to direct coal fired units. Each of these
aspects is discussed in more detail below.
Resource Utilization
GCC systems utilizing currently available combustion turbines offer a
minor but measurable improvement in heat rate over conventional coal plants
with scrubbers. However, better efficiencies projected for GCC plants with.
higher temperature turbines currently being developed, i.e., machines capable
of operating at firing temperatures above 2000°F upwards to 2600°Ff should
result in significant reductions in coal use versus direct coal-based units
of similar capacity as reflected in the range of coal consumption estimates
for GCC plants in Table 3.
270
-------
Table 1 SUMMARY OF PRESENT AND PROJECTED
GCC SYSTEM PERFORMANCE
1978 FEDERAL EMISSION CONTROL REQUIREMENTS
Coal Fired Texaco GCC
Plant 2000°F Turbine
Heat Rate, 9900 9500
BtuAWh
Texaco GCC BGC Slagger GCC
2600°F Turbine 2600°F Turbine
8460 7920
Capital Require- 900
ment,
860
830
690
30-Year Levelized
Cost of Elec-
tricity ,
mills AWh
57.5
51.1
47.9
41.3
Basis: mid-1978 dollars; high-sulfur Illinois coal; coal cost $1.00/million
Btu; 70% capacity factor.
271
-------
Table 2 ECONOMIC COMPARISON OF TEXACO GASIFICATION-BASED
POWER SYSTEMS USING CURRENT (2000° F) COMBUSTION TURBINES
WITH CONVENTIONAL COAL-FIRED STEAM PLANTS
EMPLOYING WET SCRUBBING OF STACK GASES.
1978 Federal
Emission Controls
Projected mid-1980's
Emission Controls
Coal Fired Texaco GCC Coal Fired Texaco GCC
Heat Rate, BtuAWh 9900 9500
Capital Requirement/
SAW 900 860
30-Year Levelized Cost
of Electricity,
mills AWh 57.5 51.1
9950
1180
69.0
9680
900
52.9
Basis: mid-1978 dollars; high-sulfur Illinois coal; coal cost
$1.00/million Btu; 70% capacity factor.
Emission Controls
sulfur
particulates
N0x
waste water
coal ash
1978
85% removal
0.03 lbs/106 Btu
0.6 lbs/106 Btu
mid-1980's
95% removal
0.02 lbs/106 Btu
0.2 lbs/106 Btu
zero discharge
special handling
272
-------
Table 3 RELATIVE ENVIRONMENTAL EFFECTS/RESOURCE REQUIREMENTS
1000 MW POWER PLANTS
Coal Consumption -> lbs,/kWh
Limestone Required - Ibs./kWh
S02 Emissions - ppm
NOX Emissions - ppm
Particulate Emissions - Ibs./lO Btu
Make-up Water - gal./kWh
Land Required - acres
PC Boiler
with Wet Scrubber
0.80
0.12-0.15
80-400
300-500
0.03
0.6-0.65
1200-2400
GCC
Plant
0.64-0.77
-
50-225
40-90
<0.02
0.45-0.55
200-500
Note: Solid wastes, consisting of sulfur and inert slag, produced in GCC
plants in significantly lower quantity than troublesome scrubber
sludge produced in coal fired unit.
273
-------
Also, while power produced in conventional coal plants is derived from
steam turbine generators, a large part of the electricity output of GCC plants
is developed directly from fuel combustion energy with the remainder being
produced in a steam cycle. Accordingly, make-up water requirements (by far
the major part of which supports the cooling water system for the steam tur-
bine generator condenser) are significantly less in GCC plants.
Sulfur Emissions and Disposal Land Required
The range of sulfur emissions cited in Table 3 is based on single stage
sulfur removal from high and low sulfur coal based systems for both, the coal
fired boiler and gasification combined cycle power plant. Sulfur emissions
can be reduced at additional expense by adding a second stage of stack gas
scrubbing to the coal fired boiler plant or by several mechanisms in the
coal gasification based plant. EPRI economic evaluations have shown that in-
cremental sulfur removal from gasification based systems is less expensive
than from coal fired boiler plants. Additionally, gasification based systems
will produce elemental sulfur and inert slag, potentially saleable byproducts,
while the coal fired boiler produces a much larger volume of waste sludge
which contributes significantly to the additional disposal land required for
the latter option.
In coal or oil combustion, NOX is produced by two mechanisms , the oxida-
tion of nitrogen in the fuel (.fuel NOX} , and oxidation of nitrogen in the
combustion air (thermal NOX) . Fuel NOX can account for up to 75% of the total
NOX emissions from a coal fired plant. This is not the case with coal gasifi-
cation based power plants because the coal-bound nitrogen leaves the gasifier
as either N2 or NHj which is scrubbed out in all commercial or proposed pro-
cesses. The issue then becomes one of controlling thermal NOX by limiting
temperature via steam/water injection and/or phased combustion techniques.
At Texaco 's Montebello pilot plant, EPRI has burned medium Btu gas i.n exist-
ing and developmental gas turbine combustors with promising results Cat at-
mospheric pressure). , A 70 to 80% reduction in NOjj emissions over conventional
pulverized coal fired power plants should be achievable with gasification-
combined cycle power plants.
Parti culates
There will be for various reasons, minimal particulate emissions to the
atmosphere from gasification based power plants. Gasification systems, specif-
ically those supported by EPRI, propose at least two sequential intensive gas
scrubbing steps. Isokinetic sampling at Texaco ''s Montebello pilot plant and
the Westfield Development Centre of the British Gas Corporation has failed
to detect any significant particulates after scrubbing. For combined cycle
systems, particulate levels in gas turbine fuel must be minimized to pre-
vent erosion of deposition on gas turbine blades. For mechanical integrity
of these systems, if for no other reason, particulates will be minimized.
Soot formation can occur in pulverized coal fired systems and oil fired
systems, especially during transients or upsets. Soot formation is not ex-
pected to be a problem with coal gas based systems because of the burning
characteristics of the gas and better controllability of the fuel/air ratio.
274
-------
Non-Leachable Slag
EPRI actively supports 3 gasifiers, all of which are slagging gasifiers,
that is, they are operated above the melting temperature of the coal mineral
matter so it is extracted in the form of a glassy inert frit. This slag-
ging mode of operation has two distinct advantages:
1. Operating at higher temperature speeds the gasification reactions lead-
ing to greater throughput per reactor and reduced waste of reactants
(e.g., gasification steam).
2. Slag is environmentally more acceptable than ash.
The EPA proposed Waste Extraction Procedure (.among several others1 has
been performed on the slags produced in all three gasifiers which EPRI sup-
ports, BGC/Lurgi Slagger, Texaco, and Combustion Engineering. Although the
slags were produced from a variety of coals, the •maximum concentrations of
toxic elements in the leachate, or often the •minimum limits of detection with
the available equipment, are shown in Table 4, In no case did the trace
element concentration in the leachate approach the EPA proposed criteria for
hazardous wastes which is 100 times the drinking water standard.. When more
sensitive detection equipment was used, the actual concentrations were most
often much lower than those shown in the table. Those elements with pro-
posed limits greater than 5000 ppB have been omitted from the table since in
all cases, their concentrations in the leachate actually comfortably met the
drinking water standard.
One preliminary comparison has been made between a gasifier slag and
fly ash from a coal fired boiler based on coals with similar ash composi-
tions. This effort was conducted by Oak Ridge National Laboratory under con-
tract to EPRI and examined the leachates on solid wastes from a conventional
wet bottom slagging boiler and the Combustion Engineering pilot plant gasifier.
The fly ash leachate generally had 10 to 1000 times greater concentrations of
toxic elements than the gasifier slag leachate (the narrowest margin was 2
timesl. The slag from slagging gasifiers therefore appear to be environmen-
tally tolerable, certainly more so than fly ash.
GASIFIER SELECTION FOR ELECTRIC POWER APPLICATIONS
Coal gasification is almost as old as the industrial revolution itself,
serving a wide variety of industrial applications from steel, refining, chem-
icals, to fuel and power production. Perhaps it is for this reason there are
so many coal gasification processes currently under development. A recent Oak
Ridge National Laboratory survey lists almost 100 such projects.
A first priority at the outset of the EPRI gasification program was the
establishment of criteria for selection of those processes most likely to
meet the requirements of the power industry. Coincidentally, objective cri-
teria were required to evaluate the status of process development for each.
concept and to assess the risk and benefit involved at each scaleup stage.
The attached Table 5 summarizes EPRI Program Criteria for scaleup to the demon-
stration size of 1000 tons/day of coal per unit, a size judged sufficient
for subsequent commercial deployment.
The electric power industry emphasizes the need for plant reliability
and availability. Therefore, simplicity of design with inherent ease of main-
tenance is very desirable. The preferred gasification process should be
275
-------
Table 4 GASIFIER SLAG LEACHING TESTS
Proposed EPA Gasifier Slag
Limit Leachates
ppb ppb
As 5000 < 200
Cd 1000 < 10
Pb 5000 < 140
Mn 5000 < 250
Kg 200 < 2
Se 1000 < 80
Ag 5000 < 20
Cr 5000 < 20
276
-------
Table 5 CRITERIA USED IN COAL GASIFICATION
TECHNOLOGY SELECTION FOR SCALE UP
TO DEMONSTRATION SIZE C-1000 TPD COAL)
IN THE ELECTRIC POWER INDUSTRY
• Simplicity
• Feedstock flexibility
• Complete carbon conversion
• Absence of troublesome byproducts
• Compatibility with power generation requirements
• Existence of an operating pilot plant of greater
than 100 tpd coal capacity
• Proof Cdirect experimental evidence1 of all
essential aspects of the process with regard to
the above criteria including waste heat recovery
and gas clean up
277
-------
flexible with regard to coal feed properties and should be able to convert
all the carbon to gas. Incomplete conversion or the formation of byproduct
tar gives rise to additional processing complications, disposal,problems
and the potential for greater environmental intrusion. The process must also
be compatible with the power generating system needs. This implies a rapid
response rate for ease of load change, a wide operating range, and a relatively
constant heating value of the product gas throughout the operating range and
during transients. For scaleup to demonstration size, all essential aspects
of the process should have been experimentally proven on a large pilot plant
of 100 tons/day capacity (so that eventual scaleup is less than tenfold).
Since the gasifier is only one part of a large system, such a pilot plant
should also verify the technical concepts for the waste heat recovery and
gas clean up systems.
When these criteria of simplicity, flexibility, cleanliness,etc. are
examined against the known characteristics of the three main types of gasi-
fier - moving bed (both dry ash and slagging), fluid bed and entrained systems,
it is clear that entrained systems come closest to meeting the desired cri-
teria. Coincidentally three such systems - the Texaco, Shell-Koppers and Com-
bustion Engineering, have each progressed to an advanced state of development
and pilot plants greater than 100 tons/day coal capacity are currently being
operated for each of these technologies. Each of these developments is able
to draw on a background of commercial gasification experience, and each of
these organizations plans to scale up the pilot plant to commercial size
demonstration units of about 1000 tons/day coal capacity.
Each of these three entrained systems offer distinct environmental advan-
tages in their demonstrated complete carbon conversion, production of a dense
inert slag, and absence of tar and other troublesome byproducts.
The currently commercial Koppers-Totzek process has similar environmental
advantages although low throughput, as yet incomplete carbon conversion and
atmospheric pressure indicate higher costs than the other three entrained
systems referred to above.
The current commercial Lurgi moving bed gasifier operates with dry ash
removal, and excess steam is injected at the bottom to keep the ash below
slagging temperature. This excess steam requirement reduces the thermal
efficiency and produces large volumes of contaminated water which require
treatment. The British Gas Corporation (BGC) is developing a slagging ver-
sion of the Lurgi gasifier at Westfield, Scotland. By operating at the higher
slagging temperature , essentially only the steam for the gasification reac-
tion is required. The steam consumption and overall efficiency is greatly
improved, and the waste water treatment requirements markedly reduced.
Both dry ash and slagging versions, being countercurrent devices, oper—
ate at lower outlet temperatures and the outlet gases thereby contain tars,
oils and phenols. The slagging version provides a means for their subsequent
gasification by injection into the slagging region, so no net tar production
will result. Lurgi is also working on various recycle schemes to consume the
tars and liquors.
The existence of tars does create additional processing and increased
safety and housekeeping requirements. However, such a choice can be justified
if the overall economics justify the extra costs for environmental accepta-
bility. Processes operating in the slagging region do offer the opportunity
278
-------
for recycle and consumption of streams with fuel value, and a Tseans of
recycle of contaminated water streams (perhaps with coal added as a slurry)
so as to capture the minerals in the slag.
The only currently commercial fluid bed gasifier, the Winkler, has
historically suffered from four problems - feeding caking coal, tar production
(with bituminous coals), high carbon in the ash, and inability to consume fines.
The 'U1 Gas and Westinghouse small pilot plants(< 1 ton/hour) seem to have
been able to solve the caking coal and tar production problems at least in
short runs. By operating with a specially designed ash — agglomerating zone
at the bottom, ash low in carbon has been observed, however, full consumption
of fines has yet to be demonstrated. With the smaller scale of current ex-
perimentation, we judge the scale-up risks, particularly with the ash agglom-
erating zone,to be greater than with the entrained systems. In addition there
is still some concern as to whether tar formation can be avoided during the
load change and start up conditions expected for a gasifier operating in a
power plant.
EPRI TEST RESULTS FROM COAL GASIFICATION PILOT PLANTS
The tests conducted to date on coal gasification pilot plants give rea-
son for optimism that environmentally acceptable commercial power plants can
be designed to economically meet current and proposed emission standards.
However, it must be admitted that in many cases the configuration of the
pilot plants and the short run lengths inevitably associated with pre-commer-
cial facilities, do not lead to results directly translatable to larger con-
tinuously operating plants with full economic use of recycle steams
At EPRI the overall program is aimed at obtaining process and environ-
mental data on several gasification processes judged to be at a stage of
development where commercial deployment can reasonably be projected in the
1980's. These studies are planned, wherever possible, at larger pilot plants
(e.g., BGC/Lurgi at Westfield, Texaco at Oberhausen, and Combustion Engineer-
ing at Windsor, Connecticut), during runs of sufficient length to accommodate
appropriate recycle of process streams.
Comparison of the environmental impact of various coal technologies in
the trace element area is particularly difficult.
Coal is variable, not only from mine to mine in a large deposit, but
even within a given mine, particularly with regard to variation in the -mineral
matter content.
To obtain consistent comparisons of direct coal firing, fluid bed com~
bustion and coal gasification presents a great challenge requiring an
extremely rigorous set of long term tests on the technologies with careful
•monitoring of feedstocks. Too often comparisons are made with different coals,
unrepresentative plants, short runs, etc.
EPRI has supported and is supporting test programs on the BGC/Lurgi,
Combustion Engineering and Texaco technologies. We are also working with.
Shell-Koppers, All of these processes produce the ash as a dense slag and
offer recycle opportunities.
279
-------
BGC/Lurgi Slagging Gasifier
Being a slagging gasifier, the BGC/Lurgi Slagger produces all the coal
mineral matter as an inert glassy frit. Under the DOE's high Btu demonstra-
tion program, tests on U.S. coals were conducted at BGC's Westfield pilot plant
to determine performance and to characterize emissions. Based on the slag
leaching test results, the EPA in Ohio (proposed site of the Demonstration
Plant) has agreed that the slag is a non-hazardous waste.
The Slagger is a countercurrent moving bed gasifier, and therefore tars
are present in the raw product gas. As indicated by the Kosovo tests (the
subject of a paper to be presented later in this meeting), the presence of tars
dictates that a great deal of attention must be paid to plant design and pro-
cedures to prevent worker exposure to these compounds. The Slagger can in all
cases easily accommodate complete gasification (destruction) of these tars as
successfully demonstrated under the EPRI test program (on Pittsburgh No. 8
coal) at the Westfield 350 tpd pilot plant in late 1979. The tars are there-
fore only a plant internal recycle stream and need not intrude into the outside
environment. Another advantage of the Slagger over the dry ash Lurgi type gas-
ifiers tested at Kosovo is that the Slagger normally consumes. 80-90% less
steam, dramatically reducing the hydrocarbon-saturated wastewater stream. Con-
ventional wastewater treatment of this stream to acceptable limits hence be-
comes a much more manageable endeavor. Also, since this stream is so small
the possibility exists of using it to slurry finely ground coal to an entrained
gasifier such as Texaco thus utilizing all the hydrocarbon content of the feed
coal and further simplifying the task of water treatment.
EPRI's economic evaluations of the BGC Slagger show it to be very promis-
ing and therefore worth the extra effort needed to deal with the tars in an
environmentally acceptable manner. The Pipeline Gas Demonstration Plant planned
for Ohio will hopefully verify this acceptability without reducing its
economic viability. An extensive environmental program has already been
specified for this project.
Combustion Engineering
The C-E gasifier has most of the previously cited environmental advantages
of entrained gasifieisover coal fired boilers including non-leachable slag,
no detected hydrocarbon production, minimum particulate, NOX, SO. effluents,
and reduced waste disposal land requirements. Since it operates at atmos-
pheric pressure, the C-E gasifier is economically attractive for oil or natur-
al gas fired boiler retrofit to conserve these valuable resources. In such
applications, however, water consumption would be as great as that in a con-
ventional coal-fired boiler plant. Combined cycle power plants based on the
C-E gasifier also appear competitive with direct coal firing, with advantages
of reduced water consumption and relatively low cost sulfur removal.
A comprehensive program is planned under EPRI sponsorship to measure
gaseous emissions plus liquid and solid effluents from the Process Development
Unit (PDU). gasifier at Windsor, Connecticut. At a design capacity of 120
tons of coal per day, it is currently the largest operating gasifier in the U.S.
An effort is underway by Oak Ridge National Lab CORNL) to compare wastes
from the gasification process with those of a direct coal-based power plant
using similar coal feedstock. The first results are very tentative because
the gasifier has not achieved well-balanced full-scale operation; nevertheless,
280
-------
they are very encouraging. For example, solids leaching tests on gasifier
slag point to very low concentrations of selected metals relative to pro-
posed standards. Results of combustion plant bottom ash were comparable.
However, the fly ash showed 10 - 1000 times the concentration of some toxic
elements. This appears consistent with expectations of an environmentally
acceptable solid waste from high temperature, entrained-flow gasifiers, i.e.,
in the form of chemically inert slag particles.
In the EPRI-funded effort Radian Corporation is preparing to conduct an
extensive sampling program to assess both organic and inorganic emissions,
with emphasis on potentially hazardous components. The methodology developed
here may also form the basis for future environmental assessment of other
prominent gasification technologies.
Texaco Process - Montebello Pilot Plant
In the wake of the 1973 oil embargo, Texaco undertook a concerted effort
to advance the development of its coal gasification process. This technology
had been first tested in the 1950's as an outgrowth of Texaco's successful
partial oxidation process for producing synthesis gas from heavy oils and
natural gas. In the last 5 or 6 years a large number of coals and other
solid feedstocks, including petroleum coke and coal liquefaction residue, have
been tested with considerable success in a 15 tpd pilot plant at the Montebello
Research Laboratory near Los Angeles. Among these tests, particularly in the
most recent 2 year period, have been efforts which have emphasized in signifi-
cant detail the environmental aspects of the process. The equipment configura-
tion at Montebello is shown in the attached flow sketch, Figure 1.
In a continuing set of EPRI-sponsored runs at the Montebello unit utiliz-
ing Illinois No. 6 coal as the feed and employing as the oxidant both oxygen
and, alternatively, oxygen-enriched air (.35% 02! , very encouraging operational
and environmental results have been obtained. The Texaco gasifier was shown
to be particularly responsive, reacting essentially instantaneously to rapid
changes in throughput. The product gas composition remained virtually un-
changed at various load levels and even during fast transients. One major
inherent environmental advantage of the Texaco process over most other gasi-
fiers was confirmed as expected in that no undesirable liquors or tars were
produced. These byproducts, when formed in other processes, usually appear
in the waste water streams, creating a substantial removal and disposal prob-
lem. At the high reaction temperature of the Texaco gasifier (2300 to 2800°F)_,
such condensable materials are unstable and are destroyed.
The SelexoiS/sulfur removal system, when operating within its design
specification, removed upwards of 98 percent of the H2S in the gas. The only
other significant sulfur species present was COS, measured in the feed gas to
the SelexoJC)unit at about 5 percent of the H2S level, and 50 percent of this
COS was removed. It is believed that if required, the COS level could be fur-
ther reduced by catalytic hydrolysis to J^S ahead of the acid gas absorber.
It should be noted that the SelexoiS^ process installed at Montebello is
among the acid gas removal alternatives likely to be preferentially applied in
eventual commercial gasification-combined cycle plants due to its selectivity
in removing I^S versus C02- For gas turbine applications the latter compound,
C0_, can remain in the gas and contribute, in the form of increased -mass flow,
to the total energy developed,
281
-------
IV)
00
r>o
Process
Wator
Slurry
Preparation
Screen
Gaa
Cleaning
Gaslfler
I
-------
In the EPRI-funded test runs, particulate levels in the product gas were
essentially negligible, i.e., less than O,l mg per normal cubic meter. The
ammonia level in the gas was less than 1 ppm. In addition to the product gas,
analytical data were gathered in the EPRI runs to determine the constituents
of various other plant streams, including the presence and nature of trace
materials. With the exception of benzene, organic compounds on the EPA priority
pollutant list were not detected in the effluent and recycle water streams at the
10 ppb level. Benzene was detected at a level of less than 20 ppb in the recycle
water. No polynuclear aromatics (PNA's) which appear on the EPA priority pol-
lutant list were found in the slag or particulates. Leaching tests conducted
on the slag indicated all trace metals found in the leachate fell at least
a.n order of magnitude below the one hundred times EPA drinking water standard
proposed for implementation of the Resource Conservation and Recovery Act. In
fact, all but three trace metals actually met the drinking water criteria, and
these three were present at less than ten times the drinking water standard.
A similar level of environmental analysis and testing to that discussed
above has been conducted by Texaco at the Montebello facility on a western coal,
Kaiparowits. Reference No.2 in the list at the end of this paper contains a
.detailed discussion of coal, gas, water, slag, and slag leachate compositions
in both the EPRI-sponsored Illinois No. 6 coal tests and the Kaiparowits coal
tests.
Larger Texaco Pilot Scale Facilities
Extensive testing, including substantial environmental analysis, is planned
to be carried out in larger Texaco gasification facilities now operating or
scheduled to commence operation soon. EPRI is proceeding with plans to conduct
during the next few months testing of Illinois No. 6 coal in a 150 tpd Texaco
unit in West Germany. These runs will be of similar scope to the oxygen-blown
runs performed at Montebello and the coal has been procured from the same mine.
This larger unit, operated at Oberhausen by Ruhrchemie (a European chemical
firml to produce synthesis gas for a chemical feedstock, has achieved consider-
able success in a planned test program on German coals since its start-up in
early 1978. Unlike the Montebello pilot plant, the Ruhrchemie facility is
equipped with a waste heat boiler, a key component required for efficient gas-
ification-combined cycle power applications. This factor (versus direct water
quench for cooling of the gas as employed at Montebello), along with the larger
equipment sizes in the German unit, should increase the relevancy of the en-
vironmental -measTirements taken to the projected performance of commercial scale
Texaco-based GCC plants. It is intended to perform a careful analysis of the
EPRI results from Oberhausen when available to clearly identify the reasons
for any significant difference from the Montebello tests, i.e., effects of
scale-up, dissimilarities in equipment design or configuration, differing oper-
ating conditions, etc.
Another Texaco gasifier, having a capacity of about 200 tpd of coal is
being readied for start-up by TVA at Muscle Shoals, Alabama. This plant,
designed to produce a medium-Btu gas as feedstock for ammonia synthesis, was
the subject of a paper presented earlier at this meeting. It is understood
that a comprehensive environmental program is planned for the TVA unit, which
utilizes a direct water quench for cooling of the product gas and, accordingly,
should be reasonably representative of a number of other industrial applica-
tions of the Texaco gasifier,
283
-------
COOL WATER DEMONSTRATION PROJECT
A number of major energy technology developers and supporters, including
EPRI, are proceeding with a project to design, construct, and operate a demon-
stration scale (commercial size equipment) GCC power plant at Southern Califor-
nia Edison's existing Cool Water generating station. The demonstration unit
will integrate a 1000 tpd Texaco coal gasifier with a 100 MW combined combus-
tion turbine-steam turbine electric generating system. The plant flow scheme
is depicted in Figure 2 and the project is presently in the beginning stages
of detailed design. A preliminary estimate of the product gas composition
based on the conceptual design of the Cool Water facility is provided in
Table 6. The makeup of the clean gas presented in the table reflects the de-
sign criteria of 97 percent removal of the sulfur in the raw gas based on a
feed coal containing 0.7% sulfur by weight. Similar (and higherl levels of
sulfur removal are quite readily achievable in plants feeding higher sulfur
coals through appropriate selection of design options within one of several
commercially available acid gas removal processes.
The preliminary expected emissions from the Cool Water plant are shown
in Table 7. The projection of S02 emissions is based on the clean gas compo-
sition in the previous table. It should be noted that the NOx emissions shown,
which correspond to approximately 43 ppm, reflect compliance with the plant
permit conditions which apply to the area in California where the plant is to
be situated. This criteria is significantly more strict than the federal New
Source Performance Standard for stationary gas turbines which limits NGx emis-
sions to 75 ppm. To achieve the required low- NOX emissions level the project
intends to employ gas saturation/steam injection prior to combustion, along
with the use of advanced combustor design undergoing development concurrent
with the design effort for the plant facilities.
The good performance anticipated regarding particulate emissions is a
result of effective water scrubbing of the product gas which, is carried out
as an integral part of the Texaco gasification process. The use of enclosed
storage and dustsuppression techniques in the coal receiving, transfer, and
preparation areas will, in addition, provide appropriate control of potential
emissions from these areas.
In the gasifier process section all but a relatively small amount of the
water will be recycled internally. The small amount of process blowdown will
be routed along with cooling tower blowdown and other minor power plant aqueous
effluents to a lined evaporation pond located on-site. The slag produced will
also be stored on—site in an impervious lined storage area, at least until such
time as sufficient data has been collected to confirm that, as expected, this
material is non-hazardous and alternate off-site disposal Cor practical use)
can be pursued.
Sulfur produced in the plant as a by-product will be stored at the facility
unless and until an application has been developed for it.
The Cool Water project has already received the required State environ-
mental permit from the California Energy Commission CCEC1_, The conditions of
the permit granted by the CEC require that an extensive environmental monitor-
ing and surveillance plan be carried out during the plant operations and test
period. The details of this plan are currently being developed.
284
-------
Coal
preparation
Gasifier
Water
02
Air
VSlag
Gas
cooler
Air
separation
plant
Steam
Superheated
steam
Electric
power
Steam
turbine
Particulate
scrubber
t Boiler
j feedwater
Heat
recovery
steam
generator
t Exhaust
gas
Boiler
feedwater Electric
power
Sulfur
removal
Acid gas
Fuel
gas
Sulfur
recovery
Sulfur I
Gas turbine
To Unit 1
boiler
Air
Figure 2 Block flow diagram for Cool Water project
285
-------
Table 6 COOL WATER GCC DEMONSTRATION PROJECT
PRELIMINARY ESTIMATED GAS COMPOSITIONS (DRY)
FROM A CANDIDATE WESTERN DESIGN COAL
Vol. Percent
Component Raw Gas Clean Gas
H2 33.61 35.94
CO 48.22 51.51
C02 17.38 11.86
CH4 0.09 0.10
N2 + Ar 0.54 0.58
H-S 0.15 13 ppmv
COS 0.01 40 ppmv
286
-------
Table 7 COOL WATER EXPECTED EMISSIONS
Lbst/106 Btu CCoall
SO2 0.04
NOX 0.14
Particulates Q.OO5
Notes:
1. Emissions based on performance calculations for a candidate (western)
design coal,
2. Aqueous effluent intended to be routed to lined evaporation pond.
3. Solid wastes (slag) to be stored at site.
287
-------
A one-year monitoring program to provide additional data regarding the
present local environment in the vicinity of the plant site is nearly com-
plete. The data from this effort, undertaken to comply with regulations
promulgated for implementation of the Prevention of Significant Deterioration
(PSD I provisions of the Clean Air Act Amendments, will be submitted to the
EPA to support the recently prepared project application for a PSD permit.
SUMMARY
The data from existing pilot plants enables us to identify the species,
i.e., compounds, present in the various gasification process streams. These
species would not be expected to change in scaled-up commercial facilities.
What remains unclear, however, is the concentration at which these substances
will appear in commercial plants employing recycle of certain materials and
other design dissimilarities for continuous economic operation.
The promise of the data obtained so far strongly suggests that process
schemes to meet present and future emissions and effluent standards can be
economically achieved with coal gasification combined cycle power plants.
Nevertheless the detailed long term environmental impacts and full achieve-
ment of the above promise can only be obtained by continuous long term
operation of a commercial sized Cand configured! demonstration plant. It is
with this very much in mind that EPRI together with Southern California Edison,
Texaco, G.E, and Bechtel have commenced engineering the 100MW gasification
combined cycle demonstration plant at Cool Water.
REFERENCES
1. "300 Btu Gas Combustor Development Program - Phase 1", EPRI Report AF-1144,
Research Project 1040-1, United Technologies Corp., August, 1979u
2. W. G. Schlinger and G. N. Richter, Texaco Montebello Research Laboratory,
paper entitled "An Environmental Evaluation of the Texaco Coal Gasifica-
tion Process", presented at The First International Gas Research Confer-
ence, Chicago, Illinois, June 9-12, 1980.
3. "Preliminary Design Study for an Integrated Coal Gasification Combined
Cycle Power Plant", EPRI Report AF-880, Research Project 986-4, Ralph
M. Parsons Co., August, 1978.
288
-------
COS-H2S RELATIONSHIPS IN PROCESSES PRODUCING
LOW/MEDIUM-BTU GAS*
Michael B. Faist, Robert A. Magee, and Maureen P. Kilpatrick
Energy and Process Chemistry Department
Radian Corporation
8500 Shoal Creek Blvd.
Austin, Texas 78758
ABSTRACT
The chemical aspects of the distribution of sulfur between H2S and
COS in the product gas from the gasification of coal are examined. Comparing
actual gasifier measurements with equilibrium computations we find that the gas
stream becomes frozen corresponding to equilibrium values at high temperature,
most likely corresponding to the reactor exit. This implies a sulfur distribu-
tion with a higher COS concentration than one may expect. The conversion of COS
to H2S occurs mainly by COS hydrolysis, which is very slow at low tempera-
tures. Finite rate studies indicate that an effective catalytic COS hydrolysis
rate constant of 10~1' to 10"^° cm-Vmol sec will allow the reaction to
reach >95% equilibrium in small enough residence time to allow reasonable
reaction vessel sizes.
It is found that the achievable ^S/COS equilibrium ratio is deter-
mined from the product of the locally frozen H20/C02 ratio and the COS
hydrolysis equilibrium constant. The governing parameters for the ^0/002
equilibrium ratios are the temperature, pressure, and the gas stream (H/C) and
(0/C) ratios. The higher the (H/C) ratio and the lower the (0/C) ratio the
larger the H20/C02 ratio and thus the larger the H2S/COS ratio. Moreover,
raising the (H/C) ratio and lowering the (0/C) ratio also increases the achiev-
able CH^ equilibrium concentration from a catalytic methanation module.
*Supported by the Environmental Protection Agency, Industrial Environmental
Research Laboratories/Research Triangle Park under contract EPA 68-02-3137.
289
-------
COS-H2S RELATIONSHIPS IN PROCESSES PRODUCING
LOW/MEDIUM-BTU GAS
I. INTRODUCTION
The production of gaseous and liquid fuels from domestic coal has a
high priority in the overall U.S. energy policy. Of the technologies used to
produce these fuels from coal, gasification and indirect liquefaction are com-
mercially available, and therefore, will be the first generation plants con-
structed in the U.S.
One of the largest process and environmental concerns associated with
gasification and indirect liquefaction technologies is the removal and ultimate
fate of sulfur compounds formed during the gasification of coal. Sulfur com-
pounds will poison downstream methanation and synthesis catalysts and will pre-
sent a potential environmental and health problem if emitted to the atmosphere
at certain levels.
The two primary sulfur compounds formed during coal gasification are
H2S,and COS. Of these, the amount of COS in relation to I^S is of primary
concern because of the following reasons:
• Gaseous sulfur compounds are usually removed by an acid gas
removal (AGR) process (i.e., Rectisol, Selexol, etc.). COS is
less soluble than E^S in physical AGR solvents; therefore, more
energy is required to remove COS from the product gas stream to
levels required for downstream processes (i.e., <5 ppm reduced
sulfur).
• Because of the relative solubility, when a selective AGR
operation is used, COS will distribute itself differently than
H2S in the AGR tail gases.
• Certain sulfur recovery processes (e.g., Stretford) will not
remove COS from AGR tail gases and more expensive sulfur recovery
processes may be required to reduce sulfur emissions from the
plant.
Based on the above reasons, COS can be removed from gas streams; however, it is
more difficult to remove than H2S. In order to design AGR and sulfur recovery
systems it is important to identify and understand the effect of the parameters
which control the distribution of sulfur between H2S and COS in gasifier
technologies.
290
-------
The conversion of COS to H2S is limited by the hydrolysis reaction,
COS + H20 ^. H2S + C02 . (I)
This reaction is sufficiently slow that equilibrium levels cannot be achieved.
However, catalysts exist^"^ which increase the rate of (I) and test modules
are being prepared. The scope of the present study is to investigate the rela-
tionship of H2S and COS in various gasifier technologies. Comparisons between
model computations and actual gasifier measurements lends an understanding of
the systematics to aid in future designs. Both equilibrium and finite rate
considerations are included.
The data base^"!! used for comparison is characterized in Table 1.
As can be seen the gasifiers represent a wide diversity in gasifier technology,
coal classification, and operating conditions. Table 2 shows the measured con-
centrations of the major species as well as the ^S and COS levels contained
in the raw product gas stream. These are the values to be used in comparisons
with model calculations.
II EQUILIBRIUM COMPUTATIONS
The equilibrium concentration of molecular species at a given tempera-
ture and pressure may be calculated by minimizing Gibbs Free Energy constrained
by the conservation of mass for each element. We have performed such calcula-
tions for each gasification system using as input the amounts of total carbon,
hydrogen, oxygen, nitrogen, and sulfur present from the measurements of the
product gas streams. The data base consists of the Gibb's Free Energy of over
70 molecular species from the JANAF handbook.12,13
Figures 1 and 2 show typical results from such calculations. Figure 1
corresponds to the C02 Acceptor^ and Figure 2 to the Wilputte-Chapman^.
The bars on each plot show the measured levels (with 10% uncertainty) of each of
the species. Figure 1 illustrates that the C02 Acceptor is able to maintain
its equilibrium as the gas cools to about 1000K where the reactions become
frozen. Although the Wilputte-Chapman results show a similar effect, the agree-
ment is not as definitive. The CO, H2, and CIfy are in equilibrium corres-
ponding to approximately 900K while the 1^0 is not in the same temperature
range. This is most likely due to an imprecise H20 measurement. Of the I^S
and COS, the COS measurement is much higher than equilibrium would predict at
any temperature. However, this difference is only a factor of 3 and for these
small concentrations, the deviation is considered to be reasonable. In general,
we conclude that at least the major gaseous species (H20, C02, CO, H2, and
CH4) are frozen at equilibrium values corresponding to temperatures in the
900-1300K range.
291
-------
TABLE 1. GASIFIER CHARACTERIZATIONS
Sita
Glen Gery
Fort Snelling
Riley Morgan
Ho Is ton
Rapid City
Montebello
Hanna
Type
Wellman-Galusha
Wellrasn-Galusha
Riley Stoker
Wilputte-Chapaian
C02 Acceptor
Texaco
UCG
Technology
Fixed-Bed (Thick)
Fixed-Bed (Thick)
Fixed-Bed (Thin)
Fixed Bed (Thin)
Fluid ized Bed
Entrained Bed
In-Situ
Coal
Anthracite
Lignite
Lignite
Subbituminous
Sublignite
Subbituminous
Subbituminous
Gas
Low-B tu
Low-Btu
Lotv-Btu
Low-Btu
Med-Btu
Med-Btu
Low-Btu
Pressure
(atm)
1
1
1
1
10
24
5
Flowrate
(scfs)
45
30
80
390
20
800
55
Identifier
GG
FS
RS
we
CA
T
UCG
TABLE 2. PRODUCT STREAM COMPOSITIONS3
ro
vo
Gasifierb
GG
FS
RS
we
CA
T
UCG
N2
(vol %)
48.5
37.6
33.9
50.9
6.0
0.3
47.1
H2
(vol %)
15.3
12.4
13.2
13.2
40.7
34.0
14.4
CO
(vol %)
24.0
21.1
20.3
17.9
11.7
43.8
11.4
CH4
(vol %)
0.22
0.77
0.77
1.4
8.8
0.029
2.6
H20
(vol %)
5.9
19.6
25.6
7.0=
24.7
0.47d
11.7
C02
(vol %)
5.2
7.6
5.3
7.7
7.1
21.1
11.8
E2S
(ppmv)
649
892
860
228
1000
1264
2584
COS
(ppmv)
87
115
95
25
7.5
48
84
.Only major species, H-S and COS compositions given.
Identifier from Table 1.
Estimated from partial data.
Assumed value corresponding to saturation at 100F. This value is a lower
bound to the H?0 level in the gas stream.
much higher.
The actual value is probably
-------
CO
O
§
oc
u.
UJ
oc
O
£
UJ
O
u
I
500
1000
TEMPERATURE (K)
1500
Figure 1. Plot of the Calculated Equilibrium Major Gas Species and the
H2S-COS Distributions as a Function of Temperature for the CO,
Acceptor Gasifier.
10% uncertainty).
Bars indicate actual measured levels (witfi
293
-------
o
DC
U.
UJ
O
z
<
s
DL
O
UJ
0.
500
1000
TEMPERATURE (K)
1500
Figure 2. Plot of the Calculated Equilibrium Major Gas Species and the
H S-COS Distributions as a Function of Temperature for the
Wilputte-Chapman Gasifier. .Bars indicate actual measured
levels (with 10% uncertainty).
294
-------
Figure 3 gives the calculated values for the H2S/COS ratio (by vol-
ume) for each of the gasifiers as a function of temperature assuming the system
maintains equilibrium at all temperatures. It should be noted that the measured
H2S/COS ratios for only the C02 Acceptor (CA) and the in-situ (UCG) gasi-
fiers correspond to I^S-COS equilibrium at any temperature; all others show
actual levels much lower than their equilibrium level. This is a clear indi-
cation that if equilibrium could be achieved between H2S and COS much more of
the sulfur would be in the form of ^S, especially at lower temperatures.
If H2S and COS were at equilibrium then reaction I shows that the
H2S/COS ratio is directly related to the H20/C02 ratio by the equilibrium
constant, Kj, namely,
(1)
cos / \ co2 / Ki
Since Kj is monotonically increasing with decreasing temperature as shown in
Figure 4, the larger the H20/C02 ratio is the larger the H2S/COS ratio
will be. Figure 5 shows the behavior of the equilibrium H20/C02 ratio with
changing temperature. Again bars indicate the actual measurements. Note that
the H20/C02 ratios form a family of curves related by the H/C ratio by
weight. As may be expected, the higher the H/C ratio the greater the ^O/
C02 ratio.
Now, if a catalytic module were added to increase the rate toward
equilibrium of reaction I, and since the H2S and COS are present in very low
concentrations compared to H20 and C02, H2S/COS equilibrium would be ob-
tained without significantly affecting the ^0 and C02 concentrations. Here
the equilibrium H2S/COS ratios will not be as in Figure 4 but will have the
form
/H S\
\COS/
COS
where the constant in Equation (2) is the frozen value of H20/C02- Figure 6
shows the possible equilibrium values achievable for the gasifiers studied here.
These are simply Kj(T) multiplied by the actual (H20/C02> ratio of each
gasifier* The equilibrium values of H2S/COS = R* are plotted on the left hand
axis. If only 90% of equilibrium were reached, i.e., H2S/COS = 0.9R*, then
the fraction of sulfur as H2S is H2S/(H2S + COS) = 0.9R*/(0.9R* + 1). The
right hand axis is scaled to this fraction. Therefore, if the module achieved
90% equilibrium at 500K nearly all gasifiers would yield >99.9% of all sulfur as
H2S.
295
-------
10,000.
1000.
I
g 100.
10.
1.0
CA
500
1000
TEMPERATURE (K)
1500
Figure 3. Plot of the Calculated H.S-COS (by volume) Ratio Corresponding
to Total System Equilibrium for Each Gasifier. Identifiers are
as in Table 1. Bars indicate actual measured levels (with 10%
uncertainty).
-------
1Q5
104
102
10
500
1000
TEMPERATURE (K)
1500
Figure 4. Plot of the COS Hydrolysis Equilibrium Constant as a Function
of Temperature .
297
-------
10.0
T I r
I H/Cfw
H/C (w/w)
1.0
I
O
0.1
0.01
500
1000
TEMPERATURE (K)
1500
Figure 5. Plot of the Calculated H20/C0 (by volume) Equilibrium Ratio
for Each of the Gasifiers. Identifiers are as in Table 1.
Bars indicate actual measured levels (with 10% uncertainty).
298
-------
10,000.
500
1000
TEMPERATURE (K)
1500
Figure 6. .Plot of the Achievable H2S/COS (by volume) ratio for Each of the
Gasifiers Assuming a Frozen 1^0/CO. Ratio Corresponding to Measured
Levels and COS Hydrolysis Equilibrium. Identifiers are as in
Table 1.
299
-------
It is clear that the greater the H20/C02 ratio the greater the
achievable H2S/COS ratio. Therefore, it is worth considering which parameters
determine the H20/C02 ratio. Both H20 and C(>2 are major species in the
gas phase and as such they will only be affected by the other major species. Of
the major elements present (C, H, 0, and N) only the C, H, and 0 will affect the
H20/C02 ratio. Moreover, since we are only interested in a ratio, only the
(total H/total C) and (total 0/total C) ratios in the gas stream are important
to the equilibrium. Figure 7 shows the correlation of the gasifiers between the
0/C and H/C ratios by weight, designated (0/C)W and (H/C)W, respectively.
The (0/C)W ratio for each gasifier (except the CC>2 Acceptor) is empirically
related to the (H/C)W ratio by
(0/C)W = 7.6 (H/C)W + 0.88 . (3)
The (0/C)W ratio is much lower in the C02 Acceptor due to the removal of
C02 to form CaC03 in the fluidized bed, and the absence of Q£ in the input
stream.
Using the relationship of Equation (3) the H20/C02 equilibrium
ratio is uniquely determined from the (H/C)W ratio. Separate equilibrium com-
putations were performed for atmospheric pressure considering only H, C, and 0
with various (H/C)W ratios and Equation (3). The result for the H20/C02
ratio are presented in Figure 8. Comparing Figure 8 to Figure 5, we find the
H20/C02 equilibrium ratio to be identical when conditions are the same.
Moreover, even when conditions are very different, such as the C02 Acceptor,
the H20/C02 ratio is in agreement within approximately 25% for temperature
greater than 800K. Therefore, if one knew the (H/C)W ratio and approximated
the temperature at which the H20/C02 becomes frozen (in most cases 1000-
1200K) the achievable I^S/COS equilibrium ratio could be estimated from
Figures 4 and 8 using Equation (2).
111. FINITE RATE CONSIDERATIONS
From the previous section, it is clear that at lower temperatures
nearly all of the sulfur would exist as H2S if equilibrium for reaction I
could be obtained. If a catalyst is used, the equilibrium is unaltered, only
the rate at which the equilibrium is attained is increased. Several catalysts
have been partially investigated^"^ which enhance the hydrolysis of COS; how-
ever, rates are ill-defined and catalytic poisoning has not beeen well charac-
terized. Nevertheless, it is useful to understand the effect of various rate
constants on the design of catalytic COS hydrolysis process modules.
300
-------
i
u
6
1 7
/
. RS
FS '
UCG
CA
GG WC
V
-(O/C)W = 7.6(H/C)
H/C (w/w)
0.6
Figure 7. Plot of the Gas Stream (_0/C) to (H/C) Correlation for Each
of the Gasifiers. Identifiers are as in Table 1.
301
-------
10.0
P = 1 ATM.
H/C(W/W)
0.10
1.0
o
0.1
0.01
I
500
1000
TEMPERATURE (K)
1500
Figure 8. Plot of the Calculated Equilibrium H^O/CO Ratio as a Function
of Temperature and at 1 atm. for Several CH/C.) Ratios.
(0/C) Ratios determined from Equation (3).
w
302
-------
Consider kf and kr as the effective forward and reverse rate con-
stants for Reaction I, respectively. Then the rate of change of COS is given by
dn
where n^ is the density of the ith species in moles/cm^. Now, by conserva
tion of sulfur species
ntotal,S " nCOS + n
- "
COS
= "cOS
where the superscript "o" and asterick indicate, respectively, the initial and
equilibrium values. Using Equation (5) in Equation (4) and recognizing that
Kj = kf/kr, Equation (4) may be rewritten
d.n
COS
dt - - a ncos + e '
(6a)
where
n
co
(6b)
and
= k (no + no \
r '•••H S COS; n
As discussed in Section II, H20 and C02 are major species and remain
unchanged by any redistribution of sulfur species, e.g. , reaction I. Therefore,
the H20/C02 ratio will be constant during the approach to the H2S-COS
equilibrium. Using this, a is time independent and may be written as
a = k
01 k
—
R*
(6bf)
where R* is the equilibrium ratio,
solution to Equation (6a) is given by
= H2S*/COS*. Finally, the
n - n*
COS COS
- n* )e
COS CQSJ
~at
with a given by Equation (6b').
303
-------
Defining an extent of equilibrium, Y , by
R \s/ncos
"Y = R* ~n* To* ' (8)
H2S' COS
and after considerable manipulation, we find
** " **
R*
~ at
where Yn corresponds to the initial value of Y-
Toward obtaining residence times to reach a given extent of equili-
brium, Equation (9) may be rearranged as
(Y + R*) (1 - Yn)
at = In -7 :
a-Y)
Now using the ideal gas relationship for the total gas phase density (n), and
nH20 = xH20n» where X^o is tne H2° mole fraction, a [cf. Equation
(6b* )] is given by
a = . (n)
where P and T are the pressure and temperature, respectively. Substituting
Equation (11) into Equation (10) , we find
-77
, 1.36 x 10
If t=r is the time to reach 90% of equilibrium then Y= 0. 9 and the right hand
side is a given value depending on the achievable equilibrium ratio R* and the
initial value YQ.
304
-------
Table 3 presents these values for a wide range of YQ and R* for
Y= 0.85, 0.90, 0.95. As can be seen, the entries are relatively independent of
Y and R*, and all entries are well represented by
VokfFT/T -
1.5 ± 1 x 10 22 Y = 0.85
2.0 ± 1 x 10~22 Y = 0.90 (13)
3.0 ± 1 x 10~22 Y = 0.95
In fact, all three categories may be summarized by
XR Q kf PT/T = 2.0 ± 2 x 10~22 (14)
or, for a given process with a given rate constant, the reaction time necessary
to achieve >95% of equilibrium is
TR > 4 x ID'22 T/X^ P kf . (15)
Here, we have used the conservative upper limit for the constant. The fact that
these constants are all very similar in magnitude is just a reflection of the
nature of first order kinetics. That is, these constants represent the driving
force toward equilibrium and the further the system is from equilibrium
initially, the faster the approach to equilibrium, providing similar times to
reach the desired extent of reaction. Now, the required residence time in a
reactor (reaction time) is related to the reactor volume, V, and the actual gas
flowrate, F, by
T = V/F = 300 V P/F T , (16>
R o
where FQ is the flowrate at 300K and 1 atm. Therefore, Equation (15) may be
rewritten
f > 1.33 x ID'2* T2/^ P2 kf . 07)
o 2
305
-------
TABLE 3. EQUILIBRIUM DRIVING FORCES3
A**
10-*
10-3
10-2
10-1
0.5
0.7
IQ-*
10-3
10-2
10-1
0.5
0.7
10-3
10-2
10-1
0.5
0.7
aEntries
10°
1.71
1.71
1.69
1.57
0.96
0.53
2.00
2.00
1.99
1.87
1.26
0.82
2.49
2.49
2.48
2.35
1.74
1.31
correspond to
101
2.45
2.45
2.43
2.30
1.53
0.87
2.95
2.95
2.94
2.81
2.04
1.38
3.82
3.81
3.80
3.67
2.90
2.24
1022 XH2(
102
Y = 0.85
2.57
2.56
2.55
2.42
1.63
0.94
Y = 0.90
3.11
3.11
3.10
2.97
2.17
1.48
Y = 0.95
4.05
4.05
4.03
3.90
3.11
2.42
) kfPT/T.
103
2.58
2.58
2.56
2.44
1.64
0.94
3.13
3.13
3.12
2.99
2.19
1.49
4.07
4.07
4.06
3.93
3.13
2.43
10*
2.58
2.58
2.56
2.44
1.64
0.94
3.13
3.13
3.12
2.99
2.19
1.49
4.07
4.07
4.06
3.93
3.13
2.43
306
-------
Equation (17) may be thought of as a design criterion for a process module. It
relates the necessary volume of the module to the governing parameters, Figure 9
shows a log-log plot of V/FO vs kf for each of the gasifier conditions with
a process module temperature of 500K. V/FQ values above the line correspond
to a sufficiently sized process module for a given effective rate constant to
achieve 95% equilibrium. The two horizontal dashed lines correspond to large
scale systems (flowrates of 3000 SCF/sec) with modules of 1000 and 100 ft-*.
For these parameters, the catalytic rate must be kf ^ 10~1'-10~1" cm-V
mol-sec to handle all gasifiers. The noncatalytic gas phase rate constant is
not known but is estimated to be 10~26-10~2^ cm3/molsec at 500K. This
would correspond to an activation energy of approximately 15000K. Since cataly-
tic enhancement is thought to reduce the activation energy to approximately
3000K, this type of catalytic module would appear encouraging.
IV. EQUILIBRIUM REVISITED
In the previous section the governing parameters and their relation-
ship to the process module were determined. With them, once the effective
hydrolysis rate constant is determined, an optimal module may be designed. This
model presents the parameters necessary to reach a desired fraction of the
equilibrium H2S/COS ratio. This ratio is determined by the gasifier operating
conditions. As noted earlier, the I^S/COS equilibrium ratio is directly
related to the COS hydrolysis equilibrium constant by the frozen H20/C02
ratio in the gas stream. Since the value of the H2S/COS ratio is so important
to the attainable sulfur redistribution in the process module, a few points
should be noted regarding this ratio and any effect on the gaseous product fuel.
Although the minimization of Gibbs Free Energy is a numerically effi-
cient and general method of obtaining the equilibrium compositions, often the
more explicit method of solving equilibrium constant expressions can lead to
insights obscured by the above technique. In a gasifier, the major molecular
participants are H2, CO, CH4, H20, and C02* Therefore, there are only
five conditions necessary to determine the concentrations of these species.
These are the three elemental conservation equations and two additional chemical
equilibrium equations. Namely
H = 2H2 + 2H20 + 4CH4 II
0 = H20 + CO + 2C02 III
C = CO + C02 + CH4 IV
H20 + CO = C02 + H2 V
3H2 + CO = H20 + CH4 VI
The two chemical equations are the water-gas shift (V) and methanation (VI)
reactions.
307
-------
10
1000 , 3000
-\ 100 . 3000
-25
logkf
Figure 9. Log-Log Plot of V/F vs. k for Each of the Gasifiers Corresponding
to a Process Module Temperature of 500K. Area above the Lines
Corresponds to Conditions such that the Reaction will Achieve
Equilibrium. Below the lines the ratio is too slow for the
reaction to proceed.
308
-------
Now, the equilibrium of the methanation reaction is such that at high
temperatures the equilibrium is totally shifted to the left, with no City
present. At lower temperatures, equilibrium is with the CH4 formation,
however, rates became too slow to achieve the equilibrium. Since Ctfy is a
more economical fuel, often a methanation module is added to convert the H2
and CO to CE^. Therefore, it is important to understand the equilibrium over
the entire range of temperatures.
The equilibrium is naturally divided into three temperature regions
denoted by A, B, and C. Only in region B are all five molecular species
present. The molecular distribution of major species within the regions are:
A: CH4, H20,C02 (T ^ 700K)
B: CH4, H20, C02, H2, CO (700^ T^ HOOK)
C: H20, C02, H2, CO (T £ HOOK)
Therefore, since the molecular species are reduced in regions A and C, only B
requires the entire (II-VI) set of equilibrium conditions. In regions A and C,
the conditions become
H = 2H20 + 4CH4 II1
region A: 0 = H20 + 2C02 III1
C = C02 + CH4 IV
and
H = 2H2 + 2H20 II"
region C: 0 = H20 + CO + 2C02 III"
C = CO + C02 IV"
H20 + CO = C02 + H2 V
In region A the molecular distribution of major species is trivially
determined from the conservation equations.
309
-------
The solution (per mole of carbon) in region A is
'CH*\ 3 /H\ 3 /0\ 1
' - +
C I 2 \C/ 16 \CJ 2
w w
H\ .3/0
8
= 1 !fH\ _3_ /£\ (18c)
2 2 \C/ 16 \c)
'w 'w
Therefore, as (H/C)W is increased, the yield of CH4 and H20 is increased
and C02 is decreased, while as (0/C)W is increased the yield of CIfy is
decreased with I^O and C0£ being increased. Note that there is no pressure
or temperature dependence within this region.
Region C has a temperature dependence due to the addition of the water
gas reaction (V). However, since there is no change in the number of moles
during this reaction there is no pressure dependence throughout this region.
The solution for the molecular species within this region is
1(°) _
4 Vc/
w
co9
2| (19a)
C(A ' < ' -' (19b)
/ \ *•• / >
u \ /pn \
S ' <® -K§) +1 *^ ,
7 w x 'w \ / »
/C09\ 3,
with I—£1 _ G(T) [ H^(T) - 1 ] , (19d)
\ ^ /
/H\ "^ /n\
G(T) = [ 6(f) - f (l-V(c).. +13/2(1-V •
(19e)
where
w
and H(T) = 1 + 71=^2 |f (§) " 1]
v L w -J
(19f)
Table 4 gives the values of Ky for several temperatures.
310
-------
TABLE 4. EQUILIBRIUM CONSTANT FOR
+ CO = C02 +
T (K)
(C02) (H2)
(H20) (CO)
1600
1500
1400
1300
1200
1100
1000
900
800
700
600
500
0.3360
0.3899
0.4645
0.5718
0.7337
0.9936
1.445
2.315
4.246
9.472
28.44
138.0
311
-------
As seen from Figures 1 and 2, the H20/C02 ratio is most likely to become
frozen at temperatures corresponding to region C (or perhaps region B). In
region C the H20/CC>2 ratios is given by,
H-0 » i , \ *_j /
2 \ L4 w j , (20)
co2 / (co2/c)
Here, an increase in (H/C)W [with constant (0/C)W] implies an increase in
H20 at the expense of C02 and thus an increase in the H20/C02
equilibrium. Another useful simplifiction within this region is obtained when
YJJ = 1. This condition corresponds to a temperature of approximately HOOK.
Here the H20/C02 ratio is easily found from
H20\
~- = 6 (£) T - HOOK . (21)
co2 ) \c /w
Region B is the only one which requires the full set of equilibrium
conditions, namely the addition of the methanation reaction. Since this reac-
tion decreases the total number of moles,the corresponding equilibrium constant
carries a factor of P^. Therefore this is the only region which will show a
pressure dependence as well as a temperature dependence.
Figure 10 shows a replotting of Figure 9 with the three temperature
regions indicated by vertical dashed lines. The accuracy of Equations (18-21)
is related by the plotted points within each region. The open circles corres-
pond to Equation (18), the solid circles correspond to Equations (19 and 20),
and the open squares correspond to Equation (21). This figure and the above
discussion illustrate that for most temperatures and pressures in the gasifica-
tion of coal, the equilibrium distribution of the major species may be predicted
without the need for more elaborate computations. Examining these relation-
ships, the governing parameters are found to be the temperature, pressure, and
the (H/C)W and (0/C)W ratio. Moreover, using Equations (18-21) it is
possible to obtain a set of conditions which will give a desired equilibrium
distribution of the sulfur species. In the following section, we will examine
the gasifier as a whole and discuss the effect of these parameters on the
overall quality of the product gas.
V. CONCLUDING REMARKS
The gas phase chemistry of a gasifier has been studied with particular
attention to the major species and their influence on the equilibrium distri-
bution of sulfur between H2S-COS and the size of the process module needed to
achieve the desired extent of equilibrium. One important conclusion is that the
312
-------
10.0
1.0
I
CM
o
O
I
0.1
0.01
500
1000
TEMPERATURE (K)
1500
Figure 10. Plot Similar to Figure 9 Showing the Three Temperature Regions
(see text). Open circles correspond to the analytical expressions
of Equation (18), solid circles correspond to Equations (19 and
20), while open squares correspond to Equation (21).
313
-------
residence time is essentially independent of the initial and final H2S/COS
ratios. Therefore, there are no module design criteria which depend on the
desired sulfur redistribution. The attainable H2S/COS ratio is completely
determined by the local H20/C02 ratio and the COS hydrolysis equilibrium
constant.
The H20/C02 ratio is controlled by the water-gas reaction at high
temperatures (>1100K) and by the water-gas and methanation reactions at inter-
mediate temperatures ( 700-1 100K). As the gas stream is quenched upon exiting
the gasifier reactor these reactions become very slow and the H20/C02 ratio
becomes frozen corresponding, most likely, to its equilibrium value at the tem-
perature of the reactor exit. Although, this temperature may be between 700 and
HOOK, (i.e., the pressure dependent region), the adjoining temperature regions
are pressure independent. Therefore, we expect that the H20/C02 ratio is
not strongly dependent on pressure. This has been born out for the gasifiers
considered in the present study.
Apart from temperature and pressure, the parameters which govern the
H20/C02 equilibrium ratio are the (H/C)W and (0/C)W ratios. In general,
increasing the (H/C)W and decreasing the (0/C)W ratios increases the I^O/-
C02 ratio which in turn increases the H2S/COS equilibrium ratio. It is
important to note that the affect of increasing the (H/C)W and decreasing the
(0/C)W ratios also increases the equilibrium Ctfy yield. Therefore, attempt-
ing to improve the sulfur distribution not only does not lower the attainable
yield from the methanation module but actually increases it.
Although, from the above discussion, it would appear that every effort
should be made to increase the (H/C)W ratio and decrease the (0/C)W ratio,
this is only true within bounds. The gasification of coal requires fairly high
temperatures. Moreover, the overall gasification reactions,
C + H20 = CO + H2
CO + H20 = C02 + H2
C + C02 = 2CO ,
are endothermic. Thus, if heat is not continually supplied the temperature will
drop and gasification will cease. This heat is produced from the combustion
zone where some of the carbon is oxidized to C02- Now, the (H/C)W ratio
may be increased by introducing more steam but this will increase the (0/C)W
ratio as well. In order to decrease the (0/C)W ratio the air (or oxygen) flow-
rate must be decreased. However, decreasing the air will cause less combustion
and therefore lower the reaction zone temperature. In actuality, increasing the
steam flowrate, will increase the endothermic gasification reactions, resulting
in lower temperature. Therefore, an increase in steam flowrate must be accom-
panyed by an increase in air (or oxygen) flowrate to maintain temperature.
314
-------
In summary, the major points of this study are:
• A process module with an effective catalytic COS hydrolysis rate
constant of approximately 10~17 to 10~16 cm^/mol-sec will
reach >95% of the equilibrium H2S/COS ratio in small enough
residence times to allow reasonable reaction vessel sizes.
• This resonance time is essentially independent of initial and
final H2S/COS ratios.
• The achievable H2S/COS equilibrium ratio at a given temperature
is completely determined from the product of the locally frozen
H20/C02 ratio and the COS hydrolysis equilibrium constant for
that temperature.
• The H20/C02 ratio becomes frozen at approximately 900-1200K,
probably near the reactor exit temperature.
• The governing parameters for the H20/C02 equilibrium ratios
are the temperature, pressure, and the gas stream (H/C)W and
(0/C)W ratios.
• The higher the (H/C)W ratio and the lower the (0/C)W ratio,
the larger the H20/C02 equilibrium ratio and thus the larger
the H2S/COS equilibrium ratio.
• Raising the (H/C)W ratio and lowering the (0/C)W ratio also
increases the achievable CE^ equilibrium concentration.
ACKNOWLEDGMENTS
The authors would like to thank Mr. Robert V. Collins and Dr. Gordon
C. Page for many stimulating discussions during this work. The equilibrium com-
putations were performed using the PACKAGE CODE which was developed and extended
by many people (including Michael B. Faist) at Aerodyne Research, Inc.
315
-------
VI. REFERENCES
1. E.G. Cavanaugh, W.E. Corbett, and G.C. Page, "Environmental Assessment Data
Base for Low/Medium-Btu Gasification Technoloyg: Volume II, Appendices
A-F," EPA Report No. EPA-600/7-77-125B (November, 1977).
2. A.Y. Chan and I.G. Dalla Lana, Can. J. Chem. Eng., 56, 751 (1978).
3. R.K. Kerr and H.G. Paskall, Energy Process/Can. 69, 38 (1976).
4. Z.M. George, J Catal, _35_, 218 (1974).
5. W.C. Thomas, K.N. Trede, and G.C. Page, "Environmental Assessment: Source
Test and Evaluation Report - Wellman-Galusha (Glen Gery) Low-Btu Gasifica-
tion," EPA Report No. EPA-600/7-79-185 (August, 1979).
6. M.P. Kilpatrick, R.A. Magee, and T.E. Emmel, "Environmental Assessment:
Source Test and Evaluation Report - Wellman-Galusha (Fort Snelling) Low-Btu
Gasification," Radian Report No. DCN 80-218-143-116 (April, 1980).
7. M.R. Fuchs, R.A. Magee, and D.A. Dalrymple, "Environmental Assessment:
Source Test and Evaluation Report - Riley-Morgan (Riey Stoker) Low-Btu
Gasification," Radian Report (in preparation).
8. Gordon C. Page, "Environmental Assessment: Source Test and Evaluation
Report - Chapman Low-Btu Gasification," EPA-600/7-78-202 (October, 1978).
9. Radian Corporation, "Environmental Characterization of the C02 Acceptor
Process, Book II: Data Summary," Radian Report No. DCN 78-200-154-13
(March, 1978).
10. Radian Corporation, "Environmental Characterization of the Texaco Coal
Gasification Pilot Plant," Radian Report No. DCN 79-216-288-03 (October,
1979).
11. R.M. Mann, R.A. Magee, W.A. Williams, and C.J. Thielen, "Production Gas
Monitoring of the Hanna In-Situ Coal Gasification Tests," Radian Report
(in preparation).
12. D.R. Stull and H. Prophet, "JANAF Thermochemical Tables," 2nd Edition, Nat.
Stan. Ref. Data Ser., Report No. 37, National Bureau of Standards (1971),
Washington, D.C.
13. D.R. Stull and H. Prophet, "JANAF Thermochemical Tables," 2nd Edition, Nat.
Stan. Ref. Data Ser., Report No. 37, National Bureau of Standards (1971),
Washington, D.C.
316
-------
BEHAVIOR OF A SEMIBATCH COAL GASIFICATION UNIT
by
W. J. McMichael
D. G. Nichols
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, N. C. 27709
ABSTRACT
This paper describes the transient behavior of a laboratory scale
fixed-bed gasifier operated in a semibatch mode. The operation is batch
with respect to the coal feed and continuous with respect to gas flows.
Various coals ranging from lignite to bituminous were gasified using
steam-air mixtures at 1.4 MPa (200 psia) and approximately 900°C. The
transient behavior of the reactor temperature at various coal bed depths
was examined. Test results from nine tests involving five coals are
reported. The data presented include the rate of production of various
gasification products. These include CH,, CO, R~, benzene, toluene,
xylene, H S, COS, and thiophene, as a function of run time. It was
found that the majority of the CH,, the minor hydrocarbons, and sulfur
species were evolved during coal devolatilization. These data were
analyzed using a simple kinetic model which assumes that the rate of
production of a compound at any time is proportional to the (potential)
amount of that compound remaining in the coal. This model explains the
data reasonably well during the devolatilization period. It was found
that the specific rate of production of individual species was practically
the same for all coals and gasification products considered; the ultimate
yield was dependent on coal type. The ultimate yield of (a) CH, or
benzene, and (b) sulfur species roughly paralleled the volatile and
sulfur contents of the coals, respectively.
Duane G. Nichols is now with the Conoco Coal Development Company, Research
Division, Library, PA.
317
-------
BEHAVIOR OF A SEMIBATCH COAL GASIFICATION UNIT
INTRODUCTION
The Research Triangle Institute (RTI) has performed over 40 gasification
[1 21
tests in a laboratory scale gasifier using a variety of coals. ' During
these tests, RTI has developed procedures for the sampling of the various
gasifier process streams and for identifying and quantifying potential environ-
[3]
mental pollutants found in these streams.
The coal gasification tests were performed in a semibatch reactor where
the experiments are batch with respect to the coal and continuous with respect
to gas flows. The gasifier is approximately 6.6 cm I.D. and its 60 cm active
length is surrounded by a three zone furnace. During a gasifier run, the
gasifier was initially heated electrically to the desired gasification tempera-
ture of about 950°C with the desired air and steam flow passing through the
gasifier. The air flows varied from 5.0 to 15.0 standard liters per minute
(slpm) and steam varied from 5.0 to 18.0 slpm. After reaching gasification
temperatures, the coal was batch-fed to the gasifier with the charge ranging
from approximately 1.2 to 1.6 kg. The coal size was 8 x 16 mesh, and the
charge was supported by a porous ceramic plate which also acted as the gas
distributor.
The coal was charged to the gasifier at room temperature and, consequently,
cooled the gasifier well below the initial temperature. This behavior is
shown in Figure 1. Recovery of the temperature took about 30 minutes, and the
rate of increase in the average bed temperature after coal drop appeared to be
proportional to the difference between the average final temperature and
instantaneous average bed temperature. It was found that after the recovery
period, the temperature profiles in the coal bed closely matched the initial
temperature profile and were dominated by the furnace except in the combustion
zone of the bed.
The gasification tests were characterized by two distinct periods of
operation: (1) the initial stage after the coal drop during which devolati-
lization of the coal occurred (surge period), and (2) a steady-state period
which followed the surge and was the stage where coal gasification took place
resulting in a fairly steady product gas composition.
318
-------
1100
71 minutes after coal drop
1000
Initial Temperature Profile
300
0 10 20 30
Height Above Distributor, in.
Figure 1. Temperature Profile in the Batch Gasifier
Run 25
319
-------
Figures 2, 3, and 4 show the time-dependent nature of a typical gasi-
fier test in which Illinois No.6 bituminous, Wyoming subbituminous and
North Dakota lignite coals were gasified. The composition of the coals
and gasification conditions, are shown in Tables 1 and 2, respectively. It
can be seen from these figures that production of methane and other minor
hydrocarbons is greatest during the initial stage of the gasification test
or during coal devolatilization. The production rate of these components
fall almost two orders of magnitude from their initial rates during the
surge period. A more complete description of the production rate-time
characteristics of the semibatch gasification of the five coals in nine
[4]
tests have been presented elsewhere.
Based on the data in Figures 2, 3, and 4 and additional data pre-
F41
sented by McMichael et al., the following observations can be made
about the rate of pollutant and product production as a function of time:
1. The production of pollutants and CH, in the product gas usually
surges to a high rate just after the coal drop, and drops quickly
as the bed temperature rises. A majority of the minor components
and CH. are formed in the first 25 to 30 minutes of the run.
[i
After this time the product rate decreases.
2. For the bituminous coals and the Montana subbituminous coal the
rate of H~ production increases during the initial stages of
gasification during devolatilization. This could be a conse-
quence of (a) increasing bed temperatures at the beginning of
the run resulting in increasing H~ formation from the steam-
carbon reaction, and (b) decreased availability of reactive
carbon as coal devolatilization proceeds, thus more H« appears
in the gas. Hydrogen formation peaks early in the run, and the
rate of formation decreases fairly steadily over the remainder
of the run. This steady decrease is probably due to the
decrease in the density of carbon in the bed with time.
3. For a steady flow of steam and air, the rate of production of CO
approximately parallels the H~ production.
4. For Illinois No.6 bituminous coal, the rate of CO- production
reaches a maximum in the initial stage of the gasification run
and then decreases or remains fairly constant. The Western
Kentucky coal also shows this trend except the production rate
increased sharply at oxygen breakthrough. For the subbituminous
and lignite coals, CCL production reaches a maximum during
devolatilization and then quickly drops to a minimum at about
25 minutes into the run. After this minimum the production
rate increases steadily over the length of the run. The C09
increase is usually accompanied by a slow decrease in the
rate^of CO production. The reason for this could be that as
the density of carbon in the bed decreases through gasifi-
cation, more CO is burned in the gas phase.
320
-------
100 200
Run Time, minutes
SULFUR COMPOUNDS IN RAH GAS
'.2
\
i io"^ L • Thi°°hene
100 200 300
Run Time, minutes
6ASIFIEB OPERATING CONDITIONS
100 200
Run Time, minutes
Figure 2. Gasifier operating conditions and production rate of various com-
pounds as a function of run time - Run 23, Illinois No.6 bituminous
coal.
321
-------
'"F-V--^*—.
co,
MAJOR COMPONENTS IN RAW GAS
MINOR HYDROCARBONS IN RAW GAS
Run Time, minutes
50 100
Run Time, minutes
SULFUR COMPOUNDS IN RAH GAS
a io-J
Run Time, minutes
£ 600
3
I
GASIFIER OPERATING CONDITIONS
5.0 |
3
Run Time, minutes
Figure 3. Gaslfier operating conditions and production rate of various com-
pounds as a function of run time Run 33, Wyoming subbiturainous
coal.
322
-------
MINOR HYDROCARBONS IN RAM GAS
\Xylanes
50 100
Run Time, minutes
SO 100
Run Time, minutes
SULFUR CONFOUNDS III RAW GAS
I -
120".
\
v'
', \ TMophene
*
a 600
9- 400
GA5IFIER OPERATION CONDITIONS
Run Time, minutes
I !
1 I
Figure 4. Gasifier operating conditions and production rate of various com-
pounds as a function of run time - Run 36, North Dakota lignite.
323
-------
TABLE 1. ANALYSIS (AS RECEIVED) OF FUELS GASIFIED
CO
ro
Fuel
Illinois No. 6
Bituminous
Montana
Rosebud
Subb ituminous
Wyoming
Subbituminous
North Dakota
Lignite
Western
Kentucky
No. 9
Bituminous
Sulfate
Volatile Fixed Organic
Moisture Ash Matter Carbon Pyritic
% % % % Total S
0.00
1.83
1.24
5.31 11.03 34.16 49.50 3.07
0.17
0.21
0.21
21.19 8.86 31.56 38.39 0.59
0.07
0.08
0.40
15.56 6.31 38.30 39.83 0.55
0.01
0.54
0.01
29.63 6.39 28.57 35.41 0.56
0.05
2.69
1.70
7.03 7.83 38.78 46.36 4.44
Carbon Hydrogen Oxygen Nitrogen
% % % % FSI
66.35 5.32 12.71 1.525 3.5
53.95 6.87 28.53 1.20 0.0
56.80 5.94 30.02 0.38 0.0
46.82 9.85 35.65 0.73 0.0
67.36 5.58 13.71 1.08 4.0
-------
TABLE 2. SUMMARY OF OPERATING CONDITIONS FOR THE RTI SCREENING TESTS
01
Steam (g)
Air (g)
Coal (g)
Air/Coal
Steam/Coal
Air/Steam
T °C
max*
16
Illinois
No. 6
3704
1350
1569
0.86
2.4
0.35
941
21
Illinois
No. 6
4713
1720
1543
1.1
3.1
0.35
984
23
Illinois
No. 6
1952
3288
1594
2.1
1.2
1.8
1020
41
Western
Kentucky
1390
3060
1250
2.5
1.1
2.2
1034
25
Montana
748
2482
1491
1.7
0.50
3.4
1006
33
Wyoming
500
2097
1396
1.5
0.36
4.2
1010
35
Wyoming
527
2461
1420
1.7
0.37
4.6
790
36
North
Dakota
639
1939
1444
1.3
0.44
3.1
916
43
North
Dakota
422
2022
1458
1.4
0.29
4.8
914
*Time averaged maximum bed temperature.
-------
5. The rates of production of benzene, toluene, and xylenes parallel
each other. In general, benzene has the highest rate of produc-
tion and the xylenes the lowest. Each has a high initial pro-
duction rate. The rate decreases rapidly during devolatilization
by one to two orders of magnitude.
6. The production of H2S and COS is at a maximum during devolatili-
zation and falls off rapidly near the end of this period. After
devolatilization, H/?S and COS appear to follow the production of
C02- This is probably due to two modes of sulfur release from the
coal. The first is during devolatilization when sulfur-containing
compounds are being rapidly evolved from the coal. Decomposition
of these compounds results in COS and H2S. In the second mode
after devolatilization, sulfur is released by oxidation of the
char matrix. Upon release the sulfur species react with H2, CO,
or C02 giving rise to H^S and COS. Thus the production rate of
H2S and COS follows that of C02 since it is indicative of oxidation.
7 - Methanethiol and thiophene are produced primarily during coal
devolatilization. For each compound the production rate starts at
a high initial value and falls below detection limits within 25 to
50 minutes after the coal drop.
The yield of potential environmental pollutants in the gasifier product
gas over the length of the gasification runs has been computed for the RTI
gasifier by integrating the rate of production with respect to time. These
yields have been compared by Green, et al. to yield data reported in the
literature for larger scale, continuous gasifier. An example of this is
shown in Table 3. It can be seen that for a majority of the components
reported that the data from the RTI gasifier appears to bracket the data
from the continuous gasifier even though the continuous gasifiers represent
a range of gasifier operation from fixed- to fluidized-bed. Analysis of
data from semibatch operation is difficult due to the unsteady nature of
operation. Recently RTI has been operating its gasifier in a continuous
feed mode and analysis of this data is now underway.
The initial production rates of methane and minor hydrocarbons during
the devolatilization of the coal as shown in Figures 2, 3, and 4 can be in-
terpreted in several ways. One way is in terms of the Gregory-Littlejohn
equation. For a constant heating rate this equation predicts a straight
line on a semilog graph of rate of production of volatiles versus time.
This equation could perhaps be applied to the individual components making
up the total volatile yield.
326
-------
TABLE 3. POLLUTANT PRODUCTION IN RAW MOISTURE-FREE PRODUCT GAS FROM
GASIFICATION OF NORTH DAKOTA LIGNITE
Pollutant
Hydrogen Sulfide
Carbonyl Sulfide
Thiophene
Methylthiophene
Dimethylthiophene
Methanethiol
Benzene
Toluene
Xylene
Ammonia
Air-Blown
Synthane (Mercer County)
yg/g coal
9.4E3
7.6E2
<3.8E1
<4.4E1
<5.0E1
3.4E1
4.8E3
5.8E2
1.9E2
NA
C0« Acceptor
tVelva)
pg/g coal
2.1E3
9.7E1
NA
NA
NA
NA
NA
NA
NA
5.5E3
GFETC (Velva)
yg/g coal
1.5E3
1.3E2*
NA
NA
NA
8.5E1**
NA
NA
NA
NA
RTI Range
Beulah Zap (Mercer County)
yg/g coal
1.7E3-2.6E3
1.7E2-2.9E2
3.8EO-5.7E2
1.3E1-3.7E1
1.3EO***
1.3E1-7.8E1
2.0E3-5.3E3
1.1E3-2.1E3
2.4E2-7.6E2
5.3E1-1.7E2
co
ro
*Includes CS .
**"thiols."
***C2-thiophenes.
-------
Another way to interpret data of the type shown in Figures 2, 3, and 4
involves the use of a rate expression. The most commonly used kinetic
approach is to assume that the rate of evolution of a volatile species is
proportional to the potential amount of that species remaining in the coal.
dV.
dF - \ (\ - vi ) <«
where k. = the rate constant, min
V. = the yield of the ith volatile component, s£/kg coal.
V = the ultimate yield of the ith volatile component, sA/kg coal.
°°i
t = time, min.
Assuming isothermal conditions, Equation (1) can be integrated subject
to V. = 0 at t = 0 to give
V - V. = V e~kifc (2)
oo 2. °°
Substituting Equation (2) into (1) gives
dV
1 = k,Vm e'V (3)
dt i
i
Taking the log of Equation (3) yields
dV.
In -r-i = In (k.V } - k.t (4)
dt \ i °°. / i
Equation (4) predicts that a semilog plot of the rate of production of
a volatile species versus time should yield a straight line with the slope
equal to the negative of the rate constant and the intercept equal to the
product of the ultimate yield and the rate constant. A substantial number of
product rate-time curves determined in RTI's gasification experiments, can be
interpreted in terms of Equation (4) if the rate constant, k., is viewed as
an average constant over the period of the linear data. This can be done if
the rate constant is not a strong function of temperature such as would be
the case in diffusion-controlled processes.
328
-------
A kinetic analysis has been made of the rate data for nine gasifi-
cation tests using Equation (4). The results of this analysis are shown in
Table 4. This table presents average results for individual species for
an initial rate period for each type coal gasified. The ultimate yield
values shown have been normalized to a unit coal basis.
The following observations can be drawn from Table 4.
1. The average ultimate yield of City for Illinois No.6 coal is
approximately 2.7 scf CH^/lb coal maf which is in good agreement
with a value of 2.4 scf CH^/lb coal maf which would be obtained
by extrapolating the data for the SYNTHANE gasifier to 200 psig.
2. The kinetic parameters for the initial rate period are for the
most part fairly consistent within a given coal type. For
example, for Wyoming coal the rate constants range from 0.149 to
0.173 min"-*-. In the worst case (Illinois No.6 coal), the rate
constants vary by a factor of four which is still in fair agree-
ment considering the assumptions made in the analysis and errors
involved in computing production rates. Wyoming subbituminous
coal appeared on the average to have the highest specific rate of
product formation (i.e., largest rate constants) of any of the
coals tested.
3. The values of the rate constants for the different coals and each
component are close to each other with a simple average constant
being approximately 0.10 min~l.
4. Examination of the average ultimate yields for the various coals
in Table 4 shows that the bituminous coals have the greatest
potential for the production of CH^ and C^Ef, as well as the
sulfur-containing species. The potential for sulfur species
production appears to roughly parallel the sulfur content of the
coal except for COS in the case of Illinois No.6. However, only
one value of the ultimate COS yield could be computed out of the
three Illinois runs, and this may not be representative. Of the
lower ranked coals, the Wyoming subbituminous coal had the highest
potential for CIfy and C6H6 product with ultimate yields of these
components approximately on the same order as the Illinois No.6
bituminous coal.
CONCLUSIONS
Screening tests in which several types of coal were gasified have been
considered in this paper. Major emphasis has been placed on the analysis
of temperature histories in the gasifier bed and transient production rates
of the maj.or gas products, minor hydrocarbons, and selected sulfur-containing
species.
The temperature in the bed was found to be dominated by the gasifier
furnace when the furnace was in operation. The rate of increase in the
average bed temperature in the gasifier after the coal drop appeared to be
329
-------
TABLE 4. AVERAGE KINETIC PARAMETERS FOR THE INITIAL RATE PERIOD
Volatile
Species
CH4
C6H6
H2S
COS
Thiophene
Kinetic Parameters
Illinois No. 6 Coal
V Voo>
min sJl/kg coal
0.080 141.0
0.088 3.05
0.047 11.5
0.036* 0.027*
0.130 0.47
Western Kentucky*
kl' V«>
min s£/kg coal
0.155 243.0
0.095 4.28
0.101 10.2
0.107 0.17
0.192 0.17
Montana*
kl> Vco>
min s£/kg coal
0.103 63.3
0.092 1.32
0.104 0.93
0.062 0.13
0.104 0.015
Wyoming
V V=o»
min s£/kg coal
0.149 121.0
0.165 2.55
0.164 1.80
0.173 0.071
0.149 0.0093
Zap North Dakota
k-., V ,
1' oo'
min s£/kg coal
0.064 67.8
0.108* 0.70*
0.087 1.16
0.057 0.077
0.046 0.0057
oo
GO
o
*Data available for only one gasification test.
k = rate constant for the initial kinetic period.
V = ultimate yield.
-------
be proportional to the difference between the average final temperature and
the instantaneous average bed temperature.
According to the Gregory-Littlejohn equation, the coal bed temperature
should have a significant effect on evolution of total volatile material.
At a constant heating rate the Gregory-Littlejohn equation predicts that a
semilog graph of the devolatilization rate as a function of time should be
linear during the initial stages of the gasification test. This behavior
was observed for the evolution of individual components such as methane,
benzene, minor hydrocarbons, and sulfur species indicating the possibility
of developing a Gregory-Littlejohn type of equation for each volatile species.
A simple kinetic model, which has been widely used in the literature in
one form or another, was applied to rate-time data for selected chemical
components. This model assumes that the rate of formation of a species is
proportional to the potential amount of that species remaining in the coal.
The model involves two parameters: (1) the ultimate yield of the species,
and (2) a proportionality (kinetic rate) constant. It was found that the
kinetic rate constant was roughly the same for all species and all coals
with a simple average of the constants being 0.10 min
The average ultimate yield for each coal for a given species was
dependent on the chemical species and coal type. The ultimate yield of
methane and benzene approximately paralleled the volatile content of the
coal and yield of sulfur-containing components paralleled the sulfur content
of the coal. The potential for the evolution of sulfur-containing compounds
into the gas was found to be an order of magnitude less for the subbituminous
and lignite coal than for the bituminous coals.
ACKNOWLEDGEMENT
Support of this work from the U.S. Environmental Protection Agency,
Fuel Process Branch, Research Triangle Park, North Carolina, under Grant No.
R804979 is gratefully acknowledged.
REFERENCES
1. Cleland, J. G., et al. "Pollutants from Synthetic Fuels Pro-
duction: Facility Construction and Preliminary Tests." EPA-600/
7-78-171, U.S. Environmental Protection Agency, Research Triangle
Park, N. C. (August 1978).
331
-------
2. Cleland, J. G., et al. "Pollutants from Synthetic Fuels Production:
Coal Gasification Screening Test Results." EPA-600/7-79-200, U.S.
Environmental Protecton Agency, Research Triangle Park, N. C.
(August 1979).
3. Gangwal, S. K., et al. "Pollutants from Synthetic Fuels Production:
Sampling and Analysis Methods for Coal Gasification." EPA-600/7-
79-201, U.S. Environmental Protection Agency, Research Triangle
Park, N. C. (August 1979).
4. McMichael, W. J., et al. "Pollutants from Synthetic Fuels Production:
Behavior of a Semibatch Coal Gasification Unit." RTI/1700-08S,
Research Triangle Institute, Research Triangle Park, N.C. (June
1980).
5. Green, D. A., et al. "Pollutants from Synthetic Fuels Production:
Laboratory Simulation of Coal Gasifiers." RTI/1700/00-09S,
Research Triangle Institute, Research Triangle Park, N. C. (August
1980).
6. Gregory, D. R., and R. F. Littlejohn. "A Survey of Numerical Data
on the Thermal Decomposition of Coal." British Coal Utilization
Research Association Monthly Bulletin, 29, No.6, 137 (1965).
7. Anthony, D. B., and J. B. Howard. "Coal Devolatilization and
Hydrogasification." AIChE Journal, 22, 625 (1976).
332
-------
CARBON CONVERSION, MAKE GAS PRODUCTION,
AND FORMATION OF SULFUR GAS SPECIES
IN A PILOT-SCALE FLUIDIZED BED GASIFIER
by
M. J. Purdy, J. K. Ferrell,
R. M. Felder, S. Ganesan, and R. M. Kelly
ABSTRACT
The steam-oxygen gasification of a pretreated Western Kentucky
No. 11 bituminous coal was carried out in a pilot-scale fluidized bed
gasifier. This paper describes the experiments and summarizes meas-
ured carbon conversions, sulfur conversions, make gas production
rates, and the results of material balance calculations on total mass
and major elements (C, H, 0, N, and S). The development of a single
stage kinetic model for the gasifier is outlined, and correlations of
the experimental results using this model are presented. Quantities
of sulfur gas compounds formed in the gasifier at different operating
conditions are summarized and a first analysis of these results is
presented.
333
-------
INTRODUCTION
Since 1976, the Department of Chemical Engineering at North Caro-
lina State University has been engaged in a research project on coal
gasification sponsored by the U. S. Environmental Protection Agency.
The facility used for this research is a small coal gasification-gas
cleaning pilot plant. The overall objective of the project is to
characterize the gaseous and condensed phase emissions from the gasif-
ication-gas cleaning process, and to determine how emission rates of
various pollutants depend on adjustable process parameters. Specific
tasks to be performed are:
1. Identify and measure the gross and trace species concentra-
tions in the gasifier product streams.
2. Correlate measured emission levels with coal composition and
gasifier operating variables.
3. Perform material balances around the gasifier, raw gas clean-
up system, and acid gas removal system, and determine the ex-
tent to which selected species are removed from the synthesis
gas in each subsystem.
4. Correlate measured extents of conversion and removal effici-
encies for various species with system operating variables.
5. Evaluate and compare the performance characteristics of al-
ternative acid gas removal processes.
6. Use results to develop models for the gasification and gas
cleanup processes.
A complete description of the facility and operating procedures
is given by Ferrell et al., Vol I, (1980), and in abbreviated form by
Felder et al. (1980). A schematic diagram of the Gasifier and Parti-
culates, Condensables, and Solubles (PCS) removal system is shown in
Figure 1. The Acid Gas Removal System (AGRS) is an integtral part of
the facility, but will not be discussed here.
In the initial series of runs on the gasifier, a pretreated West-
ern Kentucky No. 11 coal was gasified with steam and oxygen. A com-
puter program was written to reduce the operating and analytical data
for a run to manageable proportions and to perform material balance
calculations. In addition, a single-stage model for the gasifier was
formulated and used to correlate the results of the char gasification
runs. This paper outlines the data processing program, describes the
modeling and model parameter estimation procedures, presents the char
gasification results and comparisions with model predictions, and pre-
sents a preliminary analysis of the formation of sulfur gases in the
gasifier.
334
-------
FIGURE 1
GASIFIER - PCS SYSTEM
co
CO
en
N2 Purge
Coal Feed
Hopper
PIC r Preaaure Indicator
and Controller
S = Sample Port
Char
Receiver
Circulation
Pump /""*
| N2 Purge
N2, O2
Steam
Plant Water
Filter
Mlat
Eliminator
Heat
Exchangers
AGRS
Drain
-------
DATA REDUCTION COMPUTER PROGRAM
A complete description of the data reduction program is given by
Ferrell et al., Vol II, (1980). The program takes as input the reac-
tor temperature profile and pressure, bed dimensions, solid feed pro-
perties (sieve analysis, density, settled bed density, proximate and
ultimate analyses), feed rates of coal, steam, oxygen and nitrogen,
removal rate of char, reactor leak rate, gas flow rate at the PCS sys-
tem outlet, masses of coal fed, spent char collected, cyclone dust
collected, ultimate analyses of the spent char and cyclone dust, chro-
matographic analyses of the gases exiting the cyclone and the PCS sys-
tem, pressure drop across a 20-inch segment of the bed, various feed
and effluent flow meter calibration temperatures and pressures, and
results of trace element and wastewater constituent analyses.
The output of the program contains the following components:
1. Reactor specifications, including the average bed temperature
and pressure, the apparent bed density and void fraction, and
the bed expansion factor.
2. Solid feed properties, including coal type, solid particle
and settled bed densities, as-received moisture content,
average feed particle diameter, and proximate and untimate
analyses.
3. Feed rates of coal, steam, oxygen, and nitrogen, selected
feed ratios and inlet conditions, superficial gas velocity,
solids holdup, and space times for both gases and solids.
4. The make gas flow rate and chemical composition.
5. Production rates of fuel components and the heating value of
the make gas.
6. Carbon, steam, and sulfur conversions.
7. Material balances on total mass, and on carbon, hydrogen, ox-
ygen, nitrogen, and sulfur.
8. An energy balance.
9. Results of water analyses.
10. Results of trace element analyses and trace element material
balances.
An example of the partial output for a run made on January 22,
1980, is shown in Table 1 „
336
-------
Table 1
wwtwwwwttwmwwtmww
i »
I NCSU DEPARTMENT OF CHEMICAL ENGINEERING t
* t
* FLUIDIZED BED COAL GASIFICATION REACTOR t
* *
mmtmtmmmmtmmmmmtm
RUN GO-44B 1/22/80 11215-14:30
REACTOR SPECIFICATIONS
PRESSURE =101.6 PSIG ( 801,7 KPA)
TEMPERATURE = 1699.8 DEG.F ( 926.5 DE6.C)
BED HEIGHT = 38,0 IN, (0,97 METERS)
BED DIAMETER = 6,0 IN, (0,152 METERS)
ESTIMATED BED VOIDAGE = 0,74
SOLIDS HOLDUP = 18,4 LB ( 8.3 KG)
FEED RATES AND RATIOS
34.69 LB/HR (15,74 KG/HR)
55.85 LB/HR (25.33 KG/HR)
= 10.10 LB/HR ( 4,58 KG/HR)
( 2.87 KG/HR)
( 6,42 KG/HR)
STEAM/CARBON = 1,31 MOLES STEAM/MOLE C
02/CARBON = 0.13 MOLES 02/MOLE C
N2/02 = 0.71 MOLES N2/MOLE 02
COAL
STEAM
OXYGEN
NITROGEN = 6.32 LB/HR
PURGE N2 =14.16 LB/HR
ELEMENTAL MATERIAL BALANCES ! FLOWS IN LB/HR
MASS C H 0 N S
COAL
GASES
TOTAL INPUT
CHAR
DUST
GASES
UASTEUATER
TOTAL OUTPUT
34.7
66,4
121,1
21,8
1.8
96,2
0,0
119,8
28,44
0,00
28,44
18,10
1,20
8,99
0,00
28,29
0,16
6,25
6,41
0,08
0,01
6,43
0,00
6,52
1,37
59,70
61,06
0,53
0,23
59,88
0,00
60.64
0.05
20.47
20.52
0.08
0.01
20.43
0.00
20,52
0.918
0.000
0.918
0.412
0.029
0.426
0.000
0.866
2 RECOVERY 98,92 99,53! 101,SI 99.3Z 100,02 94.32
EXPERIMENTAL MODEL
CARBON CONVERSION (PERCENT)
COMBUSTION
GASIFICATION
TOTAL
DRY HAKE GAS FLOW RATE (SCFM)
HEATING VALUE OF SUEET GAS (BTU/SCF)
EFFLUENT FLOU RATES (LB/HR)
CO
H2
CH4
C02
N2
H2S
31.6
11.7
296.0
8.48
0,94
0.66
17.79
20,43
0.434
14.0
18.7
32.7
12,0
286,1
8,67
1,00
0,41
19,33
20,48
0.297
337
-------
GASIFIER MODEL
To aid in the analysis of the char gasification runs, a mathemat-
ical model of the fluidized bed gasifier was developed. The model
takes as input the average reactor bed temperature and pressure, bed
dimensions, feed rates of coal, steam, oxygen, nitrogen, and purge ni-
trogen, solids holdup, ultimate analysis of the feed coke and spent
char, and values of three adjustable model parameters, the relative
reactivity of the coke, the CO/C02 distribution coefficient, and the
water gas shift reactivity parameter.
MODEL DEVELOPMENT AND ASSUMPTIONS
The model treats the gasifier as a single perfect mixer, with the
following six reactions taking place:
C + H20 = CO + H2 (1)
C + 2H2 = CH4 (2)
2C + H2 + H20 = CO + CH4 (3)
CO + H20 = C02 + H2 (4)
C + 1/202 = CO (5)
C + 02 = C02 (6)
Reactions 5 and 6 are the oxidation steps required to supply heat
for the remaining reactions. These two reactions are assumed to occur
instantaneously in a zone of negligible volume separate from the ga-
sification zone. All oxygen in the feed gas is assumed to be consumed
to form CO and C02 according to the relation
C + a02 = (2-2a)CO + (2a-l)C02 (7)
where "a", the combustion product distribution parameter, is an ad-
justable parameter. A value of a =0.5 indicates that all CO is
formed, while a value of a = 1.0 indicates that only C02 is formed.
Reactions 1, 2, and 3 are the reactions by which Johnson (1974)
at the Institute of Gas Technology correlated gasification kinetics.
Reaction 1 is the conventional steam-carbon reaction. Reaction 3 is
assumed to be an independent reaction, although it is attainable as a
linear combination of 1 and 2.
The correlation used by Johnson to describe the carbon conversion
is given by
338
-------
r = fLkT(l-fc)2/3exp(-bfc) (8)
where r is the rate at which the carbon is gasified, kjis the sum of
the rate constants for Reactions 1,2, and 3, f-is the fractional car-
bon conversion and b is a kinetic parameter which depends on gas com-
position and pressure. Expressions for k,, k?, and k_are presented by
Ferrell et al., Vol II. (1980). ' * J
The relative reactivity factor f is determined from
fL = f0 exp(8467/T0) (9)
where T is the maximum temperature to which the char has been exposed
prior to gasification. The relative reactivity factor, f , which is
an adjustable parameter whose values depend on the particular char
used, has values ranging from 0.3 for low-volatile bituminous coal
chars to about 10 for North Dakota lignites (Johnson, 1974).
Reacton 4 is the water gas shift reaction, often assumed to be at
equilibrium in gasification processes. Results to be described indi-
cate this may be a bad assumption, leading to the necessity of incor-
porating shift kinetics into the model. The rate expression used is
that given by Wen and Tseng (1979)
r4 = 1.6652 X 104V(l-e)f exp(-25147/T) P 6 (10)
where
V = bed volume
G
e
f = adjustable shift reactivity parameter
W9 (varies from char to char)
K/i = equilibrium constant
= [CO] - [H2][C02]/[H20][K4]
= bed void fraction
The equilibrium constants for the water gas shift reaction and
for reactions 1, 2, and 3 were taken from Lowry (1963), and were fit
to the equation
Ln (KE) = (a0/T) + 3] (11)
by least-squares analysis (Alexander, 1978).
A complete description of the model development and the reactor
simulation computer program is given by Ferrell et al., Vol II,
(1980).
339
-------
CHAR GASIFICATION RESULTS
A total of 56 runs have been completed using a Western Kentucky
No. 11 coal char as feed stock. The first 13 of these runs were used
primarily for the development of operating and sampling procedures,
and refinement of analytical methods. The data from gasifier runs
GO-14 through GO-56 have been collected and reviewed, and a complete
analysis of these runs is presented by Ferrell et al., Vol II, (1980).
MASS BALANCES
An example of a single page output from the previously described
data processing program is shown as Table 1. Criteria for acceptance
of a run were arbitrarily chosen following inspection of the mass bal-
ance results. A run is judged acceptable if the total mass recovery
is within 5% of 100%, and if the worst of the recoveries of elements
C, H, and 0 are within 8%. Based on these criteria, 22 of the 34 runs
reviewed are acceptable, and are designated by crossed circles in the
figures. Points with filled circles are for runs with total mass re-
coveries within 5% and worst element recoveries within 6%. Open cir-
cles are used for all other runs.
TEMPERATURE EFFECTS
The effect of the average bed temperature on the dry,
nitrogen-free make gas flow rate is shown in Figure 2. For the points
shown, the molar steam to carbon ratio varied from 0.92 to 1.15. The
plot indicates that the make gas flow rate is highly sensitive to the
average bed temperature, with scatter due mainly to the small steam to
carbon ratio differences and differing feed rates. The high sensitiv-
ity makes determination of the average bed temperature crucial for
good model predictions.
STEAM TO CARBON EFFECTS
The effect of the steam to carbon ratio on the make gas flow rate
is shown in Figure 3. At any given temperature the effect of increas-
ing the steam rate at a given carbon input is to increase the make gas
flow rate. A side benefit to operating with relatively high steam to
carbon ratios in the fluidized bed gasifier is a reduced tendency for
the char to clinker.
SULFUR CONVERSION
Measured sulfur conversion, assumed to equal the carbon conver-
sion by the model, is plotted vs carbon conversion in Figure 4. In
most cases the sulfur conversion is greater than the carbon conver-
sion. Studies are currently under way to put the sulfur gas evolution
340
-------
FIGURE 2
THE EFFECT OF THE AVERAGE BED TEMPERATURE
ON THE MAKE GAS FLOW RATE (DRY, N2 FREE)
18
16
0)
0)
14
Molar Steam to Carbon
Ratio of 0.92 to 1.15
o
o
t/)
. 12
Ol
-(->
(0
o:
o
£ 10
o
0)
1C
1680
o
o
o
1720
1760
1800
Average Bed Temperature, °F
1840
1880
341
-------
FIGURE 3
THE EFFECT OF THE STEAM TO CARBON RATIO
ON THE MAKE GAS FLOW RATE (DRY, N FREE)
O)
01
Average Bed Temperature
19 I" of 1869-1882° F
18
17
fe 16
CO
ft
a;
to
* 15 L O
o
(O
is
O
14
O
13
1.00 1.04 1.08 1.12 1.16 1.20 1.24
Molar Steam to Carbon Ratio•
342
-------
80
70
I 60
c
o
50
O
C
D.
40
FIGURE 4
COMPARISON OF PERCENT SULFUR CONVERSION
TO PERCENT CARBON CONVERSION
o
O
O
O
o
30
20
O
o
20
30
40.
50
60
Percent Carbon Conversion
70
343
-------
on a firmer theoretical foundation.
EVALUATION OF MODEL PARAMETERS
In its present form, the model has three adjustable parameters:
1. the char reactivity, f
2. the combustion product distribution parameter, a, which spec-
ifies the split between CO and C02 in the products of the
combustion stage of the gasification
3. the water gas shift reactivity parameter, f
These parameters were evaluated by using a Pattern Search algor-
ithm to minimize a function of the sum of squared deviations between
predicted and measured values of gasifier performance variables. This
analysis gave the following values:
1. f0= 0.50
2. a = 0.95
3. fw= 0.0000099
The value of a, when substituted into Eg. 7, indicates that 90%
of the carbon oxidized forms C02 and 10% forms CO. An equation by Ar-
thur (1951) predicts values of 0.57 at 1400 F to 0.52 at 2000 F, while
several gasification studies have assumed a = 1.0.
Johnson (1975) developed a correlation for char reactivity
^ = 6.2 y (1-y) (12)
where y is the dry, ash free carbon fraction in the original raw coal.
Eq. 12 predicts a value of f = 1.1, which is larger than that deter-
mined in this study. The difference may be due to the differences in
the microbalance used by Johnson and the fluidized bed of this study.
The value of £,„= 0.0000099 indicates that the shift reaction rate
is approximately five orders of magnitude less than the rate obtained
in catalytic shift reactors. Wen and Tseng (1979) used a shift reac-
tivity value of 0.00017 in modeling the gasification of a bituminous
coal by the SYNTHANE process. The larger value used by Wen and Tseng
may be attributed to the differences between the coal of their study
and the char used in this study.
Due to the simplicity of the model, it is also likely that the
effects of factors not specifically accounted for in the model have
influenced the optimal values of the three model parameters. The va-
344
-------
lues of the parameters found as described above appear to be reason-
able, and are probably a fair representation of what actually happens
in the fluidized bed gasifier.
MODEL RESULTS
Using the optimal parameter values, the model was run for gasif-
ier runs GO-14 through GO-56. A representative model output is shown
for run GO-44B in Table 2. Plots of predicted vs measured values of
carbon conversion, dry make gas flow rate, and sweet gas heating value
are shown in Figures 5-7. The reasonably close proximity of most
points to the 45 degree line is gratifying in view of the crudeness of
the model. The proximity of the points corresponding to the "best"
runs (from the standpoint of satisfying mass balances) is even more
satisfying.
For each run, the ratio
K = tC02]lH2J/lCO]£H2Oj (13)
was calculated, where [ ] is the mole fraction of the evaluated spe-
cies in the product gas. This quantity would equal the water-gas
shift equilibrium constant at the reactor temperature if this reaction
proceeded to equilibrium, A plot of the predicted vs experimental va-
lues of this ratio, K, is given in Figure 8. The substantial degree
of scatter may be attributed to the simplicity of the model, and
equally to the fact that the mole fractions which are the constituents
of K are interdependent, so that an experimental error in one of them
affects the values of the others.
The significance of this plot emerges when it is compared with
Figure 9, which shows the values of K predicted assuming shift equili-
brium. This assumption leads to the overprediction of K by as much as
a factor of two, and lends support to the conclusion that the shift
reaction should not be assumed to proceed to equilibrium.
FORMATION OF SULFUR GASES
One of the objectives of gasifier runs GO-43 through GO-59 was to
investigate the production of sulfur gas species in the fluidized bed
reactor. A summary of results is given in Table 3.
The coal char used in this study has a very low volatile matter,
less than 2%, and it is very likely that most of the sulfur is present
as pyritic sulfur. For this case, it has been postulated that during
345
-------
Table 2
mttmmmmmmmmm
* t
* WELL-HIXED CHAR GASIFICATION *
t *
* HODEL RESULTS I
* t
mwmmmmmmmmtt
60-448 1-22-80 11S15-14J30
REACTOR SPECIFICATIONS
BED PRESSURE(PSIG) 101.60
BED TEMPERATURE(F) 1699.80
SOLIDS HOLDUP(LB) 18.40
BED HEIGHT(IN) 38.00
BED DIAMETERUN) 6.00
BED VOIDAGE 0.74
FEEDRATES(LB/HR)
INLET CHAR 34.69
STEAM 55.85
OXYGEN 10.10
NITROGEN 6.32
HYDROGEN 0.00
PURGE N2 14.16
HODEL PARAMETERS
PRETREAT TEMP(F)
CHAR REACTIVITY
COMBUSTION EXTENT
SHIFT REACTIVITY
FEED CHAR ANALYSIS^ PERCENT)
2000.00
0.5000
0,9500
9.900E-06
CARBON
HYDROGEN
OXYGEN
NITROGEN
SULFUR
ASH
82.00
0.50
3.90
0.10
2.60
10.80
DRY GAS FLOU RATE (SCFM)
STEAM CONVERSION
CARBON CONVERSION
COMBUSTION
GASIFICATION
TOTAL
ASH CONTENT OF CHAR
CHAR REMOVAL RATE (LB/HR)
HODEL EXPERIMENTAL
12.04 11.73
0.171 0.153
0.140
0.187
0.327
15.24
23.07
0.316
12.00
21.80
GAS COMPOSITION (MOLE PERCENT)
MODEL EXPERIMENTAL
CO
H2
CH4
C02
N2
H2S
COS
H20
6.76
10.85
0.56
9,59
15.96
0.19 *
0.00
56.08
6.60
10.11
0.89
8.82
15,91
0.28
0.01
57,38
(* ESTIMATED)
346
-------
FIGURE 5
PREDICTED VS. EXPERIMENTAL CARBON CONVERSION
70
% Carbon Conversion
60
50
c
o
u
O)
Q.
r- 40
CD
TJ
O
30
20-
20
30 40
50
Experimental
60
70
347
-------
FIGURE 6
PREDICTED VS. EXPERIMENTAL DRY MAKE GAS FLOW RATE
22
Dry Make Gas Flow Rate (SCFM)
20
o
18
O
c
o
O
•5
01
O.
OJ
-o
O
16
O
14
12
10
10
12
14
Experimental
16
18
20
348
-------
FIGURE 7
PREDICTED VS. EXPERIMENTAL HEATING VALUE OF SWEET GAS
360
Heating value of Sweet Gas (Btu/SCF)
320
280
c
o
U
0)
v
•o
I
240
200
120
349
-------
FIGURE 8
PREDICTED VS. EXPERIMENTAL K VALUE
0.8
C0
K =
0.7
0.6
o
o
OJ
Q.
O)
•O
O
0.5
0.4
s>
0.3
Y Y
YCOYH20
0.2
0.2
O
0.3
0.4
0.5
O
o
o
o
0.6
O
0.7
Experimental
350
-------
FIGURE 9
PREDICTED VS. EXPERIMENTAL K VALUE
ASSUMING SHIFT EQUILIBRIUM
0.8
K =
0.7
0.6
-------
TABLE 3
CONCENTRATIONS OF SULFUR GASES IN REACTOR EFFLUENT
Run
No.
43
44
45
46
47
48
49
51
55
56
57
58
59
Bed
Temp.
F
1794
1678
1671
1790
1785
1778
1799
1777
1708
1800
1778
1771
1803
Reactor Effluent Concentrations ppm
H2S
6229
6510
3433
5478
5071
6912
7052
6711
8931
8924
8098
5111
8470
COS
211
283
266
222
272
312
403
299
465
410
388
362
306
CS2
2.27
2.44
7.92
1.56
1.97
3.30
3.80
1.56
2.95
1.58
1.58
1.36
1.61
Methyl
Mercap-
tan
X
X
X
X
X
X
X
X
X
X
X
X
X
Thiophene
X
N.D.
X
X
X
X
X
X
N.D.
X
X
X
X
x - Less then 1 ppm
N.D. - Not detected
TABLE 3 CONTINUED
EQUILIBRIUM CONSTANTS FROM EXPERIMENTAL DATA
Run
No.
43
44
45
46
47
48
49
51
55
56
57
58
59
Reactor Effluent Concentrations
CO
16.80
6.60
4.22
12.77
13.89
12.88
15.42
10.98
9.28
10.73
12.74
11.68
10.10
co2
12.36
8.82
6.81
9.08
9.86
10.79
16.03
15.11
12.61
12.75
14.87
15.80
14.60
H2
11.21
10.11
8.27
13.82
15.16
15.77
13.68
19.06
15.24
16.98
18.05
19.84
17.15
H20
43.60
57.38
39.27
33.14
29.89
46.62
36.25
41.49
50.05
48.95
41.92
38.38
47.68
Equilibrium
Constants
Kl
6.4
3.5
2.2
6.8
6.2
5.1
7.7
8.2
4.8
5.7
7.4
5.8
8.5
K2
33.7
15.0
6.6
22.8
17.1
18.1
19.7
12.9
11.7
13.8
14.7
8.3
16.3
352
-------
steam-oxygen gasification the gas-solid reactions form mainly hydrogen
sulfide. The gas phase reactions then tend to bring the compounds
C02, H20, H2, H2S, and COS to an equilibrium mixture.
The two gas phase reactions of most importance involving H2S and
COS are:
COS + H20 = H2S + C02 (14)
COS + H2 = H2S + CO (15)
The equilibrium constants for these two reactions are defined as
follows:
K! = [H2SJ[C02J/£COSJ[H20] (16)
K2 = [H2S][COJ/[COSJ[H2J (17)
where, due to the stoichiometry of the reactions, the brackets may in-
dicate any convenient concentration. Ideal gas behavior is assumed.
A survey of the literature yielded several sets of equilibrium
data for the above reactions, and several predictions based on thermo-
dynamic data. Since there were substantial differences amoung the
sources of data, predictions of the two equilibrium constants as fun-
tions of temperature were derived from the data given in Reid et al.
(1977). A least squares fit of the literature data, and the predicted
curve from the data of Reid are shown in Figures 10 and 11.
Also shown on Figures 10 and 11 are calculated values of the
equilibrium constants from the data in Table 3. Figures 12 and 13
show the experimental data on a expanded scale and a comparison of our
data with the literature values given in Kohl and Riesenfeld (1979).
Although there is considerable uncertainty in determining the
correct value of the equilibrium constants, and some inaccuracy in the
experimental data, it appears that the sulfur compounds H2S and COS
are in equilibrium with the major gases at the exit of the fluidized
bed, and that the distribution of the sulfur gases between H2S and COS
can be predicted if the sulfur conversion is known.
353
-------
FIGURE 10
THE EQUILIBRIUM CONSTANT KI
CO
en
50
40
30
20
10
1100
1300
Calculated from Thermodynamic Data
From Volkov and Ruzagkin
From Gattow and Bohrendt
From Terres and Wesemann
Experimental NCSU
1500
Temperature °F
1700
1900
-------
CO
en
en
30
CM
20
10
FIGURE 11
THE EQUILIBRIUM CONSTANT 10
50
40
1 Calculated from Thermcdynamic Data
2 From Gibson
3 From Kohl and Riesenfeld
• Experimental NCSU
1100
1300
1500
Temperature °F
1700
1900
-------
FIGURE 12
COMPARISON OF EXPERIMENTAL VALUES OF K-, WITH DATA OF KOHL AND RIESENFELD
10
CO .
en I
o-i
8
FROM KOHL AND RIESENFELD
O
O
O
•^x
0
O
O
0
O
0
0
0
_ ©
_ (H2SXC02)
1 " (COSXH20)
0
1660
1700
1740
1780
1820
Temperature at Top of Bed °F
-------
50
FIGURE 13
COMPARISON OF EXPERIMENTAL VALUES OF K0 WITH DATA OF KOHL AND RIESENFELD
40
_ (H2SXCO)
2 ~~ (COSXH2)
30
0
GO
en
K
20"
10-
0
FROM KOHL AND RIESENFELD
©
00
o
0
"0"
0
O
oL
0
1660
1700
1740
1780
1820
Temperature at Top of Bed
-------
REFERENCES
1. Alexander, D. W., Ph. D. Thesis, Department of Chemical
Engineering, N. C. State University, Raleigh, N. C.,
(1978).
2. Arthur, J. R., "Reactions Between Carbon and Hydrogen",
Trans. Faraday Soc0, 47, 164 (1951).
3. Felder, R. M., R. M. Kelly, J. K. Ferrell, and R. W.
Rousseau, "How Clean Gas is Made from Coal", Env. Science
and Tech., Vol 14, 658, (1980).
4. Ferrell, J. K., R. M. Felder, R. W. Rousseau, J. C.
McCue, R. M. Kelly, and W. E. Willis, "Coal
Gasification/Gas Cleanup Test Facility: Vol I. Description
and Operation", EPA-600/7-80-046a, (1980).
5. Ferrell, J. K., M. J. Purdy, R. M. Felder, and J. C.
McCue, "Coal Gasification/Gas Cleanup Test Facility: Vol II.
Data Processing and Reactor Modeling Programs, and Char Ga-
sification Studies", EPA, (1980).
6. Johnson, J. L., "Kinetics of Bituminous Coal Char Gasifica-
tion with Gases Containing Steam and Hydrogen", Advances in
Chemistry Series, No. 131, (1974).
7. Johnson, J. L., "Relationship Between the Gasification Reac-
tivities of Coal Char and the Physical and Chemical Proper-
ties of Coal and Coal Char", Presented at American Chemical
Society, Division of Fuel Chemistry Coal Gasification Sympo-
sium, Chicago, (1975).
8. Kohl, A. L. and F. C. Riesenfeld, "Gas Purification", 3rd
Ed., Gulf Publishing Co., (1979).
9. Lowry, H. H., ed., "Chemistry of Coal Utilization", John
Wiley and Sons, Inc., New York, (1963).
10. Reid, R. C., J. M. Prausnitz, and T.K. Sherwood, "The
Properties of Gases and Liquids", 3rd Ed., McGraw-Hill Book
Co., (1977).
11. Terres, T., and H. Wesemann, Angewandte Chemie, 45, 795,
(1932)o
12. Volkov, V. P., and G. I. Ruzaykin, "Petrogenetic Problems
in Relation to Methods of Calculating Gas Equilibria", Geo-
chemistry International, 6, 773, (1969).
13. Wen, C. P., and H. P. Tseng, "A Model for Fluidized Bed
Coal Gasification Simulation", Presented at 72nd Annual AIChE
Meeting, San Francisco, (1979).
358
-------
MODDERFONTEIN
KOPPERS-TOTZEK
SOURCE TEST RESULTS
J. F. Clausen
C. A. Zee
TRW Systems and Energy
One Space Park
Redondo Beach, CA 90278
ABSTRACT
A source test program was conducted at a Koppers-Totzek (K-T) coal
gasification facility operated by AECI Limited at Modderfontein, Republic
of South Africa. The EPA's interest in the K-T process stems from the
fact that the process economics and demonstrated commercial reliability
make it a very viable prospect for some U.S. applications. The responsi-
bilities for sampling, analysis, and engineering descriptions of the
Modderfontein plant were shared between TRW and GKT, Gessellschaft fur
Kohle-Technplogie mbH of Essen, Federal Republic of Germany. GKT is the
wholly owned subsidiary of the German-based parent company which is the
developer and licensor of the K-T process. EPA's phased approach for en-
vironmental assessment was followed. Level 1 and Level 2 data were col-
lected along with priority pollutant screening data. Much of the effort
was focused on wastewater streams. The wastewater treatment, consisting
of a clarifier and settling pond, was adequate to produce a final discharge
that had lower pollutant levels than the fresh input waters supplied to
the plant. The complete data are presented in this paper along with brief
descriptions of the K-T process and the Modderfontein plant. The purpose
of the Source Test and Evaluation was intended as an initial effort and
was somewhat limited in scope.
359
-------
MODDERFONTEIN KOPPERS-TOTZEK
SOURCE TEST RESULTS
INTRODUCTION AND SUMMARY
TRW, under contract 68-02-2635 to the Environmental Protection Agency,
and at the direction of Project Officer William J. Rhodes, is performing
the environmental assessment of high-BTU gasification and indirect lique-
faction technologies. A major portion of this environmental assessment
project is to obtain data on commercial operating facilities through
Source Test and Evaluation (STE) programs. The ultimate objective of each
STE program is to obtain the data necessary to: 1) evaluate environmental
and health effects of waste streams or streams that may potentially be dis-
charged in plants designed for U.S. sites, and 2) allow subsequent evalua-
tion of the equipment available or required for controlling these streams.
This paper describes an STE program that was conducted on a Koppers-
Totzek (K-T) coal gasifier plant operated by AECI Limited in Modderfontein,
Republic of South Africa. The EPA's interest in the K-T process stems
from two principal factors: first, in the national drive to supplement
liquid and gaseous fossil fuels through coal conversion, process economics
dictate that the more viable conversion products will be those having the
highest unit retail value. The K-T process represents one of the prime
candidates for converting raw coal into the intermediate synthesis gas
needed to produce these high-value products. Secondly, the K-T process
has a lengthy history of successful application to a variety of foreign
coals and promises to be equally adaptable over the range of American
coals. This factor is particularly important in view of the contrasting
lack of demonstrated commercial reliability on the part of the develop-
mental U.S. gasifiers, and is viewed in a very positive light by both
conversion project financiers and program managers.
The K-T process operates on an entrained bed principle. It utilizes
a high temperature, atmospheric pressure reaction fueled by a continuous
co-current input stream of coal, oxygen, and steam. The gasification
360
-------
reactor vessel is a horizontal, ellipsoidal, double-walled steel chamber
with a refractory lining. Two gasifier designs are available. The two-
burner gasifier design utilized at Modderfontein has a burner head located
on each end of the ellipsoid as illustrated in Figure 1. The four-burner
gasifier resembles two of the two-burner gasifiers which intersect one
another at a 90° angle. A burner head is located at each of the ends of
the two intersecting ellipsoids. The gasifier operates with a flame
temperature of 2000 C (3650°F) or more and a gas outlet temperature of
about 1400° to 1600°C (2550° to 2900°F). The major constituents of the
gasifier output stream are carbon monoxide and hydrogen.
All of the K-T gasification facilities in operation as of 1978 were
used entirely to make synthesis gas as an input stream for the production
of ammonia. The Modderfontein plant, illustrated in Figure 2, was com-
missioned in 1974 and has a design production rate of 1000 tonnes per day
of ammonia. It utilizes a High Volatile B, Bituminous coal that is high
in ash content (20%) and low in sulfur (1.0%).
The STE program was carried out as a joint effort between TRW and
GKT. TRW's initial review of the Modderfontein plant resulted in the
identification of 25 streams as necessary to the comprehensive STE goals.
Of these 25 streams, nine were selected for testing as a result of discus-
sions between GKT and TRW in which streams considered proprietary, not
applicable, or otherwise restricted were eliminated from the list. The
STE thus became limited in scope and focused on the nine available streams.
Further STE programs are anticipated in the future which will serve to
provide basic characterization data on K-T generated wastes so that control
technology requirements for facilities built in the U.S. can be identified
early in the planning stages. It is not intended that any data presented
in this paper of future data resulting from tests at Modderfontein be used
for the purpose of either promoting or criticizing specific process designs
or operating practices of that facility. It should be stressed that each
K-T plant is unique and that numerous design options exist for pollutant
reduction within the process depending upon customer requirements.
361
-------
RAW PRODUCT
GAS TO WASTE
HEAT BOILER
LOW
PRESSURE
STEAM
COAL
STEAM
OXYGEN
BURNER
COOLING
WATER BOILER
FEED
WATER
— R
v, 1
BOILER FEED WATER
LOW PRESSURE STEAM
FRESH
INPUT WATER
WASTEWATER
FROM SLAG
QUENCHING
QUENCHED SLAG
CONVEYOR REMOVAL
FROM PLANT
Figure 1. Koppers-Totzek Gasifier
362
-------
co
en
CO
1 RAW COAL
2 DRY. MILLED COAL
3 COAL FINES
4 RECYCLED COAL
CONVEYING GAS
B WASTE GAS
6 PURGE GAS
7 COAL DUST
S LOW PRESSURE STEAM
FROM GASIFIER WATER
JACKET
9 HOT SLAG
10 QUENCHED GASIFIER
SLAG
11 SLAG QUENCHING
WASTE WATER
12 POKE HOLE GAS
13 RAW GAS
14 STEAM CONDENSATE/
RECYCLED BOILER
FIED WATER
IS RAW GAS AFTER
BLOWER
18 INPUT WATER (CW)
17 COMPRESSED RAW GAS
IIHCN FREE RAW GAS
IS SULFUR FREE PRODUCT
GAS
20 COMPRESSED SULFUR
FREE GAS
21 SHIFTED PRODUCT GAS 27 PURGE WATER
22 CO2 FREE PRODUCT QAS 28 NITROGEN WASH TAIL GAS
23 NH3 SYNTHESIS FEED GAS 29 METHANOL
24 NHj SYNTHESIS FEED 30 RECYCLE METHANOL
(COMPRESSED) 31 C02 RICH METHANOL
26 RECYCLE GAS 32 DILUTED RECTISOL CONDENSATE
2S SPENT CATALYST 33 TAIL GAS
34 COj RICH ACID GAS
SBH-S RICH ACID GAS
M HjS RICH METHANOL
37 H,* RICH METHANOL
38 TAIL GAS
39 HCN WASH CONDENSATE
40 COMPRESSORS CONDENSATE
41 ELECTROSTATIC
PRECIPITATOR
WASH WATER
42 WATER SEAL WASTE WATER
43 DISINTEGRATOR WASH
WATER
44 WASHER COOLER SLOWDOWN
4E CLARIFIER FEED
«INPUT WATER (PSE)
47CLARIFIER EFFLUENT
4B RECYCLED WASH/WATER
49 SETTLED CLARIFIER SOLIDS
SO SETTLING POND DISCHARGE
WATER
51 SETTLING POND SLUDGE
SI BOTTOM ASH SLURRY
S3 FLUE GAS TO DRY COAL
Figure 2. Diagram of Number 4 Ammonia Plant at Modderfontein
-------
APPROACH
The nine streams included in this STE along with their stream
numbers which correspond to Figure 2, are as follows:
• Solids
• Coal Dust Feed/7
• Gas Streams
• Raw Product Gas/15
• Tail Gas from H2S Absorber/38
t Tail Gas from C02 Absorber/33
t Aqueous Streams
• Input Water (Purified Sewage Effluent)/46
• Input Water (Cooling Water)/16
• Settling Pond Effluent/50
• Compressor Condensate Wastewater/40
• Diluted Rectisol Condensate Wastewater/32
The basic approach was to perform a comprehensive organic and inor-
ganic characterization of these nine streams per the EPA procedures for
Level 1 and Level 2 environmental assessments and for Priority Pollutants
(1, 2, 3). The Level 1 methods provide a broad semi-quantitative survey
from which constituents found to be present at levels of potential
concern are selected for further quantitative examination, Level 2. The
Priority Pollutant screening consists of analyses for a specific list of
129 pollutants of concern to the EPA.
The sampling and analysis responsibilities for the K-T facility test
were divided between TRW and GKT. GKT performed all of the sampling and
most of the on-site analyses during a three week period in November 1979.
TRW arranged to have the remaining time-critical analyses performed by a
local South African laboratory (McLachlan & Lazar pty LTD) and to have
portions of the coal feed and aqueous process stream samples shipped
back to TRW for analysis.
Level 1 Analysis
Most of the Level 1 analyses that are time critical were performed by
GKT (i.e., all gas analyses and most wastewater quality tests). The only
wastewater quality tests remaining were nitrates and BOD, which were then
handled by McLachlan & Lazar in Johannesburg. Replicate analysis of a
364
-------
few of the species measured by GKT were also performed by the local lab
for quality assurance. The methods used by GKT and the commercial lab-
oratory were for the most part comparable to U.S. methods and were accep-
table for source evaluations. The analysis of organic materials and trace
metals was performed by TRW on preserved aliquots of the aqueous stream
samples that were shipped back to the U.S. The methods used for the Level 1
analyses were taken from the EPA-IERL/RTP procedures manual (1).
Level 2 Analysis
Level 2 analyses of the aqueous Modderfontein samples consisted of
atomic absorption techniques (AAS) for Fe and Mn, and a high performance
liquid chromatography (HPLC) technique for polynuclear organic material
(POM) compounds. These two metals and the POM compounds were selected
on the basis of comparing the Level 1 data to the EPA's discharge multi-
media environmental goal values (4), thus determining the potentially
hazardous species present which warranted further investigation, and by
examining which Level 2 data requirements had not already been met by
either the wastewater quality or priority pollutant analyses.
The AAS techniques were standard methods (5). The HPLC technique
for POMs utilized a reverse phase, quarternary solvent system for separation
of three-ring and larger POM compounds. Both UV and fluorescence detec-
tors were used in tandem in order to yield corroborative data for the
identification and quantisation of the compounds present. Further qual-
itative data for POM identification was obtained by collecting the HPLC
fractions and analyzing them by GC/MS.
Priority Pollutant Screening Analysis
The analyses for organic priority pollutants were done in three phases.
Volatile, acid extractable non-volatile and base-neutral extractable non-
volatile organics were tested in accordance with the EPA procedures
manual (3). The samples were analyzed by computerized gas chromatography-
mass spectrometry (GC/MS) using an INCOS data system. A computer program
was used to reduce the data. The results were manually examined and if
necessary, modified. The thirteen priority pollutant metals (i.e.,
Ag, As, Be, Cd, Cr, Cu, Hg, Pb, Mn, Sb, Se, Tl, and Zn) were analyzed by
a combination of flame and flameless atomic absorption techniques in
accordance with the EPA protocol (3).
365
-------
Source Analysis Model
All of the data obtained from this STE were used in the EPA's Source
Analysis Model/IA, which compares the measured concentrations of the con-
stituents analyzed to the EPA's Discharge Multimedia Environmental Goals
(6). This model calculates discharge severities based on the constituent
concentrations alone (total discharge severity) and on the concentrations
combined with the stream flow rate (weighted discharge severity). This
approach is being used uniformly by all of the EPA's contractors in the
coal conversion area and thus provides a consistent basis for evaluating
STE data.
RESULTS
Coal Feed Stream
The results of the proximate and ultimate analysis on the coal feed,
shown in Table 1, show that the sample may be characterized as Bituminous,
High Volatile B coal. When compared to must U.S. coals it is found to be
very high in ash content and low in sulfur. A trace element survey, more
precise determinations of the major minerals present and other measure-
ments were also performed. This data will be included in the Source Test
and Evaluation Report currently in preparation for the EPA.
Gas Streams
All gas analyses were performed by GKT and the data obtained are
shown in Table 2. The raw gas results reflect the average composition
from all five gasifiers (the stream was sampled at a common line leading
to the gas holder) after the gas has been water-washed for particulate
removal. A description of the major reactions that take place in the raw
gas washing stages is as follows:
• NH , HCN, S02, and to a small degree H2S and CQ^, are dissolved
in the wash water.
• H~S is eventually converted to S,,03~, S04~, and insoluble metal
sulfides due to the pH, temperature, and flyash content of the water.
• HCN reacts with the sulfur compounds to form SCN~ and with the
iron content of the flyash to form insoluble complexes.
• Additional oxidation reactions occur which are catalyzed by the
flyash involving NH3, S03=, S203=, CM", and SCN".
366
-------
TABLE 1
PROXIMATE AND ULTIMATE RESULTS
FROM COAL ANALYSIS
Proximate
% Moisture
% Ash
% Volatile
Analysis
As Received
1.49
19.60
27.52
% Fixed Carbon 51.39
Btu/lb.
% Sulfur
100.00
10853
0.99
Ultimate Analysis
% Moisture
% Carbon
% Hydrogen
% Nitrogen
% Chlorine
% Sulfur
% Ash
% Oxygen (di
As Received
1.49
64.41
3.72
1.12
0.01
0.99
19.60
ff) 8.66
100.00
367
-------
Table 2. GAS ANALYSIS DATA
Parameter/ Units
Flowrate, Mm /hr
H20 g/Nm3
H2 Vol.* (dry)
CO
co2
N2/Ar*
CH4 «
H2S mg/Nm (dry)
COS
cs2
so2
NH3
HCN
NOX
Mercaptans "
Raw Gas
103,600
54
28.2
59.1
10.9
1.8
<0.1
6,300
740
450
14
57
76
28
<1
Tail Gas from
H2S Rewash
Column
13,700
5
<0.1
1.9
52.6
45.5
<0.1
t
t
t
<3
39
62
<1
<1
Tail Gas from
C02 Stripper
48,800
5
<0.1
0.3
84.3
15.4
<0.1
<1
<3
<10
<3
3
8
<1
<1
* By difference
t Not determined
368
-------
The main components in the water-washed gas are then H20, CO, C02, H2,
and N2. Data on hydrocarbons contained in the raw gas stream were not
obtained due to problems with on-site analytical instrumentation, but low
concentrations would be expected due to the high temperature of the K-T
gasification reaction.
The two tail gas streams from the Rectisol module consist primarily
of C02, the nitrogen used for methanol stripping, small amounts of CO and
H20 and traces of NH3 and HCN. During the test period, plant operating
data indicated that temperature control in the Rectisol unit was not
working properly with the result that sulfur species levels in the H2$
stripper tail gas were outside design specifications and were not typical
of normal Rectisol unit operation. Therefore sulfur species data on this
tail gas stream are not included in Table 2. A design value of less than
2 ppm total sulfur is quoted by GKT.
Use of the SAM/IA model, which assesses the potential health and
ecological effects of discharge streams based on chemical constituents,
yielded the calculated Total Discharge Severity (TDS) and Weighted Dis-
charge Severity (WDS) values shown in Table 3. In the tail gas stream
from the H2S absorber, CO, HCN, and NH3 are present at levels of potential
concern; and in the tail gas from the C02 absorber, CO and NH3 are of
concern.
Table 3. SUMMARY OF SAM/IA TDS AND WDS RESULTS FOR GAS STREAMS
TDS and WDS Values
Total Discharge Severity (TDS)
Health-Based
Ecology-Based
Weighted Discharge Severity (WDS)
Health-Based
Ecology-Based
Tail Gas from
H2S Rewash
5.6 E + 02
2.9 E + 02
2.1 E + 03
1.1 E + 03
Tail Gas from
C02 Stripper
7.6 E + 01
3.4 E + 01
1.0 E + 03
4.6 E + 02
Aqueous Streams
The results of the Level 1 standard wastewater analyses performed
jointly by GKT, TRW and McLachlan & Lazar are summarized in Table 4. The
369
-------
Table 4. WASTEWATER QUALITY TEST DATA
Parameter/Units
0
Flowrate, mJ/hr
pH
TSS, mg/L
TDS,mg/L
Hardness, mg/L
Conducti vity,pmhos/cm
BOD, mg/L
COD, mg/L
TOC, mg/L
NH3, mg/L
CN~, mg/L
SCN", mg/L
H2S, mg/L
S2°3~» "ig/L
S03=, mg/L
S04=, mg/L
Input
Water
(PSE)
215
6.8
<1
1,580
450
2,300
5
16
31
73
0.2
2.1
<1
<1
<1
580
Input
Water
(CW)
34
8.5
8
1,460
620
1,900
4
24
16
3
1.2
2.1
<1
<1
<1
850
Settling
Pond
Effluent
230
8.7
<1
1,560
540
2,100
4
4
5
33
0.2
1.8
<1
<1
<1
730
Process Waters
Compressor
Condensate
9.1
8.1
6
220
53
5,800
550
600
140
940
8.9
14
49
6.3
<1
53
Rectisol
Condensate
3.9
8.6
45
1,520
620
1,900
800
1,600
590
38
2.8
120
2.8
17
<1
500
370
-------
settling pond effluent, the only aqueous stream discharged by the plant,
appears from the data to be quite similar to the input waters. This would
seem to indicate that any aqueous pollutants contributed by the gasifica-
tion process are esentially removed in the settling pond.
The results of the Level 1 survey for organics, shown in Table 5,
Table 5. LEVEL 1 ORGANIC SURVEY DATA
Stream/ Flowrate
Input Water (PSE)/215 m3/hr
Input Water (CW)/34 m3/hr
Settling Pond Effluent/230 m3/hr
Process Streams
Compressor Condensate/9.1 m3/hr
Rectisol Condensate/3.9 m3/hr
Volatiles
(mg/L)
0.04
<0.01
0.05
0.01
0.49
Non-
Volatiles
(mg/L)
0.68
0.88
0.06
3.83
33.4
Total
Organics
Ong/L)
0.72
0.88
0.10
3.84
33.9
indicate that the total organic loading was low and that the material
present was primarily nonvolatile (BP >100°C). Examination of the non-
volatile material by infrared (IR) spectroscopy indicated that the
classes of compounds present in all of the samples are primarily saturated
hydrocarbons along with some esters. There was also some IR evidence of
low levels of aromatic hydrocarbons present in the compressor condensate
and Rectisol unit samples. Examination of the nonvolatile samples by solids
probe low resolution mass spectroscopy (LRMS) yielded additional infor-
mation regarding the classes of compounds present. The intensity of the
mass spectra peaks were used to assign relative concentration factors
(100 = major, 10 = minor, 1 = trace) to the compound classes identified.
The LRMS results are summarized in Table 6. The mass spectra data confirm
the IR data indicating the presence of aliphatic hydrocarbons, esters,
and traces of aromatics. Traces of phenols, cresols, and alcohols also
appear in many of the samples. Significant levels of elemental sulfur
(Sg) are also seen because of its appreciable solubility in the solvent
used for these extractions (methylene chloride).
371
-------
Table 6. ORGANIC COMPOUND CLASS DATA
Stream
Compound Class
Contribution to
Total Organics
Input Water (PSE)
Input Water (CW)
Settling Pond Effluent
Process Streams
Compressor Condensate
Rectisol Condensate
Esters (phthalates)
Nitro Aromatic Hydrocarbons
Primary Alcohols
Esters (phthalates)
Primary Alcohols
Secondary Alcohols
Aliphatic Hydrocarbons
Esters (phthalates)
Unsaturated Alkyl Hal ides
Ketones
Sulfur (S8)
Ethers
Esters (phthalates)
Phenols
Chlorinated Phenols
Chlorinated Cresols
Polynuclear Organic Materials
(POMs)
Carboxylic Acids
Aliphatic Hydrocarbons
Sulfur (S8)
Polynuclear Organic Materials
(POMs)
Phenols
Esters (phthalates)
Major
Minor
Minor
Major
Major
Major
Minor
Minor
Trace
Trace
Major
Major
Minor
Trace
Trace
Trace
Trace
Trace
Major
Minor
Trace
Trace
Trace
372
-------
The Level 1 inorganic survey of the aqueous samples consisted of a
spark source mass spectroscopy (SSMS) analysis. The data
indicated that, based upon elemental composition, the settling pond effluent
is quite similar to the input waters. Similarity between these streams
based upon standard wastewater parameters was previously noted. The only
trace elements that show an increase from input water levels to settling
pond effluent levels are cesium, strontium, barium, gallium, and molybdenum.
This is in general agreement with the trace element analysis of the coal.
Other elements (i.e., aluminum, iron, and manganese) actually show a
significant decrease in the settling pond effluent compared to the input
water.
As is mentioned in the analytical approach, the Level 1 data were
compared to the EPA's Discharge Multimedia Environmental Goals (DMEGs)
using the SAM/IA model in order to determine which species were present
at levels of potential concern and were thus candidates for further
investigation. Those species determined to be of interest were then com-
pared to the priority pollutant list. It was found that most of the
Level 2 data requirements would be satisfied by the priority pollutant
analyses and that the only additional determinations needed were the
quantisation of Fe and Mn in most of the samples and quantisation of
polynuclear organic material (POM) compounds in the Rectisol condensate
samples. It is thus appropriate to discuss the Level 2 and priority
pollutant results together as a coordinated analytical effort.
The organic priority pollutant data are summarized in Table 7. The
results show that very few of the 116 organic priority pollutant compounds
were found. Those that were present were mostly at very low concentra-
tions. The level of concern specified by the EPA's Effluent Guidelines
Division is 10 yg/L.
The results of the HPLC analysis for POMs performed on the methylene
chloride extracts from the two Rectisol unit samples indicated that each
extract contained essentially the same POMs at very similar levels.
Eleven distinct POM compounds were detected. Comparison of retention time
data as well as relative response ratio for the two detectors with similar
data for available standards enabled the positive identification and
quantisation of five compounds, Table 8. Those compounds which overlap
373
-------
Table 7. ORGANIC PRIORITY POLLUTANT DATA
Sampl 1 ng
Day
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Nov. 12
Nov. 19
i
Stream Description/Stream Number
Input Water—Purified Sewage Effluent
Input Water— Cooling Water
Settling Pond Effluent
Settling Pond Effluent
Combined Condensates from #1—4 Compressors
Combined Condensates from #1 — 4 Compressors
Condensate from Rectl sol Unit
Condensate from Rectl sol Unit
Priority Pollutant Compounds Found
Base/Neutral Fraction
Compound
Nitrobenzene
1,2,4-Trichlorobenzene
Isophorone
Bis (2-Ethylhexyl)phthalate
01 -n-octyl phthalate
.Butyl benzyl phthalate
None Detected
Butyl benzyl phtha 1 a te
Naphthalene
Naphthalene
Diethyl phthalate
Di-n-butyl phthalate
Butyl benzyl phthalate
Naphthalene
Fluorene
Anthracene plus phenanthrene
Fluoranthene
Pyrene
Butyl benzyl phthal ate
Acenaphthalene
Dimethyl phthal ate
Fluorene
Diethylphthalate
Anthracene plus phenanthrene
Fluoranthene
Pyrene
Chrysene
ug/L
T
T
T
T
T
T
T
T
T
T
6.0
T
T
T
T
6.3
25
T
T
T
1.0
T
4.6
19
97
34
Acid Fraction
Compound
None Detected
None Detected
None Detected
None Detected
4-Chloro-m-Cresol
Phenol
Pentachlorophenol
None Detected
Phenol
2,4-Dimethylphenol
pg/L
2.3
T
T
T
T
Volatiles
Compound
None Detected
Chloroform
None Detected
Chloroform
None Detected
Chloromethane
Bromomethane
Chloroform
Chloroform
Chloroform
yg/L
T
T
7.8
49
T
T
T
<
T = Trace (
-------
Table 8. LEVEL 2 POM DATA
Compounds Identified
Fluoranthene
Pyrene
1,2-Benzofluorene
1,2-Benzanthracene
Benzo(k)fluoranthene
11/12/79
Rectisol
Condensate
24 yg/L
32 yg/L
15 yg/L
23 yg/L
2 pg/L
11/19/79
Rectisol
Condensate
17 yg/L
25 yg/L
15 yg/L
16 yg/L
2 yg/L
with the priority pollutant screening (i.e., fluoranthene and pyrene) are
more accurately quantitated by the HPLC technique. The priority pollutant
screening also identified a four-ringed compound as chrysene which in
the HPLC analysis was determined to be 1,2-benzanthracene (also four-
ringed).
HPLC fractions were collected and analyzed by gas chromatography/
mass spectrometry (GC/MS) to obtain molecular weight data on the remain-
ing unknown compounds. The five unidentified POMs are believed to be
present at levels less than 30 yg/L based on the HPLC peak areas. They
had molecular weights of 230 (1), 242 (2), and 252 (2).
The health-based DMEGs for the identified POM compounds range from
670 yg/L to 24,000 wg/L, while ecology-based DMEGs are 100 yg/L for all
five of these POMS. Comparison of data and DMEGs shows that the levels
measured would not be considered to be of concern.
It should be noted that the very toxic POM benzo(a)pyrene was one
of the standards used in this analysis. None of the HPLC peaks matched the
retention time and response ratios for B(a)P. Thus the unidentified
compounds with MW 252 are clearly some other POM with the identical
molecular weight.
The priority pollutant metals screening involves the analysis of
13 elements each of which has its own level of concern. These elements
and the corresponding levels of concern which have been defined by the
EPA are: Ag - 5 ppb, Tl - 50 ppb, Sb - 100 ppb, As - 25 ppb, Se - 10 ppb,
Zn - 1,000 ppb, Pb - 25 ppb, Cd - 5 ppb, Ni - 500 ppb, Be - 50 ppb,
Cu - 20 ppb, Cr - 25 ppb, and Hg - 1 ppb. The results obtained from atomic
375
-------
adsorption and emission spectroscopy analyses for these thirteen elements
plus the two elements (Fe and Mn) quantitated for Level 2 requirements
are presented in Table 9. The data show that the process waters (compres-
sor condensate and Rectisol unit samples) frequently exceed the levels of
concern particularly for Se, Zn, Cu and Hg. However, as was noticed in
the Level 1 SSMS inorganic survey, the only aqueous discharged stream
(settling pond effluent) is relatively clean compared to both the process
streams and the input waters (purified sewage effluent and cooling water).
Overall reduction in trace element levels across the plant were observed
for Sb, As, Zn, Pb, Ni and Ca.
All of the data obtained on the aqueous streams were evaluated
using the SAM/IA model to assess the potential health and ecological
effects of the streams. Of particular interest is the discharged stream,
the settling pond effluent. The TDS and WDS values obtained for this
discharge as compared to the input streams supplied as process water to
the plant, are summarized in Table 10. The fact that the health-based
values for the aqueous input and discharge streams reflect a potential
concern is due mainly to Mn and Fe and to a lesser extent phosphorus. The
ecology-based values are entirely due to phosphorus. The ecology DMEG
value for phosphorus and its various anions is extremely low (0.5 yg/L)
and thus easily becomes the most significant value obtained in the SAM/IA
calculations. However ecology-based severity values >1 were also obtained
for Cd, Cu, Mn, Ni, Pb, S, Zn, and phthalate esters in the input water
streams and Cd, Mn, "Ni, and S in the settling pond discharge stream.
The reduction in both TDS and WDS values for the effluent versus the
input water appears to be due to a decrease in the concentrations of the
phthalate esters, phosphorus, Cu, Pb, and Zn. These and other constituents
as well appear to be transferred to the settling pond sludge.
376
-------
Table 9. INORGANIC PRIORITY POLLUTANT AND LEVEL 2 DATA
Element
Antimony
Arsenic
Beryllium
Cadmi urn
Chromi urn
Copper
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
Iron
Manganese
Concentration, ppb
Input
Water
(PSE)
10
33
<0.5
1.3
<5
78
50
0.5
180
<2
<1
<5
660
<100
1,300
Input
Water
(CW)
<3
<5
<0.5
<0.5
7
43
28
<0.2
<10
<2
<1
<5
3,500
700
<50
Settling
Pond
Effluent
<3
9
<0.5
<0.5
6
6
<5
<0.2
<10
3
<1
<5
<100
140
720
Process Waters
Compressor
Condensate
<3
<5
<0.5
<0.5
6
31
19
250
<10
3,500
<1
<5
270
1,200
<25
Rectisol
Condensate
<3
11
<0.5
<0.5
7
90
13
23
190
26
<1
<5
2,600
4,000
50
377
-------
Table 10. SAM/IA RESULTS FOR AQUEOUS STREAMS
Stream
Discharge Water—Settling
Pond Effluent
Input Water—Purified
Treated Sewage
Input Water — Cooling
Water
Total Discharge
Severity (TDS)
Health-
Based
6.1 E + 00
9.8 E + 00
6.7 E + 00
Ecology-
Based
1.9 E + 02
1.6 E + 04
4.2 E + 03
Weighted Discharge
Severity (WDS)
Health-
Based
3.9 E + 02
5.9 E + 02
6.4 E + 01
Ecology-
Based
1.2 E + 04
9.6 E + 05
4.0 E + 04
CONCLUSIONS
The limited source test program conducted at the Modderfontein
facility has provided some of the key data needed for the environmental
assessment of Koppers-Totzek based synthetic fuels plant which may be
built in the United States. The data obtained do not indicate that any
special problems should be encountered in controlling the process effluents
to environmentally acceptable levels for plants built in the U.S. For
example, the wastewater treatment at Modderfontein, consisting of a clari-
fier and settling pond, was adequate to produce a final discharge that had
lower pollutant levels than the fresh input waters supplied to the plant.
Relatively steady state conditions were realized during the test
period,thus most of the samples taken were generally representative of
typical plant operation. This in turn indicates that the data can
reliably be used as intended. Nearly full design capacity was obtained
throughout the test period. All collection of samples and associated
operating data occurred at production rates of between 102,000 and 104,000
normal cubic meters per hour (Mm /h) of dry raw gas and the gasification
plant operated in a very stable manner with no process upsets.
378
-------
ACKNOWLEDGMENTS
TRW wishes to acknowledge GKT Gesellschaft fur .Kohle-Technologie mbH
for their interest and willingness to participate in this effort thus
making the field tests possible. Some of the key individuals from GKT were
Mr. Herbert Stempelmann, Dr. Gerhard Preusser, and Dr. B. Firnhaber.
The cooperation of AECI Limited in allowing their plant to be tested is
also gratefully acknowledged. Further acknowledgment goes to Mr. Robin
Lazar and the Staff of McLachlan & Lazar (pty) LTD, for their assistance
in the analysis, preservation, and shipment of samples for TRW. TRW also
acknowledges the assistance and guidance of their EPA Project Officer,
Mr. William J. Rhodes.
REFERENCES
1. IERL-RTP Procedures Manual: Level 1 Environmental Assessment
(Second Edition), EPA-600/7-78-201, October 1978.
2. EPA/IERL-RTP Procedures for Level 2 Sampling and Analysis of
Organic Materials, EPA-600/7-79-033, February, 1979.
3. Sampling and Analysis Procedures for Screening of Industrial
Effluents for Priority Pollutants, EPA-EMSL, Cincinnati, Ohio,
Revised April 1977.
4. Multimedia Environmental Goals for Environmental Assessment,
Volumes I—IV, EPA-600/7-7-136 and EPA-600/7-79-176, November 1977
and August 1979.
5. Standard Methods for the Examination of Water and Wastewater,
Fourteenth Edition; APHA, AWWA, WPCF; Washington, DC.
6. SAM/IA: A Rapid Screening Method for Environmental Assessment,
of Fossil Energy Process Effluents, EPA-600-7-78-015,
February 1978.
379
-------
AN ENVIRONMENTALLY BASED EVALUATION OF
THE MULTIMEDIA DISCHARGES FROM THE KOSOVO LURGI
COAL GASIFICATION SYSTEM
By
K. J. Bombaugh, W. E. Corbett, K. W. Lee and W. S. Seames
Radian Corporation, Austin, Texas
ABSTRACT
The U.S. Environmental Protection Agency and the government of Yugoslavia
have jointly sponsored a cooperative environmental data acquisition program.
This program has focused upon a commercial-scale medium-Btu Lurgi gasification
facility which is currently operating in the Kosovo region of Yugoslavia. The
objective of this program was to characterize the uncontrolled discharge
streams associated with the Kosovo facility in order to gain insight into
control technology needs for future U.S. Lurgi plants. The Kosovo study was
undertaken because the Lurgi process has a significant potential for future
use in the United States.
In the Kosovo test program, the most environmentally significant compo-
nents in the plant's key feed, product, and discharge streams were identified
and quantified. Also, selected in-plant process streams were sampled and
analyzed to gain insight into how specific pollutants distributed themselves
among the plant's gaseous, aqueous, and solid discharge streams. The EPA's
Source Analysis Model/lA was used to identify and prioritize the pollutants
found in the plant's discharge streams.
The results of the Kosovo test program indicate that there are many
gaseous, aqueous, and solid discharge streams from a Lurgi gasification
facility which have the potential to significantly impact the environment.
The key pollutants identified in the plant's gaseous discharge streams
included reduced sulfur and nitrogen species (^S, mercaptans, HCN, and
ammonia), hydrocarbons (benzene), and CO. Key pollutants in the Phenosolvan
wastewater included phenols, cyanides, sulfides, and total organics. Effec-
tive controls for the waste streams containing these pollutants will be
essential to minimize the environmental problems associated with Lurgi
gasification technology.
In general, trace elements were not found to be a significant problem at
Kosovo. The dry gasifier ash met the RCRA Extraction Procedure test criteria
for nonhazardous wastes. Trace organics, particularly polynuclear aromatic
compounds which are likely to be present in streams containing tar aerosols,
should be given attention in the development of controls for U.S. Lurgi
facilities.
380
-------
AN ENVIRONMENTALLY BASED EVALUATION OF
THE MULTIMEDIA DISCHARGES FROM THE KOSOVO LURGI COAL
GASIFICATION SYSTEM
An international program sponsored by the Industrial Environmental
Research Laboratory (IERL) of the U.S. Environmental Protection Agency, is
being conducted in the Kosovo region of Yugoslavia to characterize potential
environmental problems associated with Lurgi gasification technology. The
study, conducted over a three year period, was a cooperative endeavor
between scientists from Yugoslavia and EPA/Radian. The program was
undertaken because the Lurgi gasification process has significant potential
for use in the United States.
The purpose of the Kosovo study was to characterize the uncontrolled
discharges from a commercial Lurgi facility. This was done to gain insight
into the environmental control needs for future U.S. Lurgi gasification
plants. The test program was conducted in four phases whose objectives
were:
Phase Objective
I Identify and quantify major and minor pollutants
in the plant's discharge streams.
II Identify and quantify trace pollutants in the
plant's discharge streams.
Ill Characterize ambient air pollutants in the
plant's vicinity.
IV Measure fugitive emission rates in the
plant.
The program schedule is shown in Figure 1. Negotiations for this
cooperative program were initiated in 1974. Testing, which was initiated in
1977, has been carried out in six individual campaigns over a three year
period. Phase I results were reported previously (Ref. 1 through 3).
Documentation of the results from Phases II and III will become available in
1981. Testing for fugitive emissions (Phase IV) was completed in August,
1980, and the results are currently being evaluated.
381
-------
1974 1975 1976 1977 1978 1979 1980
Pretest Negotiations
Stream Selection
Pretest Analyses
Site Specific Test Plans
Test Execution
Phase I - Campaign 1
Phase I - Campaign 2
Phase I - Campaign 3
Phase II
Phase III
Phase IV
Preparation Bi Milestones . < Data Analysis
Figure 1. Kosovo test program schedule.
Waste
Gases — ' * Fl««
Fines to
Steam and Power
1 t L
L» L_>
Run-ol-mlne Coal Dried, __ Ges Crude^
Uoal ^ Preparation sized Coal ^ Production ^
Wastewater °2
Gas
Liquor '
'
Flue Gases
^
llsol
Clean *
Gas
,
Purlllcallon
i
Gas
Distribution
-.HjtoNHj
Synthesis
Medium
^ Blu Gas
to Pipeline
[Naphtha
Tar/011
Separation
^
Phenolic
, Water
Phenosolvan
>
Tars & „
Ol
a '
Pher
^
Byproduct
Storage
By-Prdducls
.to Steam and
' Power
Generation
ols
Waslewaler
Figure 2. Simplified flow diagram of the Kosovo coal gasification plant.
382
-------
This paper presents an overview of the Phase I and II test results.
These results address the major, minor, and trace pollutants found in the
plant's key process and discharge streams. An assessment of the severity of
the plant's gaseous, aqueous, and solid discharges is included. This
assessment is based upon the use of the EPA-IERL's Source Analysis Model/lA
(SAM/1A). This model prioritizes pollutants based on their potential for
causing adverse health effects.
Plant Description
Detailed descriptions of the Kosovo coal gasification plant were
provided in previous publications (Ref. 1 through 3). A brief plant
description is included here to facilitate understanding of the results.
The Kosovo Lurgi gasification facility is an integral part of a large
mine-mouth industrial complex. A simplified flow diagram is shown in Figure
2. The gasification plant consumes dried lignite and produces two primary
products: a medium-Btu fuel gas having a net heating value of approximately
14 MJ/nP @ 25°C (360 Btu/scf), and hydrogen which is used as an ammonia
synthesis feedstock. Several hydrocarbon by-products including light tar,
medium oil, naphtha, and crude phenol are also produced.
Run-of-mine coal which contains around 50 weight percent moisture is
dried by the Fleissner process (high temperature steam soak) to around 25
weight percent moisture and sized to select particles between 6 and 60 mm in
diameter. Typical feed coal properties are presented in the results
section. After sizing,1the dried coal is fed to the Lurgi gasifiers where
it reacts with oxygen and steam at 2.5 MPa (25 atm) pressure. The crude
product gas is cooled and then cleaned to remove acid gases prior to its
transportation by pipeline to the utilization site. In the cooling step,
tars, oils, naphtha, and phenolic water are condensed and removed from the
gas. In the acid gas removal step, H2S and C02 are removed by sorption
into cold methanol. The rich methanol is regenerated by depressurization
and heating. The E^S-rich waste gas released by the regeneration step is
sent to a flare while the C02~rich waste gas is vented directly to the
atmosphere. Tar and oil are separated from the phenolic water by
decantation after which the water soluble organics (crude phenols) are
removed from the wastewater by extraction with diisopropyl ether. Four
liquid by-products: naphtha, medium oil, light tar, and crude phenol are
collected in storage tanks and used as fuels. Ammonia, removed from the
phenolic water by steam stripping, is vented to the atmosphere.
Figure 3 shows the design flow rates of the plant's major inlet and
outlet streams. These flow rates, are based on design conditions with five
of six Lurgi gasifiers in operation. As indicated in Figure 3, the plant is
designed to produce 25 Mg (65,000 m3 @ 25°C) of product gas for every 80
Mg of dried coal consumed.
383
-------
Rectisol
Acid Gases
(H2S-Rich and CO2-Rich)
(45)
1
Dried Coal (80)
Steam (65)
02(14)
Kosovo
Gasification
Plant
T
GasifierAsh
(14)
Heavy Tar
(.5)
I
Waste-
waters
(68)
Clean
Product
Gas
(25)
•>• Light Tar (2.2)
Medium Oil (1.3)
Naphtha (.7)
Phenols (.4)
Ammonia (1)
Figure 3. Major stream flow rates in the Kosovo
gasification plant (megagrams/hr).
Slcain
• - indicates sampling point
Figure 4. Kosovo coal drying section.
384
-------
The Kosovo plant is smaller than proposed first generation U.S. Lurgi
gasification plants, but it contains many of the process units which are
likely to be employed in those plants. For this reason, the plant is
considered to be representative of the Lurgi facilities likely to be built
in the U.S. in the near future.
While many of the process units employed at Kosovo are representative
of those proposed for use in future U.S. Lurgi facilities, the environmental
control practices followed at the Kosovo plant are not. Thus, while the
discharges that enter the environment at Kosovo are not representative of
those that would be encountered in similar U.S. facilities, the types of
control problems facing U.S. Lurgi plant operators will be similar to those
found at Kosovo. A study of the waste and process streams at the Kosovo
plant should aid U.S. plant designers in developing the process
modifications and control schemes necessary to achieve U.S. standards of
environmental protection.
Test Rationale
The Kosovo gasification plant contains approximately 70 streams which
have a significant potential for adversely impacting the environment.
However, since the cost of characterizing such a large number of streams was
considered prohibitive, during Phase I, approximately 50 streams were
surveyed. In this survey, the major pollutants present in the process and
uncontrolled discharge streams were identified. Based on these results, a
limited number (20 to 30) of streams were selected for detailed study in
Phase II. i
Process and discharge streams were selected for study for one or more
of the following reasons:
• high discharge rate,
• significant pollutant concentration,
• needed for trace pollutant fate determination, and/or
• provided useful process information.
Figures 4 through 10 show simplified flow schemes of the primary
process units of the Kosovo plant. Streams selected for Phase II testing
are identified in these figures.
385
-------
Coal Room Venl
Coal
H.P. Coal Loci***
Vent to Flare
h
Qasirier
Start-up Vent
Gas Liquor
to Tar
Separation
Ash Lock Vent
Hot, Dry Ash
to Quench Bath
- Low Pressure
- High Pressure
- indicates sampling point
Figure 5. Kosovo Gas Production section,
H2S-Rich Waste Gas
to Flare
Crude
Product
Gas
Condensate
-Product
Naphtha
CO2-Rlch Waste
Gas Vent
Clean
Product
Gas
• - indicates sampling point
Figure 6. Kosovo Rectisol section.
386
-------
Flash Gases
To Flare
Tar Separation,
Section Waste Gas
Gas Liquors
From Gas
Production
Section
Gas Liquor
from 2nd
Stage Coolers
Gas Liquor
From 1 st •
Stage Coolers
1
r
d —
s
k
— *
Tar
T • Medium Oil Tank Vent
Medium
Oil
Separator
Separator
*-
Medium
Oil
Tank
t
k By-Product
v Medium Oil
Phenolic
Water
Tank
fc Phenolic
r Water to
Phenosolvan
1 • Tar Tank Vent
Tar
Tank
fc By-Product
* Light Tar
Heavy Tar
• - indicates sampling point
Figure 7. Kosovo Tar Separation section.
Inlet
Water
NH3
Recovery
Phenolic Water
From Tar
Separation
r t
> NH4OH
H3 Stripper Vent
Outlet
Water
Storage
•
Degassing
•
NH3
Stripping
Wastewater
t
Steam
j not in service during testing
Crude Phenol
Tank Vent
By Product
Crude Phenol
- indicates sampling point
Figure 8. Kosovo Phenosolvan section,
387
-------
i Naphtha Storage Tank Vent
Light
Tar
^
r
-
Medium
Oil
1
r
-*
Naphtha
^
r
-
Phenols
^
r
— t
( >
|NH4OHJ
!
To
Steam/Power
Generation
i
L
1 not in service during testing
- indicates sampling point
Figure 9. Kosovo By-Product Storage section.
H2S-Rich Waste Gas
(from Rectisol)
i
H.P.
Coal Lock Gas
I
Tar Separation
Waste Gas
I
Combined Gases
to Flare
I M
o Flare
• - indicates sampling point
Figure 10. Kosovo Flare Feed system.
388
-------
Stream Parameters; The Phase I and Phase II characterization efforts
addressed the following parameters:
Gaseous Streams
Flow rate
Particulate concentration
Gas composition
Condensible organics
Trace elements
Aqueous Streams
• Water quality parameters
• Trace elements
• Organic constituents
Solids
• Proximate analyses
• Ultimate analyses
• Trace elements
• Leachate analyses
Liquid By-Products
• Bulk composition
• Trace elements
Sampling and Analytical Methods; With the exception of the condensible
organics analysis, all gas stream characterization work was performed
on-site. The methods used for gaseous sampling and analysis are listed in
Table 1. Liquid and solid analyses were performed where applicable, with
either EPA or ASTM standard methods. These methods are identified and
discussed elsewhere (Ref. 3). New methods, developed specifically to
characterize sulfur- and nitrogen-containing organic compounds in liquid
by-products will be reported separately.
Data Evaluation - Source Analysis Model I/A
The Source Analysis Model I/A (SAM/1A) is a procedure developed by
EPA-IERL for evaluating discharge stream data. Its principle strength is
that it makes possible the reduction of pollutant discharge data to a common
numerical base so that discharges can be ranked or prioritized.
389
-------
TABLE 1. SAMPLING AND ANALYTICAL METHODS
Parameter
Collection Method
Analytical Method
CONDENSIBLE HYDROCARBONS:
Condensible Hydrocarbons
Benzene, Toluene, and
Xylene
Gas stream cooled to 0°C and
resulting condensate trapped in
impingers. The remaining condensible
hydrocarbons trapped on XAD-2 resin.
Vapors trapped from gas stream by
activated carbon.
Organic material extracted
from condensate and resin
with CH2C12- Extract analyzed
with gas chromatography/mass
spectrometry.
Vapors solvent extracted from
carbon and analyzed by GC
with flame ionization detector,
CO
«3
o
GASEOUS SPECIES BY GC:
Fixed Gases (CO, H2, C02,
N2, 02,
Hydrocarbons Ci - Ce, Ce
Benzene, Toluene, and
Xylene
Sulfur Species (H2S, COS,
CS2, S02, Mercaptans)
Sample was heated, filtered and dried
then compressed into silanized glass
bombs for analyses.
Sample was heated, filtered and dried
then compressed into silanized glass
bombs for analyses.
Sample was heated, filtered and dried
then compressed into silanized glass
bombs for analyses.
Gas chromatograph with thermal
conductivity detector.
Gas chromatograph with flame
ionization detector.
Gas chromatograph with flame
photometric detector.
(Continued)
-------
TABLE 1 (Continued). SAMPLING AND ANALYTICAL METHODS
Parameter
Collection Method
Analytical Method
GO
10
PARTICIPATE:
Suspended Particulate
Suspended Particulate
Plus Condensibles
TRACE ELEMENTS;
Non-Volatile Elements
(Be, Cd, Co, Cr, Cu, Mo,
Ni, Pb, Sr, Tl, V, Zn)
Volatile Elements
(Hg, As, Sb, Se)
Iron and Nickel Carbonyls
OTHER GASES:
Ammonia
EPA Method 5, gas filtered at 250°F
out of stack.
EPA Method 17, gas filtered at duct
temperature in stack.
Condensation and collection in a
series of water filled impingers.
Two impingers with 10% HNOs followed
by two impingers with 10% NaOH.
Two impingers with 10% HNOs followed
by two impingers with 10% NaOH.
Two fritted impingers with 3% HC1.
Two fritted impingers with 0.1 N
H2SOit.
Gravimetric.
Gravimetric.
Filtration, extraction with
CH2Cl2, Gravimetric.
Dissolution, AA with Graphite
Furnace.
Dissolution, AA with Hydride
Generation.
AA with Graphite Furnace.
Distillation into boric acid
and back titration with
sulfuric acid.
(Continued)
-------
TABLE 1 (Continued). SAMPLING AND ANALYTICAL METHODS
GO
tQ
ro
Parameter
Collection Method
Analytical Method
Hydrogen Sulfide
Hydrogen Cyanide
Phenols
Two fritted impingers with 0.1 N
cadmium acetate.
Two fritted impingers with 0.1 N
cadmium acetate followed by two
fritted impingers with 0.1 N NaOH.
Two fritted impingers with 0.1 N
NaOH.
Iodine addition and back
titration with thiosulfate.
Distillation and titration
with silver nitrate.
Spectrophotometric determina-
tion by reaction with
4-aminoantipyrine.
-------
The SAM/1A model is based upon the use of discharge multimedia
environmental goals (DMEG's) to compute Discharge Severity (DS) values
(Ref. 4). DMEG's are concentration levels below which the discharged
component is of low concern for its potential effects on either human health
or the ecology. Thus, it is a "target value" for components in discharge
streams. DMEG's have been defined for many substances representing 26
classes of organic compounds (Ref. 5). Target levels have been defined in
terms of their effect on both human health and ecology for discharges to the
three environmental media: air, water, and soil. DMEG (Air/Health) values
for 16 components whose concentrations were measured in this study, are
shown graphically in Figure 11. A reciprocal of DMEG is plotted since DS is
the product of concentration and 1/DMEG as defined below:
Qg = Measured Concentration of a Pollutant
DMEG of that Pollutant
Since the DMEG allows the severity of different compounds to be related to a
common numerical base ("multiples of the target value"), a stream's total
discharge severity (TDS) can be determined by summing the DS values for all
components in that stream:
TDS = £DS.
The TDS value provides a basis for comparing uncontrolled discharge streams,
and, therefore, provides a basis for identifying the most severe (highest
TDS) streams.
Discharge severity is a concentration - based value that does not take
into account the quantity of mass emitted. Used alone it cannot define the
environmental effects of a discharge because such effects are related to
both quantity and severity. With the SAM/1A Model, the environmental
significance of a pollutant in a given discharge stream is defined by its
Weighted Discharge Severity (WDS):
WDS = F • DS
where F = Stream Flow Rate;
and further, the environmental significance of that discharge stream is
defined by its Total Weighted Discharge Severity (TWOS):
TWOS = F • £ DS = F • TDS
By comparing discharge streams within a given medium, such as gaseous,
aqueous, or solid, the stream with the highest TWOS value may be selected as
the most environmentally significant.
393
-------
M&E Mercaptans
C5, Toluene, Xylene, COS
E** -7
-6 -5 -4
Log10(Nm3/yg)
-3
*Methyl and Ethyl Mercaptans
E** = Exponential (E-5 = 10~5)
Figure 11. Key Kosovo gaseous pollutants in order of
severity (1/DMEG).
394
-------
Results and Discussion
The results obtained during the Kosovo study consist of stream
composition and flow rate data. The data presented and discussed in this
section were selected from Phases I and II as "best values" based on
engineering and analytical judgment. The results discussed here are for
the streams selected for detailed examination in the Phase II test program.
Gaseous Streams; Test data for gaseous streams are presented in Tables 2
and 3. In Table 2, the concentration data are given in molar concentration
units (vol % or ppmv) while in Table 3, these data are expressed in mass
concentration units (yg/m^). Oxygen and nitrogen analyses were included
in the fixed gas analyses for quality control. Samples showing abnormal
levels of Q£ and N2 (indicating an air leakage into the sample) were
resampled.
The data in Table 3 were used to calculate the mass discharge rate from
each stream for each major pollutant. Table 4 summarizes the streams having
the highest concentration and those having the highest mass flow for each
type of pollutant measured. As this table shows, a single stream, such as
the ammonia stripper vent, can be the source of several pollutants at
comparatively high concentrations. The table also indicates that the t^S-
rich waste gas and C02~rich waste gas streams are of concern because of
the high flow rates of these streams. In addition, the by-product tank
vents (naphtha storage tank, medium oil tank, phenolic water tank) are
significant because of high pollutant concentrations.
Figure 12 shows a graphic representation of the mass flow rate of the
major gaseous pollutants. As shown, Cj to Cg+ hydrocarbons and sulfur
species pollutants are produced in the largest quantities. Most of the
sulfur species are sent to the flare, whereas most of the ammonia and
phenols are discharged directly to the atmosphere. The Cj to G£+
hydrocarbons are well distributed among most of the flare feed and
uncontrolled discharge streams.
Discharge severity values accent pollutants of greatest concern in
terms of their potential to cause adverse health or environmental effects.
Figure 13 illustrates the relationship between DS values and pollutant mass
concentration data for the major pollutants in the coal lock vent discharge.
Note that BTX (benzene, toluene, and xylene) and mercaptans, which are at
relatively low concentrations (Figure 13A), emerge as pollutants of high
concern when the severity of the discharge is investigated (Figure 13B).
395
-------
TABLE 2. KOSOVO GASEOUS STREAM COMPOSITION DATA
PLANT SECTION;
SAMPLE POINT;
Dry Gaa Flow Rate
(mj/gasifier-hr @ 25°C)
Temperature (°C)
Moisture Content (Z)
Molecular Ut. of Dry Gas
GAS
1
3.2
Low Pressure
Coal Lock Vent
21
56
44
23.5
PRODUCTION
3.3
Gaslf ier
Start-up
Vent
70
33.1
— |
3.6
High Pressure
Coal Lock (Flare
Feed Stream)
230
54
11
24.9
I
7.1
HzS-rlch
Waste Gaa (Flare
Feed Stream)
3600
12
3.9
43.0
KECT1SOL
7.2
COz-rlch Waste
Gas Vent
3600
19
5.1
42.2
7.3
Crude Gas
(Process Stream)
18,800*
22
2.5
21.9
|
7.4
Product Gas
(Process stream)
13,100*
4.1
10.3
Composition (Dry Baals)
Fixed Gaaes (Vol X)
HZ
Oj
Hz
CB«
CO
CO;
Sulfur Species (ppmv)
H2S
COS
CH3SH
CzHsSH
Hydrocarbons (Vol Z)
CzHt
C2H,
Ci's
Ci, 's
Cs's
ct+
Aromatic Species (ppmv)
Benzene
Toluene
Xylene & Ethylbenzene
Phenols
Higher Aromatlcs
Nitrogen Species (ppmv)
NUi
1ICN
37
0.27
0.18
8.6
14.6
36.5
13,000
110
420
220
0.22
Tr
0.14
0.05
Tr
0.12
760
220
75
5.7
2400
600
0.09
4.5
42
1.6
14
34
6300
110
490
240
0.15
0.05
0.08
0.03
0.007
0.09
90
10
Tr
630
11.000
2.900
32
0.24
0.14
10.5
12
42
3500
120
460
210
0.42
Tr
0.25
0.11
0.01
0.08
550
100
38
2.5
or
170
0.11
Tr
Tr
4.3
1.1
88
45,400
420
2100
780
0.82
Tr
0.63
0.32
0.04
0.21
110
8
NF
Tr
2200
200
Tr
Tr
Tr
1.2
Tr
94
39
62
8.5
4.4
1.6
Tr
0.28
Tr
Tr
NF
1.0
Tr
Tr
NF
4.6
13
38.1
0.36
0.64
11.5
15
32
6000
97
590
200
0.47
0.04
0.19
0.074
0.044
0.064
750
230
100
Tr
3.3
320
60
0.44
0.38
16
22
0.02
NF
0.17
1.1
1.0
0.15
Tr
Tr
Tr
Tr
0.03
Tr
Tr
Tr - Trace • 0.01 vol. Z COT fixed |««ea. 1 pp«v for «11 othera.
HF - Hoc Found - !••• than • trac«.
• - Dulgn Value.
- - Ho Data Available.
(Continued)
396
-------
TABLE 2 (Continued). KOSOVO GASEOUS STREAM COMPOSITION DATA
PLANT SECTION:
SAMPLE POINT!
Dry Gas Flow Rate
(m'/gasifier-hr 8 25'C)
Temperature (*C)
Moisture Content (Z)
Molecular Ht. of Dry Gas
Composition (Dry Basis)
Fixed Gases (Vol Z)
H,
02
N;
CH,
CO
C02
Sulfur Species (ppmv)
H2S
COS
CH3SH
C2H,SH
Hvdrocarbons (Vol Z)
C2Ht
C2Ht
Cs's
C»'s
C5's
C6+
Aromatic Species (ppmv)
Benzene
Toluene
Xylene & Ethylbenzene
Phenols
Higher Aromatics
Nltroeen Siecies (ppnv)
NHs
HCN
1
13.1
Tar Tank
Vent
0.55
52
14
29.1
Tr
19
77.5
0.16
Tr
0.86
6900
110
390
240
Tr
—
0.01
Tr
Tr
0.37
2000
960
220
57
2.2
2600
130
TAR
13.3
Medium Oil
Tank Vent
1.7
42
8.4
32.5
Tr
0.45
1.1
7.6
5.9
56
26,000
96
5200
2100
0.34
Tr
0.30
0.25
O.O9
2.4
7650
1400
140
110
19
57
SEPARATION
13.6
Tar Separation
Waste Gas(Flare
Feed Stream)
28*
40
7.7
39.0
11
Tr
Tr
3.5
1.1
77.5
9000
120
2500
1600
0.33
Tr
0.41
0.41
0.09
1.3
9600
1200
150
4.2
4.9
i 9 , inn
M
1
13.7
Phenolic Hater
Tank Vent
5.5
76
42
34.4
Tr
13
39
0.2
NF
35
12,600
41
2100
7200
10.02
0.02
0.02
0.006
1.8
11,000
2300
280
Tr
3.1
12,000
38
PHENOSOLVAN
1 1
14.5
NHj Stripper
Vent
260
91
76
32.7
NF
_
Tr
NF
55
19,500
NF
290
100
J
)
Tr
Tr
Tr
NF
Tr
Tr
6200
418,000
4800
BY-
PRODUCT
STORAGE
1 1 1-
15.3
Naphtha
Storage
Tank Vent
4.5
32
5
33.3
NF
2.6
84
NF
NF
0.85
NF
NF
2600
9700
u
}
0.01
0.07
0.08
5.3
37,600
1900
60
Tr
NF
1100
FLAKE SYSTEM
I |
20.1
Combined Gas
to Flare
1330
21
2.5
41.7
Tr
0.10
0.21
6.2
1.9
88
10,600
250
2500
190
0.77
Tr
0.65
0.38
0.04
0.06
640
215
33
Tr
NF
100
Tr - Trace - 0.01 vol. Z for fixed gases, 1 ppmv for all others.
NF - Not Found - less than a trace.
* • Design Value.
- - No Data Available.
397
-------
TABLE 3. COMPONENT CONCENTRATIONS IN KOSOVO GASEOUS STREAMS
CO
10
oo
PLANT SECTION:
SAMPLE POINT:
Component ((Jg/mj @ 25°C)
Fixed Gases
H2
02
N2
CH4
CO
C02
Sulfur Species
H2S
COS
CH3SH
C2H5SH
Hydrocarbons
C2H6
C2H4
C3's
C4's
C5'B
C6+
Benzene
Toluene
Xylene 4 Ethylbenzene
Phenols
Nitrogen Species
NH3
HCN
Dry Gas Flow Rate
(m3/gaslfler-hr @ 25°C)
NF - Not Found
Tr - Trace
* <* Design Value
GAS
3.2
Low Pressure
Coal
Lock Vent
3.05E07
3.53E06
2.06E06
5.64E07
1.67E08
6.56E08
1.81E07
2.70E05
8.25E05
5.57E05
2.70E06
Tr
2.52E06
1.19E06
Tr
4.22E06
2.43E06
8.38E05
3.25E05
2. 19E04
1.67E06
6.62E05
21
PRODUCTION
3.6
High Pressure
Coal Lock Vent
(Flare Feed Stream)
2.64E07
3.14E06
1.60E06
6.88E07
1.37E08
7.55E08
4.87E06
2.95E05
9.04E05
5.33E05
5.16E06
Tr
4.50E06
2.61E06
2.95E05
2.82E06
1.76E06
3.76E05
1.65E05
9.61E03
NF
1.88E05
230
RECTISOL
7.1
H2S-Rlch
Waste Gas
(Flare Feed Stream)
9.06E04
Tr
Tr
2.82E07
1.26E07
1.58E09
6.32E07
1.03E06
4.13E06
1.98E06
1.01E07
Tr
1.14E07
7.60E06
1.18E06
7.39E06
3.51E05
3.00E04
NF
Tr
1.53E06
2.21E05
3,600
7.2
C02-Rlch
Waste
Gas Vent
Tr
Tr
Tr
7.84E06
Tr
1.69E09
5.43E04
1.52E05
1.67E04
1.12E04
1.97E07
Tr
5.04E06
Tr
Tr
NF
3.19E03
Tr
Tr
NF
3.20E03
1.44E04
3,600
7.3
Crude
Product Gae
3. 14E07
4.70E06
7.32E06
7.54E07
1.71E08
5.81E08
8.35E06
2.38E05
1 . 16E06
5.08E05
5.77E06
4.58E05
3.42E06
1.76E06
1.30E06
2.25E06
2.39E06
8.66E05
4.34E05
Tr
2.30E03
3.53E05
18,800*
7.4
Clean
Product Gas
4.94E07
5.75E06
4.35E06
1.05E08
2.52E08
3.60E05
NF
4.17E02
2.16E03
2.54E03
1.84E06
Tr
Tr
Tr
Tr
1.06E06
-
-
-
Tr
Tr
-
13,100*
(Continued)
-------
TABLE 3 (Continued). COMPONENT CONCENTRATIONS IN KOSOVO GASEOUS STREAMS
CO
io
10
PLANT SECTION:
SAMPLE POINT:
Component (ug/m3 @ 25°C)
Fixed Gases
"2
°2
N2
CH4
CO
C02
Sulfur Species
H2S
COS
CH3SH
C2H5SH
Hydrocarbons
C2H6
C2H4
C3's
C4's
C5's
C6+
Benzene
Toluene
Xylene & Ethylbenzene
Phenols
Nitrogen Species
NH3
HCN
Dry Gas Flow Rate
(m3/gaslfier-hr 9 25°C)
NF - Not Found
Tr " Trace
* - Design Value
13.1
Tar
Tank Vent
Tr
2.48E08
8.87E08
1.04E06
Tr
1.55E07
9.61E06
2.70E05
7.66E05
6.09E05
Tr
-
1.80E05
Tr
Tr
1.30E07
6.38E06
3.61E06
9.54E04
2.19E05
1.81E06
1.44E05
0.55
13.2
Medium Oil
Tank Vent
Tr
5.88E06
1.26E07
4.98E07
6.75E07
1.01E09
3.62E07
2.36E05
1.02E07
5.33E06
4.18E06
Tr
5.40E06
5.94E06
2.65E06
8.45E07
2.44E07
5.27E06
6.06E05
4.24E05
1.32E04
6.28E04
1.7
TAR SEPARATION
13.6
Tar Separation
Waste Gas
(Flare Feed Stream)
9.06E06
Tr
Tr
2.29E07
1.26E07
1.40E09
1.25E07
2.94E05
4.91E06
4.06E06
4.05E06
Tr
7.39E06
9.74E06
2.65E06
4.58E07
3.06E07
4.52E06
6.51E05
1.62E04
1.34E07
7.05E04
28*
13.7
Phenolic Water
Tank Vent
Tr
1.70E08
4.46E08
1.31E06
NF
6.29E08
1.75E07
1.01E05
4.13E06
1.83E07
2.46E05
-
3.60E05
4.75E05
1.77E05
6.34E07
3.51E07
8.66E06
1.21E06
Tr
8.35E06
4.20E04
5.5
PHENOSOLVAN
14.5
Ammonia
Stripper
Vent
NF
-
-
Tr
NF
9.89E08
2 . 72E07
NF
5.70E05
2.54E05
Tr
-
Tr
Tr
Tr
NF
-
-
Tr
2.38E07
2.91E08
5.30E06
260
BY-PRODUCT
STORAGE
15.3
Naphtha
Storage
Tank Vent
NF
3.40E07
9.61E08
NF
NF
1.53E07
NF
NF
5.11E06
2.46E07
Tr
-
1.80E05
1.66E06
2.36E06
1.87E08
1.20E08
7.15E06
2.60E05
Tr
NF
1.21E06
4.5
FLARE
SYSTEM
20.1
Combined
Gas
to Flare
Tr
1.31E06
2.40E06
4.06E07
2.17E07
1 . 58E09
1.48E07
6.14E05
4.91E06
4.82E05
9.46E06
Tr
1.17E07
9.03E06
1.18E06
2.11E06
2.04E06
8.09E05
1.43E05
Tr
NF
1 . 10E05
1,330
-------
TABLE 4. MAJOR KOSOVO DISCHARGE STREAMS BASED ON POLLUTANT CONCENTRATION AND MASS FLOW RATE
HIGHEST CONCENTRATION
GREATEST MASS FLOW RATE
O
O
Pollutant
Direct
Atmospheric Discharges
Total Plant*
Direct
Atmospheric Discharges
Total Plant*
CO LP Coal Lock Vent
GI - Cg Naphtha Storage Tank Vent
BTXt Naphtha Storage Tank Vent
Total Medium Oil Tank Vent
Sulfur Naphtha Storage Tank Vent
Species Phenolic Water Tank Vent
H2S Medium Oil Tank Vent
COS LP Coal Lock Vent
Mercaptans Naphtha Storage Tank Vent
Phenols Ammonia Stripper Vent
NH3 Ammonia Stripper Vent
HCN Ammonia Stripper Vent
LP Coal Lock Vent
Naphtha Storage Tank Vent
Naphtha Storage Tank Vent
H2S-Rich Waste Gas
H2~S-Rich Waste Gas
H2S-Rich Waste Gas
Naphtha Storage Tank Vent
Ammonia Stripper Vent
Ammonia Stripper Vent
Ammonia Stripper Vent
LP Coal Lock Vent
C02~Rich Waste Gas Vent
Phenolic Water Tank Vent
Ammonia Stripper Vent
Ammonia Stripper Vent
C02~Rich Waste Gas
Ammonia Stripper Vent
Ammonia Stripper Vent
Ammonia Stripper Vent
Ammonia Stripper Vent
H2S-Rich Waste Gas
H2S-Rich Waste Gas
Tar Separation Waste Gas
H2S-Rich Waste Gas
H2S-Rich Waste Gas
H2S-Rich Waste Gas
H2S-Rich Waste Gas
Ammonia Stripper Vent
Ammonia Stripper Vent
Ammonia Stripper Vent
*Includes both direct discharge and flare feed streams.
tBenzene, Toluene, and Xylenes.
-------
Figure 12A.
Hydrocarbons"
Sulfur Species
Ammonia
Carbon Monoxide
Phenols
Benzene (BTX)'
Hydrogen Cyanide
Mass Discharge Rate
(9/hr)
Figure 12B.
CT-CS Hydrocarbons"
Sulfur Species
Ammonia
Carbon Monoxide
Phenols
Benzene (BTX)'
Hydrogen Cyanide
Mass Discharge Rate
(g/hr)
Figure 12C.
Ct - C6 Hydrocarbons"
Sulfur Species
Ammonia
Carbon Monoxide
Phenols
Benzene (BTX)'
Hydrogen Cyanide
Mass Discharge Rate
(g/hr)
*BTX = Benzene, Toluene, Xylenes
**Excluding Benzene
A. Plant-Wide Discharge and Flare Feed Streams
B. Discharge Streams Only
C. Flare Feed Streams Only
Figure 12. Total mass flow rate in Kosovo Gaseous Streams.
401
-------
Figure 13A.
Figure 13B.
CO
CI-G,
BTX
H2S
COS
MERCAPTANS
PHENOLS
NH,
HCN
E* 01 23456789
Logio (yg/m3)
Mass concentration of pollutants
in LP coal lock vent
*Exponential (E05 = 105)
1 1 1 1 1
01234
Logio (DS)
Discharge Severity of pollutants
in LP coal lock vent
Figure 13. Comparison of mass concentrations with calculated discharge
severities in the low pressure coal lock vent discharge
stream.
402
-------
Discharge severity values for the individual pollutants and total
stream discharge severity values for the plant's key gaseous streams are
listed in Table 5. Figure 14 shows a comparison of these total stream
discharge severities for the seven uncontrolled discharge streams examined
during Phase II. From this comparison it is evident that the discharge from
the naphtha storage tank vent is several hundred times more severe (DS on
the order of 70,000) than the discharge from the C02~rich waste gas vent
(DS on the order of 200). However, when the flow rates of the respective
streams are taken into consideration, the two streams have comparable TWDS
values as is illustrated in Figure 15. The relationship of flow rate and
TDS to TWDS is illustrated well in Figure 15 for the seven uncontrolled
streams. Since the bar graphs are plotted on a log scale, the sum of the
logs of the flow component and the TDS component equals the log of the TWDS.
From this plot, streams can be prioritized according to flow rate, DS, or
TWDS:
• Largest Stream (highest flow rate) - C02~rich
waste gas vent.
• Most Severe Stream (highest TDS) - Naphtha
storage tank vent.
• Most Environmentally Significant Stream (highest TWDS) -
Ammonia stripper vent.
Figure 15 illustrates why the very large stream with a low TDS value
(C02~rich waste gas) and the very small stream with a high TDS value (tar
tank) are both environmentally significant (TWDS values are comparable).
Pollutant WDS values from the seven uncontrolled discharge streams are
shown in Figure 16. This prioritization indicates that, of the pollutants
discharged from the Kosovo plant, ammonia and sulfur species (I^S and
mercaptans) are the most environmentally significant (highest WDS values).
Particulates in Gaseous Streams; Particulate loadings were measured in six
gaseous discharge streams. Except for the coal room vent (a dry stream),
all measurements were made by the wet impinger method. In this method,
particulates are collected as three fractions:
• filterable solids,
• dissolved solids, and
• tars and oils (condensible organics).
403
-------
TABLE 5. DISCHARGE SEVERITY DATA FOR KOSOVO GASEOUS DISCHARGE STREAMS
PLANT SECTION:
SAMPLE POINT:
Component Discharge
Severities
Fixed Gases
cm
CO
C02
Sulfur Species
H2S
COS
CH3SH
C2H,SH
Hydrocarbons
C2H6
c2m
C3's
C4's
C5's
C6 +
Aromatic Species
Benzene
Toluene
Xylene and
Ethylbenzene
(as xylene)
Phenols (as Phenol)
Nitrogen Species
NH3
HCN
Total Stream
Discharge Severity
GAS PRODUCTION
3.2
Low Pressure Coal
Lock Vent
1.70E01
4.20E03
7.30E01
1.20E03
6.13E-01
8.30E02
5.60E02
4.42E-01
2.01E-04
2.80E-01
8.50E-01
8.41E-03
1.20E01
8.10E02
2.20EOO
7.40E-01
1.20EOO
9.30E01
6.02E01
7.88E03
3.3
Gastfier
Start Up Vent
3.20E-00
4.00E03
6 . 80E01
5.84E02
6.13E-01
9.62E02
6.10E02
3.02E-01
1.60E-01
5.10E-01
5.90E-01
9.10EOO
9.60E01
7.60E-01
1.30E02
4.30E02
2.91E02
7.19E03
3.6 1
High Pressure Coal Lock
(Flare Feed Stream)
2.10E01
3.43E03
8.40E01
3.24E02
6.70E-01
9.03E02
5.33E02
8.50E-01
2.01E-04
5.00E-01
1 . 90E-00
8.41E-01
8.04EOO
5.90E02
9.90E-01
3.74E-01
5.10E-01
1.70E01
5.92E03
RECTISOL
1 7.1
HjS-Rich Waste Gas CO
(Flare Feed Stream)
8.53EOO
3.14E02
1.76E02
4.21E03
2.34EOO
4.12E03
2.00E03
1.70EOO
2.01E-04
1.30EOO
5.42EOO
3.40EOO
2.11E01
1.20E02
7.90E-02
8.50E01
2 . OOE01
1.11E04
7.2 1
2-Rich Waste
Gas Vent
2.44EOO
2 . 90EOO
1.88E02
3.61EOO
3.34E-01
1 . 70E01
1.11E01
3.22EOO
2.01E-04
5.60E-01
1.70E-03
8.41E-03
1.10EOO
9.94E-03
1.40E-02
1.77E-01
1 . 30EOO
2.32E02
TAR
1 13.1
Tar Tank
Vent
1.72E-01
2.90EOO
1.71EOO
6.40E02
6.13E-01
7.70E02
6.10E02
2.01E-04
2.00E-02
1.70E-03
8.41E-03
3.72E01
2.12E03
9.51EOO
2.20E-01
1.16E01
1.00E02
1.30E01
4.31E03
SEPARATION
13.3 1
Medium Oil
Tank Vent
1.50E01
1.70E03
1.12E02
2.41E03
5.40E-01
1.02E04
5.33E03
6.84E-01
2.01E-04
6.00E-01
4.24EOO
7.60EOO
2.41E02
8.13E03
1.40E01
1.40EOO
2.22E01
7.32E-01
5.71EOO
2.82E04
Dry Gas Flow Rate 21.0
(m3/gasifier-hr(225°C)
Total Weighted 1.65E05
Discharge Severity
(m3/gasifier-hrl?25°C)
230
1.36E06
3600
3.99E07
3600
8.37E05
0.55
2.20E03
1.7
4.79E04
(Continued)
-------
TABLE 5 (Continued). DISCHARGE SEVERITY DATA FOR GASEOUS DISCHARGE STREAMS AT KOSOVO
PLANT SECTION:
SAMPLE POINT:
Component Discharge
Severities
Fixed Gases
CHi,
CO
C02
Sulfur Species
H2S
COS
CH3SH
C2H5SH
Hydrocarbons
C2H6
C2H,,
Ca's
C»'s
C5's
C6+
Aromatic Species
Benzene
Toluene
Xylene & Ethylbenzene
(as xylene)
Phenols (as Phenol)
Nitrogen Species
NH3
HCN
Total Stream Discharge
Severity
Dry Gas Flow Rate
(m3/gasifier-hr @ 25°C)
Total Weighted Discharge
Severity (m3/gasif ier-hr (
TAR SEPARATION
13.6
Tar Separation Waste
Gas (Flare Feed Stream)
7.00EOO
3.14E 02
1.56E 02
8.40E02
6.70E-01
4.10E03
4.91E03
6.64E-01
2.01E-04
8.20E-01
7.00EOO
7.60EOO
1.30E02
1.02E04
1.20EQ1
1.50EOO
8.50E-01
7.44E02
6 . 40EOO
2.06E04
28
1 25°C) 7.66E05
13.7 '
Phenolic Water
Tank Vent
4.00E-01
7.00E01
1.20E03
2.30E-01
4.12E03
1.82E04
4.02E-02
4.00E-02
3.40E-01
5.10E-01
1.81E02
1.20E04
2.30E01
2 . 80E-00
2.04E-05
4.63E02
3.81EOO
3.67E04
5.5
2.02E06
PHENOSOLVAN
1 14.5 '
Ammonia Stripper
Vent
2.00E-02
1.10E02
2.00E03
6.30E02
2.80E02
9.5E-03
2.00E-04
1 . 70E-03
8.41E-03
l.OOE-02
2.30E-06
1.40E03
1.61E04
5.31E02
2.07E04
260
5.39E06
BY-PRODUCT STORAGE
Naphtha Storage
Tank Vent
1.70EOO
NF
5.10E03
2.50E04
2.01E-04
2.00E-02
1.20EOO
6.73EOO
5.33E02
4.00E04
1.90E01
5.91E-01
5.30E-05
1.10E02
7.08E04
4.5
3.19E05
FLARE SYSTEM
I 1
Combined Gas
to Flare
1.23E01
5.43E02
1.76E02
9.84E02
1.40EOO
4.91E03
4.82E02
1.50EOO
2.01E-04
1.32EOO
5.42EOO
3.40EOO
1.00E01
6.80E02
2.12EOO
3.30E-01
5.30E-05
1.00E01
1.22E04
1330
1.62E07
-------
o
en
Naphtha Storage Tank Vent
Phenolic Water Tank Vent
Medium Oil Tank Vent
Ammonia Stripper Vent
L.P. Coal Lock Vent
Tar Tank Vent
COz-Rich Waste Gas Vent
2 3
Logio (TDS)
Figure 14. Key Kosovo gaseous discharge streams in order of decreasing TDS.
-------
CO2-Rich Waste Gas
NHs Stripper Vent
L.P. Coal Lock
Naphtha Storage Tank
Phenolic Water Tank
Medium Oil Tank
Tar Tank
E -1
Flow Component
TDS Component
i
6
Total Weighted Discharge Severity Log10 (TDS) + Log10 (Flow in m3/hr)
Figure 15. TWDS for key Kosovo gaseous discharge streams.
Ammonia
Sulfur Species
Phenol
BTX*
Hydrogen Cyanide
Carbon Monoxide
C-i-Ce Hydrocarbons
(Excluding Benzene)
E
Weighted Discharge Severity Log10(DS • Flow) in m3/hr
'Benzene, Toluene, and Xylenes
Figure 16. Most significant gaseous pollutants (plant-wide) in uncontrolled
discharge streams.
407
-------
The particulate data are shown in Table 6. This discussion will focus
on the results from the impinger collections and particularly on those
collected from the LP coal lock vent. This stream is emphasized because of
the potential environmental significance of the particulates that it
transports.
As indicated below, a major portion of the particulate catch from most
gaseous streams consisted of condensed organics (tars and oils):
Tars and Oils
Stream (Wt % in Particulates)
LP coal lock vent 90
Gasifier start-up vent 95
HP coal lock vent 69
Tar separation waste gases 72
Combined gas to flare 76
Analytical results are not yet available from these collections; however,
by-product analysis data can be used to make judgments about the
significance of these particulates. For example, the LP coal lock vent
discharge contained 8.1E06 yg/m3 of particulates of which 7.3E06 Vg/nr
were tars and oils. In order to provide an estimate of the PNA content of
the particulates in this stream, it was assumed that the PNA concentrations
in the condensed organic fraction of the particulates (tars and oils) were
the same as the PNA concentrations in the by-product tars and oils. Table 7
shows the concentrations of several of the most severe PNA's contained in
the light tar and the medium oil. Using the following data:
Concentration Mass Flow
LP Coal Lock Vent yg/m3 g/hr
Total Particulate 8.1E06 1.7E08
Tars and Oils in Particulate 7.3E06 1.5E08
Benzo(a)pyrene based on BaP in tar 1.5E03 3.2E04
Benzo(a)pyrene based on BaP in medium oil 0.5E03 1.0E04
the calculated concentration level of benzo(a)pyrene in the LP coal lock
vent discharge is in the range of 500 to 1,500 yg/m . This level of
PNA's will increase the TWOS of the LP coal lock vent significantly. The
effects of PNA's upon the TWDS values of key streams (using the average PNA
content of light tar and medium oil) are shown in Figure 17. Note that the
increase in TDS (and TWDS) by the inclusion of the PNA data elevates the LP
coal lock vent to the same order of magnitude as the ammonia stripper vent
and identifies it as the second most environmentally significant of the
uncontrolled discharge streams at Kosovo (excluding flare feed streams -
H2S-rich waste gas, and HP coal lock vent).
408
-------
TABLE 6. PARTICULATE CONCENTRATION DATA FOR KOSOVO GASEOUS STREAMS
STREAM TYPE:
SAMPLE POINT:
Dry Gas Flow Rate
(ra3/gasifler-hr @ 25°C)
Total Parttculate.
(mg/ra3 @ 25°C)
Condensed Organics
(Tars and Otis)
Dissolved Solids
Filtered Solids
DISCHARGE STREAMS
2.2 3.2
Low
Pressure
Coal Coal
Room Vent Lock Vent
7200 21
98 8100
** 7300
** 650
** 220
FLARE FEED STREAMS
3.3
Gasifier
Start-Up
Vent
*
9450
8980
400
61
3.6
High
Pressure
Coal
Lock Vent
230
960
660
240
61
13.6
Tar
Separation
Waste Gas
28
920
660
230
29
20.1
Combl ned
Gases
to Flare
1330
410
310
54
47
* - Variable Flow Rate.
** - Dry Stream; Analysis Not Applicable.
-------
-pa
O
H2S-Rich Waste Gas
NH3 Stripper
H.P. Coal Lock
CO2-Rich Waste Gas
Tar Sep. Waste Gas
Naphtha Storage Tank
Phenolic Water Tank
Low Pressure Coal Lock
E 0 1 2 3 4 5 6
Logic (WDS) + Logio (Flow)
Due to DS
| ] Due to Flow
Approximate
addition to
IDS due to
DS of PNA's
in particulate
8
Figure 17. Total weighted discharge severity for key Kosovo gaseous streams.
-------
TABLE 7. PNA'S IN KOSOVO LIGHT TAR AND MEDIUM OIL (Pg/g)
Compound
7, 12-Dimethylbenz(a)anthracene
Benz(a)anthracene
Benzo(b)fluorene
Benzo(a)pyrene
Dibenzo(a)anthracene
3-methylcholanthrene
252 Group
Light
Tar
1,100
490
310
210
23
26
950*
Medium
Oil
62
160
120
68
7
NF
280*
*Benzo(a)pyrene concentration = 24 Percent
Aqueous Streams; The two major aqueous waste streams in the Kosovo
Gasification Plant are:
• Gasification section (quenched ash) wastewater,
which is a combination of:
- ash quench water,
- coal bunker vent gas scrubber blowdown, and
ash lock vent gas scrubber blowdown; and
• Phenosolvan wastewater.
Water quality parameters and concentration data for anions and polynuclear
aromatics (PNA's) are presented in Table 8.
Gasification section wastewaters contain a variety of pollutants
including components leached from the ash or scrubbed from the coal bunker
or ash lock vents and components which enter the system along with the ash
quench and scrubber makeup water streams. The gasification wastewater has a
high pH (due to the alkaline nature of the Kosovo ash) and significant
concentrations of dissolved and suspended solids. Other components present
(e.g., phenols, NH3) indicate that at least a portion of the makeup water
used in these systems was derived from process condensate. The presence of
phenols and NH3 in the ash lock vent gases tends to confirm this
hypothesis since it would not be expected that phenols would be present in
any of the other process streams entering the Kosovo ash lock system. The
sulfur species detected in these wastewaters were present primarily in the
form of sulfate.
411
-------
TABLE 8. KOSOVO AQUEOUS STREAM DATA
PLANT SECTION:
SAMPLE POINT:
Design Flow Rate
(m-Vgasif ier-hr)
pH
Temperature (°C)
Solids Analysis (mg/L)
Total Solids
Suspended Solids
Dissolved Solids
Water Quality Paramters
COD (as mg 02 /L)
Permanganate (mg/L)
BOD5 (as mg 02 /L)
Aqueous Composition Data (mg/L)
TOC
Total Phenols
Volatile Phenols
Free Ammonia
Fixed Ammonia
Cyanide
Nitrites
Nitrates
Pyridines
Chlorides
Fluorides
Total Sulfur
Sulfites
Sulfates
Sulfides
Thiocyanates
Thiosulf ates
PNA Analysis (mg/L)
Benz(a)anthracene
7, 12-dimethylbenz(a)anthracene
Benzo(a)fluroanthrene
Benzo(a)pyrene
3-me thy 1 cho Ian thr ene
Dibenz(a,h)anthracene
252 Group (as BaP)
GAS PRODUCTION
12.3
Quenched Ash
Wastewater
3.0
0.1 - 12.1
10,900
8,760
2,100
1,460
8,060
90
—
-
0.17
Tr
1.9
0.01
0.40
4.8
-
28
0.91
-
Tr
495
Tr
0.026
Tr
_
-
-
-
-
-
PHENOSOLVAN
14.0
Phenolic
Water
>13
9.2
60
2,320
150
1,170
18,900
14.2
9,030
4,970
2,120
-
3,510
250
<1
<1
142
_
-
-
-
<75
—
0.92
0.23
0.68
0.19
<0.004
0.02
1.26
14.11
Phenosolvan
Wastewater
13
9.6
33
1,350
1,160
190
7,910
4,040
2,350
1,470
230
130
Tr
205
0.019
Tr
11.4
-
60
Tr
84
-
110
-
<75
Tr
NF
NF
NF
NF
NF
NF
0.19
Tr Trace
NF = Not Found
- = Not Analyzed
412
-------
The Phenosolvan wastewater stream data presented in Table 8 indicates
that a significant reduction in the organic pollutant loading is achieved in
the Phenosolvan section. As expected, the phenol level was reduced
significantly (by approximately 90 percent) by treatment in this section.
It should also be noted that the concentrations of several significant PNA's
were reduced to undetectable levels. The fate of the PNA's was not
confirmed since no sample of the by-product phenol was obtained. Presumably
this by-product stream was the vehicle by which the PNA's present in the
inlet water left the unit.
Although a significant portion of the phenolic material was removed from the
inlet water by the Phenosolvan unit, a significant amount of organic matter
remained in the discharge. This assertion is supported by the following
data from Table 8:
o TOG in outlet water - 1,470 mg/L.
o Phenols in outlet water - 230 mg/L.
o Volatile phenols in outlet water - 130 mg/L.
The level of volatile phenols in the outlet water significantly exceeds
the DMEG for aqueous discharges (DS Total Phenol = 2.6E04). Since the
composition of the unextracted TOC has not yet been determined, no realistic
assessment has been made of the characteristics of the bulk of this
material. However, in laboratory tests, a relatively large fraction of the
inlet TOC (30 percent) remains in the wastewater after extraction in the
laboratory with diethyl ether and methylene chloride at pH values of 1
and 12 (Ref. 6).
jaolid Streams; Solid phase analytical results are summarized in Table 9.
The data shown for the dried coal are based upon an average of
approximately 40 different spot samples taken over a several month period.
The ash values shown are averages of approximately 20 different samples
taken over the same period.
On the average, very high carbon conversion levels were achieved
(approximately 99 percent) in the Kosovo Lurgi gasifiers. This is expected
for a highly reactive coal such as the lignite being processed at Kosovo.
The ash from the gasifiers (after quenching) has a positive heating value,
but would not be classified as ignitable and, therefore, would not require
special handling in accordance with applicable RCRA criteria for ignitable
wastes.
Heavy tar is another solid waste stream produced in the Kosovo gasifi-
cation facility. Because of the high heating value of this stream as well
as the likely presence of highly toxic organic materials, such as phenols
and PNA's, this stream would probably be consumed in an on-site steam/power
boiler or incinerator in the U.S. At Kosovo, this stream is landfilled.
413
-------
TABLE 9. KOSOVO SOLID STREAM DATA
PLANT SECTION:
SAMPLE POINT:
Ultimate Analysis (wt. %)
Moisture
Ash
Carbon
Sulfur
Hydrogen
Nitrogen
Oxygen
Chlorine
Proximate Analysis (wt. %)
Moisture
Ash
Volatile
Fixed Carbon
C02
Total Sulfur
Free Sulfur
Hydrogen
Nitrogen/Oxygen
Chlorine
Heating Values (kcal/kg)
Proximate HHV
Proximate LHV
Ultimate HHV
Specific Gravity
GAS PRODUCTION TAR SEPARATION
2.0 12.1 12.2 13.8
Dry Wet
Dried Gasifier Gasifier
Coal Ash Ash Heavy Tar
20 2.1 not analyzed (moisture free analysis)
14 94 6.6
45 1.7 56.0
0.89 0.15 0.33
3.5 0.25 7.6
1.1 0.03 0.87
16 2.3 28.6
0.01 0.04
24 2.1 34 not analyzed
14 94 59
36 6.5 6.0
27 - 1.3
2.3 - 5.7
1.2 0.15 0.09
0.35 - 0.02
3.4 - 0.38
17 - 4.2 \f
0.01 - - '
3900 27.8
3700
4100 - - 6340
0.538
= No Data Available
414
-------
Product and By-Product Streams: The compositions of the products and
by-products will affect their final uses and their resulting environmental
impacts. Data for the crude and clean product gases are presented in Tables
2 and 3. Comparing the compositions of these streams indicates that the
Rectisol unit has removed almost all of the acid gases (CC>2, ^S, and
NH3) from the product gas.
Chemical analysis data for Kosovo by-products are shown in Table 10.
Table 11 presents a comparison of some ultimate analysis data for the feed
coal, heavy tar, and liquid by-products. Table 11 indicates that the sulfur
contents of the liquid by-products become progressively higher in the
"lighter" fractions. In contrast, the trend in the nitrogen values is
reversed. These data indicate that heavy hydrocarbon by-products similar to
those generated at Kosovo, could be used to satisfy some of the on-site fuel
needs (e.g., for steam generation) of a U.S. Lurgi plant without an FGD unit
if SC>2 emissions standards consistent with those for large fossil fuel
fired steam generators were applicable.
Trace Elements: The trace element concentrations in a number of the plant's
key feed, product, by-product, and waste streams were determined to
establish whether any of these streams contained elements at concentration
levels of concern. In addition, trace element leachabilities were evaluated
for the gasifier ash to determine whether this material would be classified
as an RCRA hazardous waste. Trace element concentration data are summarized
in Tables 12 through 14. These data include both SSMS results, which
provide a semiquantitative estimate of trace element concentrations on a
broad screening basis, and AA results, which provide more accurate estimates
of the concentrations of 15 selected elements. The elements selected for AA
analysis were those which were indicated to be present at levels of
potential concern by the SSMS results or through previous experience with
gasification process waste streams.
The levels of trace elements in the discharge from the LP Coal Lock
Vent shown in Table 13 are of particular interest. The concentration of
arsenic (1,700 yg/m3) is 850 times its DMEG. Other elements in the LP
Coal Lock Vent whose concentration exceeds, their DMEG values are chromium
(DS = 2.7E02), nickel (DS = 7.8EOO), cadmium (DS - 2.7EOO), beryllium
(DS = 2.0EOO), and mercury (DS = 1.1EOO). With the possible exception of
arsenic, these elements are probably being transported in the coal dust
which is contained in the discharge. A significant level of mercury was
found in the Phenolic water (Table 13). This value (0.14 mg/L) is 14 times
its DMEG for aqueous discharge.
The completion of trace element balances were outside the scope of the
Phase II effort. However, rough calculations of trace element distributions
were performed to provide some insight into the behavior of trace elements
in a Lurgi gasification system. These results are included in Table 13.
Most of the recovered trace elements which entered the gasifier with the
415
-------
TABLE 10. CHEMICAL AND PHYSICAL DATA FOR KOSOVO BY-PRODUCTS
By-Product: Light Tar
Specific Gravity
(g/cm3)
Higher Heating
Value (kcal/kg)
Lower Heating Value
(kcal/kg)
Ultimate Analysis (wt %)
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Chlorine
Oxygen (difference)
Moisure Content (wt %)
PNA Analysis (mg/kg)
Be nz ( a ) anthra cene
7 , 12-dimethylbenz (a)anthracene
Benzo(b)fluoroanthrene
Benzo(a)pyrene
3-methylcholanthrene
Dibenz(a ,h)anthracene
252 Group (as BaP)
1.06
8910
8280
82
8.4
1.3
0.49
0.22
7.8
1.1
490
1100
310
210
26
23
950
Medium Oil
0.97
9500
9400
82
8.9
1.00
0.83
0.03
8.2
0.8
160
62
120
68
NF
6.6
280
Naphtha
0.85
9940
8925
86
9.9
0.18
2.2
2.2
NF
NF
NF
NF
NF
NF
NF
NF = not found
= no data available
416
-------
TABLE 11. COMPARISON OF ULTIMATE ANALYSIS DATA FOR
SELECTED KOSOVO SOLIDS AND BY-PRODUCTS
Component
C
H
N
S
Ash
0
Moisture
HHV**
so2***
Dried
Coal
45
3.5
1.1
0.89
14
16
20
16.3
1090
Heavy
Tar*
56
7.6
0.87
0.33
6.6
29
-
26.5
250
Light
Tar
82
8.4
1.3
0.49
0.22
7.8
1.1
37.3
260
Medium
Oil
82
8.9
1.0
0.83
0.03
8.2
0.8
40.0
420
Naphtha
86
9.9
0.2
2.2
-
2.1
-
41.6
1060
* Moisture Free Analysis
** Higher Heating Value expressed as KJ/g.
*** Expressed as ng/J assuming 100% conversion of S to S02-
NOTE-S02 Emission Limitations for Large Fossil Fuel Fired Steam
Generators (40 CFR 60D):
Coal and Solid Fuels - 520 ng/J (1.2 lb/106 Btu)
Liquid Fuels - 340 ng/J (0.8 lb/106 Btu)
417
-------
TABLE 12. A SURVEY OF TRACE ELEMENTS IN KOSOVO STREAMS ANALYZED BY SSMS
SAMPLE POINT:
Trace Element
Ag
Al
As
B
Ba
Be
Bl
Br
Ca
Cd
Ce
Cl
Co
Cr
Cs
Cu
Dy
Er
Eu
F
Fe
Ga
Gd
Ge
Ho
I
K
La
Li
Ln
Mg
Mn
Mo
Na
Nb
Nd
Ni
Np
P
Pb
Pr
Rb
S
Sb
Sc
Se
Si
Sm
Sn
Sr
Tb
Te
Th
Ti
U
V
Y
Zn
Zr
2.0
Dried Coal
(rag/kg)
NF
>1000
2
21
110
NF
NF
2
>1000
0.4
3
32
0.4
11
0.1
8
NF
NF
<0.3
2
>1000
2
NF
0.1
NF
0.5
>1000
2
1
NF
>1000
230
6
>1000
3
0.8
23
NF
780
2
0.9
5
>1000
NF
1
0.6
>1000
1
0.5
91
NF
0.4
< 2
660
< 2
8
2
1
6
12.1
Dry Gasifier Ash
(mg/kg)
NF
>1000
62
190
>1000
4
NF
17
>1000
NF
29
45
4
2
3
27
2
0.5
1
= 710
>1000
17
2
0.5
0.6
2
>1000
21
28
NF
>1000
>1000
6
>1000
10
10
180
10
>1000
9
5
35
420
2
12
< 1
>1000
9
0.8
320
0.4
< 1
9
>1000
2
67
17
33
33
12.2
Wet Gasifier Ash
(mg/kg)
NF
—
—
630
1670
NF
NF
—
—
1.2
—
—
15
240
—
76
—
—
—
—
—
37
—
—
—
—
—
NF
—
—
—
2700
30
—
—
—
180
—
—
27
—
—
—
NF
20
—
—
—
NF
4100
—
—
—
2300
—
140
39
56
180
15.2
Medium Oil
(mg/L)
NF
0.09
0.4
0.07
0.09
NF
0.01
NF
5
0.01
0.003
0.008
0.004
0.02
NF
0.5
NF
NF
NF
=0.03
2
NF
NF
NF
NF
NF
0.3
NF
0.001
<0.004
>10
0.02
0.005
0.1
NF
NF
0.03
NF
0.1
0.09
NF
NF
0.6
NF
<0.001
0.02
2
NF
0.008
0.008
NF
NF
<0.02
0.09
0.07
0.01
0.003
0.3
<0.003
14.11
Phenosolvan
Wastewater
(mg/L)
NF
0.1
0.02
0.1
0.05
NF
NF
0.009
6
NF
NF
0.08
0.003
0.005
NF
0.03
NF
NF
NF
=0.02
0.5
NF
NF
0.03
NF
0.02
1
NF
0.003
NF
2
0.01
NF
4
NF
NF
0.08
NF
0.08
0.07
NF
NF
>10
NF
<0.005
0.03
1
NF
0.009
0.02
NF
NF
<0.04
0.02
<0.03
0.003
<0.03
0.7
0.02
NF - not found
= no data available
418
-------
TABLE 13. TRACE ELEMENTS IN KEY KOSOVO STREAMS
ANALYZED BY ATOMIC ABSORPTION SPECTROMETRY
SAMPLE POINT:
Trace Element
As
Be
Cd
Ce
Cr
Cu
Hg
Mo
Ni
Pb
Sb
Se
Sr
Tl
V
Zn
1 2.
Dried
Concentration
(mg/Kg)
59
1.0
4.0
3.4
87
43
0.74
6.4
150
8.2
NF
20
190
NF
14
140
0
Coal
Mass Flow
(g/hr)
940
16
64
54
1400
690
12
100
2400
130
NF
320
3000
NF
220
2200
SOLIDS
12.1
Dry Gasifier
Concentration
(mg/Kg)
75
2.5
69
17
180
40
0.30
8.9
320
52
NF
24
370
NF
100
2.1
Ash
Mass Flow
(g/hr)
200
6.8
190
46
490
110
0.82
24
860
140
NF
65
1000
NF
270
5.7
13.
Heavy
Concentration
(mg/Kg)
16
0.29
3.7
1.5
30
6.0
0.64
0.85
21
64
3.9
2.6
41
NF
5.7
98
8 1
Tar
Mass Flow
(g/hr)
1.6
0.029
0.37
0.15
3.0
0.60
0.064
0.085
2.1
6.4
0.39
0.26
4.1
NF
0.57
9.8
NF = below detection limits
(Continued)
-------
TABLE 13 (Continued).
ro
o
TRACE ELEMENTS IN KEY KOSOVO STREAMS
ANALYZED BY ATOMIC ABSORPTION SPECTROMETRY
LIQUID BY-PRODUCT
SAMPLE POINT:
Trace Element
As
Be
Cd
Co
Cr
Cu
Hg
Mo
HI
Pb
Sb
Se
Sr
Tl
V
Zn
15.1
Light Tar
Concentration (Mass Flow
(mg/kg) (g/hr)
1.7E+01
9.0E-02
6.6E-01
NF
3.0E 00
1.6E+01
NF
NF
9.0E-00
6.8E 00
NF
1.6E 00
2.0E+01
NF
NF
2.8E401
6.8E 00
3.6E-02
2.64E-01
NF
1.2E 00
6.4E 00
NF
NF
3.6E 00
2.7E 00
NF
6.4E-01
8.0E 00
NF
NF
1. 1E-HJ1
15.2
Medium Oil
Concentration {Mass Flow
(mg/kg) (g/hr)
2.0E 00
NF
7.7E-02
2.0E-01
4.0E 00
LIE 00
2.0E-01
1.9E-01
NF
1.4E 00
NF
1.9E 00
8.6E 00
NF
NF
1.5E+01
5.0E-01
NF
1.9E-02
4.5E-02
l.OE 00
2.8E-01
5.2E-02
4.8E-02
NF
3.5E-01
NF
4.8E-01
2.2E 00
NF
NF
3.8E 00
15.3
Naphtha
Concentration [Mass Flow
(mg/kg) (g/hr)
5
1
8
5
1
1
1
9
1
6
1
7
1
.5E-01
.8E-03
.OE-04
.OE-03
.OE 01
.5E-01
.3E-01
.OE-03
.4E-01
.4E-02
.2E-02
.3E-01
NF
NF
NF
.4E-01.
8
2
1
7
1
2
2
1
2
9
1
1
2
.5E 02
.7E-04
.2E-04
.7E-04
.5E 02
.4E-02
.OE-02
.4E-03
.1E-02
.8E-03
.9E-03
. 1E-01
NF
NF
NF
.1E-02
HATER
14
Phenosolvan
Concentration
(mg/L)
l.OE-01
NF
1.4E-03
NF
2.3E-02
1. 1E-02
1.4E-01
NF
1.3E-02
1.4E-02
NF
5. OE-02
l.OE-01
NF
NF
2.8E-01
.0
Inlet Water
Mass FLow
(g/hr)
1
1
3
1
8
1
1
6
1
3
.3E 00
NF
.8E-02
NF
.OE-01
.4E-01
.2E 00
NF
.7E-01
.8E-01
NF
.5E-01
.3E 00
NF
NF
.6E 00
NF - Not Found (below detection Halts)
(Continued)
-------
TABLE 13 (Continued).
ro
TRACE ELEMENTS IN KEY KOSOVO STREAMS
ANALYZED BY ATOMIC ABSORPTION SPECTROMETRY
GASES
SAMPLE POINT:
Trace Element
As
Be
Cd
Ce
Cr
Cu
Hg
Mo
Hi
Pb
Sb
Se
Sr
Tl
V
Zn
1 3.2
Low Pressure
Lock Vent
Concentration
(mg/L)
1.7E-03
4.0E-06
2.7E-05
4.9E-06
2.7E-04
1.8E-04
5.3E-05
4.5E-05
1.2E-04
7.2E-05
NF
NQ
6.1E-04
NF
9.0E-06
1.6E-03
Coal
Mass Flow
(B/hr)
3.6E-02
8.4E-05
5.7E-04
l.OE-04
5.7E-03
3.8E-03
1.1E-03
9.5E-04
2.5E-03
1.5E-03
NF
NQ
1.3E-02
NF
1.9E-04
3.4E-02
20.1
Combined Gas
Concentration
(mg/L)
1.9E-06
NF
2.4E-07
1.7E-07
NF
5.8E-06
NF
NF
7.5E-06
l.OE-06
NF
7.2E-06
4.4E-06
NF
NF
3.1E-05
1
to Flare
Mass Flow
(g/hr)
2.5E-03
NF
3.2E-04
2.3E-04
NF
7.7E-03
NF
NF
l.OE-02
1.3E-03
NF
9.6E-03
5.9E-03
NF
NF
4.1E-02
Percentage of Amount
Found in Dried Coal Accounted
For in the Streams Listed
in this Table
22
43
298
85
35
17
23
24
36
115
-
21
34
-
123
1.5
NF = below detection limits
NQ = present but not quantifiable
-------
TABLE 14. TRACE ELEMENTS IN KOSOVO ASH LEACHATES ANALYZED BY SSMS
RCRA LKACHATE
(Acid)
Trace
Element
Al
As
B
Ba
Be
Bi
Br
Cd
Ce
Cl
Co
Cr
Cs
Cu
Dy
Er
Eu
F
Fe
Ga
Gd
Ge
Ho
I
La
LI
Mg
Mn
Mo
Na
Nb
Ni
Np
P
Pb
Pr
Rb
S
Sb
Sc
Se
Si
Sm
Sn
Sr
Tb
Te
Th
Ti
U
V
Y
Zn
Zr
Composition
(mg/L)
0.01
<0.004
0.09
3
NF
NF
<0.008
NF
NF
0.05
<0.001
0.3
0.004
0.01
NF
NF
NF
0.8
10
NF
NF
<0.001
NF
NF
NF
0.03
2
0.001
0.1
>2
NF
0.04
NF
0.02
0.008
NF
0.04
>6
<0.002
<0.001
0.01
8
NF
<0.001
4
NF
NF
<0.008
0.01
•C0.007
0.07
0.008
0.05
<0.006
D.S.
Value
1.30E-04
<1.60E02
1.91E-03
6.00E-01
-
-
0.0
-
-
3.84E-05
<1.33E-03
1.20E 00
3.33E-06
2.00E-03
-
-
-
2.10E-02
6.70E 00
-
-
<1.20E-04
-
-
-
9.10E-02
2.22E-02
4.00E-03
1.33E-03
>2.50E-03
-
1.73E-01
-
1.33E-02
3.20E-02
-
2.22E-05
0.0
<2.70E-04
<1.30E-06
2.00E-01
5.33E-02
-
0.0
8.70E-02
-
-
<1.30E-03
1.11E-04
<1.20E-04
2.80E-02
5.33E-04
2.00E-03
<8.00E-05
ASTM LEACHATE
(Neutral)
Composition
(mg/L)
2
0.01
0.1
0.05
NF
NF
0.4
NF
NF
0.7
<0.007
0.5
NF
0.03
NF
NF
NF
7
0.1
0.02
NF
0.01
NF
0.005
NF
0.07
NF
0.02
0.05
NF
0.006
0.02
NF
0.2
0.07
NF
0.09
NF
NF
<0.003
0.007
7
NF
NF
0.3
NF
NF
<0.04
0.02
<0.03
0.004
NF
0.08
NF
D.S.
Value
2.50E-02
4.00E-02
2.12E-03
l.OOE-02
-
-
0.0
-
-
5.40E-04
<9.33E-03
2.00E 00
-
6.00E-03
-
-
1.84E-01
6.70E-02
2.70E-04
-
1.20E-03
-
0.0
-
2.12E-01
-
8.70E-02
6.70E-04
-
1.81E-05
8.70E-02
-
1.33E-02
2.80E-01
-
5.00E-05
-
-
O.74E-06
1.40E-01
4.70-E02
-
-
6.52E-03
-
-
<6.34E-03
2.22E-04
<5.00E-04
1.60E-03
-
3.20E-03
"
NF - Not Found
422
-------
feed coal were found in the gasifier ash. The recovery values shown are
based upon the use of plant design flow data for the feed, by-product, and
waste streams considered. The only trace elements found in any significiant
concentrations in streams other than the dry gasifier ash are antimony and
lead (in the heavy tar), and copper (in the by-product naphtha). Very poor
calculated recoveries were obtained for most of the trace elements (on the
order of 20 to 40 percent). Zinc recoveries were particularly poor, with
less than 5 percent of the coal input zinc accounted for. These poor
recovery values are probably the result of several factors including:
actual stream flow measurements were not obtained for many of the streams,
time phased sampling was not attempted, and a statistically significant data
base was not obtained.
The largest solid waste stream generated in a Lurgi gasification plant
is quenched gasifier ash. In order to determine the leaching characteris-
tics of this material and to predict its classification under RCRA guide-
lines, a series of leaching studies were conducted. The results of these
tests, which are reported in Table 14, indicate that no trace elements were
present in the ash leachate in sufficient concentrations which would cause
this material to be classified as hazardous.
A Comparison of Discharge Streams Plant Wide: TWDS values for all major
discharge streams - aqueous, gaseous, and solid - are shown in Figure 18.
Attention is called to the flow rate units: liters per gasifier hour -
aqueous streams; cubic meters per gasifier hour - gaseous streams; and
kilograms per gasifier hour - solid streams. These are the units of the
DMEGS used. Figure 18 shows the streams prioritized in each discharge
medium according to their TWDS values.
Mass Balances for Key Species; Figure 19 summarizes the results of mass
balance calculations for carbon, sulfur, and nitrogen species in the Kosovo
plant. The amount of carbon found in key Kosovo solid, liquid, and gaseous
streams, expressed as a percentage of the carbon entering the gasifier in
the dried coal indicates that the majority of the carbon entering the system
with the dried coal leaves in gaseous streams. It is significant that there
is almost as much carbon (mainly as (X^) in the t^S-rich waste gas flare
feed stream (88 vol. % CC>2) as there is in the C02~rich waste gas stream
(94 vol. % C02). Small quantities of the inlet carbon ends up in the
gasifier ash (0.7%), aqueous wastewaters (0.3%), and the remaining gaseous
discharge streams (excluding the C02~rich waste gas stream).
Most of the sulfur leaves the plant in the I^S-rich waste gas stream.
Of the remaining sulfur, the majority appears in the by-products - naphtha
(1.5%), medium oil (1.1%), and light tar (1.1%) - and the ammonia stripper
vent (3.7%). A small percentage is discharged in the ash (1.3%), heavy tar
(0.2%), and aqueous wastewaters (0.9%). The relatively poor accountability
of the sulfur balance is probably due to variations in the input coal sulfur
content, variations in flow rate measurements, and the lack of time-phased
sampling.
423
-------
Aqueous - Log TDS+ Log (l/hr)
Phenosolvan
Ash Quench
Y///////////////////7A
Y/////////////7/7A
IND
Gaseous - Log TDS+Log (m3/hr)
Ammonia Stripper
CO2 Rich Vent
Naphtha Storage
Phenolic Water Tank
LP Coal Lock Vent
Medium Oil Tank
Tar Tank
Y/////////A
W/A
Y/////////A
frfrfr] Flow Component
I I DS Component
Solid - log TDS+log (kg/hr)
Heavy Tar
Dry Ash*
Y/////////////////A
01
Logio
* DS Based on RCRA Leachate
23456
(IDS) + Logio (Flow)
Figure 18. Total weighted discharge severity of uncontrolled
Kosovo discharge streams.
-------
ro
en
180%
CSN
Dried Coal
and
Oxygen
C S N
C S N
CSN CSN CSN CSN
CSN
CSN
Clean Liquid Solid Aqueous F|are CO2 Rich Ammonia Olher
Product By-Producls Discharges Discharges Streams Waste Qas Stripper Gaseous
Gas Vent Discharges
Qas
CSN
Total
Figure 19. Summary of Carbon, Sulfur, and Nitrogen Mass Balance Results for the Kosovo Plant
-------
Nitrogen entering in the dried coal and oxygen feed streams is
converted primarily to ammonia, hydrogen cyanide, a number of organic
nitrogen compounds, and N£- Most of this nitrogen appears in gaseous
discharge streams. A large percentage is discharged in the ammonia stripper
vent (which contains 41.8 vol. % NH^ on a dry basis).
Summary and Conclusions; The Kosovo Phase II data has corroborated
substantially the indications from the Phase I test results and has also
added significant new information about the aqueous and solid discharges
from the Kosovo plant. It has also provided significant information about
trace pollutants, both organic and inorganic. The following are some of the
more salient findings:
• All process units studied have a significant potential
for polluting the environment.
• The highest priority streams in each medium are:
H2S-rich waste gas,
Phenosolvan wastewater, and
heavy tar.
• The CC>2-rich waste gas may contain significant
levels of nonmethane hydrocarbons and mercaptans.
• PNA's make a significant contribution to the discharge
severity (DS) of tar-bearing streams (e.g., LP Coal Lock
vent and heavy tar).
The severity of the coal lock vent discharge is increased
significantly by the contribution of PNA's in the tar
aerosols.
• Benzo(a)pyrene and 7,12-Dimethylbenz(a)anthracene are the
two most significant (highest D.S. values) pollutants in
Kosovo tar.
• Trace elements appear to be less significant than
trace organics as pollutants in organic containing
streams.
• Ash leaching problems appear to be of low concern.
Concentrations of all trace elements were at least
an order of magnitude lower in the RCRA leach test
results than those levels specified in the
EP toxicity test.
• After Phenosolvan treatment, the treated process
condensate contained undetectable levels of PNA's,
but high residual organic material concentrations and
high solids concentrations.
426
-------
ACKNOWLEDGMENT
This work was sponsored by the Industrial Environmental Research Lab-
oratory of the United States Environmental Protection Agency. The authors
express their thanks to the following organizations and individuals for
their contributions to this work:
U.S. EPA - T. Kelly Janes, W. J. Rhodes
Radian Corporation - R. V. Collins, R. A. Magee, G. C. Page,
K. Schwitzgebel, E. C. Cavanaugh,
G. M. Crawford
Rudarski Institute - M. Mitrovic, D. Petkovic
Kosovo Institute - B. Shalja, Amir Kukaj, Mile Milesavljevic
REMHK Kosovo - Shani Dyla, Emlia Boti
INEP - S. Kapor
REFERENCES
1. Becir Salja and Mira Mitrovic, Environmental and Engineering Evaluation
of the Kosovo Coal Gasification Plant - Yugoslavia (Phase I), Symposium
Proceedings: Environmental Aspects of Fuel Conversion Technology IV,
EPA-600/7-79-217, April 1979, Hollywood, Florida.
2. Bombaugh, Karl J. and William E. Corbett, Kosovo Gasification Test
Program Results - Part II: Data Analysis and Interpretation,
Symposium Proceedings: Environmental Aspects of Fuel Conversion
Technology IV, EPA-600/7-79-217, April 1979, Hollywood, Florida.
3. Bombaugh, Karl J., W. E. Corbett, and M. D. Matson, Envionmental
Assessment: Source Test and Evaluation Report - Lurgi (Kosovo)
Medium-Btu Gasification, Phase I, EPA-600/7-79-190, August 1979.
4. Schalit, L. M. and K. J. Wolfe, SAM/1A: A Rapid Screening Method
for Environmental Assessment of Fossil Energy Process Effluents.
Acurex Corporation/Energy and Environmental Division, Mountain View,
California. EPA Contract Number 600/7-78-015 (NTIS Number PB 277-088).
February 1978.
5. Cleland, J. G. and G. L. Kingsbury, Multimedia Environmental Goals
for Environmental Assessment, Volumes I and II (final report).
Research Triangle Institute, Research Triangle Park, North Carolina.
Report Number EPA-600/7-77-136a, b, NTIS Number PB 276-920 (Volume II).
EPA Contract Number 68-02-2612. November 1977.
6. Collins, R. V., K. W. Lee, and D. S. Lewis, Comparison of Coal
Conversion Wastewaters, Symposium Proceedings: Environmental Aspects
of Fuel Conversion Technology V, September 1980, St. Louis, Missouri.
427
-------
AMBIENT AIR DOWNWIND OF THE KOSOVO GASIFICATION COMPLEX:
A COMPENDIUM
Ronald K. Patterson
Aerosol Research Branch
Atmospheric Chemistry and Physics Division
Environmental Sciences Research Laboratory
Environmental Protection Agency
Research Triangle Park, NC
ABSTRACT
In an attempt to obtain environmental impact data for a com-
mercial scale coal gasification facility the Environmental
Sciences Research Laboratory-RTF (ESRL-RTP) Aerosol Research
Branch, conducted a 16-d continuous ambient air study in the
Region Kosovo, Yugoslavia. Five sampling sites were established
around and ~2 km outside the fence line of the Kosovo medium BTU
Lurgi gasification complex.
Organics in total particulate matter; total and fine particle
maSjS,, inorganics, and elemental species; trace metal in size-
fractionated particles; and vapor phase organics were deter-
mined. Physical and chemical analyses were carried out on parti-
culate matter using gravimetric analysis, ion chromatography, and
scanning electron microscopy. Elemental analysis was done using
the inductively coupled argon plasma emission technique, proton-
induced X-ray emission, and combustion analysis. Both particle
catches and vapors trapped on Tenax resins were subjected to
organic analysis using gas chromatography. The chromatographic
fractions were identified and quantified usiftg flame ionization
detection, sulfur and nitrogen specific detectors, and mass spec-
trometry. A comprehensive quality assurance and quality control
program was implemented to ensure the validity of the samples col-
lected and analyzed.
A number of United States and Yugoslavian laboratories parti-
cipated in the ambient air sampling and analysis phases of this
study. This paper is a compendium of the major results and con-
clusions obtained by the participant laboratories.
428
-------
INTRODUCTION
The Environmental Sciences Research Laboratory-RTF (ESRL-RTP)
Aerosol Research Branch conducted an ambient air study near the
commercial medium BTU Lurgi coal gasification plant located in the
Kosovo Region of Yugoslavia. The objectives of the study were to
characterize the ambient aerosols and volatile organic pollutants
downwind of the Kosovo complex, to correlate specific pollutants to
the gasification plants, and to evaluate the impact of the Lurgi
gasification process on the air quality downwind of the Kosovo
complex. This study represents Phase III of the Industrial
Environmental Research Laboratory-RTF (IERL-RTP) multimedia
assessment program at the Kosovo complex.
The Kosovo Industrial Complex (Kombinat Kosovo) consists of a
coal processing facility, a coal gasification plant (six Lurgi
gasifiers), a fertilizer plant, a steam plant, a 790 MW lignite
burning power plant, and a gasification process by-product storage
area. Major activities outside the complex are lignite coal
mining, lignite ash disposal (piles) , and farming. Forty-eight
trains (27 diesel and 21 steam) pass along the southern edge of the
Kosovo complex daily. Several improved analytical techniques and
123
procedures were developed by Radian Corporation ' ' and by the Oak
Ridge National Laboratory4'5'6 in anticipation of difficulty in
differentiating the complex sources in the area.
429
-------
SAMPLING STRATEGY
Five sampling sites were located around and =*2 km outside the
fence line of the Kosovo complex. Using the stack of the steam
plant as a center reference point and Yugoslav wind direction data
for the month of May (average winds from Northeast) , the sampling
stations were deployed in a manner indicative of prevailing upwind,
downwind, and crosswind locations (see Figure 1).
Each sampling station was equipped to collect total suspended
particulate (TSP) matter for organic analysis; total (<15 ym) and
fine (<2.0 urn) particles for gravimetric, inorganic, and elemental
analysis; size-fractionated particles for elemental analysis; and
organic vapors. The sample collection equipment at each station
consisted of:
1. one 24 h HiVol sampler (1.1 m /min) using a 265 mm
diameter Gelman Microquartz filter and a HiVol motor
exhaust filtration system ;
Q
2. one 24 h Tenax vapor trap system (4 1/min) which taps
into the post-filter section of the HiVol sampling head;
3. one 6 h LoVol sampler (28 1/min) using two 47 mm diameter
Gelman Microquartz filters, one total (<15 ym) and one
fine (<2.0 ym) , preceeded by a Southern Research Insti-
Q
tute - Cyclone II ;
4. one 6 h modified Battelle cascade impactor (1 1/min);
5. one 7 d time-phased aerosol sampler1 (=2 1/min); and
6. one Sears 3 kw gasoline electric power generator posi-
tioned 40 m downwind of the sampling equipment.
430
-------
(15°)
o
SAMPLE
SITE
<•' \ CENTER
\ FERTILIZER \ REFERENCE
\ ru«mi , POINT \
^ '* \ A
V '' \ \ ^X \
\ ^ \ W
' ^^ \ ^L STEAM\
r V-" V 1 PLANT \
COAL
FIRED
UTILITY
PLANT
1 KILOMETER
(15QO|
O
SAMPLE
SITE
£3
Figure 1. Schematic of the Kosovo complex with the five
sampling sites indicated (Reference 1).
431
-------
Site No. 3 was equipped with a Bendix Aerovane (6 blade) wind
speed, wind direction, and time system. Site No. 5 was equipped
with a Climatronics meteorological station and a Datel Data Logger
II magnetic tape system which recorded wind speed, wind direction,
solar flux, barometric pressure, temperature, and time. The
meteorological data from Site No. 5 were used to calculate percent
downwind values for each site location.
Mass measurements on LoVol filters were made on a Mettler
Model ME 30/36 Electronic Microbalance. Quality assurance audits1
covering sample collection media preparation, equipment calibra-
tion and operation, initial and final gravimetric measurements,
sample storage and transport, and sample documentation were con-
ducted daily by on-site personnel representing the prime contrac-
tor, Radian Corporation. All aspects of sample collection and
handling, except quality assurance/quality control, were carried
out by Yugoslav personnel under American supervision.
Sampling began at 0000 h on May 14 and ended at 2400 h on May
29, 1979. Approximately 3000 samples were collected during the
study. The samples were distributed between several investigators
for analysis (see Table 1).
ANALYSIS STRATEGY
The objectives of the analysis program were to analyze the
aerosols and vapors collected in the vicinity of the Kosovo
complex, and to compare the ambient air results with those obtained
from the analysis of Kosovo gasification process emissions and
by-product streams. To accomplish these objectives four integrated
courses of analysis were followed: (1) physical characterization
432
-------
Table 1. SAMPLES COLLECTED AND RESPECTIVE RECIPIENTS3
ORGANIZATION
INEPb
RADIAN
ORNL
FSU
EPA/GKPB
TOTALS
HIVOL
FILTERS
42
23
22
87
ORGANIC
VAPOR
TRAPS
83
42
42
167
BATTELLE
IMPACTOR
DISC SETS
157
161
318
STREAKER
SAMPLER
SLIDES
6
12
18
LOVOL
FILTERS
316
326C
642
GRAB
SAMPLE
BOMBS
3
3
6
aFROM REFERENCE!.
bINEP (INSTITUT ZA PRIMENU NUKLEARNE ENERGIJE, BELGRADE, YUGOSLAVIA).
CTHE OREGON GRADUATE CENTER RECEIVED TWO SECTIONS FROM EACH LOVOL
FILTER IN RADIAN'S POSSESSION.
433
-------
of the aerosol; (2) carbon spteciat,ion of the aerosol; (3) inorganic
analysis of the aerosol; and (4) organic analysis of the species in
the vapor phase and adsorbed on the aerosol.
The percentage of time that each station was located downwind
from the gasification plant was of interest for the purpose of
correlating identified chemical species with their source (s). The
reduction and analysis of the Climatronics meteorological data
indicated that Site No. 4 was the predominant downwind location
(=40%) and that Sites No. 1 and No. 5 were the predominant upwind
locations (==1%) . Site No. 3 (=20%) was an intermediate location.
Samples from Sites 1, 3, and 4 received first priority for
screening and analysis.
ANALYSIS RESULTS
Physical Analysis
Gravimetric data showed that the ambient aerosol loadings
(both <15 ym and <2.0 i_im) were significantly greater downwind of
the Kosovo complex than upwind (note Figure 2). The increase was
greater for the coarse (total minus fine) aerosol fraction than for
the fine fraction. The particulate matter collected downwind of
the complex appeared to be mineral; only small amounts (<1%) of
typically spherical fly ash material were observed. The latter
result indicates that the sampling stations were located in areas
least affected by the plume of the Kosovo power plant.
Carbon Analysis
Carbon speciation analysis by Huntzicker, et al. (Oregon
Graduate Center) showed the coarse aerosol fractions from Site
434
-------
327
456
390
300
10
20 30 40 50
SITE DOWNWIND TIME, percent
60
Figure 2. A Plot of the <15 /xm and <2.0 ^m particle fractions versus percent
downwind. The lines are the linear-least-squares plot of the data (Reference 1).
70
435
-------
No. 4 (downwind) to exhibit a very strong periodicity in elemental
carbon concentration. This peaking always appeared at night.
Weaker periodicities were observed for coarse organic carbon and
total (<15 urn) carbonate carbon. Site No. 4 also exhibited a
daytime peaking trend in organic carbon concentration in the fine
aerosol fraction. An explanation for this pattern has not yet been
developed. The elemental carbon and organic carbon in the fine
fraction were weakly correlated (r = 0.36) at Site No. 4,
suggesting a multiplicity of sources and poor mixing. At Site No.
5 (upwind), the organic and elemental carbon in the fine fraction
were strongly correlated (r = 0.77). For all sites and sampling
periods, when the percent of time downwind was <5%, the correlation
coefficient was 0.63. The latter two results indicate well aged
aerosol similar to aerosol sampled at urban U.S. sites. The high
concentrations of carbonate carbon (up to 12 yC/m ) observed during
many of the high mass loading periods suggest blowing coal dust.
This is a reasonable assumption in that Kosovo lignite is rich in
12 1
carbonate. Total carbon analysis data obtained by Radian showed
that a higher percentage of carbon was collected downwind of the
plant and that the additional carbon was >2.0 ym in diameter.
Upwind, -70-80% of the carbon was in the <2.0 ym fraction.
Inorganic Analysis
Preliminary data from Boueres, et al. (Florida State Univer-
sity) on the time phase streaker sampler at Site No. 4 showed
regular daytime peaking of sulfur and iron as well as lead and
zinc. These element pairs are not synchronous but may be related
to the peaking seen by Huntzicker, a possibility now being investi-
436
-------
gated. There are some indications of photochemical activity and
sulfur transformation chemistry. However, this inference will
remain speculative until more definitive data are obtained through
a more detailed analysis of the impactor and streaker data bases.
Figure 3 shows the average background concentration of sulfur
at all sites to be on the order of 2 yg/m . With the exception of
Site No. 5, all sites show maxima in [S] occurring at different
times of the day between 1200 and 1800 h. Each maximum of [S~]
appears to be composed of a distinct peak superimposed on a smooth
(bell shaped) maximum. From the wind direction and site position
information, we hypothesize that the distinct peak may be associ-
ated with direct emission plus rapid heterogeneous transformation
within the plume. The other two components (the background and the
smooth maximum) may be associated with homogeneous nucleation, slow
heterogeneous reaction, and resuspension of particles deposited in
the soil.14
Preliminary assessment of the Kosovo samples thus far suggests
that most of the observed trace metal aerosol components were
derived from sources other than the coal gasification plant.
Radian's ' inorganic analyses also show no correlations between
concentration and percent downwind from the coal gasification
facility for any soluble (Na+, Nfit, NdZ, Cl~, and SOT) or elemental
species except total carbon (discussed above). Iron, lead, and
zinc data analyses are incomplete at this time.
Organic Analysis
The Tenax resin cartridges analyzed by Radian showed organic
species in the volatility range from benzene to pyrene. Benzene
437
-------
7
6
5
4
3
2
I
0
*
8
7
6
•V5
s 4
*""" 3
i—i
ICO ,
0
SULFUR
0
Jl
SITE I
6 12 18 24 ~ 0 6 12 18 24
1_
SITE 4
0
6 12 18 24 0 6 12 18 24
5
4
3
2
I
0
SITE 5
ALL SITES
0 6 12 18 24 0 6 12 18 24
TIME(h)
Figure 3. Histograms showing the average daily pattern of
aerosol sulfur concentrations at all sites individually and
their overall grand average. Plotted are the 2-h averages
of [S] for the 15.5 days of sampling (14-26 May 1979)
(Reference 14).
438.
-------
and toluene (and possibly other volatile species) experienced
breakthrough and were not quantifiable, but xylenes and all heavier
compounds were quantitatively collected. There is a clear distinc-
tion (with some overlap) between the organic compounds adsorbed on
the particulate matter caught on the HiVol filter and in the vapors
sorbed on the Tenax resin. The vapors spanned benzene (MW 78) to
pyrene (MW 202) . The filter samples contained polynuclear aromatic
hydrocarbons (PAH's) from naphthalene (MW 128) through the benzo-
pyrene isomeric group (MW 252).
Mass spectrometric analysis of Tenax and filters samples
succeeded in tentatively identifying more than 50 organic compounds
and isomeric groups in the ambient air downwind of the Kosovo
Industrial Complex. The list of identified compounds includes:
alkylated benzenes through C. substitution, polyaromatic hydro-
carbons (PAH's) and alkylated PAH's through benzopyrenes, linear
and heterocyclic hydrocarbons, phenols,, ketones, alkylated pyri-
dines and quinolines, alkylated thiophenes, and dibenzofuran. Some
of the volatile organic compounds detected in the ambient air were
identical to some of the compounds found in certain emissions from
the coal gasification plant (see Figures 4-7) .
Quantification by mass spectrometry and flame ionization
detection placed the maximum, individual [poncentrations of naphtha-
lene in the vapor phase and benzopyrene isomer group adsorbed on
o 3
the particulate matter at 8 ug/m and 0.08 ug/m , respectively,
when extrapolated to 100% downwind. The basis for such an extra-
polation is shown in Figure 8. Comparison of measured concentra-
tions with Ambient-Multimedia Environmental Goal (A-MEG) values
439
-------
TENAX#1022
(DOWNWIND)
10
15
20
10
15
20
25
RETENTION TIME, min.
Figure 4. GC-HECD sulfur compound profiles for a downwind Tenax
vapor trap extract (#1022, day 6, site #4) and for Kosovo medium
oil (Reference 1).
440
-------
TENAX#1044
(BLANK)
10
15
20
TENAX#1010
(UPWIND)
10
RETENTION TIME, min.
15
20
Figure 5. GC-HECD sulfur compound profiles for an upwind (#1010,day 6, site
#1) and a blank (#1044) Tenax vapor trap extract (Reference 1).
441
-------
10 15 20
RETENTION TIME, min.
Figure 6. GC-HECD nitrogen compound profiles for a downwind
Tenax vapor trap extract (#1022) and for Kosovo medium oil
(Reference 1).
442
-------
I I
TENAX#1010
(UPWIND)
10
15
20
25
TEN AX #1044
(BLANK)
RETENTION TIME, min.
Figure 7. GC-HECD nitrogen compound profiles for an upwind (#1010) and a blank (#1044) Tenax
vapor trap extract (Reference 1).
443
-------
M
E
Z 6
HI
o
o
o
ui 4
Z
UJ
a.
| 2
CO
/
X
0 10 20 30 40 50 60 70 80 90 100
SITE DOWNWIND TIME, percent
Figure 8. Correlation of organic loading in Kosovo ambient air with the percent of time the
sampling site was downwind of the coal gasification plant (Reference 1).
444
-------
indicates that certain species (e.g., benzopyrene isomer) may cause
harmful health effects. A-MEG's are target value ambient air
concentration levels below which the component is of low concern
for its potential effects.
Griest, et al. (Oak Ridge National Laboratory) used analyti-
cal procedures different from those of Radian and observed 120
vapor phase organics in the ambient air surrounding the Kosovo
Industrial Complex. The 28 major components are listed in Table 2.
The majority of the vapor phase organics were C.-C^ alkyl-substi-
tuted benzenes. Also present were diaromatics (such as naphtha-
lenes and biphenyl) and several oxygenated species (such as benzal-
dehyde, acetophenone, phenol, and the cresols). Concentrations of
individual constituents ranged from 0.02 to 9.0 yg/m , with
toluene, phenol, benzaldehyde, and acetophenone being the major
species in the vapor phase samples. Naphthalene, phenol, and the
cresols were more concentrated in samples collected downwind of the
gasifiers. Blanks were virtually featureless. (It should be noted
here that the Tenax cartridges (200) used in this study were
prepared by the Oak Ridge National Laboratory in October 1978.)
Approximately 100 aerosol phase constituents were observed in
the gas chromatographic analysis of the unfractionated filter
extracts. Filter blanks were featureless. As shown in Table 2,
the major species were C,g-Co6 n-paraffins and phthalates. In
contrast to the vapor phase organics, the particulate phase
organics appeared to be more aliphatic and approximately 2 to 3
orders of magnitude lower in concentration. N-paraffins ranged
from 1 to 40 ng/m ; the most concentrated particulate phase organic
445
-------
Table 2. TENTATIVE IDENTIFICATION AND RANGE OF CONCENTRATIONS OF VAPOR
AND PARTICULATE PHASE CONSTITUENTS IN SAMPLES COLLECTED
NEAR YUGOSLAVIAN GASIFIER6
VAPOR PHASE
PARTICULATE PHASE
TENTATIVE
IDENTIFICATION
BENZENE
n-CgH2Q
TOLUENE
n-C-|OH22
ETHYL BENZENE
m-XYLENE
p-XYLENE
o-XYLENE
CUMENE
CS-BENZENE
CS-BENZENE
MESITYLENE
CS-BENZENE
CS-BENZENE
CS-BENZENE
o-METHYL STYRENE
BENZALDEHYDE
ACETOPHENONE
NAPHTHALENE
2-METHYL NAPHTHALENE
1-METHYL NAPHTHALENE
PHENOL
o-CRESOL
BIPHENYL
INDOLE
p-CRESOL
m-CRESOL
p-ETHYL PHENOL
RANGE OF
CONCENTRATION3,
jug/m3
0.33-1.8
0.16-1.0
0.74-9.0
0.16-0.60
0.46-1.3
0.20-1.3
0.38-3.2
0.24-1.6
0.02-0.38
0.11-0.52
0.25-2.0
0.06-0.58
ND-0.51
0.21-2.2
0.10-0.81
ND-0.11
1.1-2.8
1.3-3.0
0.02-1.5
0.03-0.25
0.01-0.15
0.16-2.3
ND-1.0
0.04-0.09
0.02-0.13
ND-0.24
ND-0.36
ND-0.16
TENTATIVE
IDENTIFICATION
BIPHENYL
n-CigH4Q
PHENANTHRENE
n-C2flH42
C14-BENZENE
n-C21 H44
C-I4-BENZENE
n-C22H46
FLUORANTHENE (+ HYDROCARBON)
n-C23^48
n-C24^50
MW 256 + 274
"-C25H52
"-C26H54
BIS-(2-ETHYL HEXYDPHTHALATE
MW 226d
"-C27H56
"-C28H58
C4-QUINOLINE
n-C29^60
n-C 30^62
BENZO(b,j,OR klFLUORANTHENE
"•C31H64
"-C32H66
"-C33H68
"-C34H70
n-C35H72
RANGE OF
CONCENTRATIONS,
ng/m3
0.29-4.2
1.8-11
_b
0.44-2.0
_c
1.0-4.7
_c
8.5-28
0.93-4.1
5.4-13
1.6-8.8
—
6.2-18
3.9-16
43-120
—
19-40
13-42
_c
11-21
2.2-7.9
2.3-6.2
7.4-13
T.4-7.2
2.2-6.5
1.1-3.6
0.8-2.9
aND = NOT DETECTED.
bINCOMPLETE RESOLUTION PREVENTS QUANTITATION.
CSTANDARD NOT AVAILABLE FOR QUANTITATION.
dNOT BENZO(ghi)FLUORANTHENE.
eFROM REFERENCE 17.
446
-------
observed, bis-(2-ethyl hexyl) phthalate, ranged from 43 to 121
ng/m . Polycyclic aromatic hydrocarbons were approximately 10 as
concentrated as the paraffins. Oak Ridge results were not weighted
by percent downwind. Differences between upwind and downwind
aerosol phase organics were not as apparent as those for the vapor
phase organics. This result suggests that the vapor phase organics
are a more sensitive indicator of the gasification plant's impact.
However, further fractionation of the particulate phase organics
may reveal more substantial differences than those observed from
the profiles of the gross filter extracts.
CONCLUSIONS
Each of the Yugoslav ambient air study objectives was met.
The adverse impact on the surrounding atmosphere of the Kosovo
Industrial Complex, especially downwind, is unmistakable as
described in the following conclusions:
-Aerosols in the form of coal dust are a significant pollutant
from the coal handling operation.
-Aerosol emissions from the gasification process are over-
shadowed by aerosol emissions from coal handling.
-Ambient aerosol levels exceed the primary and secondary U.S.
National Ambient Air Quality Standards.
-Aerosols appear to be carriers of PAH's.
-The source of the PAH's in the aerosol collections is as yet
unknown but may be the flare.
-The level of benzo.(a)pyrene exceeds the A-MEG's by a factor
of 1000,
447
-------
-Even though the light organic compounds were lost during
sampling, benzene probably exceeds the A-MEG's by a factor of
10 to 100.
-Organic pollutants can be traced to the gasification plant.
-There is a broad range of organic compounds in the ambient
air. The classes include aromatic and aliphatic hydrocarbons
as well as their oxygen-, sulfur-, and nitrogen-containing
derivatives.
Even though proposed U.S. facilities will be "better con-
trolled" due to the use of state-of-the-art control technology and
U.S. regulations, this study revealed areas of special concern on
which emphasis should be placed when making decisions about the
development, control, and placement of such facilities in the U.S.
Such aspects as coal mining, processing, transport, and storage;
process by-product storage and venting; fugitive emissions of
organics throughout the process; and the storage of gasifier (and
power plant) ash should be carefully reviewed. The Kosovo complex
is a commercial scale facility, but only one tenth the size of
proposed U.S. facilities. This study suggests that it is possible
to differentiate between the emissions from a gasification plant
and those from other sources near an industrial complex, and it
also provides a unique data base for researchers as well as policy
makers.
ACKNOWLEDGMENTS
The cooperation and assistance of T. Kelly Janes, Chief of the
IERL-RTP Gasification and Indirect Liquefaction Branch, is
448
-------
gratefully acknowledged. The cooperation and interest of Yugoslav
government officials, scientists, and technicians were key to the
success of this study. Special thanks are given to the following
Yugoslav agencies:
Rudarski Institut, Belgrade
Kosovo Institut, Pristina
Kombinat Kosovo, Pristina
Institut za Primenu Nuklearne Energije, Belgrade
449
-------
REFERENCES
1. Borabaugh, K.J., G.C. Page, C.H. Williams, L.O. Edwards, W.D.
Balfour, D.S. Lewis, and K.W. Lee. Aerosol Characterization
of Ambient Air Near a Commercial Lurgi Coal Gasification
Plant: Kosovo Region, Yugoslavia. Submitted to EPA by Radian
Corporation in July 1980, in preparation as EPA Research
Report.
2. Williams, C.H., Jr., K.J. Bombaugh, P.H. Lin, K.W. Lee, and
C.L. Prescott. GC-MS Characterization of Trace Organic
Compounds in the Ambient Aerosol Associated with a Coal
Gasification Plant in Kosovo. Presented at the Second
Chemical Congress of the North American Continent, American
Chemical Society, Las Vegas, Nevada, 1980. In preparation for
publication.
3. Lee, K.W., C.H. Williams, Jr., D.S. Lewis, and L.D. Ogle. A
Comparison of the Organics Collected from the Ambient Air with
the By-Products of a Lurgi Coal Gasification Plant. Presented
at the Second Chemical Congress of the North American
Continent, American Chemical Society, Las Vegas, Nevada, 1980.
In preparation for publication.
4. Griest, W.H., J.E. Caton, M.R. Guerin, L.B. Yeatts, Jr., and
C.E. Higgins. Extraction and Recovery of Polycyclic Aromatic
Hydrocarbons from Highly Sorptive Matrices such as Fly Ash.
In: Polynuclear Aromatic Hydrocarbons: Chemistry and
Biological Effects, A. Bjorseth and A.J. Dennis, eds.
Battelle Press, Columbus, Ohio, 1980, pp. 819-828.
5. Higgins, C.E. and M.R. Guerin. Recovery of Naphthalene during
Evaporative Concentration. Oak Ridge National Laboratory,
submitted to Analytical Chemistry.
6. Higgins, C.E. Rapid Preparation of Reproducibly-Behaving
Wall-Coated Capillary Columns. Oak Ridge National Laboratory,
in preparation for publication.
7. Patterson, R.K. Aerosol Contamination from High-Volume
Sampler Exhaust. Journal of the Air Pollution Control
Association, Vol 30, No. 2, 1980.
8. Holmberg, R.W. and J.H. Moneyhun. Volatile Organics Sampling
System. Oak Ridge National Laboratory, in preparation for
publication.
9. Smith, W.B., R.R. Wilson, Jr., and D.B. Harris. A Five-Stage
Cyclone System for In situ Sampling. Environmental Science
and Technology, Vol. 13, No. 11, 1979. pp. 1387-1382.
10. Woodard, A.P., Jr., B. Jensen, A.C.D. Leslie, J.W. Nelson,
J.W. Winchester, R.J. Ferek, and P. Van Espen. Aerosol
Characterization by Impactors and Streaker Sampling and PIXE
450
-------
Analysis. In: American Institute of Chemical Engineers
Symposium Series, Vol. 75, No. 188. Control of Emissions
from Stationary Combustion Sources: Pollutant Detection and
Behavior in the Atmosphere, W. Licht, A. Engel, and S.
Slater, eds. New York, New York, 1979-
11. Huntzicker, J.J., R.L. Johnson, and J.J. Shah. Carbonaceous
Aerosol in the Vicinity of a Lurgi Gasifier. Presented at the
Second Chemical Congress of the North American Continent,
American Chemical Society, Las Vegas, Nevada, 1980. In
preparation for publication.
12. Mitrovic, M., S. Tomasic, and S. Bratuljevic. Should High-Ash
Lignite be Burned at Power Plants (Kolubara and Kosovo
Lignite)? Bulletin of Mines, Rudarski Institut. Belgrade,
Yugoslavia ]976. (Translated from Sorbo-Croatian by the
Ralph McElroy Co.)
13. Boueres, L.C.S., J.W. Winchester, and J.W. Nelson. Trace
Metal Aerosols Near a Coal Gasification Plant in the Kosovo
Region, Yugoslavia. Florida State University, in preparation
for publication.
14. Boueres, L.C.S. and R.K. Patterson. Aerosol Emissions Near a
Coal Gasification Plant in the Kosovo Region, Yugoslavia.
Presented at the Second International Conference of Particle
Induced X-Ray Emission and Its Analytical Applications, Lund,
Sweden, 1980.
15. Balfour, W.D., K.J. Bombaugh, L.O. Edwards, R.K. Patterson,
and J.C. King. Collection and Characterization of Ambient
Aerosols Downwind from a Commercial Lurgi Coal Gasification
Facility. Presented at the Second Chemical Congress of the
North American Continent, American Chemical Society, Las
Vegas, Nevada, 1980. In preparation for publication.
16. Kingsbury, G.L., R.C. Sims, and J.B. White. Multimedia
Environmental Goals for Environmental Assessment; Volume III
and IV. EPA-600/7-79-176 a/b. U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1975.
17. Griest, W.H., C.E. Higgins, J.E. Caton, and J.S. Wike.
Characterization of Ambient Vapor and Particulate Phase
Organics Near the Kosovo Coal Gasifier. Presented at the
Second Chemical Congress of the North American Continent,
American Chemical Society, Las Vegas, Nevada, 1980. In
preparation for publication.
451
-------
CHARACTERIZATION OF COAL GASIFICATION ASH
LEACHATE USING THE RCRA EXTRACTION PROCEDURE
by
Kar Y. Yu, TRW and Guy M. Crawford, Radian
ABSTRACT
Gasification ash constitutes the single largest solid waste stream from
coal gasification facilities, and its disposal is subject to regulations
promulgated under RCRA. Ashes from Lurgi gasifier, Wellman-Galusha gasifier
and Texaco gasifier were subjected to the RCRA Extraction Procedure test.
The results are reviewed in light of similar data on boiler ashes. Those
findings indicate that these materials will not be considered toxic based
on the 100X primary drinking water standard criteria.
452
-------
1.0 INTRODUCTION
The Resource Conservation and Recovery Act of 1976 directs the Environ-
mental Protection Agency to promulgate regulations to insure the proper dis-
posal of solid wastes for the protection of both human health and the environ-
ment. With the recent reemphasis on America's coal resources, coal gasifica-
tion may soon be providing a large amount of America's energy needs. As with
all non-renewable energy resources, wastes will be generated in the produc-
tion of the coal gas. Future commercial-scale gasifiers will need to be
designed, constructed, and operated to protect human health and the environ-
ment. Solid wastes in the form of slags or ashes are produced from all coal
gasification facilities. The proper disposal of these solid wastes will be
a portion of this environmental protection.
To anticipate possible problems with solids disposal, the EPA has set
forth a procedure to test the potential hazard of solid waste—the EP Toxicity
Test.
2.0 WASTE COLLECTION
Three coal gasifiers were sampled and the solid wastes subjected to the
EP Toxicity Test. The data was compared to previous extraction tests performed
on two ashes from a coal-fired boiler. To investigate the distribution of
extractable metals among different sizes of ash, the Lurgi ash samples were
divided into three size fractions; triplicates of each fraction were sub-
jected to the EP test.
2.1 The Texaco Gasifier
Coarse slag was collected at the sieve screen used to separate the coarse
slag from the slag water as the slag was blown down from the gasifier. A
composited sample was taken over a 16-hour sampling period during gasifica-
tion of a western subbituminous coal under conditions typical of a commercial
operation.
2.2 The Wellman-Galusha Gasifier
Gasifier ash was sampled as the ash was transferred from the bottom of
the gasifier to a storage bin. A dewatered composite sample was taken over
a 12-hour sampling period. Cyclone dust samples were taken from the bottom
453
-------
of cyclone by raking the solid from the trough and allowing excess water to
drain. Sampling was conducted during the gasification of a North Dakota
lignite.
2.3 Hie Lurgi Gasifier
Unquenched Lurgi ash of three U.S. coals (Rosebud, Illinois #5 and Illinois
#6) were furnished by the Peabody Company. The ashes were collected during a
trial run at the Westfield gasification facility.
2.4 The Coal-Fired Steam Station
Precipitator ash was taken from the ash silo prior to removal by truck.
Bottom ash was taken from the sluice pipe as it empties into the ash pond.
A western lignite is normal boiler feed for the station.
3.0 RCRA TESTING PROCEDURE
The prescribed procedure is designed to roughly approximate the extracting
of soluble material with rainwater. The solid is extracted with a sixteen-
fold excess of leaching solution at a pH of 5.0 for a 24-hour time period
at room temperature. Following the extraction period the sample is filtered
and the final aqueous volume is made to 20 times the sample weight. The
procedure followed is listed in Table 1. The extract is then analyzed for 8
metals which are listed in the EP and other constituents. Results are compared
with the National Interium Primary Drinking Water Standards (NIPDWS) for
eight metals:
arsenic lead
barium mercury
cadmium selenium
chromium silver
4.0 RESULTS AND DISCUSSION
Table 2 presents a comparison of the extract characteristics and the
drinking water standards. Although the coal-fired boiler and the gasifiers
operate at different conditions, the RCRA extract characteristics are in
general quite similar. When compared to the 100X primary drinking water stan-
dards, none of the wastes analyzed are considered hazardous. This result is
similar to those presented by other investigators working with different coal
gasification ashes and boiler ashes
454
-------
(2)
TABLE 1. RCRA EXTRACTION PROCEDURE
Weight lOOg solid into extractor
Add 1600 ml deionized water
Measure the pH
If less than 5.0, continue with extraction
If greater than 5.0, add 0.5N ultrex acetic acid until
pH 5.0. Check and readjust pH at intervals of 15,
30, 60, 120 minutes, if pH rises above 5.2.
Extraction by shaking or stirring for 24 hours at
20°-40°C
Filter through 0.45 micron filter
Dilute to 2000 ml with deionized water
455
-------
TABLE 2. CHARACTERISTICS OF WASTE EXTRACTS USING THE RCRA EXTRACTION PROCEDURE
Concentration, yg/1
Lurgi- Rosebud
3/8"-20 mesh
20-100 mesh
<100 mesh
Lurgi-Illinois #5
3/8"-20 mesh
20-100 mesh
<100 mesh
Lurgi-Illinois #6
3/8"-20 mesh
20-100 mesh
<100 mesh
Wellman-Galusha, ash
Wellman-Galusha, dust
Texaco, slag
Boiler bottom ash
Boiler fly ash
lOOx primary
drinking water
standard
Ag
<0.2
<0.2
<0.2
<0.2
<0.2
1.6
0.9
1.4
<0.2
<1
<1
<2
<1
2
5000
As Ba*
<1 0.5
2 1.0
3 2.3
<1 <0.2
<1 0.8
3 1.0
4 <0.2
<1 <0.2
<1 <0.2
19 1.0
33 1.0
<2 0.19
<1 0.28
5 0.44
5000 100
Cd
<0.1
1.1
2.0
52
32
26
13
5.1
4.3
<7
<1
37
<0.3-
5.3
1000
Cr
<6
<6
<6
5
3
4
3
3
<2
1
1
4
<3
16
5000
Hg
<0.4
<0.4
<0.4
<0.4
<0.4
<0.4
<0.4
<0.4
<0.4
<0.6
<0.3
<0.2
<0.2
<0.2
200
Pb Se
.<0.2 <1
1.0 <1
1.8 <1
0.9 <1
3.1 <1
4.4 3
1.3 3
1.3 <1
1.6 <1
7 14
8 6
<2 <1
<3 <1
<3 2
5000 1000
B* Cu Mn* Ni U* Zn*
0.55 2.7 3.22 34 <0.5 0.124
1.48 5.4 5.83 80 <0.5 0.157
1.85 13.3 9.25 138 <0.5 0.321
0.28 5.6 0.39 4240 <0.5 37.1
0.77 6.5 1.15 442 <0.5 28.5
0.49 5.1 2.50 441 <0.5 9.2
0.04 <2 0.28 49 <0.5 4.27
0.25 <2 0.39 56 <0.5 2.84
0.20 <2 0.71 72 <0.5 1.13
—
—
—
—
—
— — — — — —
*Values in mg/1
-------
As expected, partly due to the larger surface area and partly due to the
volatility of trace metals, the boiler fly ash contains slightly more extract-
able metals than the boiler bottom ash. For the Lurgi samples leachate metal
concentrations were observed to be inversely proportional to the particle (ash)
size for the Rosebud coal, but not necessarily for Illinois #5 or #6, suggest-
ing surface phenomena could be one of the major factors controlling the leach-
ability of metals in Lurgi gasifier ash.
As discussed before the Lurgi samples analyzed are unquenched ashes.
Quenched ash is likely to contain even less extractable metals because a por-
tion of the total extractable metals will be carried away by the quench water.
However, all proposed commercial Lurgi plants plan to recycle process waste-
water as quench water, and to achieve zero discharge (especially in the east
where solar evaporation is not feasible) it has been proposed to evaporate
the gas liquor in a forced evaporator, and to use the concentrated brine to
moisten the ash. It is uncertain whether the practice would make the ash
hazardous.
Table 3 presents the characteristics of Lurgi gas liquor, expressed in
terms of yg/g of coal; also presented in Table 3 are the leachable metal con-
tents of coal. As a worst case approach, one may assume all trace metals in
the gas liquor ends up in the RCRA leachate, i.e.
Total leachable metal = extractable metal + soluble metal
Comparing the extractable metal (from ash) and the soluble metal (from gas
liquor) data indicates that adding the soluble metal content will increase
the extractable Se by 1h times, the largest increase among all eight metals.
Even so, the leachate concentration is calculated as seen in Table 4, to be
7 pg/1, still below the 100X primary drinking water standard. The RCRA leach-
ate characteristics for Lurgi ash and boiler ash calculated based on this
worst case scenario are presented in Table 4. Again, none of the metals exceeds
the 100X drinking water standards.
Still, there are coals that contain much higher metal contents than the
coals used in these studies. Table 5 presents the characteristics of the
coals used in these studies and the maximum metal concentrations in coals
457
-------
TABLE 3. EXTRACTABLE AND LIQUOR METAL CONCENTRATION IN COALS USED IN DIFFERENT GASIFIERS AND BOILER
Gasifier - Coal
Extractable Metals Cone.*
Lurgi - Rosebud
Illinois #5
Illinois #6
Texaco - Western
Subbituminous
Wellman-Galusha
(ash) No. Dakota Lignite
(dust) No. Dakota Lignite
Boiler (bottom ash)
Western Lignite
Boiler (fly ash)
Western Lignite
Soluble Metal Cone.**
Lurgi Liquor - Rosebud
Illinois #6
Total Leachable Metal Cone. ***
Lurgi (maximum)
Boiler bottom ash
Boiler fly ash
Metal Concentration, yg/g
Ag
<0.52
4.1
3.6
<4.3
<1.4
<1.4
<4.3
8.6
0.041
0.31
4.4
<4.6
8.9
As Ba Cd Cr Hg
7.7 5.9 5.2 <15 <1.0
7.7 2.6 130 13 <1.0
10 <0.52 34 7.7 <1.0
<4.3 0.41 80 8.6 <0.43
26 1400 <9.5 1.4 <0.82
45 1400 <1.4 1.4 <0.41
<4.3 1.2 <1.3 <13 <0.86
22 1.9 23 69 <0.86
0.41 <0.01 0.26 3.5 0.15
1.1 <0.2 <0.21 <0.21 1.25
11.1 <6.1 130 <19 <2.3
<5.4 <1.4 <1.6 <17 <2.1
23 <2.1 23 74 <2.1
Pb Se
4.6 <2.6
11 7.7
4.1 7.7
<4.3 <22
9.5 19
11 8.2
<13 <4.3
<13 8.6
0.32 0.13
6.3 10.5
17 18
<19 <15
<19 19
Ash Content
12.9
10.1
9.2
10.8
6.8
6.8
21.6
21.6
en
oo
*Extractable metal cone. = 20 x RCRA leachate cone, x % ash in coal
**Soluble metal cone = liquor cone, x liquor quantity
coal feed
***Total leachable metal cone. = extractable metal cone. + soluble metal cone.
-------
TABLE 4. PREDICTED LEACHATE CHARACTERISTICS FOR LURGI ASH AND BOILER ASHES
WHEN CO-DISPOSED WITH BRINE FROM CONCENTRATING LURGI GAS LIQUOR
Metals ,
Ag
As
Ba**
Cd
Cr
Hg
Pb
Se
Lurgi Ash
1.7
4.3
<2.4
50
<7.4
<0.87
6.6
7.0
Leachate Characteristics , *
Boiler Bottom Ash
<1.1
<1.3
<0.32
<0.37
<3.9
<0.49
<4.4
<3.5
yg/i
Boiler Fly Ash
2.1
5.3
<0.49
5.3
17
<0.49
<4.4
4.4
*Conc. = total extractable metal cone. T (20 x % ash)
**Ba values in yg/ml; all other in yg/1
459
-------
TABLE 5. METAL CONCENTRATIONS IN VARIOUS COALS AND RCRA LEACHATE CHARACTERISTICS BASED ON
WORST CASE OIL
-p.
en
o
Coal Characteristics, yg/g
Rosebud
Illinois #6(5)
Western Subbituminous
(Texaco)
Lignite (Wellman-Galusha)
(6)
Maximum Cone, in Coal
Ag As
0.06 1.2
1.0
0.3 <0.9
1 6.5
0.08 120
Ba
87
320
1300
1600
Metals
Cd
0.4
<0.4
0.2
0.4
26
Cr
4
20
34
10
60
Hg
0.11
1.1
0.1
0.39
1.6
Pb
0.51
10
4
2
220
Se
0.33
1.3
1.7
1
*
8.1
Predicted Max. Leachate Characteristics, yg/1
Lurgi
Texaco
Wellman-Galusha, ash
dust
2.2 470
<270
<19
92
43000
950
340
540
3300*
4800*
<20
350
<88
7.1
<18
96
52
<32
<0.82
<0.82
240
<110
<330
<330
78
<5
<8.1
16
*Value exceeded the lOOx drinking water standards
-------
found in open literature. The Teachability characteristics of other coals
is not known, but as a first approximation one may assume the leachable metal
content is proportional to the total metal content. The predicted maximum
leachate characteristics thus derived are presented in Table 5. As the pre-
dictions indicate, only cadmium in both the Lurgi ash and Texaco slag exceed
the 100X drinking water limit. It should be emphasized that the above assump-
tion is very conservative as, undoubtedly, other factors such as mineralogy
will play a major role in controlling the leachable metals. Furthermore, it
is uncommon to encounter coals with as high a Cd concentration (26 ppm). Of
the samples analyzed by Gluskoter, et al, only about 6% had Cd values in
that range, with over 90% having less than 1 ppm Cd.
Additional data on the leachate characteristics of other coals/gasifiers
are expected to be available by next year. As an ongoing EPA program, Radian
is presently testing the ash collected from a Lurgi facility in Kosovo,
Yugoslavia, and TRW is scheduled to sample a Koppers-Totzek facility in
Modderfontein, South Africa, early next year.
5.0 CONCLUSION
The RCRA EP Toxicity Test as performed on the ashes from a Lurgi gasi-
fier, a Texaco gasifier and a Wellman-Galusha gasifier indicates these ma-
terials will not be considered hazardous wastes based on the toxicity cri-
terion alone. Based on the metal contents in the ash and in the Lurgi gas
liquor, co-disposal of the gas liquor with the gasifier ash also will not be
considered hazardous. However, Lurgi gas liquors are known to contain aro-
matic organics, some of which are priority pollutants. Unless these organics
are removed prior to co-disposal with ash, EPA may eventually list this as a
hazardous waste.
461
-------
ACKNOWLEDGMENT
The authors wish to thank the Peabody Company for supplying the Lurgi
gasifier ash samples; to Ms. Cheryl May who, at Radian, supervised the analysis
for the Texaco gasifier samples; Wellman-Galusha gasifier samples and the
boiler ash samples; and to Mr. Dave Ringwald who, at TRW, supervised the
analysis for the Lurgi sample.
462
-------
REFERENCE
1. EPA Report, Test Methods for the Evaluation of Solid Wastes, Physical/
Chemical Methods, SW-846, US EPA, 1979.
2. Federal Register, May 9, 1980.
3. Boston, C. R. and Boegly, W. J., Jr. Leachate Studies on Coal and Coal
Conversion Wastes, NTIS CONF-790571-1, 1979.
4. Tennessee Valley Authority. Draft Environmental Impact Statement, Coal
Gasification Project, 1980.
5. Ghassemi, M., et al. Environmental Assessment Report - Lurgi Coal Gasi-
fication Systems for SNG, EPA 600/7-79-120, May 1979.
6. Gluskoter, H. J., et al. Trace Elements in Coal: Occurrent and Distri-
bution, Illinois State Geological Survey, Circular 499, Urbana, IL 61801,
1977.
463
-------
COMPARISON OF COAL CONVERSION WASTEWATERS
By
Robert V. Collins,
Kenneth W. Lee, and
D. Scott Lewis
Radian Corporation
8501 MoPac
Austin, TX 78758
This paper presents the analytical results obtained from the aqueous
process condensates from an oxygen-blown, lignite-fired Lurgi gasifier, an
air-blown, bituminous-fired Chapman gasifier and a coke oven process. Re-
sults show that strong similarities exist between the two gasifier process
condensates. These similarities include both gross chemical parameters and
the concentrations of specific organic compounds. Extraction of the three
condensates using diisopropyl ether resulted in a 99+ percent removal of
total phenols and a 75 percent average removal of the total organic carbon
(TOC). Further extraction with an exhaustive technique only removed an
average of 9 percent of the remaining TOC from the two gasifier waters. The
<500 MW to >500 MW ratio was approximately two for the remaining refractory
organics. The results of a brief study using activated carbon to remove the
refractory organics indicated that the TOC levels could be further reduced,
but the levels remained relatively high. The occurrences of eight nitrogen-
containing organic species were compared using a gas chromatograph equipped
with a Hall Electrolytic Conductivity Detector in the nitrogen-specific mode.
The occurrences of phenolic species were also compared using a gas chromato-
graph equipped with a flame ionization detector. The three process condensates
contained the same phenolic and nitrogen heterocyclic compounds.
464
-------
COMPARISON OF COAL CONVERSION WASTEWATERS
INTRODUCTION
Three coal conversion process condensates were characterized as part of
Radian Corporation's overall effort to perform a comprehensive environmental
assessment of low- and medium-Btu coal gasification technology for the U.S.
Environmental Protection Agency. The overall program is being directed by the
Fuel Process Branch of EPA's Industrial Environmental Research Laboratory in
Research Triangle Park, North Carolina.
The objective of this study was to compare the composition of the con-
densates and to screen for possible steps in treatability. The three aqueous
condensates and the reasons they were chosen are as follows:
Wastewater
Lurgi (Process Gondensate)
Chapman (Recycled Process
Condensate)
Coke Oven (Process Conden-
sate Spray Down)
Rationale
Proposed for commercial plants
in the United States
Currently available in the
United States and possible
similarities in composition
to Lurgi
Extensive data available on
treatability and possible
similarities in composition
to Lurgi
PROCESS DESCRIPTIONS
The three processes will be described briefly in this section. Where
the samples orginated in the processes will be shown.
In Figure 1, a schematic diagram of the Lurgi Gasification Process is
illustrated. The main points to notice are the quench and cooling towers
which condense water along with the organic and inorganic components from the
product gas, and the separator where the aqueous layer is separated from the
465
-------
AIR
COAL
STEAM
en
VENT
GASES
RAW GAS
T
SLAG
/GASIFIER
+-COAL GAS
GAS
COOLING
TOWERS
f SEPARATION
V TANKS
rz
\
WASTEWATER
TO
PHENOSOLVAN
*- GASOLINE
MEDIUM OIL
Figure 1. Schematic diagram of the Lurgi Gasification Process.
-------
tars and oils. The Lurgi condensate was obtained from the exit point of the
aqueous layer from the separator. The plant sampled for this study was an
oxygen-blown, lignite-fired Lurgi gasification plant in the Kosovo Region of
Yugoslavia.
The Chapman-Wilputte Gasification Process is illustrated in Figure 2.
The aqueous layer after separation of the tars and oils is recirculated to
the gas quenching/cooling processes. A grab sample of the wastewater was
obtained from the aqueous layer in the separation tank. The plant sampled
was located near Kingsport, Tennessee and was equipped with an air-blown,
bituminous-fired Chapman gasifier.
The coke oven system is illustrated in Figure 3. Even though coking
may at first appear to be very different from a gasification process, there
are many similarities. The design is different from either a Lurgi or
Chapman facility but, again, as illustrated, there is a gas quenching and
cooling system to cool the gases and remove water, tars, and oils. The
quench liquor is sent to a separator where tars/oils are separated from the
aqueous layer. Part of the water layer is recirculated and the rest is
treated. The condensate sample was obtained at the point where the excess
aqueous layer exits the separator.
RESULTS AND DISCUSSION
The following subsections will detail the results of the different
types of analyses and will contain brief discussions on treatability. These
sections will include:
• water quality parameters,
• extractions of organics,
• concentrations of phenols,
467
-------
Air
Coal
Steam
Gasifier
.en
oo
Liquor
Trap
Spray
1
Product Gas
Scrubbers
T
Liquor Separator
I
By-Products
Figure 2. Flow diagram of Chapman facility.
-------
Coke
Oven
T
Coal
-K
cr>
10
Sprayers
Further
Gas
Clean Up
Separator
Tank
By-Products
-^Wastewater
Figure 3. Flow diagram of coke oven.
-------
• concentrations of nitrogen-containing organics,
• molecular weight distribution of refractory com-
pounds , and
• removal of refractories.
Water Quality Parameters
The water quality parameters for all three process condensates are
listed in Table 1. In general, the parameters are very similar for the con-
densates from the two gasification processes using two different coals (lig-
nite and bituminous). The water quality parameters for the coke oven pro-
cess condensate are generally lower than the other two process condensates.
Biological oxygen demand (BOD), chemical oxygen demand (COD) and total
organic carbon (TOG) are specific measurements where the process condensates
of the Lurgi and Chapman gasification processes are similar. The differences
among the three condensates may be caused by the types of coal being used.
For instance, the lignite from the Kosovo region of Yugoslavia used in the
Lurgi Process may contain much less phosphorous than the coal for the Chap-
man Process. Of course, differences in the process conditions may also
affect the composition of the aqueous condensate. Differences may also be
caused by Chapman recirculating the water, whereas the Lurgi does not recir-
culate it. Therefore, higher levels would be expected in the Chapman aque-
ous condensate. To test the process effects would require using the same
coal at both facilities.
Extractions of Organics
Two extraction procedures were used on the three aqueous condensates.
The first extraction procedure was designed to mimic the Phenosolvan Process
used by Lurgi to remove phenols from process wastewaters. Three volumes of
diisopropyl ether (each equal to 1/3 the sample volume) were added, one at
a time, to the aqueous condensate. The samples were then shaken vigorously
470
-------
Table 1. WATER QUALITY PARAMETERS FOR THREE COAL CONVERSION
AQUEOUS PROCESS CONDENSATES
Water Quality Parameters
(mg/*)
BOD
COD
TOC
NH3 -Nitrogen
Total Kjeldahl Nitrogen
Nitrate-Nitrogen
Total Phosphate-Phosphorous
Total Acid Hydrolyzable
Phosphate-Phosphorous
Phenol
Oil and Grease
Cyanide
Thiocyanate
Sulfide
TDS
TVDS
TSS
TVSS
Lurgi
12,200
20,200
6,490
4,340
4,010
<0.5
0.12
0.08
3,030
917
<0.02
83
<10
2,010
1,890
417
402
Aqueous Process Condensates
Chapman
15,900
28,500±1,100
9,430
8,130±90
9,420
<0.5
5.48
5.48
2,130±110
540
59±1
1,450
207+12
48,600
42,300
11
11
Coke Oven
3,420
4,860±390
6,160
2,850±0
3,160
<0.5
0.21
0.21
1,140
700
69±1
570
241±18
4,870
4,700
20
18
-------
for two minutes and allowed to stand in a separatory funnel until the layers
separated. Then the ether layer was removed.
The second extraction procedure followed the above steps except that
methylene chloride and diethyl ether substituted for the diisopropyl ether
and the aqueous layer was extracted at both pH equal to <2 and >12. This
procedure will be labeled the "analytical extraction" procedure. This proce-
dure was used to show if changes in pH and solvent would increase the amount
of organics removed from the aqueous layer.
In Table 2, the effects of the two sequential extractions on selected
water quality parameters are listed. The diisopropyl ether (DIPE) extrac-
tion eliminated greater than 99+ percent of the phenol (phenolic content)
from all three process condensates. The oil and grease measurements also
dropped below the detection level of 10 mg/JJ, for all the condensates. The
BOD, COD, and TOC values were reduced significantly by the DIPE extraction.
The exhaustive, analytical extraction did not significantly reduce the
values of the water quality parameters when applied to the waters after DIPE
extraction.
The organic carbon left in the aqueous phase after the two extractions
was classified as refractory organic compounds. These refractories are im-
portant because Phenosolvan treatment alone leaves them in the aqueous phase
and they must be addressed in further treatment steps. The relative amounts
of refractories (non-extractables) as measured by TOC are graphically illus-
trated in Figure 4. The refractories must be very polar and/or ionic in
nature since both the extraction procedures (including pH adjustment) would
not remove them.
For further characterization of the refractories, the molecular weight
distribution above and below 500 was determined by gel permeation chroma-
tography. This separation, as measured by TOC, is illustrated in Figure 5
for the aqueous condensates of the gasification processes. The relative
472
-------
-vl
CO
Table 2. EFFECTS OF THE DIPE EXTRACTION AND THE ANALYTICAL EXTRACTION (SEQUENTIAL) ON
SELECTED WATER .QUALITY PARAMETERS IN THE THREE AQUEOUS PROCESS CONDENSATES
Process Condensate
Water
Quality
Parameters
BOD
COD
TOG
Phenol
Oil & Gas
Raw
12,200
20,200
6,490
3,030
917
Lurgi (tng/£)
Chapman (mg/&)
Coke Oven (mg/£)
DIPE Analytical Raw DIPE Analytical Raw
3,080 ND*
4,940 4,270
2,010 1,894
8.9 ND
<10 ND
15,900 2,800
28,500±1,100 15,500
9,430 3,290
2,130±10 3.0
540 <10
ND
7,230
1,830
ND
ND
3,420
4,860±390
6,160
1,140
700
DIPE Analytical
727 ND
2,770 1,690
602 477
9.4 ND
<10 10
*Not Determined
-------
V////A Nonextractable
Analytical Extractable
DIPE Extractable
10,000
o>
o
o
5,000
Lurgi Chapman
Coke
Oven
Figure 4. Amounts of total organic carbon removed by the DIPE
and Analytical Extraction Techniques.
474
-------
2000
en
"B)
O 1000
>500MW
35.5%
<500MW
64.5%
HI
A?*sSc*e.Ayw
>500MW
32%
<500MW
68%
Lurgi
Chapman
Figure 5. Molecular weight distribution of refractory organics.
-------
amounts of the refractories and their molecular weight distribution are the
same within experimental error for the Lurgi and Chapman waters. This
strongly suggests that the Chapman aqueous condensate, after BIPE extrac-
tion to mimic phenol removed by Phenosolvan, can be used as a model for
treatment studies of Lurgi-produced wastewater.
Concentrations of Phenols and Nitrogen-Containing Compounds
Another indication that the aqueous process condensates are similar is
the distribution of phenolic and nitrogen-containing compounds. Most of
these compounds were removed by the DIPE extraction; therefore, an analysis
of the DIPE layer was performed.
Figure 6 compares a standard consisting of 11 phenolic compounds to the
organics extracted by DIPE from the LURGI wastewater. These chromatograms
were produced by a gas chromatograph equipped with a flame ionization detec-
tor. The shaded peaks in the DIPE extract match the retention times of the
phenolic standards. This suggests that the major portion of organics in the
Lurgi wastewater is phenols. Similar results were observed for the Chapman
and coke oven process condensates.
Table 3 contains a list of the concentrations of the phenolic compounds
found in the three process condensates. The phenolic species show a very
strong correlation even in concentrations between the two gasification pro-
cesses. Again, as in the water quality parameters, the coke oven phenolics
were found at lower concentrations than those in the gasification conden-
sates. The same species, however, were present in all three aqueous process
condensates.
Trace species in the form of nitrogen-containing compounds were analyzed
in the DIPE extracts of all three process condensates. The results of the
semiquantitative analysis are listed in Table 4. Even at trace levels, all
three aqueous process condensates contained the same nitrogen heterocyclic
compounds. Even though the data is semiquantitative, the relative
476
-------
-------
Table 3. PHENOL SPECIATION DATA FOR THE DIPE EXTRACTS OF THE
THREE AQUEOUS PROCESS CONDENSATES
Aqueous Process Condensate
Compound
Phenol
o-Cresol
m&p-Cresol
2 , 6-Dimethylphenol
2,4-Dimethylphenol
00
3 ,5-Dimethylphenol
3 ,4-Dimethylphenol
l&2-Naphthol
Lurgi
1,740+100
406+27
1,040+60
33.1+10.0
172+17
266+21
271+24
13.0+30.6
Chapman
(mg/£)
1,460+170
•420+54
1,120+120
19.1+0.2
196+27
172+24
681+82
14.5+0.3
si r\— 2
Coke Oven
(mg/£)
888+52
70.0+2.3
279+14
2.2+1.0
14 . 5+0 . 1
23.4+0.8
41.5+1.3
4.5
p-Phenylphenol
,-2
-------
Table 4. NITROGEN-CONTAINING ORGANIC COMPOUNDS IN THE DIPE EXTRACTS OF THE THREE AQUEOUS
PROCESS CONDENSATES (SEMIQUANTITATIVE DATA)
Compounds*
Pyridine
2-Methylpyridine
3-Me thylpyr id ine
Ethyl/Dime thy Ipyridines
Trimethyl/Ethylmethylpyridines
Cif-pyridines
Quinoline
Lurgi
(mg/£)
12
19
45
7
27
18
9
Aqueous Process Condensate
Chapman
(mgM)
2
5
11
1
17
17
10
Coke Oven
(mg/£)
11
11
12
1
28
14
30
compounds quantified as pyridine.
-------
concentrations of the compounds within each of the condensate extracts are
virtually identical as listed in Table 4.
Removal of Refractory Compounds by Activated Carbon
The graph in Figure 7 illustrates the removal of the refractory com-
pounds with activated carbon. TOC measurements indicated the amounts of
organics remaining in the water after the addition of varying amounts of
activated carbon. The initial amount of activated carbon (0.005 g/irJl) re-
moved most of the organic matter that could be removed. Additional amounts
of activated carbon, up to a ratio of 0.1 g activated carbon per milliliter
of wastewater, did not significantly increase the amount of refractory com-
pounds removed. The activated carbon was effective in taking out the color
species in the wastewater.
CONCLUSIONS
The following statements summarize the conclusions of this brief study.
• Water quality parameters are similar in the three
aqueous process condensates with coke oven con-
densates having lower values.
• The same phenolic compounds were found in each
process condensate. Levels of these compounds
were similar in the gasification condensates.
The coke oven condensate had lower levels of
phenols.
• The same trace nitrogen species were found in
all three condensates.
• Levels of nonextractable organics were similar in
the Chapman and Lurgi condensates.
480
-------
bJO
CO
§ 400
ctf
0)
300-
.3
g 200-
(171)
f\j
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
g Activated C/ml H2O
* Smaller Particle Size
Figure 7. Removal of organics from the extracted Lurgi wastewater by activated carbon.
-------
Treatability of gasification wastewaters:
- may not be similar to coke oven treat-
ment because of nonextractables;
- may not be sufficiently polished by
activated carbon due to high residual
TOG levels; and
- can be studied using the Chapman process
condensate as a good model for the Lurgi
wastewater.
482
-------
Session IV: ENVIRONMENTAL CONTROL
Forest O. Mixon, Jr., Chairman
Research Triangle Institute
Research Triangle Park, North Carolina
483
-------
RANKING OF POTENTIAL POLLUTANTS FROM
COAL GASIFICATION PROCESSES
by
Duane G. Nichols
David A. Green
Research Triangle Institute
P. 0. Box 12194
Research Triangle Park, N. C. 27709
ABSTRACT
Potential pollutants associated with coal gasification processes were
studied based on data from the EPA environmental assessment research pro-
gram. An environmental assessment methodology based on health and eco-
logical Multimedia Environmental Goals (MEGs) is described and applied to
product, byproduct, process and waste streams. A list of chemical species
that were measured or qualitatively identified in coal gasification streams
is given. Maximum concentrations of each quantitated species in each
medium (solid, liquid, gas, tar) are given. Production factors have been
computed and normalized on the basis of coal input rate to facilitate
comparisons. Chemical species have been ranked by potential hazard to
health and ecology. Priorities for monitoring, regulation and control
technology development may be established from these lists.
Duane G. Nichols is now with the Conoco Coal Development Company, Research
Division, Library, PA.
484
-------
RANKING OF POTENTIAL POLLUTANTS FROM
COAL GASIFICATION PROCESSES
INTRODUCTION
This study was initiated to compile the various source and laboratory
(experimental) test results on potentially hazardous species which have been
obtained under the EPA synthetic fuels environmental assessment program.
The compilation has been developed in the form of listed chemical constituents
which are ranked on the basis of their potential hazard. Since the data
represent various gasifiers, coal types, operating conditions and configura-
tions, and since the effluents are variable in their physical and chemical
nature and their quantity, a systematic approach was needed to place the
results on a common basis for comparison and/or ranking.
The information and results are needed to help provide direction to
future environmental assessment activities, to focus EPA and interagency
health/ecological effects testing on compounds and mixtures of greatest
concern, and to assist EPA program and regional offices in the establishment
of appropriate regulations, criteria, guidelines and permit policies.
The achievement and maintenance of an acceptable (or quality) environ-
ment must from a practical viewpoint involve the establishment of maximum
allowable concentrations of chemical contaminants in the air, water, and
land which constitute the natural environment. Such concentrations may be
referred to as Multimedia Environmental Goals (MEG) values. Discharge MEGs
(DMEGs) represent approximate concentrations for contaminants in source
emissions to air, water or land which will not evoke significant harmful or
irreversible responses in exposed humans or ecology when these exposures are
limited to short duration. DMEGs for human health and ecology have been
1-4
developed for use in assessing the impact of effluent discharges.
A number of coal gasification operations are currently active around
the world. Direct coal and (oil shale) liquefaction may be proved to be
technically feasible and economically acceptable in the future; these
alternatives may require special processing of the potential product to meet
acceptable market specifications, and significant costs may be incurred to
accommodate process residuals.
485
-------
In this study, the chemical analyses of coal gasification product, by-
product, discharge and process streams sampled and analyzed by the Radian
Corporation during four source testing programs have been subjected to an
environmental assessment analysis based upon multimedia environmental goals.
A similar analysis of data obtained from the laboratory coal gasification
system at Research Triangle Institute (RTI) has also been conducted.
Radian Corporation Source Tests
The Radian Corporation has conducted source tests at four operating
coal gasification facilities. Two Wellman-Galusha units located at York,
PA and Ft. Snelling, MN were sampled as well as a Lurgi gasifier in
Kosovo, Yugoslavia and a Chapman (Wilputte) gasifier located at Kingsport,
TN. A variety of products, byproducts, process streams and effluents were
sampled at the different sites. The sampling strategies did not yield data
that were directly comparable. Sampling was not meant to be exhaustive
but was designed to focus on streams of potential environmental signifi-
cance.
The Wellman-Galusha gasifier at York, PA converts anthracite coal
into fuel gas used for brick manufacturing at the Glen Gery Brick Company.
Data on five different streams were available for this study: two solid
wastes, the gasifier ash and cyclone dust, one liquid stream, the ash
sluice water and two gaseous streams, the poke hole gas and coal hopper
gas.
The Wellman-Galusha gasifier at Ft. Snelling, MN uses North Dakota
Indian Head lignite as a feedstock for low Btu gas production. Data on
seven different streams were available for this study: two solid streams,
the gasifier ash and cyclone dust, three liquid streams, the cyclone
quench water, ash sluice water, and service water and two gas streams,
the product gas and the coal bin vent gas. As no flow rate was available
for the coal bin vent gas, a limited environmental assessment approach to
gaseous effluents was taken.
The Chapman (Wilputte) gasifier at Kingsport, TN converts low sulfur
Virginia bituminous coal to low Btu guel gas. Data on four effluent
streams were available. Three solid streams—the cyclone dust, gasifier
486
-------
ash, and byproduct tar, two gaseous streams—the coal feeder vent gas and
separator vent gas and the separator liquor, a recycled aqueous stream
were sampled.
Data on 18 gaseous streams and three liquid streams sampled at the
7-9
Lurgi gasifier at Kosovo, Yugoslavia, were used in this study. This
plant converts Yugoslavian lignite to medium Btu fuel gas. Of the gaseous
streams, eight were discharges and 10 were process streams. The gaseous
discharges were the autoclave vent gas, coal bunker vent gas, CCL-rich
Rectisol gas, tar tank vent gas, medium oil tank vent gas, phenolic water
tank vent gas, degassing column gas and gasoline tank vent gas. The
cyanic water and the inlet and outlet from the Phenolsolvan unit are
aqueous process streams that were sampled. No solid stream data were
available.
RTI Gasifier Tests10'12
Data from 10 selected semicontinuous, fixed-bed tests of the RTI
laboratory gasifier were analyzed in detail. In each case the solid
gasifier ash and the aqueous condensate stream were the two discharges
sampled. Two additional streams, the product gas and the byproduct tar
(considered a solid) were also sampled. The 10 selected tests involved
steam/air gasification of North Dakota Beulah/Zap lignite, Montana Rosebud/
McKay and Wyoming Smith/Roland subbituminous coals, Illinois No.6 and
Western Kentucky No.9 bituminous coals and Pennsylvania Bottom Red Ash
anthracite.
ASSESSMENT METHODOLOGY
Multimedia Environmental Goals (MEGs) form the basis for the environ-
mental assessment methodology developed under the guidance of the Fuel
Process Branch of EPA/IERL/RTP. Each component or species is assigned
discharge multimedia environmental goal (DMEG) and ambient multimedia
1-4
environmental goal, (AMEG) values. Individual DMEG values for a sub-
stance are related to the health or ecological effects of that substance;
DMEG is the estimated concentration of the substance which would cause
minimal adverse effects to a healthy receptor (man, animal, plant) which
is exposed only once, or intermittently for short time periods. (AMEG
487
-------
values are similar except that they are based upon a continuous, rather
than single or intermittent, exposure period.
DMEG values generally carry two subscripts, be they explicit or im-
plicit. The first defines whether the value refers to air Ca)» water Cw),
or land (1); the second, whether the value refers to human health (h) or
the ecological environment Ce). In this study the health-based DMEG values
were used primarily. The ecology-based DMEG values were used only to
generate a comparative ranking of pollutants. No AMEGs were used in this
study.
Discharge severity (DS) is a measure (index) of the degree to which
the concentration of a particular substance is at a potentially hazardous
level in a discharge (effluent.) DS is dimensionless. It is computed as
the concentration of the substance in a discharge divided by the DMEG value
for that substance. DS may thus carry two subscripts, in general; one
represents the phase and the other whether the potential harmful effects
are health or ecological in nature.
Production factors based on coal input rates have been developed from
the chemical analytical data available. These production factors have the
dimensions of mass/mass; specifically, the units yg produced/g coal input
have been used. Production in all measured product, byproduct and discharge
streams is included in these figures and maxima among all sources considered
in the study were selected.
ASSESSMENT RESULTS
The complex heterogeneous nature of coal gives rise to a wide variety
of organic compounds in the streams resulting from coal conversion pro-
cesses. Table 1 lists the organic compounds identified during the four
Radian Corporation source tests as well as those identified from operation
of the RTI laboratory gasifier over the last four years. Within each MEG
category, the compounds that have been quantitated are given first, followed
by those that have been identified but not measured. In addition, a large
number of inorganic compounds and elements have also been identified.
The maximum concentrations measured in the various media are presented
in Tables 2 through 4. Because of their particular properties, tars have
been considered to be a separate medium in these tables. The concentration
488
-------
TABLE 1. ORGANIC COMPOUNDS IDENTIFIED IN COAL
GASIFICATION STREAMS
MFG
1.
Category Name.
Aliphatic Hydrocarbons
methane
ethane
propane
butanes
isobutane
alkanes >C,
methyl cycl ohexane
alkanes >C, ,
C--hydrocarDons
C.-hydrocarbons
Cg-hydrocarbons
Cg+hydrocarbons
ethyl ene
propylene
acetylene
phenyl acetylene
MEG Category Name
5. Alcohols
aliphatic alcohols
>c6
aliphatic alcohols
alkymcohols >Cg
alkylalcohols >C,,
3,5,5-trimethyl-
1-hexanol
7. Aldehydes, Ketones
acetophenone
acetaldehyde
butanal
pentanal
MEG Category Name
10. Amines
aniline
C2-alkylaniline
Cg-alkyl aniline
ami no toluene
benzofluoreneamine
methyl ami noace-
naphthylene
methybenzof 1 uorene-
amine
benzidine
1-aminonaphthalene
methyl ami nonaphthal ene
aminotetralin
diphenylamine
N-methyl-o-toluidine
n-pentane
isopentane
n-hexane
2-methylpentane
3-meChylpentane
n-heptane
n-octane
n-nonane
n-decane
n-undecane
n-dodecane
n-tridecane
n-tetradecane
n-pentadecatie
n-hexadecane
methylcyclobutane
cyclopentane
cyclohexane
dimethylcyclohexane
trimethylcyclohexane
cyclooctane
dimethyldecahydro-
naphthalene
butane
isobutene
hexene
1-pentene
2-methyl-l-butene
1,3-butadiene
pentadiene
cyclopentene
cyclohexene
cyclopentadiane
ethyne
propyne
2. Alkyl. Hal ides
dichloromethane
(artifact)
trichloromethane
(artifact)
carbon tetrachloride
(artifact)
3. Ethers
anisoles
methylanisole
diethylether
phenyl-2-propynylether
1-methoxynaphthalene
2-methoxynaphthalene
3,6-dimethoxyphenanthrene
2-methoxyfluorene
p-hexanal
n-heptanal 1 3.
n-octanal
n-nonanal
undecanal
dodecanal
benzaldehyde
dimethyIbenzaldehyde
acetone
methylisopropyl ketone
butanone
1-phenyl-l-propanone
2-pentanone
o-hydroxyacetophenone 15.
m-hydroxyac etophenone
benzophenone
9-fluorenone
benzofluorenone
dihydroxyanthraquinone
tetrahydroanthraquinone
phenanthridone
8. Carboxylic Acids and
Derivatives
phthalic acids
phthalic esters
adipate esters
phthalate esters
>Cg aliphatic esters
acetic acid
benzoic acid
benzamide
ethyl acetate
ethylbenzyl acetate
methyl benzoate
isobutyl cinnamate
dibutyl phthalate
(artifact)
diisobutyl phthalate
(artifact)
dicyclohexyl phthalate
(artifact)
9. Nitriles
cyanotoluene
(benzonitrile)
acetonitrile
cyanobutadiene
2,2'-dicyanobipheny1
489
Thiols, Sulfides, and
Disul fides
methanethiol
ethanethiol
propylenethiol
2,3,4-trithiapentane
dimethyl sulfide
dimethyl disulfide
trithiahexane
diphenyl disulfide
Benzene, Substituted
Benzene Hydrocarbons
benzene
Cp-alkylbenzene
C,-alkylbenzene
tdluene
ethyl benzene
styrene
C--benzene
C^-benzene
btphenyl
biphenylene
diphenylmethane
indan
C2-alkylindane
C^-alkylindane
methylindane
xylenes
o-xylene
m- and p- xylene
xylene and ethyl
benzene
tetrahydronaphthalene
methylstyrene
ethylstyrene
n-propylbenzene
isopropylbenzene
I,2-dimethylbenzene
t-butylbenzene
n-pentylbenzene
3,5-dimethyl-l-
isopropylbenzene
Criethylbenzene
o-ethy1toluene
m-ethyltoluene
trimethylbenzene
1,2,4-trimethyl-
benzene
1,3,5-trimethylbenzene
o-diethylbenzene
m-diethylbenzene
p-diethylbenzene
-------
TABLE 1 (continued).
MEG Category Name
MEG Category Name
MEG Category
Name
15. (Continued)
methyltetrahydro-
naphthalene
dimethyltetrahydro-
naphthalene
trimethyltetrahydro-
naphthalene
1,2,3,4-tetrahydro-
naphthalene
5,8-dimethyl-l-n-
octyl-1,2,3,4-
te Crahydronaphthalene
l-methyl-4-n-heptyl-
1,2,3,4-tetra-
hydronaphthalene
methylbiphenyl
3-me thyIb ipheny1
diphenylethane
di(ethylphenyl)ethane
stilbene(l,2 diphenyl-
ethene)
methylphenylethyne
diphenylethyne
1,2-diphenylpropane
dixylylethane
o-terphenyl
m-terphenyl
p-terphenyl
dimethylindan
pentamethylindan
methy-1,2,3-dihydro-
indene
dimethylindene
trimethylindene
16. Polychlorinated
biphenyls (PCB)
17. Dinitrotoluenes
none
18. Phenols
phenols
C^-alkylphenol
C,-alkylphenol
C^-alkylphenol
isopropylphenol
n-propylphenol
cresol
xylenol
2,4,6-trimethyl phenol
1 -naphthol
1-acenaphthol
C,,-alkylacenaphthol
C^-alkylacenephthol
cf-alkylhydroxy-
acenaphthene
Cr-alkylhydroxy-
s anthracene
C?-alkylhydroxypyrene
C^-alkylnaphthol
hydroxyacenaphthene
hydroxyanthracene
hydroxybenzof1uorene
methylacenaphthol
methylnaphthol
indanol
18. (Continued)
o-cresol
m-cresol
p-cresol
o-ethylphenol
m-ethylphenol
p-ethylphenol
o-allylphenol
m-phenylphenol
2,3-xylenol
2,4-xylenol
2,5-xylenol
2,6-xylenol
3,4-xylenol
3,5-xylenol
3-methyl-6-ethyl-
phenol
2-methyl-4-ethyl-
phenol
4-tert-butyl-o-cresol
di-t-buytl-4-ethyl-
phenol
trimethylphenol
2-hydroxynaphthalene
methylhydroxy-
naphthalene
hydroxyfluorene
20. Dinitrocresol
none
21. Fused Polycyclic
Hydrocarbons
naphthalene
higher aromatics
methy!naphthalene
1-methyl naphtha!ene
2-methylnaphthalene
C-alkylnaphthalene
anthracene
Cp-alkylanthracene
9-methyl anthracene
phenanthrene
acenaphthene
acenaphthylene
C,-aTkylacenaphtha-
i lene
Cp-alkylacena-
phthene
C,-alkylace-
naphthene
binaphthyl
methylacenaphthy-
lene
methylacenaphthene
C15H,,:3 rings
benzota)anthracene
7,12-dimethylbenzo-
(a)anthracene
methyl phenanthra-
cene
methyltriphenylene
triphenylene
C16H1Q:4 rings
3-meiHylcholanth-
rene
benzo(c)phenan-
threne
21. (Continued)
chrysene
methyl crysene
pyrene
1-methyl pyrene
dibenz(a.h)-
anthracene
benzo(a)pyrene
perylene
benzo(e)pyrene
benzoperylene
benzo(g,h,i)perylene
cyclobutadibenzene
methyldihydro-
naphthalene
ethylnaphthalene
isopropyl-
naphthalene
l-methyl-7-isopropyl-
naphthalene
l,2-dihydro-3,5,8-
triaiethylnaphthalene
2-benzylnaphthalene
dimethyInaphthalene
1,4-dimethyInaphthalene
2,3-dimethyInaphthalene
2,6-dimethylnaphthalene
trimethyInaphthalene
3-methylacenaphthalene
ethylanthracene
1-methylphenanthrene
3-methylphenanthrene
4,5-methylphenanthrene
propenylphenanthrene
trans-9-propenylphen-
anthrene
8-n-butylphenanthrene
2,7-dimethylphenan-
threne
1,2-benzanthracene
hexahydro-l,2-benz-
anthracene
methyl-1,2-benzan-
thracene
2,3-benzanthracene
(naphthacene)
3,4-benzophenanthrene
methylbenzophenan-
threne
5,8-dimethy1-3,4-benzo-
phenanthrene
9,10-benzophenanthrene
(triphenylene)
1,2,3,4-tetrahydro-
9,10-benzo-
phenanthrene
2-methyl-9,10-benzo-
phenanthrene
2-n-hexyperylene
490
-------
Table 1 (continued).
MEG Category Name
MEG Category Name
..MEG Category Name
22. Fused Non-Alternant
Polycyclic Hydrocarbons
indene
C?-alkyl indene
C3-a1kyl indene
ffuorene
methyl indene
methylfluorene
benzofluorene
(fluoranthene)
benzo(b)fluorene
benzo(a)fluorene
benzo ( k ) fl uoranthene
benzo (b ) f 1 uoranthene
indenoO ,2,3-cd)pyrene
l-methylfluorene
dimethylfluorene
1,2,3,4-tetrahydro-
fluoranthene
23. Heterocyclic Nitrogen
Compounds
pyridine
Cg-alkylpyridine
C~-alkyl pyridine
C. -a Ikyl pyridine
methyl pyridine
(picolines)
dimethylpyroline
qui no lines
C2-alkylquinolines
C,-alkylquinolines
2jmethylquinoline
acridine
C,-alkylacridine
C-alklacridine
C^-alky! benzoquinol ine
Cj-al kyl benzoquinol i ne
methyl acridine
dihydroacridine
methylbenzophen-
anthradine
benzophenanthr i di ne
benzoquinol ine
(phenanthridine)
methyl benzoqui no! i ne
indole
methylindole
carbazole
methylcarbazole
pyrrol ine
pyrrole
methylpyrrole
4-acetylpyridine
trimethylpyridine
2,4-dimethyl-6-ethyl-
pyridine
23. (Continued)
2-hydroxy-4-phenyl-
pyridine
2-hydroxy-6-phenyl-
pyridine
3,4-diphenylpyridine
benzopyridine
2,2'-dimethyl-4,4'-
dipyridyl
methyl-3-allylhydro-
indole
3-methyl-3-allydihydro-
indole
phenylindole
3-methyl-2-phenylindole
3,3'-biindolyl
isoquinoline
3-methylquinoline
6-methylquinoline
ethylquinoline
3-n-propylquinoline
4-n-propylquinoline
8-n-propylquinoline
dimethylquinoline
2,6-dimethylquinoline
methylphenylquinoxaline
4-styrylquinoline
3-methylbenzoquinoline
benzimidazole
methylbenz imidaz ole
2-ethylbenzimidazole
benzylbenzimidazole
benzothiazole
2-methyl-5-phenyl-
tetrazole
diphenyloxazole
dimethylacridine
acridone
1,2,3,4-Cetrahydro-
carbazole
3-amino-9-ethyl-
carbazole
vinylphenylcarbazole
l,4-dihydro-2,3-
benzo(b)carbazole
2-amino-4-phenyl-6-
methyl-pyrimidine
2-amino-5-chloro-4,6-
dimethylpyrimidine
4-(l,2,3,4-tetrahydro-2-
naphthyl)-morpholine
3-benzylindene phthal-
imide
24. Heterocyclic Oxygen
Compounds
methyldioxolane
benzofuran
dibenzofuran
24. (Continued)
furan
2-methylbenzof uran
3-me thy Ibenzo furan
5-methylbenzofuran
7-methylbenzo furan
3 , 3-dihydro-2-methyl-
benzofuran
dimethylbenzofuran
3 , 6-dimethylbenzof uran
dihy drome thy Ipheny 1-
benzo furan
xanthene
25. Heterocyclic Sulfur
Compounds
thiophene
2-
methyl thiophene
dimethyl thiophene
benzothiophene
t rime thy 1 thiophene
isopropyl thiophene
ethyl thiophene
2-n-propyl-5-isobutyl-
thiophene
methy Ibenzo thiophene
dimethy Ibenzo thiophene
t rime thy Ibenzo-
thiophene
benzodi thiophene
methy Ibenzodi-
thiophene
dibenzo thiophene
methy Idibenzo-
thiophene
dihydrodimethylthieno-
thiophene
dimethy Ithiaindene
thiaxanthene
Note: Compounds are listed by MEG category with those which have been quantitated followed
by those for which qualitative identifications are available.
491
-------
TABLE 2. MAXIMUM CONCENTRATIONS REPORTED FOR GASEOUS STREAMS FROM
COAL GASIFICATION (yg/m3)
Gas (Product)
Carbon Dioxide
Carbon Monoxide
Methane
Hydrogen
Hydrogen Sulfide
Benzene
Thiophene
Toluene
Ethane
Ethylene
4.7E8
3.0E8
3.6E7
2.7E7
1.7E7
3.3E6
2.3E6
1.3E6
1.3E6
9.4E5
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
Gas
Carbon Dioxide
Ammonia
C,+ hydrocarbons
Benzene
Methane
Hydrogen Sulfide
Ethanethiol
Phenols
Ethane
Methanethiol
(Discharge)
1.1E9
3.2E8
2.9E8
1.3E8
5.4E7
3.0E7
2.7E7
2.6E7
2.1E7
1.1E7
K
K
K
K
K
K
K
K
K
K
RTI = Research Triangle Institute.
K = Kosovo Gasification Plant.
492
-------
TABLE 3. MAXIMUM CONCENTRATIONS REPORTED FOR LIQUID DISCHARGES FROM
COAL GASIFICATION
Organics
Phenol
Cresols
Xylenols
2,4, 6-Trimethylphenol
1-Methylnaphthalene
2 -Me thy Inaph thai ene
Chrysene
Phenanthrene
Acenaphthene
Fluorene
2.8E6
1.5E6
3.75E5
1.8E4
4.8E2
2.2E2
1.6E2
9.6E1
5.7E1
5.7E1
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
RTI
Inorganics
Ammonia
Sulfate
Sodium
Cyanide
Sulfur
7.9E6
2.8E6
1.7E6
1.0E6
9.7E5
Thiocyanate 2 . 7E5
Calcium
Sulfite
Sulfite
Nitrate
2.2E5
4.7E4
4.7E4
1.7E4
RTI
Ft. Snlg.
Ft. Snlg.
RTI
Ft. Snlg.
RTI
Ft. Snlg.
Ft. Snlg.
Ft. Snlg.
GG
RTI = Research Triangle Institute.
GG = Glen Gery Gasification Plant.
493
-------
TABLE 4. MAXIMUM CONCENTRATIONS REPORTED FOR SELECTED COAL
GASIFICATION STREAMS (yg/g)
Solid (Discharge)
Potassium
Silicon
Iron
Aluminum
Calcium
Rubidium
Sodium
Sulfur
Magnesium
Barium
4.0E5
1.4E5
9.0E4
8.8E4
5.0E4
2.0E4
1.8E4
1.5E4
1.3E4
5.5E3
Chapman
Ft. Snlg.
Ft. Snlg.
Ft. Snlg.
Ft. Snlg.
Chapman
Ft. Snlg.
GG
Ft. Snlg.
Ft. Snlg.
Tar (Byproducts)
Xylenols
Cresols
Naphthalene
Benzofluorene
Phthalate Esters
2,4, 6-Trimethylphenol
Pyrene
Phenanthrene
Anthracene
Phenols
1.2E5
6.7E4
5.7E4
3.4E4
3.0E4
2.4E4
2.4E4
2.3E4
2.3E4
2.2E4
RTI
RTI
RTI
RTI
Chapman
RTI
RTI
RTI
RTI
RTI
RTI = Research Triangle Institute
GG = Glen Gery Gasification Plant.
494
-------
maxima are tabulated without regard to stream flow rate or potential dilu-
tion effects, as such they represent a measure of potential acute exposure
hazard. Long-term effects may be gauged more realistically by consider-
ation of actual mass emissions.
For each source considered, the mass flow rates in all product, by-
product and discharge streams were summed for each chemical species quan-
titated. These sums were then normalized by dividing by the coal input
rate for each source to obtain production factors. Process streams which
do not leave the facility were excluded from this analysis to avoid counting
the same material more than once as it moves through the gasification faci-
lity. For the 14 source compilations (four from Radian plus 10 from RTI)
maximum production factors for each chemical species quantitated were
determined. These factors are listed in Table 5 accompanied by an entry
referring to the source upon which they are based. While those values have
been normalized on the basis of coal input, it must be remembered that
different streams were sampled at different locations and different chemi-
cal analytical strategies were adopted for different samples.
Priorities for monitoring, regulation, and control technology develop-
ment may be established from a ranking of the potential hazards associated
with individual chemical species. Discharge severity can be used for this
purpose. Table 6 lists those species of potential health hazard. Discharge
severities of less than one represent minimal hazards; species in this
category have been omitted from the table. The remaining species are
ranked by the order of magnitude of their discharge severity. Primary
consideration should be given to controlling those species occupying the
highest positions on the list.
A similar ranking is presented in Table 7. Here, ecological DMEG
values have been used in the calculation of discharge severities. Con-
siderable differences in pollutant rankings occur between the two tables; a
rational approach to pollutant control would emphasize the entries of
highest discharge severity on both bases.
DISCUSSION
The processing of coal to yield gaseous fuels generates substances
which are known to be hazardous. Among the wide spectrum of products,
byproducts, process intermediates and waste streams are substances noted
495
-------
TABLE 5 . MAXIMUM TOTAL PRODUCTION FACTORS FOR CHEMICAL SPECIES DETERMINED IN MEASURED
PRODUCT, BYPRODUCT AND DISCHARGE STREAMS FROM COAL GASIFIERS
Chemical Name
Naphthol
Methylnaphthol
C2-Alkylnaphthol
Hydroxyacenaphthylene
Hydroxyacenaphthene
Methylhydroxyacenaphthene
C,-A1kylhydroxyacenaphthene
C|-A1kylhydroxyacenaphthene
Hydroxyanthracene
C5-A1ky1hydroxyanthracene
Cj-Alkylhydroxypyrene
Hydroxybenzof1uorene
Dinitrocresol
Naphthalene
C--A1kylnaphthalene
1-Methyl naphthalene
2-Methylnaphthalene
Acenaphthylene
Acenaphthene
Phenanthrene
9-Methylanthracene
Anthracene
C,,.H,7: 3 rings
BIHapnthyl
Methylacenaphthy1ene
Methylacenaphthene
C2-Alkylacenaphthene
C'-Alkylacenaphthene
C|-A1kylanthracene
Higher Aromatics
Benz(a)Anthracene
Triphenylene
Chrysene
Pyrene
C,,H,n: 4 rings
7lT2-Dimethylbenz(a)
Anthracene
3-Methylcholanthrene
Benzo(c)Phenanthrene
Methylphenanthracene
Methylchrysene
Methylpyrene
Methyltrlphenylene
Di benzo(a,h)Anthracene
Benzo(a)Pyrene
Benzo(e)Pyrene
Perylene
Benzo(g,h,i)Perylene
Benzoperylene
Fluorene
Indene
Methylindene
C, AlkylIndene
Cf Alkylindene
Benzo(a)F1uorene
Benzo(b) Fluorene
Fluoranthene
Benzofluorene
Benzo(h}Fluoranthene
Benzo(b) Fluoranthene
MEG
Category
01A
01A
01 A
01A
01A
01A
01A
01 A
01A/B
01 B
01 B
01 C
QIC
03A
03A
05A
05A
07B
08A
08D
08D
08D
09B
09B
IOC
IOC
IOC
IOC
IOC
IOC
IOC
IOC
IOC
IOC
IOC
13A
13A
ISA
15A
ISA
ISA
ISA
15A/B
ISA
15B
15B
15B/A
15B/A
15B
16A
17A
18A
18A
ISA
ISA
ISA
ISA
18A
ISA
18C
Chemical Name
Methane
Ethane
Propane
n-Butane
i -Butane
Pentanes
Cg Alkanes
>C-|3 Alkanes
Ethane & Ethyl ene
Ethyl ene
Propylene
Acetylene
Phenyl acetylene
Anisoles
Methylanisole
>C6 Aliphatic Alcohols
>C-|j Aliphatic Alcohols
Acetophenone
Phthallic Acid*
Phthallic Esters*
Adi pate Esters
>Cg Aliphatic Esters
Benzonitrile
Cyanotoluene
Aniline
Benzidine
Aminonaphthalene
Methyl ami nonaphthal ene
Aminotetralin
C2-A1 kyl aniline
C3-Alkylaniline
Benzofl uorene amine
Methyl benzof 1 uoreneami ne
Methyl ami noacenaphthyl ene
Aminotoluene
Methanethiol
C2H6S
Benzene
Toluene
Ethybenzene
Biphenyl
Diphenyl me thane
C,-A1 kyl benzene
Styrene
Xyl enes
Indan
C, -Benzenes
C4-Benzenes
Tetrahydronaphthal ene
Polychlorinated Biphenyls*
Dinitro toluenes
Phenol
Cresols
Xylenols
Trimethyl phenol
0-Isopropyl phenol
C.-A1 kyl phenol
C,-A1 kyl phenol
C^- Al kyl phenol
Indanol
Production
(uq/g coal
1.2E5
3.4E3
4.2E2
1.7E2
1.7E2
1.2E-6
4.9E1
9.2E1
l.OE-7
2.4E3
4.9E2
3.1E1
2.5E-1
8.4E2
3.5E-1
3.4E2
6.2E-2
3.2E-2
1.0E1
3.0E3
2.2E3
4.8E2
2.0E-1
1.6E-1
8.9EO
2.0E1
1.0E2
1.1E-1
9.0E1
1.0E1
2.0E1
6.0E1
2.0E1
2.0E1
4.8E-1
7.8E1
1.0E2
3.8E4
2.2E3
2.3E2
9.2E1
6.5EO
4.2EO
1.1EO
8.0E2
4.4E1
1.2E2
8.4E2
6.6E2
3.1E-2
4.5EO
1.6E3
1.6E3
1.3E3
1.7E2
1.7E2
6.8E2
1.0E2
3.8E-1
3.0E1
Factor
input)
R41
R21
R21
R21
R21
K
C
C
K
R21
R21
R21
C
C
C
C
C
C
C
C
C
C
C
C
R21
R23
C
C
C
C
C
C
C
C
C
R36
R41
R35
R35
R21
R41
R25
C
C
R35
R41
R41
R41
C
FS
FS
R35
R50
R35
R43
RSI
C
C
C
C
MEG
Categor
18C
18C
18C
18C
18C
18C
18C
18C
18C
18C
18C
18C
20B
21 A
21 A
21A
21 A
21A
21 A
21A
21A
21 A
21A
21A
21A
21A
21A
21A
21 A
21 A
21B
218
21 B
21 B
21B
21B
21 B
21 B
21B
21B
213
21 B
21 C
21 C
21 C
21 C
21 D
21D
22A
22A
22A
22A
22A
22B
22B
22B
22B
22C
22C
Production Factor
(pg/g coal input)
1.8E2
2.0E2
3.0E1
7.4E-3
3.0E1
9.0E1
1.6E2
7.0E1
1.5E2
2.0E2
2.1E2
3.5E2
3.7EO
2.3E4
5.0E2
1.4E2
3.3E2
4.3E2
1.5E2
7.6E2
5.3E2
5.9E2
2.0E-1
2.8E-1
2.8E2
6.3E1
1.2E2
5.1E1
8.0E1
6.9E-9
1.6E2
2.9E2
2.9E2
7.2E2
4.3E-1
3.3E-1
9.6E-3
2.0EO
2.1E2
5.4E2
3.8E2
1.2E2
9.3E1
1.2E2
6.9E1
8.0E1
4.8E1
5.0E1
2.6E2
4.4E2
1.5E1
3.7E1
1.4EO
8.6E1
5.6E1
1.0E3
3.8E2
5.3E1
1.0E2
C
C
C
C
C
C
C
C
C
C
C
C
C
R21
C
R25
R21
C
C
R21
R21
R41
R35
C
C
C
C
C
C
K
R21
C
C
R41
R35
FS
FS
FS
C
C
C
C
R21
R21
R21
C
R25
C
R21
R41
C
C
C
R21
R21
R41
C
R21
R21
496
-------
TABLE 5 . MAXIMUM TOTAL PRODUCTION FACTORS FOR CHEMICAL SPECIES DETERMINED IN MEASURED
PRODUCT, BYPRODUCT AND DISCHARGE STREAMS FROM COAL GASIFIERS (continued)
MEG
Category
220
23A
23A
23A
23A
23A
23B
238
23B
23B
23B
23B
23B
23B
23B
23B
23B
23B
23B
23B
23C
23C
23C
23C
24A
24B
25A
25A
25A
25A
2SB
27
28
29
30
31
32
33
33
34
35
36
37
38
39
41
42
42
42
43
44
45
46
47
47
47
47
47
47
48
Chemical Name
IndenoO ,2,3-CD)Pyrene
Pyridine
Methylpyridine
C9-Alkylpyridine
Cf-Alkylpyridine
Cf-Alkylpyridine
QOincline
Acridine
Methylquinoline
C?-Alkylquinoline
Cj-Alkylquinoline
Methylacridine
Benzophenanthri di ne
Methylbenzophenanthri di ne
C,-A1kylacridine
C^-Alkylacridine
Benzoquinoline
Methylbenzoqui noli ne
C,-A1kylbenzoqui noline
DThydroacridine
Indole
Carbazole
Methylcarbazole
Pyrroline
Benzofuran
Dibenzofuran
Thiophene
Methylthiophene
Dimethylthiophene
C2-Thiophenes
Benzothiophene
Lithium
Sodium
Potassium
Rubidium
Cesium
Beryllium
Magnesium
Rhenium
Calcium
Strontium
Barium
Boron
Aluminum
Gallium
Thallium
Carbon Monoxide
Carbon Dioxide
Carbonate
Silicon
Germanium
Tin
Lead
Ammonia
Cyanide
Nitrogen Oxide
Nitrogen Dioxide
Nitrate
Nitrite
Phosphorus
Production Factor
(ug/g coal input)
4.6E1
1.6E-1
7.1E-1
2.8EO
1.2E1
2.0E1
1.9E3
9.0E1
6.0E1
2.3E2
1.1E2
4.0E1
9.6E-2
4.8E-2
9.0E1
6.0E1
7.0E1
3.0E2
6.0E1
2.2E-1
1.9EO
5.3E1
2.0E1
4.0E-2
1.3E2
2.7E2
3.7E3
2.9E2
5.0E1
3.3E2
2.6E2
4.1E1
1.5E4
7.3E3
1.2E3
6.8EO
7.6EO
1.1E4
6.1E-1
4.4E4
1.6E3
4.7E3
1.8E2
7.5E4
8.0EO
4.8E-2
9.8E5
1.2E6
3.5E-4
4.3E2
1.1E-1
1.8E1
1.1E1
8.8E3
2.1E1
7.3EO
5.3E1
2.2E-2
5.0E-4
1.7E3
R21
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
FS
R21
R50
C
C
R25
R21
R51
R41
R41
R23
R41
FS
FS
FS
C
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
GG
R48
R48
C
C
FS
C
R50
R21
R48
C
C
R21
FS
FS
MEG
Category
48
49
50
51
53
53
53
53
53
53
53
53
54
55
56
56
57
57-
58
58
59
59
60
61
62
63
64
65
66
68
69
70
71
72
72
74
76
76
78
79
80
81
82
83
84
84
84
84
84
84
84
84
84
84
84
84
84
84
85
85
Chemical Name
Phosphate
Arsenic
Antimony
Bismuth
Sulfur
Sulfate
Sulfite
Hydrogen Sulfide
Carbonyl Sulfide
Carbon Disulfide
Sulfur Dioxide
Thiocyanate
Selenium
Tellurium
Fluorine
Fluoride
Chlorine
Chloride
Bromine
Bromide
Iodine
Iodide
Scandium
Yttrium
Titanium
Zirconium
Hafnium
Vanadium
Niobium
Chromium
Molybdenum
Tungsten
Manganese
Iron
Iron Carbonyl**
Cobalt
Nickel
Nickel Carbonyl**
Copper
Silver
Gold
Zinc
Cadmi urn
Mercury
Ceri urn
Lanthanum
Neodymi urn
Praseodymium
Samarium
Dysprosium
Erbium
Europium
Gadolinium
Hoi mi urn
Terbium
Thulium
Lutetium
Ytterbium
Thorium
Uranium
Production Factor
(ug/g coal input)
9
2
1
1
7
7
1
4
1
2
1
5
4
2
1
5
4
2
2
5
5
5
3
5
3
1
8
3
2
5
1
8
1
7
1
2
6
2
1
8
8
2
6
1
9
9
2
1
1
1
1
2
3
2
6
1
2
1
2
1
.6E1
.7E1
.3E1
.7EO
.6E3
.OE1
.2EO
.1E4
.3E3
.8E2
.7EO
.9E2
.4E1
.OE-2
.7E2
.9EO
.8E3
.8E3
.9E1
.8E-1
.DEI
.OE-2
.5EO
.OE1
.8E3
.5E2
.6E-1
.5E2
.6E1
.5E2
.4E1
.7E-1
.9E2
.6E4
.1EO
.DEI
.4E1
.OE-4
.OE2
.1E-1
.6E-4
.DEI
.9E1
.4E1
.3E1
.3E1
.5E1
.4E1
.IE!
.2E-2
.3E-3
.OE-3
.9E-3
.OE-3
.8E-2
.9E-2
.9E-2
.9E-1
.OE1
.4E1
GG
GS
C
GG
FS
FS
FS
R25
R50
R50
C
R21
FS
GG
FS
GG
R21
R50
GG
C
GG
R50
FS
FS
FS
FS
FS
FS
FS
R50
GG
FS
FS
FS
GG
FS
FS
GG
C
FS
GG
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
GG
GG
GG
GG
FS
FS
* Probable Artifact
** Inferred Concentration
C - Chapman
FS = Wellman Galusha (Fort Snelling)
£G * Wellman Galusha (Glen Gery)
K = Kosovo
R( ) - RTI '(Test Number)
497
-------
TABLE 6. RANKING OF CHEMICAL SPECIES IN COAL GASIFICATION STREAMS RELATIVE
TO THEIR ENVIRONMENTAL (HEALTH) HAZARD POTENTIAL)
Discharge
Severity
(Order of
Magnitude) Gaseous
Stream Type
Liquid Solid
Tar
100,000 benzo(a)pyrene+(C,D)
cresols(R43,D)(R50,D)
xylenols+(R50,D)
benzo(a)pyrene +(R21,P)
cresols(R51,P)
xylenols+(R43,P)
10,000 ammonia+(K,D)
benzene+(K,D)
carbon monoxide(G.D)
ethanethiol(K,D)
methanethioHK.D)
chromiunH-(R43,D)*
dibenzo(a,h)anthracene+(R25,P)
trimethylphenol(R43.P)
1,000 carbon dioxide(K,S)
hydrogen cyanide+(K,D)
hydrogen sulfide(R25,P)
phenol+(K,D)
chromium+(C,D)
7,12-dimethylbenz(a)
anthracene(F.P)
thiophene(R51,P)
100 arsenic+(F,P)
carbonyl sulfide(K.S)
dibenzo(a.h) anthracene* (F,P)
hydrogen(R21,P)
iron carbonyl**(G,D)
mercury* (F,P)
selenium+(F,)P
silver+(C,D)
uranium(C.D)
10 aluminum(F,P)
aminotoluene(C.D)
barium(F,P)
benzo(a) anthracene* (F,P)
biphenyl(F.P)
cadmium+(F,P)
calcium
carbon disulfide(R50,P)
copper+(C,D)
cresols(C,D)
C4-hydrocarbons(K,S)
Cj-hydrocarbonsfK.D)
dinitrocresols+(F,P)
iron(F,P)
lithium(F,P)
magnesium(F,P)
methane(R51,P)
naphthalene+(R25,P)
nickel+(F,P)
nitrogen dioxide(C,D)
phenanthrene+(C,D)
phosphorus(F,P)
phthalate esters*+(C,D)
polychlorinated
biphenyls (PCB)*+(F,P)
potassium(C,0)
sulfur dioxide(G.D)
toluene+(K,D)
xylenols+CR35,P)
1 aminonaphthalene(C,D)
benzo(c)phenanthrene(F,P)
beryllium+(F,P)
chrysene+(C,D)
dinitrotoluene+(F,P)
indene(C.D)
lead+(C,D)
3-methyl chol anthrene( F ,P)
nitrogen oxide(C.D)
strontium(F.P)
xylenes(R51,P)
ammonia+(R25,D)
arsenic+(R50,D)
chromium+(R50,D) ***
cyanide+(C,S)
mercury(K.S)
mercury+(G,D)
benzo(a)pyrene+(R43,D) arsenic+(R36,D)
phenol+(R43,D)(R50,D) iron(F.D)
sodium(F,D) potassium(C,D)
fluoride(C.S)
selenium+(C,S)
sulfide(G.D)
aminotoluene(C,S)
barium(G,D)
iron(G.D)
lead+(R50,D)
lithium(F,D)(C,D)
phosphorus(C.S)
sulfate(F.D)
aluminum(F,D)
barium(F,D)
beryl lium+(R50,D)
manganese+(G,D)
nickel+(R51,D)
selenium+(R43,D)
Source Gasifier
chromium+(R36,P)***
naphthol(C.P)
benzo(a)anthracene+(R25,P)
indanol(C.P)
arsenic+(R51,P)
phenol+(R51,P)
KEY
Source Stream
Classification
G Wellman-Galusha (Glen-Gery) D Discharge
F Wellman-Galusha (Ft. Snelling) P Product or Byproduct
C Chapman S Process Stream
R# RTI Run No.
K Kosovo Lurgi
antimony+(C,D)
calcium(F,D)(C,D)
copper* (C,D)
lead+(G,D)
lithium(G,D)
phosphorus(C.D)
silicon(F,D)
aminotoluene(C,P)
benzofluorenamine(C.P)
benzo(b)fluoranthene(R21 ,P)
biphenyl(R36,P)
cadmium(R51,P)
chrysene+(R25,P)
copper(C.P)
lead+(C,P)
9-methylanthracene(R21 ,P)
phenanthrene+(R21 ,P) (R25.P)
phthalate esters*+(C,P)
*Probable artifact.
**Inferred from iron concentration.
***Stainless steel laboratory reactor probably resulted in increased concentration.
^Priority pollutant (consent decree compound).
498
-------
TABLE 7. RANKING OF CHEMICAL SPECIES IN COAL GASIFICATION STREAMS RELATIVE
TO THEIR ENVIRONMENTAL (ECOLOGY) HAZARD POTENTIAL
Discharge Stream Type
(Order of
Magnitude) Gaseous Liquid Solid
Tar
1,000,000
phosphorus(C.D)
naphthalene(R21,P)+
100,000 ammonia(K.D)
benzene(K.D)
ethylene(K.S)
10,000
1,000 carbon monoxide (G,D)
hydrogen sulfide(R25,P)
toluene(K,S)+
ammonia(C,S),(R25,D)+
cyanide(C,S)+
phosphorus(C.S)
phthalates(C,S)*+
cresols(R43, 49,50,0) copper (C,D)+
phenol (R32,D)+ iron(F.D)
phosphates(K.S) mereury(G,D)+
sulfide(C.S)
xylenols(R50,D)+
cresols(R51,P)
xylenol(R43,P)+
benzidine(R23,P)'t'
phenol(R51,P)+
phthalate esters(C,P)*+
trimethyl phenol (R43.P)
acridine(R20,P)
arsenic(R21,P)+
chromium(R36,P)+
o-isopropyl phenol (R51 ,P)
100 hydrogen cyanide(K,D)+
mercury(F,P)+
vanadium(C,D)
arsenic(R49,D)+
C2-alkylphenols(C,S)
cn>omium(R26,D)+**
copper(R49,D)+
naphthalene(C,S)+
sulfite(F,D)
aluminum(F,D)
chromium(R26,D)+**
silver(F,D)+
acenaphthene(R16,P)
aniline(R20,P)
cadmium(R51,P)
copper(C,P)+
mercury(R46,P)+
selenium(R51 ,P)+
10 methane (Pv5-l,P) aluminum(F,D)
barium(G,D)
boron(C.S)
cadmium(R16,D)
calcium(F,D)
Cj-alkylphenolsfC.S)
>C6-alkanes(C,S)
iron(GSF.D)
nitrates(G.D)
selen1um(C,S)+
silver(C,S,F&G,D)+
sulfate(F.D)
thiocyanate(R21,D)
titanium(G.D)
trimethyl phenol (R21 ,D)
1 C,-alkylbenzene(C,D) alkylpyridine(K.S)
d-alkylbenzene(C,D) aniline(C.S)
ethane(K.D) C2-alkylaniline(C,S)
thiocyanate(C.D) dimethylpyridine(K,S)
lead(K,S)+
lithium(G&F,D)
mercury (K,S )+
2-methylpyridine(K,S)
3&4-methylpyridine(K,S)
pyridine(K.S)
vanadium(G,D)
zinc(K,S)+
arsenic(G,C)+
barium(F,D)
calcium(C,0)
cobalt(C,D)
manganese(C,D)'1'
phthalate esters (C.D)
potassium(C.D)
titanium(F,0)
vanadium(F,D)
antimony(C,D)+
boron(F.D)
cadmium(C,D)
lithium(G.D)
nickel (R51,D)+
selenium(C,D)+
uranium(C,D)
aminonaphthalene(C,P)
aminotetralin(C,P)
C2-alkylacenaphthol (C,P)
C2-alkylbenzoquinoline(C,P)
C2-al kyl hydroxypyrene(C ,P)
Cc-alkylhydroxyanthraceneic.P)
cobalt(R52,P)
hydroxyanthracene (C , P )
hydroxybenzof 1 uorene ( C , P )
manganese(R51 ,P)+
methylnaptithol(C,P)
naphthol(C.P)
nickel (R51, P)+
titan1um(R52,P)
acenaphthol(C.P)
antimony(R49,P)+
C2-alkylacridine(C,P)
C2-alkylnaphthol(C,P)
C2-alkylphenol(C,P)
C3-alkylacridine(C,P)
C3-alkylacenaphthol(C,P)
C3-alkylnaphthol(C,P)
C3-alkylphenol(C,P)
C3-benzoquinoline(C,P)
>Cg-aliphatic esters(C.P)
indanol(C,P)
lead(R31,P)+
methyl acenaphthol (C,P)
methylacridine(C,P)
KEY
Source Gasifier
G Wellman-Galusha (Glen-Gery)
F Wellman-Galusha (Ft. Snelling)
C Chapman
R# RTI Run No.
K Kosovo Lurgi
Source Stream
Classification
D Discharge
P Product or Byproduct
S Process Stream
*Probable artifact.
**Stainless steel laboratory reactor probably resulted in increased concentration.
Priority pollutant (consent decree compound).
499
-------
for acute and chronic toxicity as well as substances capable of causing
long-term ecological damage. Indeed, one of the major goals of low Btu
gasification is the production of carbon -monoxide, a well-known poison even
at very low levels. Trace contaminants present in coal gasification streams
include some materials considered very hazardous and some considered rela-
tively benign, as well as a large number with unquantified health and
ecological effects.
From the standpoint of potential health hazard, the gaseous pollutant
having the highest discharge severity in an individual stream is benzo(a)-
pyrene. Present at discharge severities an order of magnitude lower CIO,000)
but still extremely high were ammonia, benzene, carbon monoxide, ethanethiol
and methanethiol. The concentrations of pollutants must be greatly reduced
before any environmentally acceptable discharge can take place. Overall,
61 gaseous species were found at DS levels greater than one including 26 of
the EPA priority pollutants.
Liquid pollutants representing the highest potential health hazards
were cresols and xylenols. Technology exists for the recovery or treatment
of these compounds. Ammonia, arsenic, chromium, cyanide, and mercury were
found in liquid streams at levels two order of magnitude lower (DS = 1000)
but still require high levels of control. Twenty-one species were found in
liquid streams at discharge severities greater than one; these include 10
species on the EPA consent decree list.
In the solid streams, chromium (DS = 10,000), mercury (DS = 1,000),
arsenic, iron and potassium (DS = 100) present the most serious health
hazards. It is likely that ash and dust disposal methods will be devised
to safely handle the overall material; no element specific treatment tech-
nology is available or promising. Eighteen species were found in solid
streams at discharge severities exceeding one. These included 10 EPA
priority pollutants.
The species present in tars which represent the highest potential
health hazards, are benzo(a)pyrene, cresols and xylenols (DS = 100,000).
One order of magnitude less hazardous, dibenz(a,h)anthracene and trimethyl-
phenol were found to be present. Some use for this byproduct material,
perhaps involving combustion or gasification to produce more valuable
chemicals may be feasible, eliminating or minimizing potential human
500
-------
exposure. Twenty-two species were found in the tar at DS levels greater
than one. These included 11 EPA priority pollutants.
Potential ecological hazards were more severe in some cases than
health hazards. Among the gas streams, three species: ammonia, benzene
and ethylene were found at ecological discharge severity levels of 100,000.
Phosphorus (solid phase) and naphthalene (tar) were found to have dis-
charge severities of 1,000,000. Carbon monoxide, hydrogen sulfide and
toluene were other ecologically hazardous pollutants in the gas phase (DS =
1000). Overall, 16 species were found in the gas phase at DS levels greater
than one. (This listing includes species for which supplemental DMEG
values were assigned). These included three EPA priority pollutants.
In the liquid phase, ammonia (DS = 10,000), and cyanide, phosphorus
and phthalates (DS = 1000) were the most hazardous ecologically. Forty-two
species were found in liquid streams at DS levels greater than one. These
include 14 species on the EPA priority list.
In addition to phosphorus (DS = 1,000,000), copper, iron, and mercury
(DS = 1000) were the most ecologically hazardous species in the solid
streams. Twenty-three species were found in the solid streams at DS levels
greater than one. Of these, 10 are on the EPA priority pollutant list.
Cresols and xylenols (DS = 100,000) were found in tars at DS levels
one order of magnitude lower than naphthalene but still represent extremely
high ecological hazards. In all, 46 species were found in tars with DS
levels greater than one. These include 15 species on the EPA priority
list.
Individual chemical species within the coal gasification streams con-
sidered in this analysis have been ranked in order of their potential
hazards to health and ecology. Priorities for future monitoring and
regulatory efforts can be developed on the basis of these rankings. Pri-
mary consideration must be given to expected discharges to the environment.
Many product materials of an extremely hazardous nature can be used with
minimal opportunities for human contact or ecological damage. Similarly,
intermediates within process facilities may be more hazardous than either
the starting material or the end product when considered strictly on the
basis of chemical analysis. Actual efforts towards pollution control and
towards the development of pollution control equipment must focus on eli-
minating hazardous discharges and minimizing fugitive emissions.
501
-------
SUMMARY
The U.S. Environmental Protection Agency (EPA) has supported a number
of research programs concerned with the environmental aspects of synthetic
fuels production. An environmental assessment methodology has been applied
to chemical data obtained from sampling and analysis of products, byproducts
and effluents from a laboratory gasifier at Research Triangle Institute
(RTI). In addition, data obtained during source tests of four operating
coal gasifiers by the Radian Corporation have been similarly analyzed.
Over 400 organic chemicals have been either quantitated or identified in
samples obtained under these programs. Additionally, a large number of
inorganic compounds and nearly all of the naturally occurring elements have
been found.
Of the chemical species quantitated, 61 in the gas phase, 21 in the
liquid phase, 18 in the solid phase and 22 in the tars were found at levels
exceeding their health DMEG values in at least one sample. Other potenti-
ally hazardous species for which no DMEG values have been established may
also be present. In addition a number of species in•each phase were found
at concentrations in excess of their ecology DMEG values.
The most serious hazards in the gas phase were ammonia, benzene,
benzo(a)pyrene, carbon monoxide, ethanethiol, ethylene, and methanethiol.
In the liquid phase ammonia, cresols, cyanide, phosphorus and xylenols were
found to present the most serious hazards. The greatest hazards in the
solid phase were phosphorus, chromium, copper, iron and mercury. Based on
land DMEGs, the most serious pollutants in the tar were naphthalene, benzo(a)
pyrene, cresols, and xylenols.
REFERENCES
1. Cleland, J. G., and G. L. Kingsbury, "Multimedia Environmental
Goals for Environmental Assessment, Vol. I," U.S. Environmental
Protection Agency, EPA-600/7-77-136a, November 1977. (NTIS PB
276 919/AS).
2. Cleland, J. G., and G. L. Kingsbury, "Multimedia Environmental
Goals for Environmental Assessment, Vol.11, Meg Charts and Back-
ground Information," U.S. Environmental Protection Agency, EPA-
600/7-77-136b, November 1977. (NTIS PB 276 920/AS).
3. Kingsbury, G. L., et al., "Multimedia Environmental Goals for
Environmental Assessment, Vol. Ill, MEG Charts and Background
Information Summaries (Categories 1-12)," U.S. Environmental
Protection Agency, EPA-600/7-79-176a, August 1979.
502
-------
4. Kingsbury, G. L., et al., "Multimedia Environmental Goals for
Environmental Assessment, Vol. IV, MEG Charts and Background
Information Summaries (Categories 13-26)," U.S. Environmental
Protection Agency, EPA-6QO/7-79-176b, August 1979.
5. Thomas, W. C., et al., "Environmental Assessment: Source Test
and Evaluation Report-Wellman-Galusha (Glen Gery) Low-Btu
Gasification," U.S. Environmental Protection Agency, EPA-600/7-
79-185, August 1979.
6. Page, G. C., "Environmental Assessment: Source Test and Evalua-
tion Report-Chapman Low-Btu Gasification," U.S. Environmental
Protection Agency, EPA-600/7-78-2Q2, October 1978.
7. Salja, B., et al., "Environmental and Engineering Evaluation of
the Kosovo Coal Gasification Plant-Yugoslavia (Phase I),"
Symposium Proceedings: Environmental Aspects of Fuel Conversion
Technology, IV, (April 1979, Hollywood, FL), EPA-600/7-79-217,
September 1979.
8. Bombaugh, K. J., and W. E. Corbett, "Kosovo Gasification Test
Program Results-Part II Data Analysis and Interpretation,"
Symposium Proceedings: Environmental Aspects of Fuel Conversion
Technology, IV (April 1979, Hollywood, FL), EPA-600/7-79-217,
September 1979.
9. Bombaugh, K. J., et al., "Environmental Assessment: Source
Test and Evaluation Report-Lurgi (Kosovo) Medium-Btu Gasifi-
cation, Phase 1," U.S. Environmental Protection Agency, EPA-
600/7-79-190, August 1979.
10. Cleland, J. G., et al., "Pollutants from Synthetic Fuels Pro-
duction: Facility Construction and Preliminary Tests," U.S.
Environmental Protection Agency, EPA-600/7-78-171, August 1978.
11. Cleland, J. G., et al., "Pollutants from Synthetic Fuels Pro-
duction: Coal Gasification Screening Test Results," U.S.
Environmental Protection Agency, EPA-600/7-79-200, August 1979.
12. Gangwal, S. K., et al., "Pollutants from Synthetic Fuels Pro-
duction: Sampling and Analysis Methods for Coal Gasification,"
U.S. Environmental Protection Agency, EPA-600/7-79-201, August
1979.
13. "Environmental Review of Synthetic Fuels," U.S. Environmental
Protection Agency, Vol.2, No.4, December 1979.
503
-------
EFFECT OF SLUDGE AGE ON THE BIOLOGICAL TREATABILITY
OF A SYNTHETIC COAL CONVERSION WASTEWATER
by
Philip C. Singer, James C. Lamb III, Frederic K. Pfaender,
Randall Goodman, Brian R. Marshall, Stephen R. Shoaf,
Anne R. Hickey, and Leslie McGeorge
Department of Environmental Sciences and Engineering
School of Public Health
University of North Carolina
Chapel Hill, North Carolina 27514
Abstract
Aerobic biological processes appear to be the focal point of any overall
scheme for treating coal conversion wastewaters since a significant number of
the major constituents of these wastes are biodegradable. Accordingly,
suitable design and operating criteria for biological treatment facilities
need to be developed. The studies to be described in this paper have been
conducted using a synthetic wastewater which was formulated to be
representative, in its organic composition, of actual wastewaters from coal
gasification and coal liquefaction processes. The wastewater contains
twenty-eight organic compounds, inorganic nutrients, and pH-buffers.
The synthetic coal conversion wastewater was fed to several bench-scale
activated sludge reactors, operated at different solids retention times
(sludge ages). Effluents from the reactors were analyzed by gas
chromatography and high performance liquid chromatography to assess the degree
of removal of the various constituents in the raw feed, and to identify
reaction products following biological treatment. Additionally, acute
toxicity studies using fathead minnows were conducted to evaluate the
biological impact of the treated wastewaters on aquatic life. Acute mammalian
cytotoxicity and Ames mutagenicity analyses were also performed on the reactor
effluents to assess their potential impact on human health. This paper
presents selected results of some of these analyses.
504
-------
EFFECT OF SLUDGE AGE ON THE BIOLOGICAL TREATABILITY
OF A SYNTHETIC COAL CONVERSION WASTEWATER
INTRODUCTION
In order to evaluate the biological treatability of wastewaters
generated during the course of coal gasification and coal liquefaction, a
synthetic coal conversion wastewater was formulated and fed to several
bench-scale activated sludge reactors. The composition of the synthetic
wastewater is shown in Table 1; the basis for formulating the wastewater in
this manner has been presented previously. >^ xhe synthetic wastewater
contains twenty-eight organic compounds representing the major classes of
organics identified in actual coal conversion wastewaters, and essentially
all of the specific organic compounds which have been reported to be
present at high concentrations are included. The theoretical total organic
carbon (TOC) concentration of all the components is 4,636 mg/1. The high
concentrations of pH-buffering agents were provided in order to avoid the
operational problems reported earlier due to inadequate control of pH.
It is unlikely that pH control will be a problem in treating real coal
conversion wastewaters in view of the abundant amounts of carbonate
alkalinity in the real wastewaters.
PROCEDURES
The synthetic wastewater was made up in 200-liter batches and stored in
a stainless steel tank. Carbon-filtered Chapel Hill tap water was used as
dilution water to which the twenty-eight constituents, shown in Table 1,
were added. This was accomplished by adding appropriate quantities from
concentrated stock solutions, prepared periodically from reagent-grade
chemicals and stored under refrigeration until use. It was found that in
order to prepare some of the concentrated solutions, an organic solvent was
required to maintain solubility of the component organics. Accordingly,
methanol was employed for this purpose. The TOC attributable to the
methanol was approximately 140 to 200 mg/1. This represents a change in
procedure from that reported in an earlier paper.
A series of 25-liter biological reactors were fed the synthetic
wastewater. The wastewater was introduced into each reactor by a
variable-speed peristaltic pump. Some of the reactors were operated as
chemostats, i.e. continuous-flow, completely-mixed activated sludge systems
with no recycle of solids (biomass). For these systems, the solids
residence time (SRT) or sludge age was equal to the hydraulic retention
time (HRT). Detention times of 3, 5, 7.5, 10, 20, and 40 days were
investigated during this phase of the study. The pumps feeding the 3- and
5-day reactors were operated continuously, while the pumps feeding the
other reactors were activated by a clock which operated them for a
pre-determined period once every half-hour. The other reactors were
operated with sludge recycle, on a modified fill-and-draw basis. In these
systems, the reactors were fed continuously or intermittently as described
above, but the effluent line from the reactor was kept closed, allowing the
volume of the mixed liquor to increase. At various times, the air supply
to the reactors was turned off for a short time (usually 30 min.), allowing
505
-------
Table 1. COMPOSITION OF SYNTHETIC COAL CONVERSION WASTEWATER
COMPOUND CONCENTRATION, mg/1
1. Phenol
2. Resorcinol
3. Catechol
4. Acetic Acid
5. o-Cresol
6. p-Cresol
7. 3,4-Xylenol
8. 2,3-Xylenol
9. Pyridine
10. Benzoic Acid
11. 4-Ethylpyridine
12. 4-Methylcatechol
13. Acetophenone
14. 2-Indanol
15. Indene
16. Indole
17. 5-Methylresorcinol
18. 2-Naphthol
19. 2,3,5-Trimethylphenol
20. 2-Methylquinoline
21. 3,5-Xylenol
22. 3-Ethylphenol
23. Aniline
24. Hexanoic Acid
25. 1-Naphthol
26. Quinoline
27. Naphthalene
28. Anthracene
NH4C1 (1000 mg/1 as N)
MgS04 • 7H20
CaCl2
NaHCO.,
FeNaEDTA
PHOSPHATE BUFFER: KHJ
Na.HPO.
2 4
2000
1000
1000
400
400
250
250
250
120
100
100
100
50
50
50
50
50
50
50
40
40
30
20
20
20
10
5
0.2
THEORETICAL ETOC == 4636 mg/1
3820
22.5
27.5
300
0.34
852
2176
'H00 3340
506
-------
the solids (biomass) in the reactor to settle. A portion of the
supernatant liquor was then withdrawn from the reactor, and the volume and
solids content of the remaining mixed liquor was adjusted to provide the
desired hydraulic detention times and solids residence times. Other
details describing the design and operation of the reactors have been
reported previously. »
It should be noted that there was a significant color change in the
synthetic feed solution, from clear to amber to brown, over the several
days during which it was used to feed the reactors. Attempts were made to
evaluate possible changes in wastewater composition during this time
through periodic measurements of TOG and chromatographic scans using high
performance liquid chromatography (HPLC). No changes in TOG were detected
and the chromatographic analyses established that, while some changes do
occur, these changes appear to be minimal.
Routine sampling of each reactor was performed three times a week.
Parameters measured included temperature, pH, dissolved oxygen, mixed
liquor suspended solids (MLSS), mixed liquor volatile suspended solids
(MLVSS), and total organic carbon (TOG). Other samples were collected as
desired for the measurement of biochemical oxygen demand (BOD) and chemical
oxygen demand (COD), and for more detailed analysis including analyses for
specific organic compounds using HPLC and GC/MS, aquatic toxicity, and
assessment of potential human health effects.
RESULTS OF REACTOR PERFORMANCE
Figure 1 illustrates the failure of the biological systems to treat the
full-strength synthetic wastewater. Both the chemostat and recycle
systems, with solids retention times of 20 and 40 days, respectively,
failed almost immediately despite attempts to gradually acclimatize the
microorganisms to the wastewater. A second attempt was made by reducing
the ammonia content of the synthetic feed to 250 mg/1 as N in order to
avoid potential ammonia toxicity, but again the reactors failed.
In order to overcome the possibility of toxicity due to other
constituents of the synthetic wastewater, the synthetic feed was diluted to
25% of that shown in Table 1. Other investigators^'^ have had to resort
to similar dilution procedures in order to treat coal conversion
wastewaters biologically. The resulting diluted version has a theoretical
TOC of 1,159 mg/1, making it comparable to wastewaters used, in
biotreatability experiments being conducted by others.
Figures 2 through 6 demonstrate the performance of the 5-, 7.5-, 10-,
20-, and 40-day chemostats treating the quarter-strength synthetic
wastewater. It is obvious that the gross toxicity effects observed for the
full-strength wastewater have been overcome. The effluent TOC, in general,
decreases with increasing retention time, reflecting improved treatment
efficiency. (The influent TOC during this period of operation was measured
to be 1,040 ^120 mg/1.) It should be noted that the scales for each of the
figures are not the same, so that care must be exercised in comparing the
results. No difficulties were encountered in controlling pH due to the
high buffer intensity of the raw feed; the pH held steady at 6.9 to 7.4
compared to difficulties experienced in earlier studies.^
507
-------
en
O
03
o
O
4000
3500
3000
2500
2000
1500
1000
500
0
0
HRT=20 DAYS
SRT=40DAYS
HRT=SRT=20DAYS
25 50 75
100 0 25
TIME , DAYS
50 75 100 125
Figure 1. Failure of biological reactors to treat full-strength synthetic wastewater.
-------
600
CP
E
•»
o
o
400
200
0
300
5*200
E
o
O
100
1
1
1
0 30 60 90 120
TIME, DAYS
Figure 2. Effluent TOC from 5-day reactor.
150 180
I
I
I
0
I
110 125
1
I
140 155 170
TIME, DAYS
85 200
Figure 3. Effluent TOC from 7.5-day reactor.
509
-------
300
200
E
•»
u
00
0
0
300
o»
E
u
o
200
00
0
1
I
1
1
40
80 120
TIME, DAYS
Figure 4. Effluent TOC from 10-day reactor.
I
T
T
I
0 40
80 1 20 1 60
TIME, DAYS
160 200
200 240
Figure 5. Effluent TOC from 20-day reactor.
510
-------
300
200
o
o 100
0
110
150
I
1
190 230
TIME, DAYS
270
310
Figure 6. Effluent TOC from 40-day reactor.
-------
Attempts to treat the quarter-strength wastewater with a 3-day
residence time failed. Immediately after feeding of the 3-day reactor
commenced, the effluent TOC began to rise and within a few days approached
the influent TOC. This pattern was observed a second time, implying that
the wastewater cannot be treated with such a low solids residence time.
A closer look at the TOC data in Figures 2 through 6 shows that, in
general, reasonably steady performance was maintained for about 130 to 170
days, after which the effluent TOC increased somewhat. In fact, there
appears to be a slight upward trend in the TOC data over the entire period
of observation. Accordingly, it may be inappropriate to speak of
steady-state behavior, despite the rather consistent performance of the
reactors over this long observation period. Some of the observed
fluctuations in TOC may be attributed to mechanical difficulties which were
encountered at various times during this period of reactor operation.
These included failures of the air compressor, feed pumps, and timing
devices leading to occasional losses in the air supply and to under- and
overfeeding of the reactors, respectively. Additionally, a significant
increase in the ambient temperature began at about the 160th day of
operation and this may have severely impacted the performance of the
reactors.
Some of these TOC fluctuations ultimately became rather extreme, as
shown in Figure 7, resulting in failure of the 5-, 7.5-, and 10-day
reactors despite up to six months of relatively stable performance. The
variability in reactor behavior is clearly illustrated in Figure 8 which
depicts the performance of the 20-day chemostat for more than one year of
operation. There appears to be a six-month metastable period during which
the effluent TOC averaged about 100 mg/1, followed by another three-month
metastable period during which the effluent TOC averaged about 175 mg/1.
The last three-month period of operation is marked by wide fluctuations in
performance. These results suggest that, while dilution of the wastewater
to 25% of full-strength overcomes the gross toxicity problem associated
with the raw wastewater, treatment of the diluted wastewater by a chemostat
system, such as an aerated lagoon, even at very long detention times,
provides variable performance and is inherently an unstable system.
Accordingly, additional studies were carried out in reactors involving
sludge recycle. Figure 9 shows the results of three reactors operated at a
solids residence time of 20 days, with hydraulic retention times of 2, 5,
and 10 days. Figure 10 shows performance data covering a twelve-month
period for a second reactor with a 10-day hydraulic retention time and a
20-day sludge age. The extent of treatment, as measured by the effluent
TOC for each reactor, appears to be approximately the same, with effluent
TOCs averaging 200-225 mg/1 (slightly higher and more variable for the
2-day HRT reactor). Comparing these effluent values to the influent TOC of
the quarter-strength synthetic feed, the reactors provided an 80-83%
reduction in TOC. The major "bumps" observed in the 10-day reactors, at 35
days (Figure 9) and 225 days (Figure 10) were caused by mechanical
problems; the reactors were apparently able to overcome these operational
malfunctions and return to a steady level of performance.
The conclusions reached from the data in Figures 9 and 10 are that a
sludge age (SRT) of 20 days results in the same level of treatment,
512
-------
800
600
400
200
800
600
40°
~ 200
O
400
200
400
200
200
0
5-DAY
i i i i i i i i
7.5-DAY
i i i i i i i i
10-DAY
i i i i i i i i
_ 20-DAY
i i i i
40-DAY
i i i i i i i i
30 90 150 210 270
TIME, DAYS
Figure 7. Summary of performance of chemostats treating 25% synthetic
wastewater at different detention times.
513
-------
450
350 -
250 -
o
100
200
TIME, DAYS
300
400
Figure 8. Long-term performance of 20-day chemostat treating 25% synthetic wastewater.
-------
o
h-
600
400
200
0
500
400
300
•
"~ 200
100
0
500
400
300
200
100
0
HRT=2 DAYS
HRT=5 DAYS
HRT=IO DAYS
0 25 50 75 100 125
TIME, DAYS
150 175
Figure 9. Summary of performance of recycle reactors treating 25%
synthetic wastewater with 20-day sludge age and different
hydraulic retention times.
515
-------
on
en
1000
800
en 600
e^
g 400
200
0
0
i 1 1 1 \ 1 1 r
\ - \
\ - r
i i i i i i i i
i i i i i i
50
100
50 200 250
TIME, DAYS
300 350 400
Figure 10. Long-term performance of recycle reactor with 20-day sludge age and
10-day hydraulic retention time.
-------
i-pgardless of the hydraulic residence time, but that control of the system
is more difficult at lower HRTs, resulting in more variable performance.
The long-term results shown in Figure 10 for the recycle system compared to
the long-term results shown in Figure 8 for the 20-day chemostat
demonstrates clearly the greater stability of the recycle system. Hence,
more data on reactor performance under different conditions of operation
(SRT and HRT) need to be developed using recycle systems in order to
establish suitable design criteria for treating coal conversion
wastewaters.
However, before this objective can be considered further, the question
of toxicity of the wastewater constituents, associated with the failure of
the reactors treating full-strength synthetic wastewater (see Figures 1 and
2), needs to be addressed. It should be noted that the full-strength
reactors were started up using mixed liquor from the quarter-strength
reactors, and gradually increasing the feed concentration from 25% to 100%
strength. Accordingly, the microorganisms comprising the mixed liquor in
these reactors should have been acclimatized to the wastewater
constituents, at least at the lower dilution rate. Nevertheless, shortly
after the wastewater feed reached full-strength, failure resulted,
reflecting the accumulation of constituents in the reactor which were toxic
to the microorganisms. As indicated previously, parallel results for the
full-strength synthetic wastewater with the ammonia concentration reduced
to 25% strength indicated that ammonia alone was not the causative agent in
bringing about failure of the full-strength reactors.
In order to begin addressing the toxicity question in a systematic
manner, a full-strength phenolics feed was formulated, the composition of
which is shown in Table 2. This phenolics feed contains only the major
phenolic constituents of the 28-component synthetic wastewater (compare
Tables 1 and 2). The theoretical TOC of the phenolics feed is 3739 mg/1;
hence, the seven constituents of the phenolics feed comprise 80.7% of the
TOC in the 28-component synthetic wastewater (TOC = 4636 mg/1). It should
be noted that the full-strength phenolics feed contains ammonia at a
concentration 25% of that in the synthetic wastewater.
The full-strength phenolics wastewater was fed to a chemostat with a
solids residence time of 20 days and to a recycle reactor with a solids
residence time of 40 days and a hydraulic retention time of 20 days. The
results are shown in Figure 11. Major fluctuations in the performance of
each of the reactors are apparent. Most of these fluctuations appear to be
related to documented mechanical problems associated with the operation of
the feed system and the air supply. Again, the recycle system behaves in a
more stable manner than the chemostat. Although some of the fluctuations
were rather extreme, the reactors have recovered and have been treating the
phenolic wastewater for more than four months, providing effluent TOC
concentrations as low as 200-250 mg/1. Comparing this output to the TOC of
the raw feed, this amounts to a 94-95% reduction in TOC. The concentration
of total phenols in the treated water, as measured by wet chemical analysis
on four occasions during this period, averaged 0.22 mg/1.
These results indicate that the full-strength phenolics wastewater,
with a phenol concentration of 2000 mg/1, is biologically treatable.
Hence, the toxicity problems associated with the 28-component full-strength
517
-------
Table 2. CHARACTERISTICS OF PHENOLICS FEED
CHEMICAL
1. Phenol
2. Resorcinol
3. Catechol
4. o-Cresol
5. p-Cresol
6. 3,4-Xylenol
7. 2,3-Xylenol
CONCENTRATION, mg/1
2000
1000
1000
400
250
250
250
Theoretical TOC === 3739 mg/1 as C
NH4C1 (250 mg/1 as N) 955
MgS04 ' 7H20
CaCl2
FeNaEDTA
PHOSPHATE BUFFER:
KH2P04
K2HP04
Na2HP04 ' 7H20
22.5
27.5
300
0.34
852
2176
3340
518
-------
en
1500
1000
HRT=20 DAYS
SRT=40DAYS
o
p
500
0
HRT = SRT = 20DAYS
1000
500
0
0
25
50 75
TIME .DAYS
100
125
150
Figure 11. Biological treatability of full-strength phenolic wastewater.
-------
synthetic wastewater must be due to one of the other minor constituents in
the synthetic feed. Based upon parallel biodegradability studies of model
compounds reported elsewhere,^ leading candidates responsible for the
toxicity problems include the pyridine and quinoline species, indole,
acetophenone, and aniline. This toxicity question is being explored
further by adding various of these additional constituents to the
full-strength phenolics mixture, and feeding this "spiked" phenolic
wastewaters to different biological reactors containing acclimatized mixed
liquor from the reactors represented by Figure 11.
RESULTS OF DETAILED CHEMICAL ANALYSES AND BIOASSAYS OF REACTOR EFFLUENTS
Treated effluent from the chemostats treating the quarter-strength
synthetic wastewater were collected at various times during the course of
their operations and analyzed for residual BOD, COD, and phenols using
standard methods of analysis.7'8 Additionally, samples were subjected to
specific organic analysis by high performance liquid chromatography (HPLC)
and gas chromatography/mass spectrometry (GC/MS). Aquatic bioassays
involving algae, fish, and Daphnia, and mammalian cytotoxicity and Ames
mutagenicity analyses were also conducted as a means of assessing the
aquatic and health impacts, respectively, of the biologically-treated
wastewater. Selected results from these detailed analyses are presented
here. The results need to be interpreted with some care in view of the
variability in reactor performance discussed above.
Wet Chemical Analyses
Table 3 shows the BOD, COD, and concentration of phenols in the
effluent from the biological reactors for the days indicated. These
values, compared to the measured influent concentrations, reflect the
excellent degrees of treatment which were achieved, especially during the
times when the reactors were performing in a reasonably stable manner. It
should be noted that the concentration of phenols was measured using the
4-aminoantipyrine procedure'' which responds only to certain of the
phenolic constituents. It is apparent from Table 3 that BOD and phenols
are virtually completely removed by the reactors having a solids retention
time of at least 20 days, while COD and TOC removal does not improve to any
great extent if the SRT is increased beyond 7.5 days. There appears to be
approximately 100-160 mg/1 of TOC with a COD of about 350-450 mg/1 which is
non-biodegradable in nature.
HPLC Analysis
Table 4 presents the results of HPLC analyses of the reactor effluents
on the days indicated. Fresh samples of the reactor effluent were
collected, filtered through 0.7 urn glass fiber filters, and injected
directly into the HPLC. Separation of the wastewater components in the
samples was achieved using a 60-minute water/acetonitrile solvent gradient
on a Waters uBondapak C^g analytical column. The eluted compounds were
detected by both UV absorbance at 280 nm and fluorescence at 275 nm
excitation and 310 nm emission wavelengths. Quantitation of the individual
phenolic compounds shown in Table 4 was accomplished from the fluorescence
measurements using effluent samples spiked with various quantities of the
constituents in question. In some cases, the concentrations in the table
520
-------
Table 3. SUMMARY OF WET CHEMICAL DATA ILLUSTRATING
REACTOR PERFORMANCE. (All values in mg/1.)
DAY BOD COD PHENOLS
Raw Feed 1,780 2,830 575
5-day Reactor 126 112 670
131 - - 54
133 126 670
140 235 850
147 485 1,160
154 430 1,080 94
161 360 825
168 150 1,025
169 - - 33
175 186 940
7.5-day Reactor 164 - - 0.70
168 10 570
175 3 435
185 6 445
192 10 465
194 - - 1.16
10-day Reactor 126 5 480
133 5 430
140 5 460
154 8 460
161 9 470 0.62
168 6 410
175 6 460
185 8 380
192 6 465
198 11 400 3.3
20-day Reactor 126 3 310
133 2 370 0.43
136 - - 0.35
140 4 355
147 2 320
150 - n - 0.35
154 2 360
157 - - 0.29
161 3 350
168 2 400
175 3 420
185 2 415
192 1 385
521
-------
Table 3. (continued)
196
198 3
203
204
210 3
217
218
224
226 3
231 4
233
193
198 1
205
210 2
212
219
224
226 1
231
240
252 1
254
259 1
273 3
282
_
420
_
_
450
__
_
460
_
465
-
340
345
_
420
_
430
_
375
_
_
_
_
0.19
0.18
0.22
0.25
40-day Reactor
198 i 34^ ;
0.11
0.18
0.12
0.15
0.10
0.11
0.09
522
-------
Table 4. CONCENTRATIONS OF MAJOR PHENOLIC COMPOUNDS IN REACTOR EFFLUENTS (mg/1).
on
f\i
CO
COMPONENT
PHENOL
0-CRESOL
P-CRESOL
3,4-XYLENOL
2,3-XYLENOL
3,5-XYLENOL
2,3, 5-TRIMETHYLPHENOL
CATECHOL
RESORCINOL
TOC
RAW
FEED
500
100
62.5
162.5
62.5
62.5
10
135
12.5
250
250
1159
5
DAY
0
22
33
9
<0
<0
362
-DAY REACTOR
163
.9
.2
.6
.0
.5
.5
DAY
0
30
31
7
<0
<0
362
175
.6
.2
.4
.0
.5
.5
7. 5 -DAY
REACTOR
DAY 188
<0.2
0.2
1.0
0.6
<0.2
<0.2
182
10 -DAY
REACTOR
DAY 176
<0 . 4
0.8
2.5
1.3
<0 . 5
<0. 5
182
20 -DAY
REACTOR
DAY
<0
<0
1
<0
<0
<0
105
176
.2
.005
.4
.08
.2
.2
20 -DAY
REACTOR
DAY 185
<0. 1
<0.02
<0.01
<0.02
<0 . 1
<0. 1
155
40 -DAY
REACTOR
DAY 303
<0.13
0.036
0.007
<0.004
<0.02
<0.02
165
-------
are shown as being less than a certain value; this value represents the
detection limit of the fluorescence detector for that compound at the
sensitivity used for that sample.
The HPLC results show that the removal of the phenolics increases with
increased detention time and that phenol, resorcinol, and catechol are
almost completely removed by the 5-day reactor. The cresols are completely
removed (to concentrations less than 1 mg/1) within 7.5 to 10 days while a
retention time of 20 days is required to reduce the concentrations of the
xylenols and trimethylphenol below 1 mg/1. (it should be noted that the
HPLC fluorescence procedure utilized is not capable of distinguishing among
the various isomers of a given compound.) The HPLC results are in
accordance with the phenol results reported in Table 3 in which the wet
chemical aminoantipyrine procedure was employed.
The results in Table 4 are significant from the standpoint of reactor
performance in that they show that the major phenolic constituents of the
synthetic wastewater are removed by the biological reactors, and that the
residual TOC in the effluent from the reactors is non-phenolic in nature.
Parallel HPLC analysis using the UV detector indicates that a major portion
of the residual TOC is comprised of highly polar compounds, e.g. aliphatic
acids, presumably cellular metabolites arising from the biological
degradation of the phenolics.
Acute Fish Toxicity
Samples of reactor effluent were collected continuously, over a 24-hour
period, from the reactor overflow ports, and centrifuged and filtered to
remove suspended solids. The samples were then frozen at -20°C. The low
flow rates for some of the reactors, particularly those with long detention
times, necessitated daily collection of the effluent over a relatively long
time period until enough of the effluent could be collected to perform the
bioassay. After a sufficient quantity of sample was available, the frozen
samples were thawed and aliquots of the effluent were diluted with
dechlorinated tap water to the desired concentration. Fathead minnows
(Pimephales promelas) were used for the fish bioassay. Ten liters of each
dilution were placed in a series of 5-gallon pickle jars, and 15 fish were
added to each jar. Each test concentration was done in duplicate, so that
a total of 30 fish were exposed to each concentration.
Figure 12 is a plot showing the percent mortality of the fish exposed
for 96 hours to various dilutions of the raw feed and the various reactor
effluents. The estimated 96-hour LC50 values, i.e. the lethal
concentrations of the various wastewater samples causing death of 50% of
the fish after 96 hours of exposure, are 1.1%, 6.6%, 33%, and 51%,
respectively, for the quarter-strength synthetic feed and the 5-, 10-, and
20-day reactor effluent samples. As expected, toxicity decreases as the
extent of the biological treatment increases.
Table 5 is a summary showing the characteristics of the wastewaters
tested along with the LC50 values calculated from the results in Figure
12. The fact that the TOC concentration of the sample from the 10-day
reactor is lower than that of the 20-day reactor is attributed to the
composite nature of the samples. The samples were collected over a
524
-------
en
r\3
en
99.9-
99-
90-
cr
o
UJ
tr
LJ
a.
i i i i i 111
i i|J i i i i i y i I
1.0 10
PERCENT EFFLUENT BY VOLUME
Figure 12. Acute toxicity of raw and treated synthetic wastewater
to fathead minnows.
-------
Table 5. RESULTS OF ACUTE TOXICITY TESTS
USING FATHEAD MINNOWS
SAMPLE
RAW FEED
5-DAY REACTOR
10-DAY REACTOR
20-DAY REACTOR
TOC AT
TIME OF
COLLECTION
Day
Day
Day
149-165
149-171
149-219
TOC,
mg/1
1150
328
150
189
PHENOLS ,
mg/1
516
94
0.62
0.22
9 6 -HOUR
LC50, %
1.1
6.6
33
51
LC50,
mg/1
12
21
49
96
.7
.7
.5
.4
PHENOLS
AT LC50,
mg/1
5
6
0
0
.7
.2
.2
.11
526
-------
relatively long period of time, as noted, during which some degree of
reactor instability was observed (see above discussion). The concentration
of phenols, however, as measured by the wet chemical method, is in
accordance with expectations, i.e. lower concentrations with increasing
reactor detention times. The aquatic toxicity of the reactor effluent
seems to be more closely related to the concentration of residual phenols
and to the detention time of the reactors than to the residual TOG
concentration; the LC50 for the sample from the 20-day reactor is 51%
compared to 33% for the 10-day reactor sample despite the fact that the TOC
of the latter is lower. Hence, the concentration of residual TOC, by
itself, is not a satisfactory indicator of the aquatic toxicity of the
treated wastewater- More information as to the composition of the various
treated samples needs to be known.
Table 5 also shows the concentration of TOC and phenols at the percent
dilution corresponding to the LC50s for each of the samples. It is
apparent that the constituents comprising the residual TOC become
correspondingly less toxic as the degree of treatment, as indicated by the
detention time of the reactor, increases. Furthermore, a comparison of the
last column in Table 5 with acute fish toxicity results for phenol alone
(see Figure 13 where the 96-hour LC50 for phenol is shown to be 28 mg/1)
indicates that the resulting toxicity of each of the composite samples,
including the raw feed, cannot be attributed solely to phenol. The
residual concentration of phenols at the LC50 dilution is, in each case,
significantly less than the 28 mg/1 LC50 for phenol. Hence, the aquatic
toxicity of the treated samples must be due to constituents other than
phenol, or to synergistic effects involving phenol and other constituents.
Mammalian Cytotoxicity
In order to evaluate the effectiveness of biological treatment in
alleviating potential human health effects associated with coal conversion
wastewaters, a clonal toxicity assay employing Chinese Hamster Ovary (CHO)
cells was used to compare the relative acute toxicities of the effluents
from the biological reactors and the quarter-strength raw synthetic
wastewaters. Effluent samples from the reactors were collected,
centrifuged, aliquoted in small bottles, and stored at -80°C. Individual
aliquots of the frozen samples were thawed immediately prior to use,
filtered through a 0.2 urn Nuclepore polycarbonate filter, and diluted with
various amounts of deionized water and growth medium to obtain the desired
concentrations.
Two hundred CHO cells were plated per tissue culture dish and allowed
to incubate and attach for 3 hours in a normal cell growth medium. The
medium was then removed and the appropriate dilution of the wastewater was
added. After an exposure period of 20 hours, the test solution was
removed. The cells were washed and reincubated in normal growth medium for
7 days. At the end of this incubation period, the colonies were fixed,
stained, and counted.
Figure 14 is a plot of percent survival of the CHO cells for various
dilutions of the different reactor effluents tested and the
quarter-strength synthetic raw feed. The source of the different samples
and the day of collection are shown in Table 6. Again, it should be noted
527
-------
99.9
99
90
1 1—I—I I I I I |
T r
o:
O
UJ
o
o:
UJ
CL
50
10
1.0
O.I
96-HR.
i i i i i i i i
1.0 2.0
5.0 10 20
PHENOL, mg/L
50
Figure 13. Acute toxicity of phenol to fathead minnows.
528
-------
en
ro
vo
0 10 20 30 40 50 60 70 80
PERCENT EFFLUENT BY VOLUME
90 100
Figure 14. Cytotoxicity of raw and treated synthetic wastewater to Chinese hamster ovary cells.
-------
Table 6. RESULTS OF CHO ACUTE MAMMALIAN CYTOTOXICITY TESTS
DAY OF TOC, LC50,
SAMPLE COLLECTION mg/1 %..
Raw Feed 850 1.2
5-day Reactor 114 211 21.6
10-day Reactor 114 126 12.6
20-day Reactor 114 96 58.1
20-day Reactor 219 195 24.5
40-day Reactor 314 164 29.1
530
-------
that the variability in reactor performance results in TOC values which are
not entirely consistent with each other. For example, on two different
dates, the effluent TOC concentrations from the 20-day reactor were 96 and
195 mg/1, resulting in very different cytotoxic responses. Figure 14 shows
that, with the exception of the 10-day reactor and its corresponding TOC
concentration of 126 mg/1, CHO toxicity decreases as effluent TOC
decreases. The concentrations of each sample resulting in 50% lethality of
the CHO cells, i.e. the LC50 values, are shown in Table 6. In contrast to
the fish bioassay results, TOC appears to be a reasonably good indicator
(with the exception of the 10-day reactor sample) of mammalian
cytotoxicity. The anomalous behavior of the 10-day reactor cannot be
explained.
Ames Mutagenicity
The Salmonella typhimurium mammalian-microsomal system was used to
.analyze the potential mutagenic activity of the raw and treated synthetic
wastewater. All five Ames tester strains recommended for screening
purposes were employed in this investigation. Two of the strains (TA100
and 1535) are capable of detecting mutagens which cause base-pair
substitutions, while the other strains (TA98, 1537, and 1538) have the
ability to detect frameshift mutagens. Standard experimental procedures
for the plate incorporation assay, as outlined by Ames , were followed
with one exception: due to the low concentrations of many of the chemicals
present in the wastewater, 0.5-2.0 ml sample volumes were assayed instead
of the standard 0.1 ml of sample per plate. The volume of the top agar
overlay containing the various sample volumes was kept constant at 5.0 ml.
One-liter samples of reactor effluent were collected, centrifuged,
aliquoted into smaller volumes, and stored at -80°C. Immediately prior
to use, the wastewater was thawed and filtered through a 0.2 um Nuclepore
polycarbonate filter. Each of the effluent samples as well as the raw feed
was first examined to determine an acceptable range of sample volumes which
would not be toxic to the bacterial strains and therefore would not
preclude the mutagenicity testing.
The experimental scheme for determining the mutagenicity of the samples
involved the assay of all the samples using one strain at a time, both with
and without metabolic activation using an S-9 preparation of Arochlor
1254-induced rat liver microsomes. Positive control mutagens dissolved in
dimethyl sulfoxide (DMSO), DMSO (solvent control), and an aqueous control
were always assayed along with the wastewater samples. Mutagenicity
studies were initiated with strain TA98 which has previously been reported
to exhibit significantly increased mutation rates in the presence of the
products of coal conversion processes.
Table 7 demonstrates some of the results of the mutagenicity testing
with strain TA98. A low level of direct-acting mutagenicity was found in
the raw synthetic wastewater when assayed using 1.0 ml sample volumes per
plate. Such activity was not observed in any of the reactor effluent
samples, even when tested at 2.0 ml sample volumes. (The 5-, 10-, and
20-day reactor effluent samples were collected on Day 114 while the 40-day
reactor effluent sample was taken on Day 314.)
531
-------
Table 7. DIRECT-ACTING MUTAGENICITY OF RAW AND TREATED
WASTEWATER SAMPLES WITH STRAIN TA98
REVERTANTS/PLATE MEAN REVERSION RATIO*
Aqueous Control 31 26 32 30 (1)
1 ml Raw Feed 66 62 57 62 2.1
2 ml Reactor Effluents
5-day
10-day
20-day
40-day
1 yg Daunomycin**
DMSO***
*Mean revertants on sample plate/mean revertants on control plate
**Used as positive control
***Solvent control for Daunomycin
33
31
29
27
500
25
29
34
26
30
560
35
36
36
28
30
726
25
33
34
28
29
595
28
1.1
1.1
1 (0.93)
1 (0.98)
21.0
(1)
532
-------
Direct mutagenic activity was found in the raw wastewater with strains
TA98 and TA1537, both of which detect fraraeshift mutagens. The mean
reversion ratio with TA98 for five trials using the raw feed was 2.0 (see
Table 8). Such a two-fold increase in the number of revertants over the
control is the generally-accepted criterion for positive mutagenicity
results. The mean reversion ratio with TA1537 for three trials (not shown)
was 4.6. Results with TA1538 indicate that this strain was less sensitive
to the frameshift mutagens in the raw wastewater than strains TA98 or
1537. There were no two-fold increases in reversion ratios found for any
of the effluent samples, as demonstrated in Table 8 for the TA98 strain.
The synthetic wastewater also contains weak indirect mutagenic activity
(not shown). Such activity requires the presence of a metabolic activation
system (such as S-9 discussed above) for detection. When TA1535, a
base-pair substitution detector, was used in the presence of S-9, the mean
reversion ratio was 2.1 for three trials using the synthetic wastewater.
No such increase was apparent for the effluent samples. Results were
negative with the other commonly-used base-pair substitution strain, TA100,
for the treated as well as the raw wastewater samples.
At this point, it can be concluded that biological treatment, even with
a solids residence time of only 5 days, is capable of reducing the
mutagenic activity associated with the raw synthetic wastewater to
undetectable levels at the concentrations examined. These mutagenicity
studies are continuing.
CONCLUSIONS
Based upon model studies using a synthetic coal conversion wastewater
at 25% of full-strength and aerobic biological processes with and without
solids recycle, coal conversion wastewaters appear to be biologically
treatable. TOC, COD, and BOD removal increase with increasing solids
residence time. Phenol is virtually completely removed with a sludge age
of 5 days, while the cresols and xylenols require 7.5 to 10 days and 20
days, respectively, for removal to levels below 1 mg/1. Some difficulties
were encountered in attaining stable reactor operation and steady-state
performance was difficult to achieve. The reactors with sludge recycle
demonstrated greater stability compared to the chemostats.
The full-strength synthetic coal conversion wastewater was found to be
non-treatable biologically, presumably due to the presence of constituents
at toxic levels in the full-strength sample. The toxicants do not appear
to be any of the major phenolic components (i.e. phenol, resorcinol,
catechol, cresols, xylenols). Studies are continuing to identify the
constituent(s) responsible for the toxic behavior of the full-strength
wastewater.
Bioassays of the raw and treated quarter-strength synthetic wastewater
show that the acute toxicity of the raw wastewater to fish and to mammalian
cells is markedly reduced as a result of biological treatment and that the
reduction in toxicity increases with increasing sludge age. Additionally,
at the concentrations tested, biological treatment reduces the mutagenic
activity associated with the raw synthetic wastewater to undetectable
levels..
533
-------
Table 8. SUMMARY OF TA98 REVERSION RATIOS*
WITH RAW AND TREATED WASTEWATER SAMPLES
Without Metabolic Activation (S-9)
TRIAL
1
2
3
4
5
RAW FLED
(1.0 ml)
1.8
2.1
1.9
2.5
2.0
—KE.AL.1UK tr r IiUCilNla \/.*\J mL) —
5 -DAY 10-DAY 20-DAY 40-DAY
1.4 1.0 1.1 1.0
1.1 1.1 1.0 1.0
MEAN 2.0 1.3 1.1 1.0
*Al1 ratios based on triplicate plates/sample.
534
-------
ACKNOWLEDGED NTS
The authors would like to acknowledge the assistance of Anthony
Maciorowski, Mark Sobsey, Dave Reckhow, Gerald Speitel, Roger Rader, Bert
Krages, and Eva Hett for contributing to various parts of this study. We
are grateful to EPA for sponsoring the project, and would like to thank
Drs. Dean Smith and Robert McAllister of the Industrial Environmental
Research Laboratory of the US Environmental Protection Agency at Research
Triangle Park, NC for their assistance.
535
-------
REFERENCES
1. Singer, P.C. , J.C. Lamb III, F.K. Pfaender, R. Goodman, R. Jones, and D.A.
Reckhow, "Evaluation of Coal Conversion Wastewater Treatability," in
Symposium Proceedings: Environmental Aspects of Fuel Conversion
Technology, IV, (April 1979, Hollywood, FL), EPA-600/7-79-217, U.S.
Environmental Protection Agency, Washington, B.C. (September 1979).
2. Singer, P.C., J.C. Lamb III, F.K. Pfaender, and R. Goodman, Treatability
and Assessment of Coal Conversion Wastewaters: Phase 1, EPA-600/7-79-248,
U.S. Environmental Protection Agency, Washington, D.C. (November 1979).
3. Forney, A.J., W.P. Haynes, S.J. Gasior, G.E. Johnson, and J.P. Strakey,
Analysis of Tars, Chars, Gases and Water in Effluents from the Synthane
Process, U.S. Bureau of Mines Technical Progress Report 76, Pittsburgh
Energy Research Center; Pittsburgh, PA (1974).
4. Luthy, R.G. and J.T. Tallon, Biological Treatment of Hygas Coal
Gasification Wastewater, FE-2496-43, U.S. Department of Energy,
Washington, DC (December 1978).
5. Johnson, G.E., R.D. Neufeld, C.J. Drummond, J.P. Strakey. W.P. Haynes,
J.D. Mack, and T.J. Valiknac, Treatability Studies of Condensate Water
from Synthane Coal Gasification, Report No. PERC/RI-77/13, U.S. Department
of Energy, Pittsburgh Energy Research Center, Pittsburgh, PA (1977).
6. Pfaender, F.K., and O.K. Ruehle, "Biodegradation of Coal Gasification
Wastewater Constituents," presented at the Annual Meeting of the American
Society for Microbiology, Miami, FL (May 1980).
7. American Public Health Association, Standard Methods for the Examination
of Water and Wastewater, 14th ed., Washington, DC (1975).
8. Luthy, R.G, Manual of Methods: Preservation and Analysis of Coal
Gasification Wastewaters, EE-2496-8 US Department of Energy, Washington,
DC (July 1977).
9. Duke, K.M., M.E. Davis, and A.J. Dennis, IERL-RTP Procedures Manual:
Level I Environmental Assessment. Biological Tests for Pilot Studies,
EPA-600/7-77-043, U.S. Environmental Protection Agency, Washington, DC
(April 1977).
10. Ames, B.N., et al., "Methods for Detecting Carcinogens and Mutagens with
the Salmonella/Mammalian Microsome Mutagenicity Test," Mut. Res.,
31:347-364 (1975)-
11. Epler, J.L., et al., ":Mutagenicity of Crude Oils Determined by the
Salmonella Typhimurium/Microsomal Activation System," Mut. Res.,
37:265-276 (1978).
536
-------
TREATMENT AND REUSE OF
COAL CONVERSION WASTEWATERS
Richard G. Luthy
Department of Civil Engineering
Carnegie-Mellon University
This paper presents a synopsis of recent experimental activities to
evaluate processing characteristics of coal conversion wastewaters.
Treatment studies have been performed with high-BTU coal gasification
process quench waters to assess enhanced removal of organic compounds
via powdered activated carbon-activated sludge treatment, and to
evaluate a coal gasification wastewater treatment train comprised of
sequential processing by ammonia removal, biological oxidation, lime-
soda softening, granular activated carbon adsorption, and reverse osmosis.
In addition, treatment studies are in progress to evaluate solvent
extraction of gasification process wastewater to recover phenolics and
to reduce wastewater loading of priority organic pollutants. Biological
oxidation of coal gasification wastewater has shown excellent removal
efficiencies of major and trace organic contaminants at moderate loadings,
addition of powdered activated carbon provides lower effluent COD and
color. Gasification process wastewater treated through biological
oxidation, lime-soda softening and activated carbon adsorption appears
suitable for reuse as cooling tower make-up water. Solvent extraction
is an effective means to reduce organic loadings to downstream processing
units. In addition, preliminary results have shown that solvent
extraction removes chromatographable organic contaminants to low levels.
537
-------
TREATMENT AND REUSE OF
COAL CONVERSION WASTEWATERS
INTRODUCTION
Experiments have been performed at Carnegie-Mellon University
to characterize coal gasification process wastewaters, to evaluate
basic wastewater treatment properties, and to assess wastewater
management strategies. The purpose of this paper is to review recent
experimental activities in these areas, and to indicate directions for
future research.
COAL GASIFICATION WASTEWATER MANAGEMENT
Figure 1 presents a general schematic representation of water
streams important in coal gasification process water balances. Major
streams include those associated with the boiler and cooling tower
systems, process condensates, treatment blowdowns, and slurry/sludge
waters. Process influent water streams generally include: water for
coal slurry feed, water for direct contact gas cooling or quenching,
and water for removal and/or quenching of char, ash, or slag. Process
steam requirements include steam to gasifier and make-up steam to CO
shift reactor. Process effluents are categorized as slag or ash quench
water, raw product gas quench condensate, CO shift condensate, acid gas
removal condensate and methanation condensate. The nature and quantities
of these process water and effluent streams are highly process specific.
The disposition of these streams for particular high BTU coal gasification
processes is discussed in Luthy, et al., 19801, for the COz-Acceptor, Bi-
Gas, Hygas, Synthane, and Lurgi processes.
Specific process water treatment and distributional configurations
are also strongly dependent on the particular gasification process being
considered. Thus various water management schemes exist for different
gasification processes. Some aspects of these schemes are well understood
and have become generally accepted as necessary in achieving a process
water balance. For example, raw makeup water is typically softened and
serves as process water, as cooling water, and as supply to the boiler
feed water treatment system. In contrast some aspects of high BTU coal
gasification process water balance are unique to this industry. This is
especially true with respect to treatment and reuse of heavily contaminated
phenolic wastewaters. In this case little previous experience is
available to detail issues associated with treatment and reuse of these
wastewaters; consequently, current research interest is focused on evalu-
ation of specific treatment characteristics for purposes of engineering
design and environmental assessment. There is also much interest in
evaluating wastewater treatment characteristics in order to achieve a
product water of suitable quality for reuse in the gasification process.
538
-------
RAW WATER
SLUDGE TO •*
SOLIDS DISPOSAL
SLOWDOWN
BRINE
DESALINATION
BD
BD
CONDENSATE
POLISHING
METHANATION
CONDENSATE
BD
CIRCULATING
STEAM/CONDENSATE
SOLIDS
DEWATERING
AND DISPOSAL
i 11
TO '
DISPOSAL
SOFTENING
SLUDGE
SLUDGE
ASH/CHAR SOLID
SLURRIES
RECIRCULATING.
' SLURRY WATER
COAL
GASIFICATION
PROCESS
Ox3
QUENCH
CONDENSATE
1 ¥~\"
'=^-—,- J \CO SHIFT
! CONDENSATE
ACID GAS REMOVAL
CONDENSATE
BY PASSING
OPTIONS
DEPEND ON
SPECIFIC
WASTE WATER
CHARACTERISTICS
RECYCLE
SLURRY
WATER
DRIFT
EVAPO-
RATION
DISTRIBUTION
FOR REUSE
WITH OR WITHOUT
FURTHER TREATMENT
It
COOLING
TOWER
CIRCULATION
BD
Figure 1. Major water streams in a coal gasification process water balance.
539
-------
Considerations Regarding Water Reuse
Medium and high-BTU coal gasification processes are net consumers
of water. The ability to achieve complete water reuse may have a signifi-
cant impact on the feasibility of a commercial-scale facility, especially
for semi-arid western regions and for eastern sites not contiguous or
adjacent to large rivers. A general design assumption should hold that
all major wastewater streams be considered for reuse, including high
organically contaminated streams and saline brines. Dirty water should
be cleaned only for reuse and not for discharge to a receiving water; any
water suitable for discharge is acceptable for reuse. Returning water
to a source is not economic when water must be cleaned to satisfy stringent
environmental regulations. Furthermore, treatment for reuse is likely
to require less severe processing than treatment for discharge.
Various water management schemes exist for a given gasification
process. These depend on the exact nature of the particular waters and
on the quality constraints for which waters will be reused. Though
specific processes may differ in water management configurations, it is
apparent that the cooling tower is the most likely target for wastewater
reuse. Treatment for reduction of high ammonia and organic loadings is
necessary, while some extent of demineralization and removal of residual
organic contaminants will be necessary to achieve a water within quality
constraints governing cooling tower makeup. Minimum quality constraints
governing acceptable levels of organic contamination in cooling tower
make-up are not clearly understood and must be evaluated. Also the fate
of toxic hazardous wastewater contaminants during wastewater treatment
and during cooling tower operation must be assessed. These factors will
ultimately determine the most appropriate treatment scheme to achieve
water reuse in a cooling tower.
WASTEWATER CHARACTERISTICS AND SCALABILITY
High-BTU coal gasification processes may be divided into two general
classifications with respect to levels of organic contamination in process
condensates: 1) those processes which produce little or no phenolics,
oils, and tars, and 2) those processes which produce substantial quantities
of these materials. Among those processes which produce organic contam-
inants a further division may be made between those which are significant
producers of tars and heavy oils. General data for comparison of coal
refinery condensates are presented in Luthy, 1979.2
The production of organic contaminants during coal gasification is
related to gasifier physical configuration and operating conditions.
Processes tending to show little or no organic contamination may be
either entrained flow or fluidized bed gasifiers that operate at temper-
atures greater than approximately 1050°C (1900°F) and produce ash as
slag or agglomerates. Examples of such processes are Bi-Gas, Combustion
Engineering, Koppers-Totzek, U-Gas, and Westinghouse. Gasifiers having
high coal devolatilization temperatures, such as the C02-Acceptor process
at 830°C (1500°F), also produce a cleaner product gas which in turn
yields condensates free of organic contamination (Fillo, 19793). Other
540
-------
important gasifier operating variables which relate to production of
organics are gas residence time, coal particle size and heat-up rate,
and the extent of gas-solids mixing (Nakles, et al., 19751*). Examples
of gasification processes which produce effluents with organic contam-
ination are Hygas, Synthane, slagging fixed-bed, Lurgi, and Wellman-
Galusha.
It should be recognized that published information on coal
gasification process wastewater characterization necessisarily reflects
a difference in process scales and use of various coals. Since much
of the available data are for analysis of condensates from process
development units or pilot plants, it should be expected that any changes
anticipated between pilot plant and commercial scale gasifier operating
conditions may have significant effects on gasifier effluent production,
especially with respect to organic contamination. Thus, scalability of
pilot plant data is a major issue in evaluating coal conversion pilot
plant effluent composition and distributional trends. Factors to consider
may include coal type and pretreatment, coal-to-steam ratio, gasifier
geometry and operating parameters, and raw product gas quench system
design and operation.
Wastewater treatment experiments performed at Carnegie-Mellon
University have utilized process quench waters from the Hygas and
slagging fixed-bed coal gasification pilot plants. While these process
condensates may not be representative in a quantititative sense of
wastewaters which would be expected in a demonstration or commercial
scale process, it is anticipated that the majority of organic and
inorganic species observed in these effluents may be expected to exist
in a commercial facility, though relationships between mass emissions
and concentrations may be somewhat different. In as much as the scope
of the investigations were to obtain basic information on biological
and physico-chemical treatability characteristics of gasification
effluents, the pilot plant wastewater samples were envisioned as
providing a reasonable matrix of representative contaminants which may
be expected in presently conceived commercial facilities.
TREATMENT STUDIES WITH COAL GASIFICATION CONDENSATES
There exists only a limited number of published studies on
treatment of organically contaminated coal gasification process waste-
waters, especially for the new generation of gasification processes
under development. Most of those studies have focused on physico-
chemical treatment for reduction of tars, oils, and ammonia prior to
biological oxidation, and on basic biological oxidation characteristics
of these wastewaters. These data are largely based on experience
gained from laboratory bench-scale experimentation.
Experimental biological oxidation studies have been reported for
Lurgi coal gasification process effluent (Cooke and Graham, 19655),
Synthane (Johnson, et al., 19776-, Neufeld, et al., 19787; and Drummond,
et al., 19798) and Morgantown Energy Technology Center (METC) pilot coal
541
-------
gasification wastewaters (Sack, 19799), and H-Coal pilot coal liquefaction
effluent (Reap, et al., 197710). In addition, biological oxidation studies
have been performed with pilot coal gasification process effluents obtained
from the Hygas pilot plant operated by the Institute of Gas Technology
in Chicago, Illinois (Luthy and Tallon, 198011) and the slagging fixed-
bed pilot plant operated by the Grand Forks Energy Technology Center
(GFETC) in Grand Forks, North Dakota (Luthy, et al., 198012).
A discussion of performance data and biological oxidation kinetic
values for treatment of coal conversion wastewaters is presented in Luthy
(19792). A general conclusion from these investigations is that waste-
waters processed for removal of ammonia by steam stripping followed by
activated sludge treatment for removal of degradable organic matter will
show high removal efficiencies for BOD, COD, phenolics and thiocyanate.
Nitrification has been demonstrated in several investigations. However,
because of the nature of coal gasification process condensates, activated
sludge treated wastewater will contain relatively high concentrations of
residual organic material. This material is associated with effluent COD
and color and is characteristic of oxidation of complex phenolic wastes.
REMOVAL OF TRACE ORGANIC CONTAMINANTS
Less information is available on the trace organic composition of
coal gasification wastewaters and removal efficiencies for these compounds
during treatment. Singer, et al. (1978) summarizes organic characteri-
zation data for coal conversion effluents. Information on removal effi-
ciencies for specific organic compounds from synthetic coal conversion
wastewater mixtures is presented in Singer, et al. (197813, 19791"*).
Stamoudis and Luthy (198015) provide results of screening gas chroma-
tography/mass spectrometry analysis of Hygas and GFETC pilot plant
wastewater to determine removal efficiencies during biological oxidation.
In these investigations wastewater was pretreated by lime addition and
air stripping to reduce excess alkalinity and ammonia prior to
biological oxidation. The biological reactors were complete-mix, single-
stage air activated sludge reactors, with GFETC wastewater being treated
at 33% strength and Hygas condensate at 100% strength. General
operating parameters and performance characteristics for the biological
reactors employed for evaluation of removal efficiencies of organic
constituents are summarized in Stamoudis and Luthy (I96015). Samples
of reacter influent and effluent were prepared for GC/MS analysis by
extraction with methylene chloride using generally accepted techniques
into acid, base and neutral fractions.
It was found that approximately 99% of influent extractable and
chromatographable organic material, on a mass basis, was derivatives
of phenol and represented in the acid fraction of the influent samples.
Activated sludge processing removed most of the organic constituents,
542
-------
with compounds of the acidic fractions being removed almost completely.
High removal efficiencies were also observed for compounds in the basic
fraction, with the exception of certain alkylated pyridines. The
extent of removal of compounds in the neutral fractions was dependent
on chemical structure. Aromatic hydrocarbons containing aliphatic
substitutions and certain polynuclear aromatic compounds were only
partially removed. A general broad conclusion from this study was
biological oxidation provides good to excellent removal for most com-
pounds present in the coal gasification process wastewater.
Followup studies were conducted with GFETC slagging fixed-bed pilot
plant wastewater pretreated in the same fashion as above in order to
compare removal of organic contaminants by activated sludge and powdered
activated carbon (PAC)-activated sludge treatment. Details of the
experimental procedures and results are presented in Luthy, et al. (19801)
A high suface area PAC (Amoco PX-21) was selected for use in this
study on the basis of results from wastewater batch adsorption isotherm
testing. PAC-activated sludge treatment was evaluated at sludge ages
of twenty and forty days with PAC mixed liquor equilibrium concentrations
of 0, 500, 1500, and 5000 mg/1. The reactors were operated for an
appropriate balance period to achieve steady state operation.
Activated sludge treatment with no addition of PAC showed excellent
removal of phenolics and BOD. Phenolics were reduced to less than
1 mg/1 from influent values of 1300-1500 mg/1; BOD was reduced to
about 30 mg/1 from influent concentration of 3600-3800 mg/1. COD
removal efficiencies were 85% and 88% at removal rates of 0.37 and
0.24 mg COD removes/mg MLVSS-day at sludge ages of twenty and forty
days, respectively.
PAC-activated sludge treatment gave significantly lower effluent
COD and color with increasing equilibrium carbon concentrations. In
addition, somewhat lower effluent concentrations of BOD, phenolics,
ammonia, organic-nitrogen, and thiocyanate were achieved by PAC-activated
sludge treatment compared to activated sludge treatment. PAC-activated
sludge treatment reduced foaming problems and gave a sludge with good
settling properties. Effluent characteristics were not significantly
different for PAC-activated sludge treatment at a sludge age of twenty
and forty days. In general, PAC-activated sludge treatment in this
study gave as good or better effluent characteristics than previously
reported results with other industrial wastes. A highly nitrified
effluent was produced by PAC-activated sludge treatment at a sludge
age of forty days. This effluent appears suitable for reuse as cooling
tower make-up water with respect to macro-organic contaminants.
Samples of biological reactor effluent with sludge age of forty
days and mixed liquor PAG concentrations of 0, 500, 1500, and 5000 mg/1
were screened for base and neutral fraction organic compounds. Base
and neutral fraction capillary column chromatograms of all four reactors
543
-------
were very similar. Characterization of sixteen compounds, representing
some of those which were found not to be completely removed in the
previous GC/MS study with slagging fixed-bed wastewater, gave similar
GC flame ionization detector responses in effluent samples for all four
reactors with concentration levels of these compounds in the range of
several mg/1. These results confirmed that biological oxidation of
coal gasification wastewaters removes organic contaminants to low levels,
however PAC-activated sludge treatment does not necessarily provide
significantly lower effluent concentrations of certain trace organic
compounds under conditions in which the biological oxidation process
has been optimized. The PAC results can be explained in part on
competition adsorption between very low concentration of base and
neutral fraction compounds and very high concentration of oxidized
and/or polymeric substances resulting from biological treatment of
phenolic wastes. These later substances are similar to humic materials
and are associated with residual effluent COD and color. These
substances are removed significantly by PAC-activated sludge treatment,
and they likely compete with trace organic contaminants for adsorption
on the powdered activated carbon.
EVALUATION OF A COAL GASIFICATION WASTEWATER TREATMENT TRAIN
A sample of Hygas pilot plant Run 79 coal gasification quench
condensate has been processed through sequential wastewater treatment
unit operations to evaluate treatment technology to achieve wastewater
reuse. The unit operations investigated in this study are shown in
Figure 2 and include: ammonia removal, biological oxidation, lime-
soda softening, activated carbon adsorption, and reverse osmosis.
The raw wastewater contained approximately 0.86 meqv/1 of alka-
linity and 0.94 meqv/1 of ammonia at pH of 7.7. These results plus
batch steam stripping tests showed that approximately 97% of the ammonia
can be liberated in one unit operation without chemical addition.
Removal of the remaining fraction of ammonia will require addition of
lime or caustic. If lime is used, this will result in a significant
increase in wastewater hardness (>1000 mg/1 as CaC03). In this study,
steam stripping was simulated by liming to precipitate alkalinity and
air stripping to remove ammonia. The residual hardness in stripped
wastewater was in the same range regardless if free- and fixed-leg steam
stripping or liming and air stripping were used for ammonia removal.
Biological oxidation at a COD removal rate of 0.16 mg COD
removed/mg MLVSS-day gave 90% reduction in COD from an influent value
of 6900 mg/1, and 99% reduction in BOD from an influent value of 3500
mg/1. There was also 96% removal of thiocyanate and reduction of
phenolics to 0.7 mg/1. Biologically treated wastewater contained about
30 mg/1 BOD, 700 mg/1 COD, and 1200 mg/1 hardness (as CaC03). It was
judged that if biologically treated wastewater were to be used as
make-up to a cooling tower, that the COD was sufficiently high to
promote potentially significant biological activity, and that calcium
and sulfate levels could lead to scaling and fouling problems. There-
fore, removal of calcium hardness was evaluated by lime-soda softening,
544
-------
en
J^
en
Hygas Run 79
Quench Water
Ammonia
Removal
Biological
Oxidation
Reject Brine to
Desalination
1
J
Reverse
Osmosis
1
Activated
Carbon
Adsorption
i
i
i
t
j
Lime-Soda
Softening
Permeate to
Boiler Feed
Water Polishing
Cooling Tower
Makeup Water
Bench Scale Treatment Train to Evaluate Processing Characteristics
of Hygas Process Quench Condensate
Figure 2. Bench Scale Treatment Train to Evaluate Processing Characteristics
of Hygas Process Quench Condensate.
-------
and removal of COD was assessed by granular activated carbon treatment
of softened wastewater.
Most of the calcium hardness in biological reactor effluent
existed as non-carbonate hardness owing to the consumption of alkalinity
during biological oxidation. Thus, lime-soda softening required propor-
tionally more soda than lime. This resulted in the replacement of
residual wastewater equivalents of hardness by equivalents of sodium.
Lime-soda softening reduced wastewater hardness to practical limits
(30-40 mg/1 as CaC03). These tests also indicated that flocculation
and/or filtration would be necessary to clarify sludge formed by the
softening operation. Granular activated carbon adsorption column testing
of softened biological effluent was conducted at pH of 7, a contact time
of seventeen minutes, and a loading of about 1.2 gpm/ft2. These tests
showed that approximately 80% of COD and 95% of residual color could
be removed by carbon adsorption.
Hygas wastewater processed by ammonia removal, biological oxidation,
lime-soda softening, and activated carbon adsorption was judged to be
of sufficient quality for reuse as cooling tower make-up water. At
this time it is not possible to predict the degree of cooling tower
biological activity which may be induced by residual COD of about 100
mg/1 in carbon treated effluent, although it is suspected that a
biocidal program could control this problem.
Reverse osmosis experiments were conducted with granular activated
carbon treated wastewater. Reverse osmosis treatment with a hollow
fiber polyamide membrane produced a clear colorless product, with a
TDS level comparable to tap water. Low levels of organic contaminants
(COD = 20 mg/1) did permeate the membrane. It is believe that these
compounds were low molecular weight, and that they permeated the
membrane owing to preferential sorption at the membrane-solution interface.
Product water from reverse osmosis treatment is suitable for reuse as
make-up to a boiler feed water polishing facility.
Reverse osmosis membrane fouling was not observed in this study
under operation at 75 percent conversion. Addition of a polyphosphate
inhibitor is thought to have been at least partially responsible for
this. A decline in membrane flux did occur, but this was primarily a
result of membrane compaction. Comparison of polyamide and cellulose
triacetate hollow fiber membranes showed that the polyamide membrane
provided a higher quality product water while the cellulose triacetate
membrane provided higher flux rates.
This investigation showed that a possible treatment scheme for
reuse of phenolic coal gasification effluents may include provisions
for ammonia stripping, biological oxidation, softening, and activated
carbon adsorption. These unit processes will provide a water with
sufficient quality for reuse as cooling tower make-up water. Further
study is required to assess the possibility of excessive biological
activity and/or emissions of trace compounds to the environment as a
546
-------
result of wastewater reuse in cooling towers. Resolution of this
problem may depend on large pilot cooling tower studies and on
operational experience gathered at demonstration plants.
Reverse osmosis appears to be an attractive technique to remove
wastewater dissolved solids. If reverse osmosis is employed in
treatment system design, the resulting product water will be of
sufficient quality to be used as a boiler feedwater source. However,
further study needs to be undertaken to determine the extent of membrane
fouling that could possibly occur under long term steady state operation.
It is probably best to evaluate reverse osmosis treatment units at the
pilot scale once demonstration plants have been built.
EVALUATION OF A PROPOSED TREATMENT TRAIN FOR A DEMONSTRATION PLANT
Figure 3 shows a simplified schematic of a proposed wastewater
treatment system for a slagging Lurgi process to gasify Illinois No. 6
bituminous coal (Continental Oil Company, 197916). Wastewater treatment
at this proposed facility handles streams discharging to an oily water
sewer, Rectisol process blowdown, solvent extracted wastewater from
ammonia recovery, and sanitary wastewater. As shown in Figure 3, the
treatment train for wastewater from ammonia recovery passes to an
equalization basin and then to a dissolved air flotation unit. Waste-
water is then treated biologically in an extended aeration basin of
three days hydraulic detention time. Effluent from the biological
reactors is clarified, processed through polishing filters, and then
pumped through granular activated carbon columns for removal of residual
organics. Wastewater from the activated carbon unit is pumped to the
utilities cooling tower.
The utilities cooling tower supplies cooling water to equipment
having ordinary or carbon steel metallurgy. Makeup to the utilities
cooling tower is obtained from various sources of which blowdown from
the process cooling tower comprises the largest portion of the total.
Makeup from wastewater treatment comprises about 17% of the total demand.
The plant is designed for zero discharge of wastewater. The key units
for this are multi-stage and Carver-Greenfield evaporators. The
multi-stage evaporator concentrates an approximate one percent feed
to an approximate 30 weight percent salt solution. The condensate is
recovered in the utility cooling tower and the salt solution is
concentrated to an approximate 60 weight percent aqueous slurry. The
concentrated salt mixture is chemically fixed and trucked to a landfill.
Continental Oil Company recommended that semi-commercial evaporators
be constructed and evaluated prior to constructing large units because
no commercial experience exists with wastewater from a gasification
facility, and there may be problems with scaling and foaming.
Figure 4 shows a schematic representation of experiments in progress
to evaluate essential features of a wastewater treatment train of the
547
-------
Oily Gas Liquor
from Shift Conversion
Dusty Gas Liquor
from Gasification
ion
Gas Liquor
Oil/Tar
Separation
Gas Liquor
Gravel Filter
Phenol
Extraction
01
-p»
oo
Biological
Oxidation-
Extended
Aeration
Dissolved
Air Flotation
Equalization
Basin
Acid Gas Removal
and
Ammonia Recovery
Sand
Filtration
Activated
Carbon Unit
Make up Water to
Utilities Cooling Tower
Figure 3. Proposed wastewater management scheme for a Lurgi plant gasifying Illinois
No. 6 bituminous coal (Continental Oil Company, 1979).
-------
Slagging Fixed -Bed Wastewater
i
Trace Crganics Characterization
by 6C/MS and HPLC
i
Solvent Extraction
with MIBK
Ammonia Stripping
Organics Characterization
(GC/MS and HPLC)
Activated Sludge PAC/Activated Sludge
Organfcs
Characterization
(GC/MS and HPLC)
Organics
Characterization
(GC/MS ana, HPLC)
Granular Activated Lime-Soda Softening
Carbon Adsorption (HPLC Analysis)
(HPLC Analysis)
Figure 4. Experiments in progress to evaluate essential features of a coal
gasification wastewater treatment train.
549
-------
type discussed above. This study utilizes GFETC slagging fixed-bed
lignite wastewater without dilution. Wastewater is processed through
solvent extraction, steam stripping, and biological oxidation with and
without PAC addition. Effluent from biological oxidation with no PAC
is treated by granular activated carbon adsorption, while effluent
from the PAC-activated sludge reactor is evaluated for lime-soda soften-
ing characteristics. High pressure liquid chromatographic analyses
are being performed after each treatment step to assess removal of
polycyclic aromatic hydrocarbons. Screening GC/MS analyses are being
conducted on raw, solvent extracted-ammonia stripped, and activated
sludge and PAC-activated sludge effluent to characterize removal efficiencies
for trace organic contaminants. At this writing, experiments have been
completed through biological oxidation. Gas chromatography and GC/MS
scans have been made for raw, solvent extracted-ammonia stripped, and
PAC-actived sludge effluent. A report on the results of this investi-
gation should be available for distribution later this year.
Several representative solvents were screened for use in the
solvent extraction step. As a result of this analysis methylisobutyl
ketone was selected for use owing to its measured high distribution
coefficient for phenolics. Wastewater was processed through five
sequential extraction steps at a solvent-to-liquid ratio of .1:15. This
reduced phenolics from 5500 mg/1 to about 5 mg/1. Concomitant with
phenolics removal there was 88% reduction of COD (32,000 to 3900 mg/1)
and 89% removal of BOD (26,000 to 2900 mg/1). Preliminary evaluation
of GC/MS data suggests that there is on the order of 99%+ removal for
most organic compounds through solvent extraction and ammonia stripping.
It has been demonstrated that solvent extracted wastewater can be
processed by either activated sludge and PAC-activated sludge treatment
without the need for dilution. Additionally, solvent extracted waste-
water does not show tendency to foam excessively as observed in previous
investigations. Effluent BOD values were in the range of 30 mg/1 for
both activated sludge and PAC-activated sludge treatment. PAC treatment
showed generally better removal_efficiency for TOC, COD, ammonia-
nitrogen, organic-nitrogen, SCN", and color. Initial assessment of
GC/MS scans of extracts from activated sludge and PAC-activated sludge
treated wastewater indicates that organics are reduced to extremely
low levels, generally less than several micrograms per liter.
This work has shown that solvent extraction offers several distinct
wastewater processing advantages. Aside from recovering phenolics for
use as a fuel or chemical commodity, there is achieved a marked reduction
of trace organic compounds. If the extract is to be used for fuel,
then there is the possibility of combusting toxic/hazardous organic
compounds to thermal extinction. Solvent extraction reduces organic
loading to a biological oxidation facility, and it may also serve as
a physico-chemical treatment step to moderate shock loadings of organics.
Solvent extracted coal gasification process wastewater is easier to treat
biologically than wastewater which would otherwise contain much higher
levels of organics.
550
-------
FUTURE WORK
It is planned to continue these investigations in order to under-
stand removal efficiencies and fates of trace organic compounds
during treatment of wastewaters derived from production of synthetic
fuels. Preparations are being made to perform experiments with
slagging fixed-bed wastewater generated from conversion bituminous coal.
Data gained from this study will be used to develop a model for predict-
ing the fates of various trace organic contaminants during treatment
with special emphasis on modeling removal of trace organics during
solvent extraction. It is also proposed to conduct analogous investi-
gations with oil shale and tar sand condensates where the objective
of these studies would be to characterize and evaluate removal of
organic compounds via proposed treatment trains for demonstration
facilities.
ACKNOWLEDGMENTS
Research investigations cited in this paper have received support
from the Department of the Interior, Office of Water Research and
Technology, and the Department of Energy through the Grand Forks Energy
Technology Center and the Energy and Environmental Systems Division
of Argonne National Laboratory.
REFERENCES
R.G., J.R. Campbell, L. McLaughlin, and R.W. Walters, "Evaluation
of Treatment Technology for Reuse of Coal Coking and Coal Gasification
Wastewaters," report prepared for U.S. Department of the Interior, Office
of Water Research and Technology, currently under review, July 1980.
2Luthy, R.G., "Treatment of Coal Coking and Gasification Wastewaters,"
Paper Presented at the 52nd Annual Meeting, Water Pollution Control
Federation, Houston, Texas, 1979, to appear in Journal Water Pollution
Control Federation.
3Fillo, J.P., "An Understanding of Phenolic Compound Production in
Gasification Processing," PhD Thesis, Department of Chemical Engineering,
Carnegie-Mellon University, Pittsburgh, PA, 1979.
^Nakles, D.V., M.J. Massey, A.J. Forney, and W.P. Haynes, "Influence
of Synthane Gasifier Conditions on Effluent and Product Gas Production."
Pittsburgh Energy Research Center. Report PERC/RI-75/6. Pittsburgh,
PA, 1975.
5Cooke, R. and P.W. Graham, "Biological Purification of the Effluent
from a Lurgi Plant Gasifying Bituminous Coals," Int. J. Air and Water
Pol., Vol 9, No 3, pp 97-112, 1965.
551
-------
6Johnson, G.E., et al., "Treatability Studies of Condensate Water
from Synthane Coal Gasification," Pittsburgh Energy Technology Center
Report N. PERC/RI-77/13, Pittsburgh, PA, November 1977.
7Neufeld, R.D., C.J. Drummond, and G.E. Johnson, "Biokinetics of Activated
Sludge Treatment of Synthane Fluidized Bed Gasification Wastewaters,"
paper presented at the 175th National ACS Meeting, Anaheim, CA, March 1978.
8Drummond, C.J., et al., "Biochemical Oxidation of Coal Conversion
Wastewaters," Proceedings 87th National Meeting AIChE, Boston, MA, August 1979.
9Sack, W.A., "Biological Treatability of Gasifier Wastewater," Morgantown
Energy Technology Center Report No. METC/CR-79/24, Morgantown, WV, June 1979.
10Reap, E.J., et al., "Wastewater Characteristics and Treatment Technology
for Liquefaction of Coal Using the H-Coal Process," Proceedings of the
32nd Purdue Industrial Waste Conference, Ann Arbor Science, Ann Arbor,
MI, pp 929-943, 1977.
i:LLuthy, R.G. and J.T. Tallon, "Biological Treatment of a Coal Gasification
Process Wastewater," Water Research, Vol 14, No 9, pp 1269-1282, September 1980.
12Luthy, R.G., D.J. Sekel, and J.T. Tallon, "Biological Treatment of a
Synthetic Fuel Wastewater," Journal Environmental Engineering Division,
ASCE, Vol 106, No EE3, pp 609-629, June 1980.
13Singer, P.C., et al., "Assessment of Coal Conversion Wastewaters:
Characterization and Preliminary Biotreatability," U.S. EPA Report
EPA-600/7-78-181, Washington, DC, 1978.
•^Singer, P.C., J.C. Lamb, F.K. Pfaender, and R. Goodman, "Treatability
and Assessment of Coal Conversion Wastewaters: Phase 1," U.S. Environ-
mental Protection Agency, IERL, EPA-600/7-79-248, November 1979.
15Stamoudis, V.C. and R.G. Luthy, "Determination of Biological Removal
of Organic Constituents in Quench Waters from High-BTU Coal Gasification
Pilot Plants," Water Research, Vol 14, No 8, pp 1143-1156, August 1980.
16Continental Oil Company, "Phase 1: The Pipeline Gas Demonstration
Plant-Design and Evaluation of Commercial Plant; Volume 2. Process and
Project Engineering Design," U.S. Department of Energy, FE-2542-10
(Vol 2), 1978.
552
-------
PILOT PLANT EVALUATION OF H2S, COS, AND C02 REMOVAL
FROM CRUDE COAL GAS BY REFRIGERATED METHANOL
by
R. M. Kelly, R. W. Rousseau, and J. K. Ferrell
Acid gas removal systems are a necessary part of coal gasifica-
tion processes. Carbon dioxide must be removed from gasifier product
gas to improve the energy content of the gas and several sulfur com-
pounds must be taken out to protect downstream process catalysts as
well as reduce potential sulfur emissions.
At North Carolina State University, an integrated coal gasifica-
tion- gas cleaning test facility is being used to study the environ-
mental and process implications of several different acid gas removal
solvents. Details of the plant facilities and operating procedures
may be found in a recent EPA technical report (Ferrell et al.,
EPA-600/7-80-046a, March 1980) (1). This paper presents some of the
initial results from acid gas removal pilot plant operation, discusses
several aspects of methanol use for acid gas removal and outlines fu-
ture experimental work on this part of the process.
553
-------
INTRODUCTION
The choice of a solvent for acid gas removal in a coal gasifica-
tion process depends upon several factors. Consideration must be
given to the type of gasification scheme used, the sulfur content of
the coal, the end use of the product gas and, most importantly, the
process chosen for off-gas sulfur recovery. For both economic and en-
vironmental reasons, most large-scale coal gasification processes cur-
rently planned in the United States include some type of sulfur reco-
very unit. In general, the higher the sulfur content of the stream
being sent to the recovery unit, the more favorable the economics.
The type of solvent chosen, therefore, should exhibit some selectivity
between the product gases, the sulfur compounds, and carbon dioxide.
Both chemical and physical solvents have been considered for use
in acid gas removal systems for coal gasification. The choice of one
type of solvent over the other depends to a large extent on the par-
tial pressure of the acid gases in the gas stream to be treated.
Chemical solvents are preferred for low to moderate acid gas partial
pressures, while physical solvents would be preferred at high acid gas
partial pressures (see Figure 1). This basis of comparison reflects
only the capacity of a particular type of solvent for acid gases and
accounts neither for the selectivity between carbon dioxide and sulfur
gases nor for the effectiveness of the solvent in treating specific
sulfur compounds.
Very little information is available concerning the fate of cer-
tain sulfur compounds in either physical or chemical solvents. In a
study undertaken to evaluate sulfur emission controls for the Western
Gasification Company's coal gasification project in New Mexico, it was
estimated that 1% of the total sulfur fed to a Lurgi gasifier would
report as carbonyl sulfide. This takes on additional significance
when considering that this represents almost 2.2 tons/day of
sulfur(2). Because hydrogen sulfide and carbonyl sulfide are not ab-
sorbed/stripped with the same efficiency in most solvents, failure to
account for each compound could result in unexpectedly high sulfur em-
issions.
As part of our research program, we plan to evaluate the effec-
tiveness of both physical and chemical solvents in removing acid gases
from both gasifier product gas and synthetic gas mixtures. Also, the
build-up in the solvent of sulfur, nitrogen, and hydrocarbon species
will be monitored. The results reported here are from experiments
using a gas produced during fluidized bed gasification of Western Ken-
tucky No. 11 coal char with emphasis on the fate of H2S and COS in
the acid gas removal system.
554
-------
FIGURE 1
EQUILIBRIUM DIAGRAM
oc
u.
UJ
-I
o
2
w
<
o
PHYSICAL
SOLVENT
CHEMICAL
SOLVENT
LIQUID MOLE FRACTION
555
-------
PROCESS DESCRIPTION
Figure 2 shows a process flow sheet for the acid gas removal sys-
tem (AGRS) used in this study. It was designed to operate with four
different solvents:
1. refrigerated methanol
2. hot potassium carbonate
3. monoethanolamine
4. dimethyl ether of polyethylene glycol(DMPEG)
With minor modifications, other solvents could also be used. Feed gas
from either the gasifier or from a mixing manifold can be used in mak-
ing process measurements.
The AGRS consists of an absorber-flash tank-stripper combination
with the necessary auxilliary equipment. The flash tank can be oper-
ated at pressures ranging from atmospheric to 28 atmospheres absolute.
For good system performance, it is normally operated around 8 atmos-
pheres absolute. The absorber and stripper are both packed columns,
each containing three sections of packing, any or all of which can be
used in mass transfer studies. Both are insulated and approach adia-
batic operation. Operating ranges and column characteristics are
given in Table 1„
A refrigeration system provides sufficient cooling to feed metha-
nol to the absorber at temperatures as low as 236 K (-35 F). Inert
gas (nitrogen) is used to strip the methanol of acid gases but a rebo-
iler is available for thermally stripping (regenerating) the chemical
solvent systems.
Plant operation is monitored and regulated from a control room
using graphical displays on a video terminal and a Honeywell TDC 2000
process control conputer. Signals from 96 process sensors (tempera-
tures, pressures, flow rates, and differential pressures) are sent to
a PDP-11/34 plant data acquisition system.
All chemical analyses are done on the premises with occasional
GC/ mass spectrometry done by EPA contractors. In the future, when
the char used as gasifier feed is replaced by coal, the recirculating
AGRS solvent will be checked for hydrocarbon build-up as well as for
any trace materials of environmental or process signficance.
556
-------
FIGURE 2
ACID GAS REMOVAL SYSTEM (AGRS)
SYN GAS
on
SOUR GAS
DEHYDRATOR
OH
SOUR-GAS
COMPRESSOR
HEAT
EXCHANGER
FIC = Flow Indicator and Controller
PIC = Pressure Indicator and Controller
TIC = Temperature Indicator and Controller
S = Sample Port
SOLVENT
PUMP
-------
TABLE 1
COLUMN SPECIFICATIONS AND OPERATING RANGES FOR REFRIGERATED METHANOL
Total Packed Height
column Diameter
No. of Packed Sections
Packing Type
Packing Size
Operating Temperature
Operating Pressure
Liquid Flow Rate
Gas Flow Rate
Absorber
21.3 ft
5 inches
3
Ceramic Intalox Saddles
1/4 inch
-35 F to -10 F
100-500 psig
0.5-1.5 gpm
10-20
Stripper
21.3 ft
6 inches
3
Ceramic Intalox Saddles
1/4 inch
-10 F to 60 F
10-25 psig
0.5-1.5 gpm
2-10
558
-------
MASS BALANCE RESULTS
One of the major objectives of all initial runs was to achieve a
closed material balance around the pilot plant. This required the
ability to operate the plant at a steady state for long periods of
time. Also, accurate flow measurements and chemical analyses are ne-
cessary as are proper sampling techniques.
A considerable amount of time was spent in improving mass balance
closure so that deviations of less than 10% resulted. Because all
flow streams were measured by orifice flow meters and laminar flow
elements, calibrations had to correct for the effect of chemical com-
position on flow stream properties. To account for differences
between the gas used for calibration and the process gas, a density
correction was provided for orifice meter calibrations and a viscosity
correction was provided for laminar flow element calibrations. These
corrections were made to the flow rate measurements recorded by the
data acquisition system and reported in a run summary.
While there is still room for improvement, the mass balance clo-
sure was adequate to reach some conclusions concerning the distribu-
tion of various compounds in the system. Improvement in the current
mass balance closure will come from improved sampling techniques,
especially for sulfur species, as well as better process control to
enhance the quality of the steady state.
USE OF METHANOL AS AN ACID GAS REMOVAL SOLVENT
The choice of an acid gas removal system in coal gasification
processes requires consideration of both process and economic factors.
The residual levels of sulfur compounds and carbon dioxide, and their
disposition in the AGRS, usually serve as the bases for decision. The
options available include hot gas clean-up, direct conversion, physi-
cal and chemical solvents and no acid gas removal at all. Any process
requiring the removal of both carbon dioxide and sulfur compounds at
high acid gas partial pressures will probably use a physical solvent.
Although there are a score of proposed physical solvent processes
for acid gas removal, only a few have been proven commercially.
SELEXOL (DMPEG), developed by the Allied Chemical Corporation, and
Rectisol (refrigerated methanol), developed by the Lurgi Corporation,
are most frequently mentioned in coal gasification applications. Both
are capable of achieving high degrees of carbon dioxide and sulfur gas
removal and show sufficient selectivity for specific acid gases. The
initial part of our study focused on the use of refrigerated methanol.
Figure 3 shows the solubilities of various gases in methanol as a
function of temperature (3). This plot shows only the solubility of
559
-------
FIGURE 3
SOLUBILITY OF GASES IN METHANOL (2)
cu
=3
V)
CO
O)
10
«c
C
O)
r—
o
C
o
CO,
o
in
o
v>
to
£
o
(0
.1
.01
CH.
CO
N,
-60
-30 0
Temperature (°F)
560
30
60
-------
each gas at a partial pressure of one atmosphere and does not reflect
the thermodynamic non-idealities associated with the multicomponent
system at higher pressures. Nevertheless, there are several points
that can be made regarding the general behavior of these constituents
in methanol.
In general, all gases shown here have an increased solubility
with decreasing temperature and increasing partial pressure. Hydrogen
and nitrogen are notable exceptions. Hydrogen solubility increases
with temperature while nitrogen solubility is insensitive to tempera-
ture. The three acid gases (H2S,COS,C02) are considerably more solu-
ble than the other permanent gases and differ somewhat among them-
selves in solubility. At individual partial pressures of one atmos-
phere, the ratios of solubilities of various gases at a temperature of
-40 F are shown in Table 2. Thus, one might conclude that the acid
gases can be separated from the permanent gases and from each other
given an appropriate separation scheme. In practice, however, thermo-
dynamic factors and mass transfer restrictions make complete separa-
tion difficult.
Clearly, the evaluation of an acid gas removal system must con-
sider both the ability of the solvent to remove acid gases to suffi-
ciently low levels as well as its ability to separate carbon dioxide
from the sulfur compounds. The absorber-flash tank-stripper combina-
tion used in this study cannot be operated to remove selectively the
specific acid gases but removal efficiencies of each acid gas can be
determined over a range of operating conditions. This information
will then be used in developing a mathematical model to describe pilot
plant operation and extended to predict both removal efficiencies and
selectivity for other configurations. The necessary vapor-liquid
equilibrium information is being developed in a parallel study and
some results are already being used (4,5). Also, several pilot plant
runs using synthetic gas mixtures are being used to determine process
parameters. The final product of this study will be a computer simu-
lation package useful in evaluating several process configurations for
acid gas removal with methanol.
INITIAL RESULTS - REFRIGERATED METHANOL
Tables 3, 4, and 5 summarize some initial results of the current
research program. It should be pointed out that the objective of
these runs was not to remove as much of the acid gases as possible but
rather to evaluate the effect of changing certain process variables on
removal efficiencies. These runs represent a portion of a larger ex-
perimental program which is still in progress and will be the subject
of a future report.
561
-------
TABLE 2
RELATIVE SOLUBILITIES IN METHANOL AT -40°F (233K)
r Solubility of Gas Solubility of Gas
Solubility of H2 Solubility of C02
H2S 2540 5.9
COS 1555 3.6
C02 430 1.0
CH4 12
CO 5
N 2.5
562
-------
TABLE 3
OPERATING CONDITIONS
Absorber
Pressure (atm.abs.)
Height of Packing (ft)
Inlet Liquid Flow Rate
Inlet Liquid Temp. (°F)
Inlet Gas Flow Rate (—
Inlet Gas Temp. (°F)
Flash Tank
Pressure (atm. abs.)
Stripper
Pressure (atm. (abs.)
Height of Packing (ft)
Stripping N2 flow (
Stripping N2 Temp. (°F)
30
28.2
7.1
60.7
-34.1
16.2
54.0
7.8
1.7
21.3
0.9
75.0
35
28.2
7.1
72.1
-36.3
15.9
53.9
7.8
1.7
21.3
0.9
75.0
36
21.4
7.1
72.6
-32.4
16.4
57.5
7.8
1.7
21.3
0.9
75.0
37
31.6
21.3
71.1
-36.3
16.6
59.9
7.8
1.7
21.3
0.9
75.0
563
-------
TABLE 4
RATIOS OF ACID GAS CONCENTRATIONS IN PROCESS STREAMS
Run #
30 C02/H2S
H2S/COS
co2/cos
35 C02/H2S
H2S/COS
co2/cos
36 C02/H2S
H2S/COS
co2/cos
37 C02/H2S
H£S/COS
co2/cos
Sour Gas
27.0
21.7
585.7
34.7
17.9
622.4
23.4
18.6
435.4
29.0
18.4
533.0
Sweet Gas
30.4
16.0
486.7
25.7
12.3
316.7
6.2
17.0
105.0
15.6
13.7
213.3
Flash Gas
68.1
14.7
1004.7
80.7
13.8
1117.1
59.3
15.0
887.6
76.1
14.6
1112.1
Acid Gas
28.1
21.7
611.1
36.4
15.5
566.0
31.4
16.7
524.0
31.4
17.4
546.5
564
-------
TABLE 5
ABSORBER OPERATION
Pressure (atm absolute)
Ht. of packing (ft)
L in (# moles/hr-ft2)
G in (# moles/hr-ft2)
G out (# moles/hr-ft2)
\ in (°F)
TL out (°F)
TG in (°F)
Liquid Temperature rise (°F)
H2S in (ppm)
H2S out (ppm)
% removed
COS in (ppm)
COS out (ppm)
% removed
C02 in («)
C02 out (%)
% removed
30
28.2
7.1
60.7
16.2
11.5
-34.1
3.5
54.0
37.6
9096
476
96.3
423
32
94.3
24.6
1.5
95.8
35
28.2
7.1
72.1
15.9
11.4
-36.3
-0.6
53.9
35.7
8072
371
96.7
449
34
94.5
28.0
1.0
97.6
36
21.4
7.1
72.6
16.4
12.8
-32.4
-1.7
57.5
30.7
8918
682
94.0
476
37
94.2
20.9
0.4
98.4
37
31.6
21.3
71.1
16.6
12.1
-36.3
7.2
59.9
43.5
8631
405
96.7
475
27
95.8
25.1
0.6
98.1
565
-------
Mass balance reports for the runs listed in Tables 3, 4, 5 are
included in the Appendix.
In general, AGRS balances meet the established criteria of less
than a 10% deviation from complete closure. In cases where more than
a 10% deviation was measured, flow meter and chemical analysis prob-
lems have been cited and will be corrected in future runs. Because
solvent losses are an important consideration in using methanol, this
analysis has recently been incorporated into the research program and
results are reported in runs 35 and 37. This will be done routinely
in the future. Failure to account for methanol losses in the gas ex-
iting the flash tank and in the acid gas stream is probably a factor
in mass balance overestimation.
Calculated liquid compositions exiting each vessel are reported
as determined by difference. In the past, liquid samples between co-
lumns and at the stripper exit were taken as were samples from the co-
lumn packing. Sampling and analytical problems led to the temporary
abandonment of this practice but it will be reinstated in the future.
The liquid exiting the stripper, however, is usually sampled and ana-
lyzed for residual acid gases. A check was also made of the hydrocar-
bon content of the solvent after approximately 60 hours of operation.
No detectable hydrocarbons were found which is not a suprising result
considering the fact that char and not coal was used as a feedstock to
the gasifier (6). Future experiments call for the gasification of
coal-char mixtures where the build-up of hydrocarbons in the methanol
will be monitored and compared to the results obtained for char gasif-
ication.
The results presented here are from the clean-up of gases gener-
ated by the gasification of Western Kentucky No. 11 bituminous coal
char. This char contains very little volatile matter (less than 2.0%)
so that the sulfur gases produced will generally be the product of the
gas phase hydrolysis of H2S, the predominant sulfur gas form. This
means that most of the sulfur gases fed to the AGRS will be in the
form of H2S, COS, with small amounts of CS2. Traces of methyl mercap-
tan, ethyl mercaptan, methyl sulfide and thiophene were also found in
some gas streams but their irregular appearance prevent any quantita-
tive conclusions concerning their distribution in the AGRS. These
sulfur species are probably related to the volatile matter present in
the feed char. Present efforts include a more detailed look at the
fate of the less concentrated sulfur species.
DISCUSSION OF RESULTS
1. System Performance
The results presented in Tables 3, 4, 5 and in the Appendix re-
present system performance for a series of runs made at fairly low li-
quid to gas (L/G) ratios. These results verify the expected order of
solubility for the three acid gases in methanol and show how these
566
-------
gases distribute in the acid gas removal system. Although the system
is considerably simpler than a commercial process, it does contain the
three basic unit operations (absorption, flash vaporization, and
stripping) found in the Rectisol process.
Overall system performance can be discussed using run AM-30 as an
example. This run was made using 7.1 feet of packing in the absorp-
tion column and 21.3 feet of packing in the stripper. Because current
emphasis is on absorber operation, each of the four runs shown here
utilized the total packed height of the stripper so that esentially
clean methanol could be fed to the absorber. This was verified
through the analysis of the methanol leaving the stripper.
The mass balance report of AM-30 shows that each compound, with
the exception of C02, was within 4.0% of complete closure. The C02
balance offset can be traced to flow meter calibration problems for
the Acid Gas stream and also to failure to account for the methanol
present in this stream. This problem also appeared in runs 36 and 37
and has been corrected for future runs. A mass balance of this quali-
ty gives added significance to the results obtained especially for the
sulfur compounds. Methanol analyses of the three exiting gas streams
were not done for this run, but other runs showed negligible amounts
in the Sweet Gas with the concentration increasing for the Flash Gas
and the Acid Gas. The increased presence of methanol in these streams
was expected because they are at decreased pressure and increased tem-
perature.
The choice of the operating pressure for the flash tank is based
on several factors. The Rectisol process contains a series of flash-
ing operations designed to remove the acid gases from the solvent and
allow for some separation of the sulfur compounds from C02. In our
system, operation at moderate pressures (4.4-11.2 atm. abs.) provides
some insight into how these gases distribute. Also, flash tank opera-
tion indicates how closely our vapor- liquid equilibrium model pred-
icts actual system performance. Moderately high pressures are a
better test as to how well the VLE model handles departures from ideal
behavior. Finally, trial and error has shown that this range of oper-
ating pressures is more compatible with overall system performance;
the effect of process controller oscillation on sampling and steady
state operation is reduced.
Stripper operating pressure was 1.7 atmospheres absolute for
AM-30 and for the three other runs. In practice, stripper operating
conditions are the result of a balance between temperature and pres-
sure to minimize solvent losses and yet regenerate the solvent. The
pressure used here represents the lowest that the stripper pressure
controller could maintain and still avoid the adverse influence of
process controller oscillation. Inlet temperature to the stripper was
not controlled but will be used later to facilitate stripper simula-
tion efforts.
Since the focus of these runs in on absorber performance, column
pressure was varied along with liquid flow rate and inlet liquid tem-
567
-------
perature. Variation in Sour Gas C02 concentration introduced addi-
tional variation demonstrating the necessity for a mathematical model
in process analysis. The model is described further in the next sec-
tion.,
The temperatures measured throughout the acid gas removal system
are very important in terms of understanding the process. Since the
sampling of liquid and gas from the column packing proved to be unsuc-
cessful, column temperature profiles take on added significance in
determining mass transfer rates. Current modeling efforts rely on
comparisons of measured and predicted column temperature profiles.
This profile is indicative of the rate of C02 transfer because of the
large heat effects associated with C02 absorption in methanol.
The absorber temperature profiles are reported in the Appendix
for all four runs and were measured with sensors located on the out-
side of the absorption column wall. For all runs, temperature sensor
TT350, located at 4.8 feet above both the gas inlet and the bottom of
the packing, did not stay fastened to the column wall and is probably
inaccurate. In addition, the lowest temperature measured, TT353, at
0.3 feet, is probably located too close to the packing end and there-
fore not useful. These will be moved for future runs.
Both height of packing and height above the gas inlet are report-
ed to point out that end effects have been minimized. In earlier
runs, the gas inlet was located 7 inches below the bottom of the pack-
ing and significant end effects were observed in those runs. Because
it is important in the modeling efforts to eliminate end effects, the
bottom of the absorber was reconstructed to ensure that the mass
transfer takes place in the column packing and not above or below it.
An interesting observation can be made concerning the temperature
profile of the stripper. At the top of the column, the acid gases
flash due to the pressure reduction of the solvent entering from the
flash tank. This can be noted from the decreasing temperatures meas-
ured in the top part of the column. Further down the column, the tem-
perature begins to increase as the influence of the warm stripping ni-
trogen is felt. A lower flash tank pressure would reduce this flash-
ing effect as the pressure drop between the flash tank and the
stripper would be less.
2. Acid Gas Distribution in the AGRS
Table 4 shows the ratios of acid gas concentrations for the vari-
ous gas streams in the AGRS. The ratios of the acid gases exiting the
stripper in the concentrated Acid Gas stream are the same as those in
the entering Sour Gas stream. This is the expected result for
non-selective physical solvent systems.
Because of problems with the analysis of low levels of C02 in the
Sweet Gas stream, not much can be said of the ratios involving C02.
However, it appears that H2S is removed at a slightly higher efficien-
cy than COS when the ratios in the Sour Gas stream are compared to the
568
-------
Sweet Gas stream. This is expected because H2S has a slightly higher
solubility than COS over the temperature range used.
The Flash Gas ratios reflect the amount of C02 initially fed to
the system. Here, the ratios of C02 to H2S and COS are about twice
those found in the entering Sour Gas stream. Changing the flash tank
operating presures would improve this selectivity. This indicates
that there is the potential to concentrate the C02 fed to the system
through a flashing process. The ratio of the sulfur compounds
(H2S:COS) is again less than that found in the Sour Gas. The fact
that H2S is more soluble than COS means that proportionately less H2S
will flash upon pressure reduction.
3 . Absorber Column Performance
Table 5 contains the results associated with absorber column per-
formance for four integrated runs treating a gas produced in the ga-
sifier. An attempt was made to vary system conditions to show the ef-
fect on acid gas removal efficiencies. A comparison of the results
from these runs underline the importance of mathematical modeling to
analyze system performance.
All runs show an acid gas removal efficiency of at least 94.0%
for the range of operating conditions used. Also, only small differ-
ences in component removal efficiencies can be seen despite the
changes in packed height, liquid flow rate, and operating pressure.
The reason for this can be explained by examining the inlet gas compo-
sitions for each run and by considering mass transfer limitations.
Gasifier operation will dictate both the composition and flow
rate of the gas stream fed to the AGRS. For the four runs shown here,
the inlet gas flow rate to the absorber varied only slightly but the
C02 content of the stream varied significantly. This affects the ab-
sorber column temperature profile as the magnitude of the absorption
heat effect depends on the amount of C02 absorbed. As the temperature
increases, the amount of acid gases removed decreases„
This effect can be seen by comparing the results of runs 35 and
36 in Table 5. Although 35 was made at a higher absorber pressure and
lower inlet liquid temperature, the acid gas removal efficiencies are
approximately the same. A closer look shows that there is 7% more C02
in the entering gas stream for run 35. The increased thermal effect
tends to offset the expected increase in column removal efficiency.
Run 37, made with three times the packed height used in the other
runs, resulted in only small improvements in acid gas removal effici-
ency. This indicates that for the range of operating conditions used,
acid gas removal efficiency has reached an upper limit. improvements
could be obtained with lower inlet temperatures, higher operating
pressures and larger liquid flow rates.
The effect of changing liquid flow rates can be seen by comparing
runs 30 and 35, The increase in the liquid flow rate from 60.7
569
-------
Ib-moles/hr/sq.ft. to 72.1 Ib-moles/hr/sq.ft. improved C02 removal
efficiency by 108%. H2S and COS removal remained about the same prob-
ably because of mass transfer limitations. Future runs will be made
at higher L/G ratios to examine more completely the effect of this
variable on removal efficiency.
The results from these four runs clearly point to the need to de-
velop a mathematical model to assist with the analysis of experimental
results and provide a basis for analyzing more complicated process
configurations. Although there exists the possibility of feeding syn-
thetic gas streams to the AGRS, the most useful information comes from
runs where gasifier product gas is used. Because of the variability
associated with gasifier operation, a carefully structured experimen-
tal plan would be difficult to complete. The strategy used thus far
has been to cover a wide range of operating conditions. Then, a ma-
thematical model will be used to extend these results to process situ-
ations that cannot be studied with the pilot plant.
PROCESS MODELING
At present, mathematical modeling efforts have mainly dealt with
describing the operation of the packed absorption column for the adia-
batic case. A calculational technique first described by Feintuch and
Treybal (7,8) for packed column design has been implemented on the
computer and is currently used for analyzing runs where synthetic gas
mixtures of carbon dioxide and nitrogen are fed to the absorption co-
Iumn0 Thus far, only cases for the absorption of a single component
have been modeled but a multicomponent case is currently being devel-
oped to describe the transfer of H2S, COS, CS2, C02, H2, N2, CO, and
CH4. Additional hydrocarbons will be added to this list as the exper-
imental program moves into the gasification of coal-char mixtures„
The calculational technique described accounts for the mass and
heat transfer resistances in both the liquid and gas phases. Solvent
evaporation is also incorporated into the calculation. It is an es-
sentially rigorous solution to a highly non-linear set of partial dif-
ferential equations which treats a packed column as a true differen-
tial device without resorting to a stage -wise, tray tower analogy
(8). The method involves dividing the tower height into differential
sections and satisfying heat transfer, mass transfer, and equilibrium
relationships for each section. Experimental verification of this
technique for air-water-ammonia systems at ambient pressure and tem-
perature has been shown by Raal and Khurana (9). Feintuch (8) sug-
gests an extension of this technique to complex multicomponent systems
but no literature data are available with which to compare the re-
sults. Initial indications from our work indicate that this calcula-
tional method applies to the multicomponent system studied here.
As a first step in model development, computer simulation for the
adiabatic absorption of C02 in methanol was tried. Results for a re-
570
-------
cent synthetic gas run (AM-32) are presented in Figure 4. Here, the
liquid temperature profile in the absorber is compared to the model
prediction. Process conditions for AM-32 are shown in Table 60 Thus
far, excellent agreement between model prediction and experimental
data has been seen for column temperature profiles and removal effici-
encies. The model also predicts both liquid and gas flow rate and
composition profiles for both design and analysis approaches to packed
column performance. The model has been used for simulation of systems
containing H2S-N2- CH30H and COS-N2-CH30H. A multicomponent case is
presently being developed for the components mentioned above. An up-
coming EPA technical report will provide a more detailed description
of mathematical modelng efforts.
FUTURE EXPERIMENTAL WORK
Figure 5 and Table 7 illustrate the present scope of our research
program and plans for future work. Currently, we anticipate using a
chemical solvent following the evaluation of refrigerated methanol and
should begin this work sometime during 1981. A full evaluation of
each solvent used includes experimental runs with both crude coal gas
and synthetic gas mixtures. A computer simulation package for each
system is planned. Also, vapor-liquid equilibrium model development
will parallel all anticipated pilot plant studies. Capability to
measure both binary and multicomponent VLE information exists and has
already been utilized. This collection of information, along with an
assessment of the fate of certain trace compounds, should provide the
basis for evaluating the relative merits of the solvents proposed for
acid gas removal in coal gasification processes.
571
-------
FIGURE 4
PACKED ABSORPTION COLUMN
LIQUID TEMPERATURE PROFILE FOR
SYNGAS RUN AM-32
7 -
Computer Model Prediction
0 Experimental Data
572
-------
TABLE 6
PROCESS CONDITIONS FOR SYNTHETIC GAS RUN AM-32
Liquid Flow Rate
TL in
Gas Flow Rate
TG in
Pressure
Inlet Gas Composition
Outlet Gas Composition
t
C02 Removal Efficiency
61.05 Ib moles/hr/fr
-36.1°F
17.31 Ib moles/hr/ft2
57.4°F
28.0 Atmospheres absolute
33.73 mole percent C02
66.27 mole percent N«
0.92 mole percent COo
99.08 mole percent N2
98.10%
573
-------
FIGURE 5
AGRS RESEARCH PROGRAM
Solubilities in
Methanol
New Solvent Selection
Refrigerated Methanol Evaluation
Methanol System Performance
Packed Absorber/Stripper Modeling
(I and II)
III
Physical Properties
System Simulation}
Staged Absorber/Stripper
Model
Adiabatic Flash Calculations
-------
TABLE 7
A. Methanol System Performance
1. C02, H2S, COS and other sulfur gas removal
2. Hydrocarbons, particularly aromatics, removal and accumula-
tion in solvent
3. Thermal behavior
4. Relationship of gasifier operation to AGRS performance
5. Comparison of SYNGAS and crude coal gas operations
6. Methanol losses from absorber, flash tank and stripper
7. Solvent stability
B. Solubilities in Methanol
1. Use current VLE model (Ferrell, Rousseau and Matange, 1980)
in absorber/stripper/flash tank calculations .
2. Use current VLE model to develop methods for calculating
heats of solution
3. Obtain VLE data on COS, CS2, and other important gases, and
incorporate into VLE model
4. Modify current model to use Wilson and/or UNIQUAC equations
Co Packed Absorber/Stripper Models I, II, and III
Model I (SIMPAK): considers a three-component system in which the
carrier gas is insoluble
Model II (MCOMP): places no restrictions on number of components
or solubility of carrier gas
Model III (von Stockar method): relies on an unsteady state des-
cription of the packed column, and is believed to have better conver-
gence properties than approach of Model I and II
1. Model development for packed columns
2. Use of model in simulation of SYNGAS operation
575
-------
3. Use of model in evaluation of crude coal gas operation
4. Use of model to guide selection of AGRS operating variables
(e.g. N2 flow rate to stripper to maximize sulfur concentra-
tion of feed stream to sulfur recovery unit.)
D. Adiabatic Flash Calculation
1. Model flash tank in AGRS
2. Describe flashing process as liquid enters stripper
E. Physical Properties and Equipment Parameters
1. Document, catalog and make available all physical properties,
diffusivities and packing characteristics used in system
F. System Simulation
1. Bring all system elements together in a program to examine
unit interactions and optimize operating conditions
G. Staged Absorber/Stripper Model
1. Extension of Packed column models to staged columns to pro-
vide necessary tools for system simulation
H. New Solvent Selection
1. Begin to consider next solvent system to study (e.g. hot po-
tassium carbonate) and determine needed information to begin
evaluation
2. Determine advantages/disadvantages of potential solvents
3. Provide basis for choosing desirable features of acid gas re-
moval solvents from environmental, process, and energy consi-
derations
576
-------
REFERENCES
(1) Ferrell, J. K., R. M. Felder, R. W. Rousseau, J. C. McCue,
and R. M. Kelly, "Coal Gasification/Gas Cleanup Test Facility:
Volume I. Description and Operation," EPA-600/7-80-046a, March 1980).
(2) Beychok, Milton R.,"Sulfur Emission Controls for a Coal Gasifica-
tion Plant," Symposium Proceedings: Environmental Aspects of Fuel
Conversion Technology, II, (Hollywood, Florida - 1975),
EPA-600/2-76-149, June 1976.
(3) Ranke, Gerhard, "The Rectisol Process- for the Selective Removal
of C02 and Sulfur Compounds from Industrial Gases," Chemical Economy
and Engineering Review, Vol. 4 (1972), pp. 25-31.
(4) Ferrell, J. K., R. W. Rousseau, and D. G. Bass, "The Solubil-
ity of Acid Gases in Methanol," EPA-600/7-79-097, April 1979.
(5) Ferrell, J. K., R. W. Rousseau , and J. N. Matange, "The So-
lubility of Acid Gases and Nitrogen in Refrigerated Methanol,"
EPA-600/ 7-80-116 May 1980.
(6) Private communication with Dr. Santosh Gangwal, Process Engineer-
ing Department, Research Triangle Institute, Durham, North Carolina,
(July 1980).
(7) Treybal, R. E., "Adiabatic Gas Absorption and Stripping in Packed
Towers," Industrial Engineering Chemistry, Vol. 61 (1969), p. 36.
(8) Feintuch, H. M., and R. E. Treybal, "The Design of Adiabatic
Packed Towers for Gas Absorption and Stripping," Industrial Engineer-
ing Chemistry Process Design and Development, Vol. 17 (1978), pp.
505-514.
(9) Raal, J. D., and M. K. Khurana, "Gas Absorption with Large Heat
Effects in Packed Columns," The Canadian Journal of Chemical Engineer-
ing, Vol.51 (1973), pp. 162-167.
577
-------
APPENDIX
578
-------
AM-30
579
-------
tmmmmmmmttmmtmmmm
t
NCSU DEPARTMENT OF CHEMICAL ENGINEERING t
t
ACID GAS REMOVAL SYSTEM t
t
mmmmmmmmtumttmtmwt
RUN NUMBER A-M-30
INTEGRATED
DATE 5/28/1980
STREAM COMPOSITION (HOL X)
C02
H2S
COS
MEOH
H2
CO
N2
CH4
SOUR GAS SUEETGAS
24,600
0,910
0.042
0,000
33.170
21,060
18,500
1,640
1,460
0,048
0,003
0.000
43,190
28.480
24,890
1.950
FLASHGAS
43,200
0.634
0.043
0.000
15.240
22,720
14.750
3.400
STRIPN2
0.000
0.000
0.000
0.000
0,000
0.000
100.000
0.000
ACID GAS
71.500
2.539
0.117
0.000
0.000
1.020
24.560
0,420
t
ABSORBOT
5,918
0,220
0,010
92,764
0,619
0,204
0.202
0.064
t
FLASHBOT
5,545
0.216
0.010
93,669
0,473
0.000
I'M
t
STRIPBOT
0.000
0.000
99)498
0.502
0.000
0.000
0.000
CALCULATED
MASS BALANCE (LB-MOLES/HR)
IN
OUT
SOUR GAS STRIP N2
SUEETGAS FLASHGAS ACID GAS
TOTAL IN TOTAL OUT Z RECOVERY
C02
H2S
COS
MEOH
H2
CO
N2
CH4
0.554
0.020
0.001
0.000
0.747
0,474
0.417
0.037
TOTAL 2.253
(LB-MOLES/HR)
0.000
0.000
0.000
0.000
0.000
0.000
0.182
0.000
0.182
0.023
0.001
0,000
0.000
0.692
0.456
0.399
0.031
0,038
0,001
0,000
0,000
0,013
0,020
0,013
0,003
0,550
0.020
0.001
0.000
0.000
0.008
0,189
0.003
1.602
0.089
0,769
0,554
0.020
0.001
0,000
0,747
0,474
0.599
0.037
2.433
0.611
0.021
0.001
0.000
0.705
0.484
0,601
0.037
2,461
110.3
101.7
104.0
0.0
94.4
102,0
100,3
101,4
101.130
METHANOL-FREE BASIS
TOTAL HETHANOL LQSS= 0,000 LB-5WLES/HR - 0,000 GALLONS/HR
580
-------
RUN NUMBER A-M-30
INTEGRATED
DATE 5/28/1980
COLUMN TEMPERATURE PROFILES I MASS BALANCES
ABSORBER
P=397.19 PSIB
FLASH TANK
P= 96.75 PSI6
STRIPPER
P= 9.85 PSI6
-> SHEET
GAS
,-->
!
FLASH
GAS
.—> ACID
: GAS
NEOHROU
— 0.66 GPH ->!
(-34.13 F)
DP= 2.50 IN H20
SOUR GAS
- 13.48 SCFM ->
( 54.04 F)
.
t
•
I
1
mm
mm
xxxxxx
mm
-29.63 F
-31.77 F
4.83 F
\
:
— --— _«-N
""""""/
•tttttttt'
12.94 F
i
'
- 0.66 GPH -
(35.96 F)
•
jyyyyyiyil
STRIPPING N2~ 1.09 SCFN ->
(75.00 F)
I-
wijTB
I
itttttttti
i
i
:
i
i
i
16.05 Ft
1
14.86 F)
t
1
1
13.89 Ft
18.63 F1
1
1
t
19.21 Ft
t
19.26 Ft
t
t
t
t
t
38.06 Ft
t
t
t
t
DP= 0.46 IN H20
tttttttttt
I--TO ABSORB€R->
t
t
t
ttttttmt
581
-------
RUN NUMBER A-M-30
INTEGRATED
DATE 5/28/1980
COLUMN TEMPERATURE PROFILE
ABSORBER COLUMN PRESSURE =397.2 PSI6
TOTAL PACKING HEIGHT 7,10 FEET
PACKING USED - 1/4* CERAMIC INTALOX SADDLES
-> SHEET GAS
9,58 SCFH
tmmmtmmmt
MEOHFLOU
0.660 6PM
-34,13 F
,--.•- •••••••••\
SOUR GAS INLET
13.48 SCFM
54.04 F
-29,63 F
-31,77 F
-28,27 F
-21,50 F
-14.11 F
7,10 FT
4.79 FT
2,46 FT
1.21 FT
0,79 FT
0,31 FT
A. Aft CT
TRANSMITTER
TT350
TT351
TT352
TT353
TT354
HEIGHT ABOVE
GAS INLET
4.79
2,46
1,21
0.31
0,79
HEIGHT OF
PACKING
4,79
2.46
1,21
0.31
0.79
TEHPERATURE(F)
-29.63
-31.77
-28,27
-14,11
-21.50
582
-------
AM-35
583
-------
mtmtmmtmmmtmmmmmm
* *
* NCSU DEPARTMENT OF CHEKICAL ENGINEERING t
* *
» ACID GAS REHOVAL SYSTEM *
* I
mmmmmmmmmmmmmtm
RUN NUMBER A-H-35
INTEGRATED RUN
DATE 6/26/1980
STREAH COMPOSITION (HOL 2)
C02
H2S
COS
KEOH
H2
CO
N2
CH4
SOHR GAS SKEETGAS
28,010
0.807
0,045
0,000
33,190
20,200
15.700
2,010
0,950
0,037
0.003
0.000
45,500
27,850
23.230
2,440
FLASHGAS
42,450
0,526
0.038
1.310
4.210
23.830
13,490
4,110
STRIPN2
0.000
0,000
0,000
0.000
0,000
0,000
100,000
0,000
ACID GA!
71,900
1.970
0.127
2,910
0,000
1,630
20,750
0,690
ABSORBOT FLASHBOT STRIPBOT
5,674
0,162
0,009
93.934
0.118
0,048
0,000
0,054
5.361
0.159
0.009
94,365
0.085
0.000
0.000
0.021
0,000
0,008
0,000
99,901
0,090
0.000
0,000
0,000
CALCULATED
MASS BALANCE (LB-MOLES/HR)
IN
OUT
SOUR GAS STRIP H2
SHFETGAS FLASHGAS ACID GAS
t t
TOTAL IN TOTAL OUT Z RECOVERY
C02
H2S
COR
MEOH
R
N2
CH4
TOTAL
0,618
0,018
0,001
0,000
1:1
0.346
0,044
2,205
0,000
0,000
0.000
0,000
o'.ooo
O.'l82
0,000
0,182
0,015
0,001
0,000
0,000
0.719
0.440
0,367
0,039
1,581
0.036
0.000
0.000
0.001
0.004
olon
0,003
0,085
0,582
0,016
0.001
0,024
ktt
0.168
0.006
0.809
0.618
0.018
0,001
0,000
8:551
0,528
0.044
2.386
0.633
0.017
0,001
0,000
m
0.547
0.048
2,441
102.4
95,3
111.9
0.0
98.8
106,3
103,4
107,5
102.301
(LB-MOLES/HR)
METHANOL-FREE BASIS
TOTAL METHANOL LOSS= 0.025 LB-HOLES/HR = 0.117 GALLONS/HR
584
-------
RUM NUMBER A-H-35
INTEGRATED RON
DATE 6/26/1980
COLUMN TEMPERATURE PROFILES i HASS BALANCES
ABSORBER
P=396,61 PSI6
FLASH TANK
P= 96,75 PSIG
STRIPPER
P= 9,87 PSIG
,—> SHEET
GAS
HEOHFLOU
™ 0,79 6PH
(-36,31 F)
>
DP= 2.50 IN H20
SOUR GAS
13,19 SCFH ->
( 53,86 F)
mmm
xxxxxx
xxxxxx
xxxxxx
xxxxxx
-34.05 F
-34,65 F
0.72 F
«
tmmm
-> FLASH
GAS
ffl
mtmm
->
9.51 F
.--> ACID
! GAS
tmmm
.- 0,79 GPH -
(33.06 F)
mtmm
STRIPPING N2- 1,09 SCFH
(75,00 F)
~>
13,65 F
1?,31 F
11.05 F
16.13 F
16.08 F
16.12 F
33.15 F
DP= 0,48 IN H20
-TO ABSORBER~>
tmmm
585
-------
RUM NUMBER A-H-35
INTEGRATED RUN
DATE 6/26/1980
COLUMN TEMPERATURE PROFILE
ABSORBER COLUMN PRESSURE =.396,6 PSIG
TOTAL PACKING HEIGHT* 7,10 FEET
PACKING USED = 1/4' CERAMIC INTALOX SADDLES
-> SHEET GAS
9,46 SCFM
mmmmmmm
wm i i
-36.31 F 1
_________'•>
j
]
i
i
i
i
SOUR GAS INLET
13.19 SCFH
53,86 F
It
It
t
b __.. _«. »„„«
It
It
t
H
t
t -34,05 F
-34,65 F
-30,99 F
-25,30 F
-17,77 F
7. IA CT
4,79 FT
2,46 FT
1,21 FT
0,79 FT
0,31 FT
A.ftft CT
mwmmstmmt
TRANSMITTER
TT350
TT351
TT352
TT353
TT354
HEIGHT ABOVE
GAS INLET
4,79
2.46
1,21
0.31
0,79
HEIGHT OF
PACKING
4,79
2,46
1.21
0.31
0,79
TEMPERATURE(F)
-34.05
-34,65
-30,99
-17,77
-25.30
586
-------
AM-36
587
-------
mummmmmttmtmttmmmm
NCSU DEPARTMENT OF CHEMICAL ENGINEERING t
t
ACID GAS REMOVAL SYSTEH *
mtmtmtmmmmttmwmtmmt
RUN NUMBER A-N-36
INTEGRATED RUN
DATE 7/18/1980
STREW COHPOSITION (HOL Z)
SOUR GAS SHEET6AS FLASHGAS STRIPN2 ACID GAS ABSORBOT FLASHBOT STRIPBOT
C02
H2S
COS
KEOH
H2
CO
N2
CH4
20,900
0,892
0,048
0,000
33.440
17,030
26,190
1,270
0.420
0,068
0,004
0,000
44,310
20,580
33.040
1.680
34.170
0,549
0,038
0,000
13,870
22,020
26,490
2.830
0,000
0,000
0.000
0.000
0,000
0,000
100.000
0,000
69,770
2,225
0,133
0.000
0.000
0,970
26.500
0,360
4,465
0.182
0.010
95.037
0,000
0,214
0,093
0,000
4.302
0.180
0.010
95.411
0,000
0,098
0,000
0.000
0.000
0.023
0.000
99,947
0,000
0,030
0,000
0.000
CALCULATED
MASS BALANCE (LB-NOLES/HR)
IN
OUT
SOUR GAS STRIP N2
SUEET6AS FLASHGAS ACID GAS
t *
TOTAL IN TOTAL OUT
C02
H2S
COS
NEOH
H2
CO
N2
CH4
0,476
0,020
0.001
0.000
0.762
0.388
0,597
0.029
TOTAL 2,278
(LB-HOLES/HR)
0.000
0,000
0,000
0.000
0,000
0.000
0,182
0,000
0.182
0.007
0.001
0,000
0.000
0.787
0.366
0.587
0.030
1.776
0,019
0,000
0.000
0,000
0,008
0.012
0.015
0.002
0.055
0.518
0.017
0.001
0.000
0.000
0.007
0.197
0.003
0.743
0,476
0,020
0,001
0,000
0,762
0.388
0.779
0.029
2,455
0,545
0,018
0,001
0,000
0.795
0*.034
2.576
Z RECOVERY
114,4
88.9
99.3
0,0
104,3
99.2
102,5
117,8
104,918
METHANOL-FREE BASIS
TOTAL NETHANOL LOSS* 0,000 LB-NOLES/HR = 0.000 GALLONS/HR
588
-------
RUN NUMBER A-M-36
INTEGRATED RUN
DATE 7/18/1980
COLUMN TEMPERATURE PROFILES t MASS BALANCES
ABSORBER
P=296.85 PSIG
FLASH TANK
P- 96,85 PSIG
STRIPPER
P= 9.81 PSI6
,--> SHEET
! GAS
J
J
':
.
.
tttttttttt
HEOHFLOV
- 0,79 6PM ->
(-32.38 F)
DP= 2,50 IN H20
SOUR GAS
13.63 SCFM ->
( 57.50 F)
mm
mm
mm
xxmx
-30.20 F
-30.20 F
-1.72 F
—> FLASH
GAS
, OF)
t
->
8.99 F
t
,--> ACID
; GAS
4.44 SCFM
(77.90F)
!
mmtm
0.796PM-
(35.03 F)
STRIPPING N2- 1.09 SCFM ->
(75.00 F)
t
17.94 Ft
t
16.83 Ft
15.66 F
20.72 F
20.64 F
20.70 F
35.38 F
DP* 0.45 IN H20
-TO ABSORBER-)
mttwtt
589
-------
RUN NUMBER A-M-36
INTEGRATED RUN
DATE 7/18/1980
COLUHN TEMPERATURE PROFILE
ABSORBER COLUMN PRESSURE =296.9 PSIG
TOTAL PACKING HEIGHT" 7.10 FEET
PACKING USED = 1/4* CERAMIC INTALOX SADDLES
10.63 SCFN
tmmmmmtmt
I I
,,_^,,— __^^_,,_,,__.,_^,,^\
SOUR GAS INLET
13.63 SCFM
57.50 F
TRANSMITTER HEIGH1
GAS 1
,___._
-30.20 F
-30.20 F
-25.22 F
-18.92 F
-14.41 F
—
ttttUtttttttttttttil
ABOVE HEK
NLET PA(
7.10 FT
4.79 FT
2.46 FT
1.21 FT
0.79 FT
0.31 FT
0.00 FT
i
JHT OF TEMPERATURE(F)
>KINo
TT350 4.79 4.79 -30.20
TT351 2.46 2.46 -30.20
TT352 1,21 1.21 -25.22
TT353 0.31 0.31 -14.41
TT354 0.79 0.79 -18.92
590
-------
AM-37
591
-------
mmmmmmmmmmmmmtm
t »
t NCSU DEPARTMENT OF CHEMICAL ENGINEERING t
t t
* ACID 6AS REMOVAL SYSTEM t
mttmmmtmmtmmmmmwm
RUN NUMBER A-M-37
INTEGRATED RUN
DATE 7/25/1980
STREAM COMPOSITION (MOL Z)
SOUR GAS SHEETGAS FLASHGAS STRIPN2 ACID GAS ADSORBOT FLASHBOT STRIPBOT
C02
H2S
COS
HEOH
H2
CO
N2
CH4
25.050
0.863
0.047
0.000
38.930
18.810
14.820
1.150
0.640
0.041
0.003
0.000
54.960
23.920
19.300
1.160
47,820
0.628
0.043
0,870
11.880
22.340
12.940
3.140
0,000
0,000
0,000
0.000
0.000
0.000
100.000
0,000
71.050
2.260
0.130
4.030
0.000
0,940
21,150
0.220
5.406
0.183
0.010
93.853
0,000
0.310
0,171
0,067
4.975
0.179
0.010
94,667
0,000
0,089
0,044
0.037
0.000
0.000
0.000
99.850
0.000
0.015
0,114
0,020
CALCULATED
MASS BALANCE (LB-MOLES/HR)
IN
OUT
SOUR GAS STRIP N2
SUEETGAS FLASHGAS ACID GAS
TOTAL IN TOTAL OUT 1 RECOVERY
C02
H2S
COS
NEDH
H2
CO
N2
CH4
TOTAL
0.579
0,020
0,001
0,000
0.900
0.435
0.343
0.027
2.313
0.000
0,000
0.000
0.000
0,000
0,000
0.182
0.000
0.182
(LB-MOLES/HR)
0.011
0.001
0.000
0,000
0,925
0,402
0,325
0,020
0,050
0.001
0.000
0.001
0.012
0.023
0.013
0,003
0.589
0.019
0.001
0,033
0,000
0,008
0,175
0,002
1,683
0,104
0,830
0,579
0,020
0.001
0.000
0,900
0.435
0.525
0.027
2,487
0,650
0.020
0.001
0,000
0.937
0.434
0.514
0.025
2.580
112.2
100.6
106.6
0.0
104.1
99.7
97,9
92,5
103,732
HETHANOL-FREE BASIS
TOTAL KETHANOL LOSS= 0.034 LB-MOLES/HR = 0,164 GALLONS/HR
592
-------
RUN NUMBER A-K-37
INTEGRATED RUN
DATE 7/25/1980
COLUMN TEMPERATURE PROFILES t MASS BALANCES
ADSORBER FLASH TANK STRIPPER
P=446.65 PSIG P= 97.28 PSIG P= 9.85 PSI6
NEOHFLOU
— 0.78 CPU ->
(-36.33 F)
DP* 2.50 IN H20
SOUR GAS
- 13.84 SCFM ~>
( 59.92 F)
»---
.
.
|
m
tmttmi
-32.12 F
-31.56 F
-25.86 F
-28.45 F
-26.44 F
-27.54 F
7.19 F
> SHEET .--> FLASH
GAS ! GAS
»
*
*
*
! f^jff?
t i
*
4
*
t
•
mttmn
16.27 F
, ..»., .._\
. --•" ~-f
I
tmmro
1
- 0,78 6PM -
(39.25 F)
t
'
\
STRIPPING N2- 1.09 SCFM ~>
(75.00 F)
'
1
f
»
:
|
Mtjr?
ttmtttu
19.91 F
18.76 F
17.57 F
22.91 F
22.86 F
22.86 F
39.84 F
I
> ACID
GAS
1
\
W= 0.43 IN H20
-TO ABSORBER->
ttstttmt
tmmm
593
-------
RUN NUMBER A-M-37
INTEGRATED RUN
DATE 7/25/1980
COLUMN TEMPERATURE PROFILE
ABSORBER COLUMN PRESSURE =446,6 PSI6
TOTAL PACKING HEIGHT* 7.10 FEET
PACKING USED = 1/4' CERAMIC INTALOX SADDLES
> SHEET GAS
10.07 SCFM
1
MEOHFLOU
0.784 6PM
-36.33 F
\
________ __ \
SOUR GAS INLET
13.84 SCFH
59.92 F
kttttttttttitttttttti
-26.64 F
-27.54 F
-22.20 F
-16.03 F
-10.07 F
!
— 7.10 FT
/ t AV * I
4.7? FT
2,46 FT
1.21 FT
0.79 FT
0.31 FT
0.00 FT
twtmmmmmt
TRANSMITTER
TT350
TT351
TT352
TT353
TT354
HEIGHT ABOVE
GAS INLET
4.79
2.46
1.21
0.31
0.79
HEIGHT OF
PACKING
4.79
2.46
1.21
0.31
0.79
TEMPERATURE(F)
-26.64
-27.54
-22.20
-10.07
-16.03
594
-------
POLLUTION CONTROL GUIDANCE DOCUMENT
FOR
LOW-BTU GASIFICATION TECHNOLOGY:
BACKGROUND STUDIES
W. C. Thomas, G. C. Page and D. A. Dalrymple
Radian Corporation
8500 Shoal Creek Boulevard
Austin, Texas 78758
ABSTRACT
The Environmental Protection Agency is currently preparing a Pollu-
tion Control Guidance Document (PCGD) for low-Btu gasification (LEG) facili-
ties which use atmospheric pressure, fixed-bed gasifiers. The PCGD is intend-
ed to aid industry and government in their efforts to commercialize LEG tech-
nology in an environmentally acceptable manner. This paper presents some of
the preliminary results of background studies performed to support the devel-
opment of the LEG PCGD.
A model plant approach was used to assess the environmental control
needs for LEG facilities. The plant configuration and coal feed combinations
for which pollution controls were identified and evaluated were selected based
on existing and proposed plants in the U.S. The major variables examined were
coal feed type (anthracite, lignite, and high- and low-sulfur bituminous coals)
and degree of product gas purification (production of hot, cooled, and desul-
furized low-Btu gas). In all, eleven combinations of these variables, i.e.,
model plants, were selected for study. Each model plant had a nominal capacity
of 45 MJ/s (150 x 106 Btu/hr) of low-Btu gas.
Multimedia pollutant sources and pollutants of potential concern were
identified and quantified for each model plant. The bases for these determin-
ations were field test data and calculated emissions projections. The EPA1s
low-Btu gasification environmental assessment program was the major source of
the field test data, but results from other government and industry test pro-
grams were also used.
Control/disposal options were identified and evaluated for each
discharge stream. Factors that were considered included the need for control,
current industry practices, control equipment performance, capital investment
requirements, annual operating costs, energy impacts, and secondary environ-
mental discharges.
595
-------
POLLUTION CONTROL GUIDANCE DOCUMENT
for
LOW-BTU GASIFICATION TECHNOLOGY:
BACKGROUND STUDIES
INTRODUCTION
Over the past several years the United States has moved from a posi-
tion of energy independence to one of energy dependence. A decade ago this
country imported only about ten percent of its crude oil needs and now the
figure is around fifty percent. The amount of oil and gas produced in the U.S.
has declined slightly over this period despite a doubling of drilling activity.
The country's vast coal reserves, however, have not been developed with the
same intensity. With the changing energy picture there has been a growing
interest on the part of government and industry in the technologies that
produce clean fuels and chemical feedstocks from coal. One such technology is
low-Btu coal gasification (LEG).
The Environmental Protection Agency is responsible for ensuring that
LBG technology and other alternate energy technologies are developed in a man-
ner which protects public health and the environment. As part of that effort,
the EPA has initiated programs to assess the environmental impacts of LBG.
The EPA has developed the Pollution Control Guidance Document (PCGD)
concept to aid industry and government in their efforts to commercialize low-
Btu gasification technology in a manner that will be environmentally accept-
able. The primary purposes of a PCGD are to:
• Provide guidance to permit writers on the best control approaches
presently available at a reasonable cost for the processes under
consideration.
• Provide system developers with an early indication of EPA's as-
sessment of the appropriate multimedia environmental protection
needs for each of these processes, considering costs, so that de-
velopers can design their facilities to achieve this level of
protection (rather than add potentially more costly retrofit
controls later).
• Describe to public interest groups EPA's judgment of the best
available controls for these processes.
• Provide the regulatory offices in EPA with information useful in
developing future regulations.
The low-Btu gasification PCGD will describe the performance capabil-
ities and costs of currently available controls for LBG facilities which use
596
-------
fixed-bed, atmospheric pressure gasifiers. (This type of gasifier is believed
to be the likely candidate for near-term commercial use). The PCGD will pro-
vide guidance both for currently regulated pollutants and for sources and/or
pollutants not covered by current standards. The guidance will be based on a
coordinated evaluation of available data by EPA's research and development,
regulatory, and permitting/enforcement offices. In the PCGD, suggested levels
of environmental protection considering costs, multimedia tradeoffs, and con-
trol system reliability will be specified for all air, water, solid waste, and
product/by-product streams. The PCGD will consist of three volumes whose
contents can be summarized as follows:
• Volume I will describe the technology, identify applicable
existing regulations, and present the control guidance;
• Volume II will summarize all of the data employed and present
the baseline engineering design, waste stream characterizations
and control option evaluations; and
• Volume III (Appendices) will contain detailed data listings and
calculations which support the guidance.
This paper presents some of the preliminary results of background
studies being conducted to support the development of the LEG PCGD. Included
in this paper are: 1) a description of the technology and an identification
and characterization of its multimedia discharges (including flow rates and
factors affecting discharge characteristics); 2) an identification and evalu-
ation of available control techniques; and 3) an estimation of the capital and
annualized cost impacts of available controls.
Technology Overview
Low-Btu coal gasification technology has been commercially available
for over 60 years. In the U.S., there are currently 20 known LEG plants either
in operation, under construction, or being planned for construction in the near
future. All of the commercially operating plants use fixed-bed, atmospheric
pressure gasifiers and are generally located in the industrialized Midwest and
Northeast regions of the Country. Feedstocks used at those plants include an-
thracite, lignite, and low-sulfur «1%) bituminous coal. No high-sulfur coals
are currently in use. The only gas purification process used at most of these
plants is a hot gas cyclone for particulate removal. Tar and oil removal using
gas quenching/scrubbing is practiced at one plant and is proposed for several
future plants. Sulfur compound removal is currently practiced only at one
plant. Current end-uses of low-Btu product gas include fuel for brick and lime
kilns, process heaters, and steam boilers.
LEG systems featuring fixed-bed, atmospheric pressure gasifiers are
most suitable for relatively small applications, with fuel demands ranging from
about 8.8 to 88 MW of thermal energy (30-300 million Btu/hr). This would re-
quire using from 1 to 10 gasifiers, depending on the coal feed. Energy demands
597
-------
greater than about 88 MW (300 million Btu/hr) may be better served by gasifica-
tion systems using gasifiers with larger capacities (for example, pressurized
gasifiers).
Applicable Existing Federal Regulations
New low-Btu gasification plants will have to comply with existing
Federal regulations for 1) sources within the plant that are already subject to
regulation (NSPS); 2) the disposal of solid wastes (RCRA); and 3) ambient-based
limitations, such as National Ambient Air Quality Standards (NAAQS), Prevention
of Significant Deterioration (PSD) requirements, Water Quality Criteria, and
Drinking Water Standards which may indirectly limit the quantities or concen-
trations of compounds in specific source discharges. However, at the current
time there are no Federal regulations which apply to specific air or water dis-
charge sources within an LEG facility. In addition, products and by-products
may be subject to restrictions if they contain toxic substances.
New plants will also be required to comply with state and local regu-
lations. The guidance in the PCGD is not intended to supersede the require-
ments of any of these existing or proposed regulations.
Approach Used For Background Studies
In conducting the background studies, an inventory of waste streams
and pollutants generated in model plant facilities was prepared and an assess-
ment of the performance and costs of various control alternatives for those
streams and pollutants was made. The approaches used to develop the pollutant
inventory and to select and evaluate applicable controls are briefly described
below.
Pollutants Considered. A listing of all the currently regulated pol-
lutants which have been found in the gaseous and aqueous wastes from LEG facil-
ities is provided in Table 1. The major pollutants not listed in this table,
but which are expected to be present in an LEG system's discharges are poly-
cyclic organic matter (POM), hydrogen cyanide and ammonia in the uncontrolled
gaseous emissions, and a number of specific organic compounds which are only
covered by gross parameters such as "organic carbon" in the aqueous effluents.
Model Plants. A model plant approach was used to characterize the
potential uncontrolled discharges from LEG systems and to evaluate pollution
control alternatives for those discharges. The model plants selected represent
processing configurations currently in use or proposed for use in the U.S.
Each has similar processes in the coal preparation and coal gasification oper-
ations. They differ in the areas of coal feedstock used and the degree to
which the low-Btu product gas is purified. For the background studies, recom-
mendations were not made as to which model plant should be used, but pollution
control information for the discharges from each model plant was developed.
598
-------
TABLE 1. CONSTITUENTS IN LOW-BTU GASIFICATION PLANT WASTE STREAMS COVERED BY EXISTING
AIR AND WATER STANDARDS
Standard
Subject Pollutants Found in Discharge Streams from Low-Btu
Gasification Facilities
en
vo
10
National Ambient Air Quality Standards
New Source Performance Standards
National Emission Standards for
Hazardous Air Pollutants
Prevention of Significant Deterioration
Standards
Increments
De Minimis Levels
Effluent Limitation Guidelines
Conventional and nonconventional
pollutants
Consent decree pollutants
(toxic pollutants)
CO, N02, S02, Pb, TSP, NMHC
CO, N02, S02, TSP, Total Reduced Sulfur, NMHC
Hg, Be, Inorganic As*, Benzene*, Radionuclides*
S02, TSP
CO, N02, TSP, S02, Pb, Hg, Se, H2S, CS2, COS
Al, Ammonia, B, Ca, Fluoride, Fe, Mn, Nitrate, Organic
Carbon, P, Sulfate, Sulfide, U, BOD5, COD, pH, Total
Nitrogen, Total Suspended Solids, Color, Oil and Grease,
Settleable Solids
Sb, As, Be, Cd, Cr, Cu, Cyanides, Pb, Hg, Ni, Phenol and
phenolic compounds, Polynculear aromatic hydrocarbons,
Se, Ag, Zn
*Listed as hazardous air pollutants; no regulations promulgated.
-------
The characteristics of the coal being gasified influence the presence,
composition and flow rates of the discharges from low-Btu gasification plants.
In order to evaluate the impact of coal properties on the discharge streams,
four different coals were examined: anthracite, lignite, low-sulfur bituminous
coal, and high-sulfur bituminous coal. These feedstocks span the range of
coals and coal properties which are or might be used in low-Btu gasification
plants.
Using the data sources described below, mass balances were calculated
for a basic plant capacity of 45 MW (approximately 150 x 106 Btu/hr) of ther-
mal energy in the product gas (based on the higher heating value of the gas).
This capacity is representative of the plant sizes expected to be constructed
in the near future. The mass balances provided a consistent basis for calcu-
lating "uncontrolled" mass discharge rates.
Based upon the expected characteristics of the waste streams, pollu-
tion control processes were identified and evaluated. "Secondary" waste
streams resulting from pollution control were also defined and controls for
these streams evaluated.
Data Sources. The major source of data used in the background
studies is an EPA-sponsored environmental assessment program for low-Btu gasi-
fication technology. As part of that program, a series of field test programs
are being conducted. To date, three data acquisition programs have been com-
pleted, another is on-going and a fifth is planned for the fall of 1980.1>2>3
All test sites are either commercially operating or commercial-size demonstra-
tion units located in the U.S. Additional data sources are other government
and industry sponsored test programs.
Information used to identify and evaluate pollution control alterna-
tives was mainly obtained by technology transfer, i.e., extrapolation from
other industries with identical or similar pollution control problems. Addi-
tional technical information was obtained from process vendors, process devel-
opers, and published literature. Only limited pollution control information
was obtained from the field test programs because of the essentially "uncon-
trolled" nature of the sites tested.
PROCESS DESCRIPTION AND POLLUTANT SOURCES
Low-Btu coal gasification systems can be considered to consist of
three basic operations: coal preparation, coal gasification, and gas purifi-
cation. Each of these operations in turn consists of process modules that are
employed to satisfy the functions of the operations.
As mentioned previously, a model plant approach was used to character-
ize the potential uncontrolled discharges from LBG systems and to evaluate pol-
lution control alternatives for those discharges. Block diagrams of the three
model plants examined are shown in Figure 1. These represent all the proces-
sing configurations of plants currently operating or proposed in the U.S.
600
-------
cn
o
oal Preparation
Operation
Coal Handling
and Storage
Coal
Gasification Gas Purification
Operation i Operation i
MODEL PLANT I
Gasification
1
Partlculate
Removal
1
Hot Low-Btu
*- Product Gas
MODEL PLANT II
Coal Handling
and Storage
Coal Handling
and Storage
Gasification
Gasification
Particulate
Removal
Quenching/ Cooled Low-Btu
MODEL PLANT III
Particulate
Removal
Quenching/ Residual Tar/ Sulfur Liesuirurizea
•^ Cooling •• Oil Removal ^ Removal ^
* Product Gas
FIGURE 1. LOW-BTU GASIFICATION MODEL PLANTS
-------
The first model plant produces a hot low-Btu product gas. The only
gas purification process used is a hot gas cyclone for partial removal of
entrained particulate matter. This process configuration is typical of most of
the plants currently in operation and several plants which are proposed or
under construction.
The second model plant produces a cooled low-Btu product gas. In this
plant, a series of wet scrubbers are used to quench and cool the hot gas. This
step also removes additional particulate matter and the majority of tars and
oils present. This configuration is similar to an existing LEG plant which
uses Chapman gasifiers.
The third model plant produces a desulfurized product gas and as a re-
sult has the most extensive gas purification scheme. In addition to a hot gas
cyclone and quenching/cooling, this model plant uses an electrostatic precipi-
tator for removal of residual tars/oils and a sulfur removal process. Avail-
able sulfur removal processes can be broadly classified as 1) those that remove
sulfur compounds and directly convert them into elemental sulfur, and 2) those
that remove sulfur compounds and produce an off-gas containing the removed
sulfur species. An evaluation of these processes, including discussions with
process licensors, indicated that the direct oxidation processes are the pre-
ferred sulfur removal technique for low-Btu gas derived from fixed-bed, atmos-
pheric pressure gasifiers. While some of the other types of processes (e.g.,
the monoethanolamine process) could be used, difficulties would be encountered
in treating the sulfur species laden off-gas due to its high C(>2 content.
This conclusion is supported by the fact that all existing and proposed designs
of LEG facilities which remove sulfur species use direct oxidation processes.
Thus, for the Model Plant III systems, only direct oxidation processes are
examined for sulfur removal. For study purposes, the Stretford process was
selected as being representative of commercially available direct oxidation
processes.
Descriptions of the three basic operations, the process modules which
might be found in them, and the potential discharges from each operation are
presented in the following sections.
Description Of The Coal Preparation Operation
Fixed-bed, atmospheric pressure gasifiers require a sized coal feed.
Current practice at all commercial LEG facilities in the U.S. is to purchase
pre-sized coal, eliminating the need for on-site crushing and sizing equipment.
Future LEG facilities are also expected to purchase pre-sized coal. As a re-
sult, coal preparation requirements for these facilities will most likely con-
sist only of coal receiving and storage, and means for transporting coal from
storage to the gasifier coal feed hoppers. Some facilities though may have to
perform final, on-site sizing if fuel size degradation occurs in shipment.
Discharges from the coal preparation operation include airborne coal
dust particles from coal handling, rainwater runoff from coal storage piles,
and, if final on-site sizing is performed, small amounts of coal fines. No
test data are available on the discharges from the coal preparation operation.
602
-------
However, their physical and chemical characteristics can be estimated from data
for similar discharges from the coal mining and coal-fired steam electric in-
dustries. Coal pile runoff tends to contain high levels of suspended and dis-
solved solids (including heavy metals) and can have an acidic or alkaline pH.
Dissolved organics tend to be at negligible or non-detectable levels. Dust
from coal handling and storage consists of small coal particles.
Description Of The Coal Gasification Operation
There are six commercially available gasifiers that operate in a
fixed-bed mode and at atmospheric pressure. They are:
Chapman (Wilputte),
Foster-Wheeler/Stoic,
Riley,
Wellman-Galusha,
Wellman Incandescent, and
Woodall-Duckham/Gas Integrale.
These gasifiers produce low-Btu gas by countercurrent gasification of coal with
a mixture of air and steam.
Coal is fed to the top of the gasifier from an overhead bin through a
lock hopper and/or a rotary feeder. As the coal gravitates downward through
the gasifier, it is contacted by rising hot gases and passes through "zones" of
progressively higher temperatures before exiting the bottom of the gasifier as
ash. As the coal is heated, it undergoes a series of physical and chemical re-
actions. Sequentially, these are drying, devolatilization, gasification, and
finally combustion. Air saturated with water, i.e., steam, enters at the bot-
tom of the gasifier. The steam absorbs some of the heat released in the com-
bustion zone, which helps to maintain the combustion temperature below the coal
ash softening temperature.
With most gasifiers, ash is collected at the bottom of the gasifier in
a water sealed ash pan and removed from the unit using an ash plow. The
Wellman-Galusha gasifier however, collects the ash in an ash hopper located be-
neath the gasifier. Ash is removed by adding water to the hopper and draining
the ash slurry through a slide valve. The water also s'erves to seal the gasi-
fier internals from the atmosphere during the ash removal step.
Pokeholes are located on the top of the gasifier. Rods are inserted
through the pokeholes to measure the depth and location of the "fire" and ash
zones. These rods can also be used to break up any agglomerates formed in the
bed.
The Wellman-Galusha, Chapman, and Riley gasifiers produce a single
low-Btu gas stream that exits the top of the gasifier. The Foster-Wheeler/
Stoic, Wellman Incandescent, and Woodall-Duckham/Gas Integrale gasifiers are
two-stage gasifiers that produce two gas streams. A "clear" gas stream,
603
-------
constituting approximately one-half of the total gas production, is withdrawn
from the gasification zone (near the middle of the gasifier). As such, it
contains essentially no tars or oils. The remaining gas, which contains tars
and oils, is withdrawn from the top of the gasifier where devolatilization of
the coal occurs.
At present, very limited environmental characterization data are
available for two-stage gasification systems. From a process viewpoint, the
two-stage gasification arrangement simplifies the gas purification operation,
but it does not appear to alter materially the system's potential environmental
impacts. The background study deals specifically with single-stage gasifica-
tion systems. However, the information developed is felt to also be generally
applicable to two-stage gasification systems.
Discharges from the coal gasification operation include:
• Gaseous emissions - pokehole gases
- coal feeder gases
- transient gases
• Liquid effluents - ash sluice water
(from Wellman-Galusha gasifiers only)
• Solid wastes - gasifier ash
Coal feeder gases, pokehole gases, and transient gases generated dur-
ing start-up, shutdown, and upset conditions are essentially raw low-Btu gas.
These discharges contain primarily carbon monoxide, carbon dioxide, hydrogen,
nitrogen, and water vapor. Minor components include hydrogen sulfide, carbonyl
sulfide, ammonia, hydrogen cyanide, entrained particulates, trace elements, low
molecular weight hydrocarbons, and, if the coal feed is lignite, bituminous, or
subbituminous, higher molecular weight organics (e.g., tars and oils).
Ash sluice water from Wellman-Galusha gasifiers contains suspended and
dissolved solids, including trace elements. Negligible or nondetectable levels
of organics have been identified, with most of them being attributable to arti-
facts of the sampling and analytical procedures. The pH of ash sluice water
can vary widely, depending on the characteristics of the ash. An alkaline pH
is typical if lignite is the coal feed, while acidic or neutral pH's are typi-
cal for other coal feeds.
Ash from the gasifier is similar to bottom ash from a coal-fired boil-
er although higher levels of residual carbon are present. Data for gasifica-
tion of several coals indicate that trace elements are not leachable in amounts
which would result in classification of gasifier ash as a hazardous waste.
Description Of The Gas Purification Operation
The purpose of the gas purification operation is to remove undesir-
able constituents such as entrained particulate matter, tars, oils, and sulfur
604
-------
from the raw low-Btu gas. Depending on the concentrations of these constitu-
ents in the raw gas and on the product gas specifications imposed by the end-
use (by either process or environmental considerations), none, some, or all of
these constituents may need to be controlled. No attempt was made to evaluate
systems producing a predefined product gas quality. Instead, systems were
selected based on existing or proposed purification configurations, with the
assumption that the resulting product gas quality would be sufficient to meet
the user's needs.
Particulate Removal. Entrained particulate matter can be removed
from the low-Btu gas with cyclones, wet scrubbers, and/or electrostatic precip-
itators (ESP). Cyclones are currently used in all domestic commercial LBG
facilities.
Tars and Oils Removal. The primary means of removing tars and oils
from raw low-Btu gas is to use wet scrubbers. These include in-line sprays,
wet cyclones, and spray, tray, or packed scrubbers. Most of the commercial-
ly available sulfur removal processes have limitations on the concentrations of
tars and oils in the gas to be treated. Normally, these levels cannot be
achieved using wet scrubbers alone. Detarrers (electrostatic precipitators)
have been used with some success for residual tars and oils removal.
Tars/oils-laden water from the scrubbers is directed to a gravity sep-
arator. Here, the heavier-than-water tars/oils are separated from the water
and recovered as a by-product. The scrubber water is then cooled in indirect
heat exchangers and recycled. Some volatile organic and inorganic species are
absorbed from the low-Btu gas when it is scrubbed. These species tend to de-
sorb from the scrubber water and fill the separator vapor space. They can be
recombined with the low-Btu gas by ducting the vapor space to the low-Btu gas
line.
In order to control the buildup of dissolved solids in the recircula-
ting scrubber water and/or to maintain a water balance in the scrubbing loop, a
portion of the scrubber water is removed as blowdown. The size of this blow-
down depends on such factors as the moisture and chloride content of the coal,
the dew point of the hot low-Btu gas and the temperature to which the gas is
cooled.
Sulfur Compounds Removal. Commercially available sulfur removal pro-
cesses include those using physical solvents, chemical solvents, combinations
of physical and chemical solvents, and processes featuring removal and direct
oxidation of sulfur compounds to produce elemental sulfur.^ Physical sol-
vent, combination chemical and physical solvents and some of the chemical
solvent processes are not well suited to the removal of sulfur compounds from
an atmospheric pressure, low-Btu gas.5 Several of the alkanolamine (chemical
solvent) processes can be used, but they require moderate pressurization of the
gas in order to obtain low residual sulfur levels. Regeneration of the
605
-------
alkanolamine solvent also produces an off-gas which contains the removed H2S
and C02, and which must be further processed for sulfur recovery. Standard
means of treating these off-gases (which will contain 70-95% CC^) is to route
them to a Glaus unit. The low t^S/high C(>2 content of these off-gases can
limit the recovery efficiency of the Glaus unit and prohibit the use of a Glaus
tail gas treatment process such as the SCOT unit. Thus, while alkanolamine
processes appear to be feasible for treating low-Btu gas, technical (and
economic) considerations indicate they are a poor choice. In light of the
above factors, none of the chemical or physical solvent processes were
evaluated in the background studies for the model plant III configurations.
The direct oxidation processes do not have gas pressure limitations
and are very effective in removing t^S. These processes also convert the
removed H2S directly into elemental sulfur, thus eliminating the need for ad-
ditional treatment of an H2S-laden off-gas. However, direct oxidation pro-
cesses do not remove significant amounts of non-K^S sulfur species such as
carbonyl sulfide (COS).5 For purposes of analysis, the Stretford process was
selected as a representative example of a commercially available direct oxida-
tion type sulfur removal process.
Summary of Discharges from Gas Purification. The existence, quan-
tity, and characteristics of discharges from the gas purification operation
depend on the degree of gas purification desired. In general terms, as the
low-Btu gas undergoes additional clean-up, additional waste streams are
created. These waste streams include:
• collected particulate matter from cyclones (all Model Plants),
• scrubbing liquor blowdown (Model Plants II and III),
• by-product tars and oils (Model Plants II and III except for
anthracite feed), and
• vent gas and sulfur cake from direct oxidation
sulfur removal processes (Model Plant III).
Collected particulates or cyclone dust has a very high carbon content
and resembles devolatilized coal. Leaching tests indicate that cyclone dust is
not a toxic waste.
Scrubbing liquor blowdown contains suspended solids, dissolved inor-
ganics (including trace elements and soluble gaseous components such as H2S
and NH3), and, unless anthracite is the coal feed, dissolved organics. By-
product tars/oils derived from gasification of non-anthracite coals are pre-
dominantly organic material, but also contain ash and various trace elements.
This material has a significant energy content, and represents a fuel resource
which should be recovered.
Discharges from the sulfur removal module include vent gases from the
Stretford oxidizer and sulfur cake. The oxidizer gases contain primarily
nitrogen, oxygen, and water vapor, with minor amounts of ammonia, carbon
dioxide, and reduced sulfur compounds. Other components of the low-Btu gas
606
-------
may also be absorbed by the Stretford scrubbing liquor and released in the
oxidizer. However, this is not expected to occur to any significant extent.
Sulfur produced in the Stretford process is initially recovered as a
cake containing nominally 50% water. Dissolved in the water are Stretford
scrubbing chemicals (sodium vanadates, anthraquinone disulfonic acid, ethylene
diamine tetracetic acid, iron, carbonates, and bicarbonates) and high levels of
nonregenerable sulfur components such as sulfates, thiosulfates, and
thiocyanates.
EVALUATION OF POLLUTION CONTROL TECHNOLOGIES
Evaluations of control technologies for application to individual
waste streams were based on considerations of control efficiency, ability to
comply with emissions regulations, capital and operating costs, energy and re-
source consumption, reliability, simplicity, multi-pollutant abatement capabil-
ity, residue generation and disposal requirements, potential for recovery of
by-products, and stage of development. The above criteria were used as a basis
for comparison of candidate control technologies either used alone or in
combination with in-plant control methods or other add-on controls.
Performance data for applicable control technologies were obtained
primarily from the open literature supplemented by vendor supplied data in some
cases. The capabilities of various control technologies were not usually as-
sessed on a design-specific basis but rather upon a generalized basis derived
from test results and/or engineering studies of the subject technologies.
In many cases performance can only be estimated in terms of control of
major constituents (e.g., carbon monoxide) or gross parameters (e.g., TOG)
since often no information is available for removal efficiencies for specific
substances. Further, even in those cases where substance-specific performance
information exists for a control technology, accurate or complete characteriza-
tion of the waste streams requiring control may be lacking. In the final ana-
lysis of course, the capabilities of state-of-the-art controls for LBG facil-
ities can be accurately evaluated only by testing operating facilities. Since
these opportunities are generally not available, the performance estimates
presented here are believed to reflect the best information currently available
based on actual experience and/or engineering analysis.
Air Pollution Control
The uncontrolled gaseous emissions from LBG facilities are summarized
in Table 2. The pollutants of potential concern, factors affecting the emis-
sion characteristics, and estimated emission flow rates are also summarized in
this table. Available control techniques for these emissions are discussed
below.
607
-------
TABLE 2. UNCONTROLLED ATMOSPHERIC EMISSIONS FROM LOW-BTU GASIFICATION FACILITIES
Uncontrolled Atmospheric
Emissions
Pollutants of
Potential Concern
Factors Affecting
Emissions Characteristics
Estimated Flowrate of
Uncontrolled Emissions
CD
O
CO
Airhorne particulates
from coal handling and
storage
(All Model Plants)
Coal feeder gases
(All Model Plants)
Pokehole gases
(All Model Plants)
Stretford oxidizer vent
gases
(Model Plant III)
Startup, shutdown and
upset gases
(All Model Plants)
Particulates
CO, H2S, HCN, trace
elements, and other
low-Btu gas components
CO, H2S, HCN, trace
elements, and other
low-Btu gas components
Reduced sulfur compounds,
ammonia
CO, H2S, HCN, trace
elements, and other
low-Btu gas components
Coal type; gasifier feed size
requirements; type and condition of
coal handling, crushing and sizing
equipment
Coal feeder design and conditions;
coal composition, feed rate and
adsorption characteristics; system
pressure
Pokehole design and conditions;
poking procedures and frequency;
system pressure
Coal Composition; Stretford unit
design and operation
Startup, shutdown and upset
procedures; gasifier reliability
Not estimated, but believed to be negligible since
presized coal is received at the plant site
Anthracite: 56 m3/hr (32 scfm)
Low-sulfur bituminous: 53 m3/hr (30 scfm)
High-sulfur bituminous: 62 m3/hr (35 scfm)
Lignite: 110 m3/hr (62 scfm)
Anthracite: 38 m3/hr (22 scfm)
Low-sulfur-bituminous: 16 ra3/hr (9 scfm)
High-sulfur bituminous: 16 m3/hr (9 scfm)
Lignite: 28 m3/hr (16 scfm)
Anthracite: 220 m3/hr (130 scfm)
Low-sulfur bituminous: 280 n3/hr (160 scfm)
High-sulfur bituminous: 2000 m3/hr (1100 scfm)
lignite: 600 m3/hr (340 scfm)
Not determined, highly variable
Note: nrVhr flow is relative to 25°C and atmospheric pressure, scfm flow is relative to 60 °F and atmospheric pressure.
-------
Airborne Particulates from Coal Handling and Preparation. Most LBG
installations will receive coal that has been crushed and sized. For these
installations, no significant particulate emissions are expected and therefore,
no control is necessary. If the coal feed is crushed and sized on site, then
airborne particulates generated by these operations may be a problem. Control
techniques involve enclosing the coal unloading facility, storage bins, crush-
ing and sizing equipment and any conveying devices. These enclosures should be
vented by low pressure ducting to a central bag filter collection system. An
induced draft fan at the outlet of the bag filters would provide the necessary
air flow and ensure that any leakage would be into the system.
Coal Feeder Gases. Low-Btu gas can leak from the gasifier vessel
through the coal feeder mechanism and up into the coal bin area by passing
countercurrent to the coal flow. One method of reducing the hazards from this
emission is to collect it before it enters the coal bin area and then disperse
it to the atmosphere through a vent pipe. The top of the coal bin must be
sealed (hooded) and a pipe run from there to an elevated outside venting point.
An induced draft fan in the vent line would draw air into the coal bin through
slots in the side of the bin. Coal feeder gases which pass up through the coal
in the bin would then be swept into the vent pipe. While this control option
incurs no significant operating costs or energy requirements, it does not
decrease the amount of coal feeder gases emitted to the atmosphere.
Another, and more effective means of controlling these emissions is to
return them to the process. This strategy can be done in one of two basic
ways. One approach is to enclose the coal bin (as with the atmospheric venting
option) and run a duct" to the intake of the gasifier air blower. To provide
continuous sweeping air in the coal bin (to prevent a possible explosive mix-
ture in the bin during very low air rates), a small vent and blow-off valve
will be needed in the air blower discharge line for venting during periods of
low gasifier air requirements. A second approach involves slightly pressuri-
zing the coal bin with an inert gas. This approach prevents the passage of
low-Btu gases into the coal bin. Either of these control options can effect
almost complete (99%) control of the coal feeder gases during normal gasifier
operations.
Pokehole Gases. Low-Btu gas escapes from pokeholes during and be-
tween poking operations. Improved pokehole designs are available with closer
tolerances and positive seal valves. While effective in reducing emissions
between poking operations, this control method still allows significant quan-
tities of gases to continue to escape during the poking operation.
A second control technology is to combine improved pokehole sealing
methods with the injection of an inert gas during poking operations. The inert
gas .effectively eliminates low-Btu gas leakage. Nitrogen is a possible choice
for the inert gas but this may incur operating costs (mainly for the purchase
of nitrogen) of up to two percent of the base plant annualized costs. If
available, steam might be a more economical choice since the steam require-
ment would be less than 0.1 percent of the product gas energy.
609
-------
Stretford Oxidizer Vent Gases. For systems using the Stretford pro-
cess to produce a desulfurlzed product gas, an air blown oxidizer is used to
convert the reduced Stretford solution back to its oxidized form. A large ex-
cess of air is used in the oxidizer and released in the vent. The vent gases
consist primarily of oxygen and nitrogen plus water vapor from the Stretford
solution. Minor amounts of ammonia and carbon dioxide and other components
absorbed from the Stretford solution may also be present. This emission is not
expected to pose a significant environmental problem if adequately dispersed to
the atmosphere.
Startup, Shutdown and Upset Gases. During gasifier startup, shut-
down, and upsets, gases are produced which do not meet product specifications.
If the gas is being burned locally and the customer can safely and economically
continue to combust the gas (possibly with auxiliary firing), then this is ob-
viously a good option and really represents a "no control required" situation.
If this option is not available, then two possible control strategies may be
used. One option is to combust these gases in an incinerator or flare. This
option requires installing piping, valves, and instrumentation. A second op-
tion is to vent the low-Btu product gas line to the atmosphere through a stack.
This option could pose localized odor problems. Therefore, its viability could
be limited to those areas where adequate dispersion is attainable.
Water Pollution Control
The uncontrolled effluents from LEG facilities are summarized in Table
3. The pollutants requiring control, factors affecting the effluent character-
istics, and estimated effluent flow rates are also summarized in this table.
Most of the processes considered for treating these effluents have not been
applied to the treatment of low-Btu gasification wastewaters. Therefore,
decisions related to the applicability, performance capabilities, and costs of
controls were based upon experience gained in related industries including the
coking, petroleum refining, and electric utility industries.
Coal Pile Runoff and Ash Sluice Water. These two effluents are very
similar to their counterparts in coal-fired power plants. They contain sus-
pended solids and dissolved inorganics but negligible dissolved organics.
Treatment techniques used in the utility industry include sedimentation, clari-
fication or filtration for suspended solids removal and acid or base addition
for pH adjustment. An additional treatment step available is chemical precipi-
tation for removal of selected trace elements. Use of these techniques for
coal pile runoff and ash sluice water from LEG facilities should produce an ef-
fluent which would meet the NSPS for coal-fired power plants.
Process Condensate. Process condensate contains suspended solids and
dissolved gases, organics, and trace elements. Viable treatment techniques for
dissolved organics include activated carbon adsorption and biological oxida-
tion. Sour water strippers can be used to remove dissolved gases. Chemical
precipitation treatment can be used to reduce the levels of trace elements,
although treatment to remove organics will be the key to disposing of this
stream in an environmentally acceptable manner.
610
-------
TABLE 3. UNCONTROLLED EFFLUENTS FROM LOW-BTU GASIFICATION FACILITIES
UNCONTROLLED EFFLUENTS FROM
LOW-BTU GASIFICATION FACILITIES
POLLUTANTS OF
POTENTIAL CONCERN
FACTORS AFFECTING EFFLUENT
CHARACTERISTICS
ESTIMATED EFFLUENT FLOWRATES
Coal Pile Runoff
(Model Plants I, II, and III)
Ash Sluice Water
(Model Plants I, II, and III
which use Wellman-Galusha
gaslfler)
Process Condensate
(Model Plants II and III)
Suspended solids
(coal fines),
inorganics leached
from coal, pH
Suspended solids,
inorganics and
trace elements
leached from ash
Suspended solids,
dissolved organlcs,
inorganics, trace
elements, and gases
Coal type and conditions of
wastewater contact with coal
(e.g., residence time) will
determine waste stream com-
position. Rainfall rates and
coal storage practices will
determine flow.
Characteristics of the ash and
contact time between the ash
and sluice water will deter-
mine waste stream composition.
Quantity of ash removed from
gasifier and operator prac-
tices will determine flow.
Composition of low-Btu gas has
major influence on composi-
tion. Important factors in-
clude H2S, HCN, NH3, and
tar/oil content of gas.
Chloride content of coal feed
and moisture content of gas
determine waste flow.
Flow rate is intermittent and variable. Annual
average: 7.5 to 15 kg/min (2 to 4 gpm).
Average from 10 year/24 hour rain: 380 to 760
kg/rain (100 to 200 gpm).
Flow rate Is Intermittent, existing only when ash
is removed. This is normally 2 or 3 times per day,
per gasifier.
Average flow: 20 to 60 kg/rain (5 to 16 gpm).
Based on maintaining water balance in quench loop:
bituminous coal - 23 kg/min (6 gpm)
lignite - 76 kg/rain (20 gpm)
anthracite - periodic
Flows may be as high as 76 kg/min (20 gpra) for all
coals In order to control chloride corrosion
problems.
-------
Thus two treatment options appear to be available for treating process
condensate: one uses carbon adsorption and steam stripping while the other
uses biological oxidation and steam stripping. Chemical precipitation could be
used with either option. For both of the options, the organics removal unit is
required only if the coal feed produces tars and oils when gasified. Since
anthracite does not produce tars and oils, the treatment of condensate from an
anthracite gasifier may not require dissolved organics removal. Representative
performance criteria for two treatment options for process condensate are
summarized in Table 4.
TABLE 4. ESTIMATED PERFORMANCE CAPABILITIES CF PROCESS CONDENSATE TREATMENT
TECHNOLOGIES
Component^Untreated Effluent Treated Effluent3 Treated Effluentb
TSS 140 <10 <30
Oil and Grease 400 <10 <30
BOD 9000 ? <1000
Phenols 2000 <5 <20
TOG 5600 <700 <700
NH3 4000 <50 <50
H2S 220 <10 <10
CN~ 1100 <10 <10
Trace Elements Yes some removal0 some removal0
Unit: mg/1
a Treatment using activated carbon adsorption and steam stripping.
b Treatment using biological oxidation and steam stripping.
° Increased removals of cationic trace elements can be achieved using
chemical precipitation.
Solid Waste Management Alternatives
The solid wastes generated by low-Btu gasification facilities are sum-
marized in Table 5. Included in this table are estimated flow rates, impor-
tant characteristics (such as physical condition, energy content, potential en-
vironmental problems), and expected classification (as hazardous or nonhazar-
dous) for each waste. Management techniques for these wastes should be based
on the criteria and guidelines developed by the EPA in response to the Resource
Conservation and Recovery Act.
Coal Fines. Generally, coal fines are not expected to be a waste
produced by low-Btu gasification facilities. This is because presized coal is
normally purchased, eliminating the need for on-site crushing and sizing. How-
ever, it is possible that final, on-site sizing may be required if fuel size
degradation occurs in shipment and handling. If so, a coal fines stream will
be produced. The quantity of fines produced is difficult to estimate but
612
-------
TABLE 5. UNCONTROLLED WASTES FROM LOW-BTU GASIFICATION FACILITIES
Waste
Flow Rate
Characteristics
Expected
Classification
en
oo
Coal Fines
(All Model Plants)
Gasifier Ash
(All Model Plants)
Cyclone Dust
(All Model Plants)
Stretford Sulfur Cake
(Model Plant III)
Tars ands Oils
(Model Plants II and
III gasifying non-
anthracite coals)
This is not a waste stream
unless on-site sizing is
employed. Flow rates have
not been estimated.
800 to 1800 kg/hr
6 to 38 kg/hr
70 to 620 kg/hr
750 to 1220 kg/hr
Dry solid; heating value
same as coal feed.
Damp solid with 20 to 30%
H20; heating value: 1.4
to 8.2 MJ/kg; leachable
trace elements.
Dry solid; heating value:
25 to 28 MJ/kg; leachable
trace elements.
Wet solid with approxi-
mately 50% H20; contains
thiocyanates, thiosul-
fates, iron, vanadates,
ADA, EDTA.
Viscous liquid; specific
gravity greater than one;
heating value: 30 to 37
MJ/kg; contains organics
and trace elements.
Non-hazardous
Non-hazardous
Non-hazardous
Hazardous
Hazardous
-------
should be very small. Since coal fines have the same energy content as coal, a
desirable means of handling them is to recover their energy value. Because of
the small quantities involved, this may be practical only if an existing com-
bustor is available on-site or nearby. If resource recovery is not practical,
then the coal fines should be disposed of as a nonhazardous waste in a sanitary
landfill.
Gasifier Ash. Gasifier ash is the unreacted portion of the coal fed
to the gasifier - predominantly mineral matter but also some carbonaceous
material. After dewatering, it is a damp solid containing 20 to 30 weight per-
cent water. All available data on gasifier ash indicate that it is a nonhazar-
dous waste. As such, the most reasonable option for disposing of gasifier ash
is disposal in a sanitary landfill.
Cyclone Dust. Cyclone dust resembles devolatilized coal. It has a
carbon content as high as 90 percent and a heating value of 25 MJ/kg (11,000
Btu/lb) or higher. It is removed from the cyclones as a dry, powdery solid.
All available data indicate that cyclone dust is a nonhazardous waste and could
be disposed of in a sanitary landfill. Because of its high energy content
though, consideration should be given to recovering its fuel value.
Stretford Sulfur Cake. Elemental sulfur is produced by a Stretford
unit and recovered as a filter cake containing approximately 50 percent water.
No test data are available for this waste. However, it will contain Stretford
solution chemicals (vanadates, anthraquinone disulfonic acid salts, EDTA, and
iron) and nonregenerable sulfur components such as thiocyanates and thiosul-
fates. Because of the presence of these contaminants, Stretford sulfur cake is
suspected to be a hazardous waste. If so, the management technique for this
waste would have to comply with the Subtitle C criteria and guidelines for haz-
ardous waste disposal. Alternatively, the contaminated sulfur can be processed
to recover a saleable by-product. This option produces an effluent containing
the contaminants originally present in the sulfur cake. Reductive incineration
and high temperature hydrolysis are two techniques recently developed for
treating Stretford solution effluent, but these approaches are not proven com-
mercially.
Tars and Oils. By-product tars and oils contain a number of toxic
organics. However, due to the high specific gravity and viscosity of this
material, it is expected to have a low vapor pressure which will minimize the
release of volatile organics during storage. Operators and handlers should
take precautionary steps to minimize contact with this material. Special note
should be taken of the NIOSH proposed criteria for coal gasification plants.
Because of its significant fuel value, the logical management technique for
by-product tars and oils is resource recovery. This would involve using the
material to fire a boiler or furnace.
614
-------
SUMMARY OF POLLUTION CONTROL COSTS AND ENERGY REQUIREMENTS
In order to compare controls for cost effectiveness and to estimate
the impact of pollution control costs on overall plant costs, approximate cap-
ital and operating costs for individual control processes/equipment were devel-
oped. These costs are based primarily on factored estimates of costs contained
in non-proprietary published literature, normalized to a first quarter 1980
basis. In some cases actual vendor quotes have been used but generally, it was
beyond the scope and purpose of the background studies to develop the detailed
engineering designs necessary for cost estimation at the "firm" (approaching ±
10 percent) level. Although the accuracy of the cost estimates varies, most
are believed to be within 50 percent.
For purposes of presentation in this paper, costs for various pollu-
tion control options are given as a percent of the "uncontrolled" plant capital
and total annualized costs. This format was selected since it more clearly
indicates the magnitude of pollution control costs on overall plant costs than
would actual dollar estimates. This approach has the additional benefit of
being less sensitive to assumed economic factors such as inflation, interest
rates (cost of capital), etc.
Total annualized costs were calculated as the sum of annual operating
cost and annualized capital costs. For purposes of annualizing the capital
investment, a fixed rate charge factor of 0.175 was calculated. This repre-
sents the fraction of the total capital investment that must be assessed as
annualized capital charge.
Table 6 summarizes the capital and annualized cost impacts of pollu-
tion control for the three model plants examined. The ranges shown reflect
differences in control costs as a result of gasifying the four coals studied.
They are not intended to reflect the accuracy of the cost impacts. All cost
numbers are expressed in terms of a percent of the uncontrolled base plant
costs.
As shown in this table, the cost impacts for emission controls are
minimal. Capital costs or annualized costs do not exceed 2 percent of the base
plant cost for any emission and, most of the control costs are below 1 percent.
On a total plant basis, the emission controls are estimated to add approxi-
mately 1 to 3 percent to the base plant capital requirements and increase an-
nualized costs by 2 to 5 percent. Energy requirements for air pollution con-
trol are negligible.
The cost impacts for controlling a specific liquid effluent are great-
est for the hot gas systems and least for the desulfurized gas systems. This
reflects an increase in the base plant costs and not a decrease in the control
costs. Total plant water treatment costs tend to increase or remain approxi-
mately constant as the degree of gas purification increases. This reflects the
fact that increases in the base plant costs (the denominator used to calculate
the percentage cost impacts shown) are offset by increased treatment costs (the
615
-------
TABLE 6. SUMMARY OF ESTSIMATED POLLUTION CONTROL COST IMPACTS3
cn
Control Costs as a Percent of Base Plant Costs
Hot Gas Cooled Gas
Capital Annualized Capital Annualized
GASEOUS EMISSIONS
Coal Feeder Gases 0.8-1.0 0.9- 1.7 0.6- 0.9 0.8- 1.6
Pokehole Gases 1.0-1.2 1.1- 2.0 0.7- 1.0 0.9- 1.9
Stretford Oxidizer - - - -
Transient Gases 0.8-1.0 1.1- 1.4 0.7 1.0- 1.3
TOTAL 2.8-3.0 3.1- 5.1 2.0- 2.6 2.7- 4.8
LIQUID EFFLUENTS
-------
numerator used to calculate the cost impacts) resulting from the need to treat
additional effluents. On a total plant basis, water pollution control costs
are estimated to increase the base plant capital costs by 3 to 15 percent and
annualized costs by 1 to 9 percent. Energy requirements for water pollution
control amount to 0.6 to 2.1 percent of the energy content of the low-Btu
product gas. This is almost entirely attributable to the sour water stripper
steam requirements for treating process condensate.
Capital cost estimates were not available for the solid waste disposal
practices. The waste disposal annualized costs are dominated by the costs of
handling gasifier ash, with the only other significant costs being those as-
sociated with sulfur cake disposal. (For the high sulfur bituminous coal case,
sulfur disposal costs are dominant). Cost factors used for disposal of wastes
were $21 and $71 per metric ton for nonhazardous and hazardous wastes, respec-
tively. Although $71 per tonne is a relatively high estimate for hazardous
waste dispoal, it may not truly reflect the costs associated with disposing of
very small quantities of hazardous wastes. For small quantities, the relative
impacts of capital costs and administrative costs (in terms of dollars per
tonne disposed) can be very large.
Energy requirements for disposing of solid wastes are minimal and are
estimated at 0.2% or less of the low-Btu gas energy content. The energy re-
quirements are mainly fuel for haul trucks and earthmoving equipment.
The total plant pollution control cost impacts are estimated to range
from approximately 6 to 17 percent of the base plant capital investment and
from 9.5 to slightly over 18 percent of the base plant's annualized costs.
617
-------
REFERENCES
1. Page, Gordon C. Environmental Assessment: Source Test and Evaluation
Report—Chapman Low-Btu Gasification. EPA-600/7-78-202, PB-289 940. Radian
Corp., Austin, TX, October 1978.
2. Thomas, W. C., K. N. Trede, and G. C. Page. Environmental Assessment:
Source Test and Evaluation Report—Wellman-Galusha (Glen Gery) Low-Btu
Gasification. EPA-600/7-79-185, PB80-102551. Radian Corp., Austin, TX,
August 1979.
3. Kilpatrick, M. P., R. A. Magee, and T. E. Emmel. Environmental Assessment:
Source Test and Evaluation Report—Wellman-Galusha (Fort Snelling) Low-Btu
Gasification. EPA-600/7-80-097, PB80-219330. Radian Corp., Austin, TX,
May 1980.
4. Cavanaugh, E. C., W. E. Corbett, and G. C. Page. Environmental
Assessment Data Base for Low/Medium-Btu Gasification Technology. Volume
I. Technical Discussion; Volume II. Appendices A-F. EPA-600/7-77-125a,
b, PB 274 844/AS, V. I., PB 274 843/AS, V. II. Radian Corp., Austin, TX,
November 1977.
5. Thomas, W. C. Technology Assessment Report for Industrial Boiler
Applications: Synthetic Fuels. EPA-600/7-79-178d. Radian Corp., Austin,
TX, November 1979.
618
-------
DEVELOPMENT OF A
POLLUTION CONTROL GUIDANCE DOCUMENT
FOR INDIRECT COAL LIQUEFACTION
by
Kimm Crawford
TRW Environmental Engineering Division
One Space Park Drive
Redondo Beach, California 90278
and
William J. Rhodes
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
and
William E. Corbett
Engineering Division
Radian Corporation
Austin, Texas 78758
ABSTRACT
Synfuels present both an opportunity and a problem for EPA in terms of
developing a new environmentally acceptable industry. The opportunity is for
EPA to encourage environmental controls to be incorporated/developed as an
integral part of the first plantdesigns rather than as "add on" technology in
an existing industry. The problem is that an adequate data base for pro-
mulgation of defensible regulations for synfuels plants does not now exist and
will likely not exist until after the first plants have been constructed and
operated for some period of time. EPA has responded to this situation with
the "Pollution Control Guidance Document (PCGD)" concept, in which the best
thinking of the various EPA R&D program and regional offices is to be provided
to permitters and to industry in the form of "guidance" for an interim period
rather than as regulations.
The Indirect Liquefaction (IL) PCGl) is one of the first such documents
which EPA is preparing with the technical support of various contractors.
TRW, Radian, Versar and RTI are involved in the preparation of the data base
for the first technical draft of the ILPCGD.
This paper summarizes the technology basis for control levels identified.
619
-------
DEVELOPMENT OF A POLLUTION CONTROL GUIDANCE DOCUMENT
FOR INDIRECT COAL LIQUEFACTION
The production of transportation fuels from domestic coal to displace
fuels derived from imported petroleum has high priority in the overall U.S.
energy policy. Since indirect liquefaction (IL) is the only commercially
demonstrated means of producing transportation fuels from coal, this technology
is likely to be among the first to be employed for synthetic fuels production
in the United States.
The Environmental Protection Agency (EPA) is responsible for ensuring
that the designs of first generation synthetic fuel technologies provide for
adequate protection of the environment. To serve this need and to avoid
costly delays in the commercialization of a process due to uncertainties con-
cerning environmental control requirements, EPA developed the Pollution Con-
trol Guidance Document (PCGD) approach. This paper summarizes the data base
that has been developed for the preparation of the PCGD for Lurgi-based IL
technology. EPA's technical support contractors in this effort are TRW,
Radian, Versar, and RTI.
The approach for the ILPCGDs was to develop a series of model plants
based on Lurgi, Texaco, and Koppers-Totzek (K-T) gasification using methanol,
Fischer-Tropsch (F-T), and Mobil M-gasoline synthesis. These technologies
are considered commercial or near-commercial. Majpr and minor constituent
material balances were established for integrated model plants using three
U.S. coals (Montana Rosebud subbituminous, Illinois No. 6 bituminous, and
North Dakota lignite) in order to provide estimates of the volumes and load-
ings of various waste streams which would be generated. Waste stream con-
stituents covered by the PCGD include both conventional/criteria/consent decree
pollutants and currently unregulated substances (e.g., POM).
The PCGD data base includes an identification and evaluation of various
pollution control options, based on the expected capabilities of available
technologies, for all major gaseous, aqueous, and solid waste streams gen-
erated in an integrated facility. This paper presents several of the control
620
-------
options developed in the data base. The control options are based on con-
siderations of the volume and toxicity of the specific waste stream, costs,
safety, reliability, degree to which controls have been demonstrated, intra-
and intermedia tradeoffs, and site specific factors.
The major sources of data used in the Lurgi data base for defining the types
and characteristics of uncontrolled indirect liquefaction plant waste streams
are (1) data obtained as part of an EPA sponsored environmental test program
of a Lurgi gasification facility at Kosovo, Yugoslavia; (2) data obtained as
part of an Energy Research and Development Administration (ERDA, now DOE)
sponsored program involving the gasification of American coals in a Lurgi
gasifier at Westfield, Scotland; (3) data obtained as part of an American
Natural Gas, Inc. sponsored program involving gasification of North Dakota
lignite at the SASOL plant in South Africa; (4) data provided to EPA by South
African Coal and Gas Corp. Ltd. (SASOL); and (5) data contained in various per-
mit filings and environmental impact statements for proposed Lurgi-based SNG
and indirect liquefaction facilities in the U.S.
Data sources employed for development of model plant/process configura-
tions were primarily engineering studies of the technology sponsored by DOE,
EPA, and EPRI. Data sources which served as the basis for the analysis of
pollution control applicability and costs include the above engineering
studies, studies conducted by TV A, various permit filings, technical informa-
tion obtained from pollution control equipment vendors and process developers,
and published literature. Much of the information on controls is derived
from applications in related industries such as petroleum refining, natural
gas processing, by-product coke production, electric utilities, and coal
preparation.
The configurations of the model plants were based on designs of Lurgi
plants which are either proposed or currently in operation. Auxiliary proc-
esses considered were those which would render a facility essentially self-
sufficient in energy (one which would need only run-of-mine coal, raw water,
and various chemicals and catalysts as inputs). A plant size corresponding
to 1 x 10 Btu/day (2.5 x 10 kcal/day) of total product was selected as
representative of the first plant(s) which may be built. This corresponds
to about 7000 bbls/day (1200 Nm3/day) gasoline plus 50 x 10 SCF (1.3 x 106
Nm ) of substitute natural gas per day (co-produced in the case of Lurgi
621
-------
gasification). This is approximately the size of the first phase facility
planned by American Natural Resources for their North Dakota SNG project.
Figures 1 and 2 are simplified flow diagrams of the main process train
and auxiliary operations associated with integrated Lurgi IL facilities.
System operations include coal preparation, coal gasification, gas purifica-
tion and upgrading, crude product synthesis and separation, and product up-
grading. Nonpollution control auxiliary processes include process cooling,
product storage, raw water treatment, steam and power generation, and oxygen
production. The major waste streams identified for facilities depicted in
the figures are listed in Table 1 along with the primary constituents/para-
meters of concern for each waste. The remainder of this paper will focus on
control options for these major streams in Lurgi-based facilities. Note that
no fundamentally new problems are believed to apply to K-T or Texaco gasifi-
cation which do not also apply to Lurgi gasification, although differences
do exist in the relative quantities of wastes/waste constituents which are
generated. Indeed, K-T and Texaco gasification may be somewhat less com-
plicated than Lurgi since the former gasifiers generate fewer organics (other
than methane and formic acid) which would eventually become components of
waste streams. The organics in Lurgi wastes present some of the more diffi-
cult pollution control problems.
Gaseous Waste Streams
Figure 3 summarizes the primary control options for Lurgi acid gases.
Indicated in the figure are both selective and nonselective Rectisol* acid
gas removal (AGR) ; that is, separate removal of CO and H_S from product gas
generating an H S-rich stream and a CO -rich stream or combined removal gen-
2, &•
erating only one dilute H S stream. The primary goal of selective AGR is to
produce a more concentrated sulfur-bearing stream for sulfur recovery allow-
ing either the use of Glaus technology or the reduction in a Stretford plant
size (and thus reduced cost). Since selective AGR is significantly more
expensive than nonselective AGR, it is economically justified only if cost
savings are realized in sulfur recovery/pollution control. If, for environ-
mental reasons, the CC>2-rich stream from selective AGR cannot be directly
discharged to the atmosphere (with perhaps incineration), then treatment
*Rectisol is a Lurgi-licensed acid gas removal (AGR) process and would be
used with all Lurgi gasifiers in the U.S.
622
-------
FISCHER
TROPSCH
SYNTHESIS
LIQUIDS
RECOVERY
^
W
METHANATION
— w
LPG
GASOLINE
DIESEL FUEL
HEAVY OIL
ALCOHOLS
SNG
RAW
COAL
ro
u>
COAL
PREPARATION
•i
w
r: ACICI^ATIAM
VjMol r IUM 1 1 WIN
^
QUENCH AND
DUST REMOVAL
^
SHIFT
CONVERSION
ACID GAS
REMOVAL
METHANOL
SYNTHESIS
METHANOL
RECOVERY
MOBIL M-
GASOLINE
SYNTHESIS
TO SNG OR FUEL
METHANOL
TO SNG, LPG OR FUEL
GASOLINE
Figure 1. Simplified Flow Diagram of Indirect Coal Liquefaction Facilities
-------
FLUE GAS
AIR
AIR
SEPARATION
PLANT
STORM
WATER"
HOLDING
POND
en
ro
RAW
WATER"
RAW WATER
TREATMENT
MAKE-UP
WATER
BOILER
FEEDWATER
STEAM AND POWER
GENERATION
STEAM
BOTTOM
ASH
BOILER FEEDWATER
MAKE-UP TREATMENT
BRINES
BRINE
CONCENTRATOR
BOILER
BLOWDOWN
EVAPORATIVE
BRINES
SLUDGES
COOLING
TOWER
EVAPORATION
DRIFT
•> COOLING TOWER
BLOWDOWN
METHANOL
GASOLINE
DIESEL OIL
HEAVY OIL
KETONES
HEAVY
ALCOHOLS
PRODUCT
STORAGE
Figure 2. Auxiliary Operations Associated with an Indirect Coal Liquefaction Facility
-------
OPTION I.
TO
ATMOSPHERE
TO
ATMOSPHERE
RAW
LURGI -
GAS
COM-
BINED
ACID
GASES
CT)
rv>
en
STRETFORD
STRETFORD
OXIDIZE!
VENT
1
INCINERATION
(BOILER)
BEAVON
FLUE GAS
DESULFURI2ATION
kTO
* ATMOSPHERE
.TO
ATMOSPHERE
ENRICHMENT
(AOIP)
GAS | 1
H2S LEAN
GAS
INCINERATION
(BOILER)
FLUE GAS
DESULFURIZATION
.TO
ATMOSPHERE
BEAVON
.TO
ATMOSPHERE
INCINERATION
(BOILER)
FLUE GAS
DESULFURIZATION
TO
' ATMOSPHERE
OPTION ii.
TO
ATMOSPHERE
TO
RAW .
GAS
r
SELECTIVE
RECTISOL
NON
SELECTIVE
RECTISOL
1
CO? RICH
GAS
H2S RICH
GAS
COM-
BINED
ACID
GASES
1 — >
i
AMINE
ENRICHMENT
(ADIP)
OXIDIZER
INCINERATION
M CLAUS
—
|
t
TO
ATMOSPHERE
SCOr * ATMOSPHERE
INCINERATION » FLUE GAS
(BOILER) DESULFURIZATION
TO
' ATMOSPHERE
Figure 3. Options for Control of Lurgi/Rectisol Acid Gases
-------
TABLE 1. MAJOR WASTE STREAMS IN AN INTEGRATED INDIRECT LIQUEFACTION FACILITY
WASTE STREAMS
GASEOUS STREAMS
• ACID GASES (INCLUDING STRIPPING AND DEPRESSURIZATION
GASES)
• BOILER FLUE GASES
• TRANSIENT WASTE GASES
• FEED LOCKHOPPER VENT GASES
PRIMARY CONSTITUENTS/PARAMETERS OF CONCERN
GASEOUS STREAMS
• REDUCED SULFUR AND NITROGEN COMPOUNDS, HYDROCARBONS
• SULFUR DIOXIDE, PARTICULATES, NITROGEN OXIDES
REDUCED SULFUR AND NITROGEN COMPOUNDS, HYDROCARBONS,
CARBON MONOXIDE, PARTICULATES, POLYCYCLIC ORGANIC
MATERIAL
CTi
IN5
CXl
• CATALYST REGENERATION/DECOMMISSIONING OFFGASES
AQUEOUS STREAMS
• RAW GAS QUENCH AND ACID GAS REMOVAL UONDENSATES
• ASH QUENCH SLOWDOWN
• SYNTHESIS WASTEWATERS
• WASTEWATER TREATMENT BRINES
SOLID WASTES/SLUDGES
• GASIFIER ASH
• BOILER ASH
• FGD SLUDGES AND BRINES
• WASTEWATER TREATMENT BRINES
• BIOSLUDGES
• SPENT CATALYSTS
• SULFUR DIOXIDE, PARTICULATES, CARBON MONOXIDE,
TRACE ELEMENTS
AQUEOUS STREAMS
• ORGANIC COMPOUNDS, SUSPENDED SOLIDS, CYANIDES AND
THIOCYANATES, AMMONIA, TRACE ELEMENTS
• DISSOLVED AND SUSPENDED SOLIDS, TRACE ELEMENTS
• ORGANIC COMPOUNDS
• DISSOLVED AND SUSPENDED SOLIDS, TRACE ELEMENTS
SOLID WASTES/SLUDGES
• SOLUBLE SALTS, TRACE ELEMENTS
• SOLUBLE SALTS, TRACE ELEMENTS
• SOLUBLE SALTS, TRACE ELEMENTS
• SOLUBLE SALTS AND ORGANICS, TRACE ELEMENTS
• SOLUBLE ORGANICS, TRACE ELEMENTS
• TRACE ELEMENTS
-------
costs for this stream would likely make the selective AGR option unattractive
and designers may revert to nonselective modes.
Option I in Figure 3 consists of Stretford or Claus sulfur recovery
followed by tail gas treatment (TGT) for residual sulfur removal and hydro-
carbon control, in the Claus cases, enrichment of the H S feed stream may
be required or desired and an amine (ADIP) system is indicated in the figure.
The ADIP offgas and the Claus offgas both receive TGT prior to atmospheric
discharge; the CO rich gas from selective AGR is directly discharged to the
atmosphere. TGT technologies include incineration/FGD (e.g., WeiIman-Lord)
and catalytic reduction H S recycle (e.g., Beavon).
The Option II alternatives consist of either Stretford sulfur recovery
followed by incineration for hydrocarbon control or Claus sulfur recovery
followed by SCOT TGT. Neither Claus without sulfur TGT nor direct incinera-
tion followed by flue gas desulfurization is considered adequate under Option
II since neither of these controls achieves the same levels of total sulfur
emissions compared to Stretford or Claus/SCOT. Note that the alternatives in
Figure 3 represent the range of controls envisioned by all conceptual and
proposed Lurgi gasification projects in the U.S. which have been identified
to date.
Table 2 summarizes the estimated costs and energy requirements for control
of acid gas in integrated facilities. The cost data represent the least expen-
sive system in each option but assume no credit for energy recovery from incin-
eration of Lurgi gases. Total annualized costs range from 3.8 to 5.7% of base
plant costs for sulfur recovery with TGT compared to 2.3 to 4.0 for sulfur re-
moval only (Stretford). Energy requirements of control of acid gases vary
from essentially zero to 1.9% of plant input energy, depending primarily on
the extent of heat recovery practiced during incineration. Recovered energy
could exceed that required to operate the sulfur control systems.
Options for the control of boiler flue gas emissions correspond to the
levels defined by electric utility NSPS (Option I) and large industrial boiler
NSPS (Option II). Table 3 summarizes the S02, particulates , and NOX options.
For gaseous and liquid fuels derived from coal (e.g., tars, oils, phenols,
naphtha, low Btu gas), the same limits apply as to the petroleum or natural
gas fuels.
627
-------
TABLE 2. RELATIVE COSTS AND ENERGY REQUIREMENTS FOR CONTROL OF ACID GASES
(AS PERCENT OF BASE PLANT COST OR ENERGY INPUT)
Low Sulfur Coal
Total
Capital Annual Energy
High Sulfur Coal
Total
Capital Annual Energy
Option I
(Sulfur removal
plus tail gas
treatment)
Option II
(Sulfur removal,
minimum or no
tail gas
treatment)
3.2
3.8 0 -0.84
1.6
2.3 0 - 0.8
5.3
3.0
5.7 0 - 1.9
4.0 0 - 1.8
TABLE 3. CONTROL OPTIONS FOR COAL BOILER S02, PARTICULATE, AND NOX EMISSIONS
Option I
g/106 cal (lb/106 Btu)
Option II
g/106 cal (lb/106 Btu)
SO,
Particulates
NOX Lignite &
bituminous
coals
Subbituminous
coals
Lurgi
byproducts
2.16 (1.2)
and 90% control unless
emissions less than
1.09 (0.6) in which
case 70% required
0.054 (0.03)
1.1 (0.6)
0.88 (0.5)
1.1 (0.6)
2.16 (1.2)
0.18 (0.10)
1.26 (0.7)
628
-------
Costs associated with a representative FGD system (Wellman-Lord) applied
to a coal- and Lurgi-byproduct-fired boiler are estimated in Table 4. Annual-
ized costs of the FGD systems amount to 2.4 - 3.9% of base plant costs, depend-
ing on the boiler size, coal sulfur content, and degree of SO2 removal attained.
Energy requirements for the example FGD units range from 2.9 to 5.8% of the
boiler heat input, or 0.4 to 0.6% of total plant input energy. Note that
incremental costs for FGD sulfur removal are about $ll-15/lb ($24-33/kg)
while incremental costs for sulfur recovery FGT sulfur removal are about $20-
30/lb ($44-66Ag) • Thus, it may be less expensive to design for lower emis-
sions at the boiler rather than lower emissions from sulfur recovery opera-r
tions if minimum overall sulfur emissions control at least cost is a defined
goal and is environmentally acceptable.
Table 5 summarizes the control options for smaller volume waste streams
in Lurgi indirect liquefaction facilities. Generally, the controls for
these streams consist of incineration with or without additional SCL and/or
particulate control.
Aqueous Waste Streams
Figure 4 presents the major options evaluated for control of gasification
and synthesis wastewaters. Lurgi wastewaters (gas liquors) are treated for
tar/oil separation, phenol removal (Phenosolvan), and ammonia removal as
basic steps in all cases. Further treatment would consist of biological or
chemical oxidation for bulk organics removal and chemical precipitation and
carbon absorption for trace elements and refractory organics removal when
discharge to surface waters is the wastewater disposal method (Option I).
When "zero discharge" to surface waters is to be practiced, treatment would
consist of volume reduction via use of cooling towers, evaporators, and/or
incinerators. Biological oxidation may precede the cooling tower concentra-
tion step. Ultimate disposal of residual brines may be via underground
injection (Option II), surface impoundment (Option III), and ash quenching
(Option IV).
The "zero discharge" options involve various tradeoffs with air emis-
sions (cooling tower evaporation/drift) or solid waste disposal (leaching of
organics or trace elements in surface impoundments or landfills). In the
case of codisposal of brines with ash, the combined waste may be rendered haz-
ardous due to the residual organics or trace elements contained in the brine.
629
-------
TABLE 4. SO2 EMISSIONS, COSTS, AND ENERGY REQUIREMENTS ASSOCIATED WITH
BOILER/WELLMAN-LORD FGD SYSTEMS
Low Sulfur
(Rosebud)
High Sulfur
(Illinois No. 6)
Sulfur
Removal
(%)
70
80
80
90
S02
Emissions
(kg/106 kcal)
0.88
0.58
0.98
0.51
Costs
Capital
(%)**
2.6
4.0
2.5
3.2
Annual
($/kg S
(%)** Removed)
2.7 9.7
3.9 12.0
2.4 9.2
3.6 12.0
Energy***
Requirements
(%)
2.9
3.2
5.2
5.8
*Coal to boiler
**Percentage of uncontrolled base plant costs1
***As percentage of coal fed to boiler
TABLE 5. CONTROL OPTIONS FOR SMALL VOLUME LURGI WASTE GASES
Feed Lock
Vent Gases
Transient
Waste Gases
Catalyst
Decommissioning offgases
Option I
Option II
Recompression/
recycle or use
as fuel for
high pressure
gases, incin-
eration of low
pressure
residuals
Discharge of
residuals via
low energy
scrubber
Incineration
with SO2 and
particulate
control
Incineration,
short term dis-
charge of high
oxygen content
waste gases
Incineration with SO2
and particulate
control
Incineration
630
-------
+ ACTIVE CARBON
REGENERATION
| OFF-GAS
STRIPPING GASES
NH3 1 fc.NH,
RECOVERY 1 ^ *
LIQUOR * SOLVHT ^ STEAM
HCN RICH EXTRACTION "* STRIPPING j m n H k
PHENOLS AGR STILL
BOTTOMS — '
f T WASTPWATPH-
MQRII .M VUASTFWATFR —
TO
INCINERATION
CONTROL OPTION I
SURFACE DISCHARGE
H CHEMICAL 1 J CARBON 1 ^
PREC,P,TAT,ON| ^ ABSORPT.ON | *
r OXIDATION | | |
BIOSLUDGE-* 1 ^SLUDGES + SORBENT
. fc-
CONTROL OPTION II
DEEP WELL INJECTION
H BIOLOGICAL
OXIDATION '
* DISCHARGE
DISCHARGE
T DRIFT T T GASES
. COOLING TOWER .1 BRINE 1 J™" 1
> CUNCENTKA- Bf LUfvuLlUIKA- 1 'M INtlNcRATION 1 W
'TION H TION 1 ^\ \
BIO COOLING 1 \
EFFLUENT TOWER ^CONCENTRATED
SLOWDOWN 1 RECOVERED BRINES
TBIOSLUDGE TWATER
CONTROL OPTION III
SURFACE IMPOUNDMEN
H BIOLOGICAL
OXIDATION
BIOSLUDGE
CONTROL OPTION IV
CO-DISPOSAL WITH ASH
H BIOLOGICAL )
OXIDATION |
1
A EVAPORATION/DRIFT
COOLING TOWERJ
^TION | r
,
A EVAPORATION/ A AFLUE GASES
ADRIFT AOFFGAS J
^1 1
ICOOLING TOWERI ^1 BRINE 1 1 "I"—— 1 r
HTION 1 T TION |
1 ' »
RECOVERED 1 *
1 DEEP WELL
|BJ fCpYlflU
1 EVAPORATION
PONDS
EVAPORATION
PONDS
1 TO ASH
QUENCH
TO ASH
QUENCH
Figure 4. Control Options for Lurgl-based Indirect Liquefaction Plant Wastewaters
-------
Table 6 summarizes the estimated costs and energy requirements for the
water pollution control technologies depicted in Figure 4. Although treat-
ment costs are highly coal-, gasifier-, and synthesis-case specific, these
estimates indicate the relative contribution of various unit processes to
overall costs. The basic treatment steps, phenol removal, ammonia removal,
and biological oxidation, constitute 40 to 80% of total treatment costs (or
about 3.1% of the base plant annualized costs). Carbon absorption/chemical
precipitation is seen as a less expensive route than forced evaporation or
surface impoundment for further treatment. The data also indicate that the
basic treatment processes also contribute a large fraction of the total energy
requirement for water pollution control, with further treatment contributing
heavily only with incineration. The use of the cooling tower as a "precon-
centration" step has been assumed in the estimates in Table 6; hence treat-
ment of wastewaters by forced evaporation, incineration, or surface impound-
ment without prior volume reduction could dramatically increase the costs of
water pollution control.
Solid/Hazardous Wastes
Options for the disposal of solid wastes generated by the subject faci-
lities are determined both by the characteristics of the waste and by the
local environment providing candidate disposal sites. The general operation
performance standards for various hazardous waste disposal methods are cur-
rently being drafted by EPA's Office of Solid Waste. These standards, based
on "best engineering judgment," are expected to largely define the practices
for and site-specific factors to be considered in the treatment/disposal of
hazardous (and in many cases nonhazardous) wastes. Thus, for purposes of
PCGD development, the focus has been on providing a data base for the classi-
fication of indirect liquefaction wastes based on their characteristics.
Perhaps the most important waste from the standpoint of volume in the
subject facilities is gasifier ash. Several papers presented at this sym-
posium have provided data on the leaching characteristics of ash from a
variety of gasifiers and coal types. Generally, these data suggest that
gasifier ash is not expected to be hazardous based upon the RCRA Extraction
Procedure* test. Thus, this material will likely be handled in a manner
*Refers to the Extraction Procedure defined in 40 CFR 261.
632
-------
TABLE 6. TYPICAL COSTS AND ENERGY REQUIREMENTS OF WATER POLLUTION CONTROL
TECHNOLOGIES
Phenosolvan
NH3 Stripping
Biological oxidation
Chemical precipitation
Carbon adsorption
Forced evaporation
Incineration
Deep well injection
Evaporation ponds
Cost*
Capital
1.2
0.9
1.4
0.5
0.3
1.3
0.3
0.2
7.1
Annual
1.4
0.6
1.1
0.4
0.2
1.1
0.3
-
4.3
Energy**
Requirements
1.3
2.9
0.1
0.04
0.01
0.2
0.9
-
-
*As percentage of uncontrolled base plant costs
**As percentage of total base plant coal energy input
633
-------
similar to boiler bottom ash and FGD sludges in the electric utility industry.
Limited data indicate that when such wastes are to be disposed of in surface
mines that placement should be in "V-notch" areas of the spoil pile rather
than in the pit bottom to minimize leaching.
Two important wastes are potentially generated by wastewater treatment
(WWT) brines from evaporators or incinerator scrubbers and sludges from bio-
logical treatment. In the case of the former, codisposal with gasifier or
boiler ash is commonly proposed (codisposal with some type of solid material
would be required in any case since RCRA guidelines prohibit the disposal of
free flowing liquids in landfills). Codisposal of WWT brines with ash is
believed to render the ash hazardous if the organics are not previously des-
troyed by incineration or wet oxidation. However, if the organics in the
brine are destroyed prior to codisposal, available data indicate that the
ash/brine mixture would be classified as nonhazardous according to the RCRA
Extraction Procedure test. Thus, a tradeoff may exist between WWT costs for
organics destruction and solid (hazardous) waste disposal costs for hazard-
ous vs. non-hazardous disposal. WWT brines may also be disposed of in sur-
face impoundments or by underground injection consistent with RCRA require-
ments. In the later case, organics in the waste may have to be destroyed
prior to injection to prevent plugging of the accepting formation.
Biosludges from WWT would likely be considered a hazardous waste under
RCRA. Options for disposal include landfarming, incineration with air pollu-
tion control, landfill or mine disposal, and surface impoundment. Dewatered
sludges may be beneficially utilized by landfarming in conjunction with
revegetation of surface mine spoil overburden.
Several types of spent catalyst wastes are generated in indirect lique-
faction facilities, including those from shift synthesis (methanol, F-T,
Mobil), methanation, and air pollution control (Claus, Beavon). Wastes such
as spent shift catalyst are expected to be hazardous due to their inherent
metal content as well as other toxic elements derived from coal. Wastes
such as Mobil-M (a zeolite material) and Claus (Bauxite) spent catalysts are
not believed to be hazardous, but data are lacking on RCRA leach character-
istics or other toxicity information. Many of the catalyst materials can be
economically recycled for their metal values, particularly when the costs of
disposal as hazardous waste are set as the point of reference.
634
-------
Table 7 summarizes the total estimated costs and energy impact of pollu-
tion control for the options presented. The data indicate that air pollution
control can add up to 14% of base plant annualized costs, water pollution
control up to about 9%, and solid/hazardous waste disposal up to 3.3%, or up
to 26% for controls in all media.
Energy requirements for pollution control range from 4.4 to almost 11%
of plant input energy, with water pollution control contributing over 60% of
the requirement. The differences in energy requirements between the control
options are not especially large.
635
-------
TABLE 7. SUMMARY OF TOTAL COSTS AND ENERGY IMPACTS FOR POLLUTION CONTROL IN
AN INTEGRATED FACILITY
Pollution Control
Technology
Air
Water
Solid Waste
Total Percent
of Base Plant
% of Total
Annual! zed
Option I Option
9.1-14
3.7 - 8.
2.6 - 3.
15.4 - 25
.1 5.8 -
5 3.1 -
3 1.8 -
.9 10.7 -
Costs
II
11.7
7.5
2.3
21.5
% of Plant
Option I
1.6 - 2.8
3.0 - 8.0
0.06 - 0.08
4.7 - 10.9
Energy Reqmts.
Option II
1.4 - 2.5
3.0 - 7.9
0.04 - 0.06
4.4 - 10.5
636
-------
INITIAL EFFORT ON A POLLUTION CONTROL
GUIDANCE DOCUMENT; DIRECT LIQUEFACTION
J. E. COTTER, C. C. SHIH, B. ST. JOHN
TRW, INC.
REDONDO BEACH, CA 90278
(ABSTRACT)
Development of the pollution control guidance document (PCGD) for direct
coal liquefaction is preceding in parallel with the permitting and construction
of the first demonstration-size liquefaction plant, the SRC-II unit in Ft.
Martin W.V. In addition to the SRC-II process, the PCGD will provide guidance
for the other major liquefaction technologies: SRC-I, H-Coal, and Exxon Donor
Solvent.
The control technology guidance will be related to baseline designs
prepared for each of the four liquefaction processes, sized at 100,000 bbls/day
production. The baseline designs are composed of material balance flowsheets and
uncontrolled waste stream calcuations, using plant configurations which are
most likely to occur in future commercial size plants. Variations of the
baseline designs will be considered if they affect control decisions. A
range of feed coals have been selected for the baseline cases, with at least
one common coal type that could be used by all four processes. The present
effort is focused on identification of the pollutants of concern using pilot-
plant test data from coal liquefaction developers, DOE, and EPA sponsored
testing programs. These data will be evaluated with a variety of engineering
analysis methodologies, so that the subsequent examination of control options
can be carried out.
The range of control options--air, water, solid waste—will be selected
from those methods that have a known track record in related industrial
applications, such as petroleum refining, coke ovens, and mining.
The control technologies will be charaterized parametrically according to
the inlet stream compositions and quantities, and their percentage release of
specific pollutants. Finally, the cost of control will be developed according
to the same parameters, with a range of costs obtained depending on the com-
plexity and efficiency of control.
637
-------
INITIAL EFFORT ON A POLLUTION CONTROL
GUIDANCE DOCUMENT; DIRECT LIQUEFACTION
DIRECT COAL LIQUEFACTION PROCESSES
The Direct Liquefaction PCGD will be based on those liquefaction processes
that are the closest to commercialization. The SRC-I, SRC-II, H-Coal and
Exxon Donor Solvent (EDS) processes are all at an advanced stage of pilot-
plant development, and the SRC-I and SRC-II processes will be expanded to
demonstration size units in the next few years. Although other "second
generation" direct liquefaction processes are in bench-scale development, they
will not be ready for commercialization until the early 1900's. The current
status of the advanced development processes are:
• The SRC-I process is being tested in a 50 tons/day pilot plant
at Fort Lewis, Washington, and in a 6 tons/day process develop-
ment unit at Wilsonville, Alabama. Preliminary designs for a
demonstration plant, to be located near Newman, Kentucky, were
completed on July 1979. The demonstration plant is designed to
produce the equivalent of 20,000 barrels of oil per day, and is
scheduled to be completed by 1984. Current plans call for en-
largement of the facility to produce the equivalent of 100,000
barrels of oil per day in 1990.
• The SRC-II process is also being tested in the pilot plant at
Fort Lewis, Washington. Preliminary designs for a SRC-II
demonstration plant, to be located at Fort Martin, West
Virginia, were completed in July 1979. The demonstration plant
is designed to process 6,000 tons of coal per day to produce
the equivalent of 20,000 barrels of oil per day. Completion of
the plant is scheduled for 1984.
• The EDS pilot plant at Baytown, Texas, started up on June 24,
1980. This plant has a capacity of 250 tons per day of coal
feed to produce approximately 600 barrels per day of synthetic
liquid fuel. A 70 tons per day Flexicoking unit at the same
site is planned to be completed in the second quarter of 1982.
The design of a demonstration plant could begin as early as
the fourth quarter of 1982, leading to a start-up date of about
1988.
• The H-Coal pilot plant at Catlettsburg, Kentucky, has been
operational since June 1980. This plant has a capacity of 600
tons per day of coal feed. Support work in a 3 tons per day
process development unit is also continuing. Groundbreaking
for a commerical plant in Breckinridge, Kentucky, is planned
for 1983. The commercial plant is expected to start production
as early as 1987.
638
-------
SRC-I PROCESS^
The SRC-I is a process for concerting high-sulfur, high-ash coals to
a low-sulfur and substantially ash-free solid fuel. In the SRC-I process
(Figure 1), feed coal is pulverized and slurried in a process-derived
solvent. This slurry is then pumped to reaction pressure (2000 psig),
mixed with hydrogen-rich recycle gas, and then heated to reaction temperature
in a fired-heater. Within the fired-heater, coal dissolution is accomplished
and hydrogenation reactions begin. At the exit of the fired-heater, hot
hydrogen makeup gas from a hydrogen makeup area is added to the slurry, and
the mixture is sent to the dissolver.
The dissolver effluent is flashed. The raw gas is sent to gas purifica-
tion, and the slurry containing unconverted coal and ash from the low-pressure
flash is sent to a vacuum column, where process solvent and lighter compo-
nents are removed from the SRC slurry. The SRC ash slurry is then sent to
solvent deashing unit, where it is separated into SRC and ash concentrates.
The ash concentrate, consisting of ash and unreacted coal, and some
residual SRC, is gasified with steam and oxygen. The syngas produced, after
shift conversion and acid gas removal, is converted to hydrogen and sent to
the dissolver unit as makeup. The major portion of the SRC concentrate is
solidified into the primary final product, solvent refined coal.
SRC-II PROCESS^
The SRC-II process is designed to produce low-sulfur liquid fuel from
high-sulfur bituminous coals. As shown in Figure 2, raw coal is pulverized,
mixed with a recycle slurry stream from the process, and then pumped together
with recycle and makeup hydrogen through a preheater to a dissolver operated
at high temperature and pressure. The coal is first dissolved in the liquid
portion of the recycle slurry and then largely hydrocracked to liquids and
gases. Much of the sulfur, oxygen, and nitrogen in the original coal is
hydrogenated to hydrogen sulfide, water, and ammonia, respectively. The
rates of these reactions are increased by the catalytic activity of the un-
dissolved mineral residues. The recycle of a portion of the product slurry
contributes substantially to the process by increasing the concentration of
catalytic mineral residue in the reactor.
639
-------
VENT OASES
HYDROGEN
SLURRY
MIX
TANK
COAL
SLURRY
HYDROGEN
HYDROCARBON
.OAS
HYDROGEN
RECOVERY
AND GAS
DESULFURIZATION
SULFUR
WATER
RAW OAS
FIRED
PREHEATER
A
aoo°f
REACTOR
SOLVENT RECYCLE
COAL
GASIFIER
AND
SHIFT
CONVERTER
STEAM
SOLIDS
SLURRY
SOLID/LIQUID
SEPARATION
OXYGEN
ASH
SOLVENT
RECOVERY
UNIT
COAL
SOLUTION
PRODUCT
LIGHT OIL
MOLTEN SRC TO SOLIDIFICATION
SRC-I PROCESS
Figure 1.
-------
DNICD
PULVERIZED
COAL
rUBIFHD HYDHQOEN HICYCli
VENT OASES 111 PIPELINE
SLURRY
/IIXING TANK
VAPOR-LIQUID
SEPARATORS
I
/
/
/
V.
1
•
I
J
1
MAKEUP if
HYDROQENJ |
SLURRY
PREHEATER
m
A
— 1
REACTOR
(DISSOLVER)
SSO'F
t.SOOpst
j"
{
PRODUCT
SLURRY
f
I I
SHIFT CONVERSION
AND PURIFICATION
OXYGEN
PLANT
•TEAM
GASIFIER
V
•LAO
=n Ml
CRYOGENIC
SEPARATOR
ACIO GAS
REMOVAL
LPO
SULFUR
WATCH
LIGHT
DISTILLATE
FRACTIONA7OR
VACUUM TOWER
RESIDUE SLURRY
SRC-II PROCESS
Figure 2.
-------
The dissolver effluent is separated into gas, light hydrocarbon liquid
and slurry streams using conventional flashing and fractionation techniques.
A portion of the mineral residue slurry and hydrocarbon liquid from the
separation area is recycled to blend with the feed coal in the slurry prepar-
ation plant. The balance of the mineral residue slurry is vacuum flashed
to recover the fuel oil product.
The dissolver area gas stream (consisting primarily of hydrogen, light
hydrocarbons, and hydrogen sulfide) is treated for liquid hydrocarbons and acid
gas removal, and the major portion of this gas is then recycled to the process.
Makeup hydrogen for the process is produced by the gasification of mineral
residue slurry to produce synthesis gas, followed by shift conversion.
Liquid products from the main process area are refined in the fraction-
ation section into naphtha, light fuel oil, and heavy fuel oil. Various by-
product liquid and gas streams are treated further in the gas plant to produce
propane, butane, and pipeline gas. Secondary recovery plants are provided
to recover ammonia, tar acids and sulfur.
EDS PROCESS f3)
The Exxon Donor Solvent (EDS) is a noncatalytic process that liquefies
coal by the use of a hydrogen donor solvent obtained from coal-derived
distillate. The donor solvent transfers hydrogen to the coal, thus, promoting
the liquefaction of coal.
In the EDS process (Figure 3), ground coal is slurried with the recycle
donor solvent. The slurry is heated by a fired-heater, and preheated hydrogen
is added. The liquefaction reaction is carried out in a tubular reactor at
800-900 F and 2000 psig. Products from the liquefaction reactor are sent to
several stages of separation units for recovery of gas, naphtha, middle dis-
tillate, and bottoms comprised primarily of unreacted coal and mineral matter.
Solid and liquid products are separated by distillation.
642
-------
GO
FUEL
GAS
(FOR
PREHEATERS)
1
GAS
CLEANING
VACUUM
FLASH
SEPARATOR
VACUUM
:RACTION-
ATING
TOWER
TO SULFUR
RECOVERY
AMMONIA
NAPHTHA
MIDDLE
DISTILLATE
VACUUM
SLURRY
BOTTOMS
COAL
PREPARATION
AND
DRYING
FLEXICOKER
MIXING
VESSEL
PREHEATER
PREHEATER
CATALYTIC
SOLVENT
HYDROGENATOR
DONOR SOLVENT
(RECYCLE)
EXXON DONOR SOLVENT PROCESS
Figure 3.
-------
The heavy vacuum bottoms from distillation are fed to a FLEXICOKING
unit with air and steam to produce additional distillate liquid products and
a low Btu fuel gas for process furnaces. In the FLEXICOKING unit, essen-
tially all organic material in the vacuum bottoms is recovered as liquid
product or combustible gases.
Hydrogen for in-plant use is produced by steam reforming of light hy-
drocarbon gases. An alternative method for hydrogen production is partial
oxidation of the heavy vacuum bottoms or of coal.
H-COAL PROCESS^
The H-Coal process is a catalytic hydro!iquefaction process that converts
high-sulfur coal to either a low-sulfur boiler fuel or to a refinery syncrude.
In this process (Figure 4), coal is dried and crushed, then slurried with
recycled oil and pumped to a pressure of 2000 atm. Compressed hydrogen is
added to the slurry, and the mixture is preheated and charged continuously to
the bottom of the ebullated-bed catalytic reactor. Upward passage of the inter-
nally recycled reaction mixture maintains the catalyst in a fluidized state
(catalyst activity is maintained by the semi continuous addition of fresh catalyst
and the withdrawal of spent catalyst). Typical mixing temperature entering the
reactor is 600° to 700° F.
The vapor product leaving the top of the reactor is cooled to condense
the heavier components as a liquid. Light hydrocarbons, ammonia and hydrogen
sulfide, are absorbed and separated from the remaining gas, leaving a hydrogen-
rich gas which is recompressed and recycled to be combined with the input slurry.
The liquid-solid product, containing unconverted coal, ash, and oil, is fed into
a flash separator. The bottoms product containing solids and heavy oil is
further separated with a hydroclone, a steam stripper, and a vacuum still.
The gas and liquid products (hydrocarbon gas, hydrogen sulfide, ammo-
nia, light and heavy distillates, and residual fuel) may be further refined
while heavy distillate is recycled as the slurry medium.
644
-------
HYDROGEN RECYCLE
FEED
COAL
1
SLURRY
PREPARATION
H.O
DRYER
J3 AND
GRINDER
in
I
I
SLURRY
PUMP
GAS
TREATMENT
AND
SEPARATION
CONDENSER
L
FUEL GAS
SULFUR
WATER
NH,
I
EBULLATED-
BED
CATALYTIC
REACTOR
LIGHT
DISTILLATE
HEAVY
DISTILLATE
FLASH
SEPARATOR
COMPRESSION
t
HYDROGEN
PRODUCTION
r
ASH
^
SOLIDS
LADEN
RESIDUE
LIQUID/SOLID
SEPARATOR
1
\\
UNDERFLI
STILL
HEAVY
^DISTILLATE
••RESIDUAL
FUEL OIL
H-COAL PROCESS
Figure 4.
-------
APPROACH TO PROCESS CHARACTERIZATION
A methodology has been established that uses a baseline design for each
process, sized at 100,000 bbls/day net equivalent of product liquids, fuel
gases, and coal-replacement solid products. The design and pilot-plant ex-
perience of the several liquefaction processes has been limited to certain
types of feed coals, so that the guidance document will have to recognize that
expected variations in proposed liquefaction plant feed coals will be limited
to an experience range. This will be particularly critical for the non-
catalytic SRC-I and SRC-II processes, which depend on the catalytic properties
of constituents found in bituminous coals for adequate yields. At least two
feed coals will be used in the PCGD analysis for each given liquefaction
process, with Illinois No. 6 grade being common to all processes. Initial
baseline design concepts are being prepared and submitted for comment to
the developers of the four liquefaction processes. In most cases, commercial
design concepts of these process developers are somewhat of a moving target,
and it is generally recognized that the baseline design cases will not neces-
sarily represent a particular final design configuration. The process developers
will be asked to confirm that proposed baseline designs represent a feasible
plant configuration, and to estimate the impact that various design options may
have on the waste stream characteristics of a baseline case. The goal of this
preparatory step is to provide a process description that EPA permit reviewers
can reasonably compare with submitted applications.
The initial baseline designs, including material balances and flowcharts
which identify the major and minor stream constituents at key points, are
being prepared by incorporating pilot plant test results and engineering estimates
with commerical-plant design cases that have been released by each process
developer. A critical feature of these analyses will be the validation and
interpretation of pilot-plant test data. Determinations will be made as to
whether these data were obtained under steady-state conditions, using standard-
ized sampling and analysis techniques. The uncontrolled constituents in
each waste stream ( gaseous, liquid, or solid) have to be estimated in these
baseline design cases in order to realistically evaluate control technology
646
-------
requirements. A substantially inaccurate estimate could lead to either inade-
quate control technology specifications or unnecessary pollution control invest-
ment requirements.
The major gaseous emission streams requiring control include the following;
• Fugitive dust emissions from coal storage
• Fugitive dust emissions from coal and slag handling
• Fugitive hydrocarbon emissions from valves, flanges, and seals
• Fugitive hydrocarbon emissions from product and byproduct storage
t Off gas from coal dryer
• Acid gases containing H2S, C02,COS, CS2, and mercaptans and NH3
from sour water stripping units
• Flue gas from process heaters
• Flue gas from steam plant
• Flue gas from power plant
• Evaporation and drifts from cooling towers
An essential element of these uncontrolled stream charaterizations is the
fugitive vapor emission category. A very limited amount of ambient organic
vapor sampling has been conducted at the SRC-II pilot plant at Ft. Lewis.
Although this sampling and analysis effort cannot be directly extrapolated to
full-scale plants because of operations which are unique to the pilot
plant, the measurements offer some insight into the ability of heavy organics
(e.g., POM) to disperse into the surrounding atmosphere as a result of small
vapor emissions.
The major wastewater streams requiring control include the following:
• Sour process wastewater from vapor washes, condensers,
fractionator overhead drums, sulfur recovery plant, and
coal slurry mixing operation
• Cooling tower blowdown
t Boiler blowdown
t Coal pile runoff
• Oily water runoff from processing areas
• Miscellaneous small wastewater streams
Untreated wastewater characterizations will be derived from measurements
conducted by process developers, EPA, and DOE sampling and analysis efforts.
Some judgements will have to be made concerning the effects of coal feed
647
-------
characteristics and process operating configurations on these measurement
values. Most of these measurements have focused on process wastewater (or
"sour water", following refinery terminology). Other anticipated sources of
wastewater include coal pile and area runoff, cooling tower blowdown, and
discharge from dust collection and conveying use. These other categories are
analagous to related discharges from coal handling and other industrial
operations.
Solid waste discharges will include gasifier slag (from hydrogen syn-
thesis), spent catalysts, wastewater and raw water treatment sludges, and
possibly non-salable byproduct residues. Some limited amount of leaching
tests have been done to characterize gasifier slags and some residue material,
but more work will have to be done before a determination can be made as
to the possible characterization of these wastes as non-hazardous or hazardous,
CONTROL TECHNOLOGY EVALUATION
EPA permit reviewers will be faced with a range of possible control
technologies connected with direct liquefaction process designs. To help
the permit reviewers in their examination of submitted plans, a number of
best-available-control-technology (BACT) options will be evaluated for each
potential waste stream for each of the four major liquefaction processes.
In addition, two levels of control effectiveness will be included. The
evaluation of each control technology will include the efficiency of pollu-
tant removal from a stream, multipollutant removal capability, installed and
operating cost, reliability, turndown ratio, sensitivity to process stream
conditions, energy consumption, and any other operating history information
such as maintenance requirements.
A primary air pollution control concern in liquefaction processes is
the treatment of acid gases generated in the liquefaction reactor, from sour
water stripping, and in gasification of residiuum streams to make hydrogen.
A typical process design method for removing C02 and H2S constituents from
these streams is some form of absorption, such as DEA, Selexol, or Benfield
processes. The H2S-rich gas stream stripped from the absorbing liquid
constitutes the acid gas stream requiring further control. Representative
648
-------
acid gas stream compositions are shown in Table 1. These streams can be
subjected to two stages of sulfur removal. Concentrated (20-70%) HgS streams
will be handled by a process technology that does bulk sulfur removal. The
Claus sulfur recovery process is the most likely candidate for this job,
based on a long history of refinery and gas processing experience, but
investigations are underway to evaluate Stretford process applicability with
high H2S concentrations. Residual sulfur removal options are numerous; some
technologies accept Claus tail-gas directly and hydrolize S02 to H2S, others
require oxidation of H2S in the stream to S02- The PCGD evaluation will
evaluate many combinations of control technology types to establish BACT
performance and cost ranges.
An example of a number of combinations is shown in Table 2, using two
bulk-sulfur removal options, three residual sulfur removal options, and a
final incineration step option (for potential trace organic removal and
oxidation of trace sulfur to S02).
649
-------
TABLE 1. REPRESENTATIVE ACID GAS STREAMS FROM DIRECT LIQUEFACTION
SOURCE
cr>
en
O
1
c:
o
2
o
c
o
o
c
01
4J
(/I
o
Stripper off gas
from process gas
treating
H2S 75
C02 20
CO Trace
COS Not determined
Stripper off gas
from syngas
purification
30
50
10
.0003
Sour water
stripper
offgas
25
50
-
-------
TABLE 2
Bulk-S Removal
Options
Combinations*
1
2
3
4
5
6
Claus
Stretford
t
t
0
Residual-S Removal Incineration
SUDT7Wellman-
Beavon SUPERSCOT Lord
•
t
•
An additional combination will be examined for streams containing very low
H2S ( or COS, CSg etc.) concentrations, since these may be directly incinerated.
Both capital and operating costs will be determined according to the
standardized guidelines prepared by IERL/RTP^ '. The impacts on other media
for any of the pollution control technologies will also be quantified; the acid gas
gas treatment systems above will produce spent catalysts as well as minor
liquid purge streams. A substantial non-hazardous solid waste quantity will
require disposal planning if the recovered sulfur is not salable. Wastewater
treatment guidance is expected to emphasize the stripping of ammonia and
H2S from sour water streams, and the absorption of phenols. The sequence
of these byproduct recovery steps may be significant to recovery efficiency.
Subsequent treatment steps will be selected to minimize the release of
trace organics and heavy metals to the environment. Investigations of "zero
discharge" evaporative methods are currently being compared with more con-
ventional biological treatment technologies. A high degree of water reuse
will be emphasized no matter what treatment method is used.
651
-------
The impact on solid waste handling and management requirements may be
substantial, depending on the control options recommended for wastewater
treatment and air pollution control technology. The cost and stringency
of solid waste management practices will be greatest for wastes designated
as hazardous under RCRA definitions.
REFERENCES
(1) Tao, J. C. and A. F. Yen; Environmental Control Systems of the SRC-I
Demonstration Plant, Second DOE Environmental Control Symposium,
Reston, Va., March 1980.
(2) Sehmalzer, D. K. and C. R. Moxley. Environmental Control System for the
SRC-II Demonstration Plant. Second DOE Environmental Control Symposium,
Reston, Va., March 1980.
(3) Green, R. C., Environmental Controls for the Exxon Donor Solvent Coal
Liquefaction Process, Second DOE Environmental Control Sumposium,
Reston, Va., March 1980.
(4) Gray, J. A., H-Coal Pilot Plant Environmental Controls, Second DOE
Environmental Control Symposium, Reston, Va., March 1980.
(5) A Standard Procedure for Cost Analysis of Pollution Control Operations,
Vol. 1. EPA-600/8-79-013a, June 1979.
652
-------
APPENDIX: ATTENDEES
653
-------
ATTENDEES
FUEL CONVERSION TECHNOLOGY, V SYMPOSIUM
September 16-19, 1980
Chase-Park Plaza Hotel
St. Louis, MO
Oi
cr
Alexander
AUred
Al.maula
Altschuler
Andrews
Antizzo
Applewhite
Aronson
Aul
Ayer
Azevedo
Baker
Barnett
Barrs
Batty
Bee
Bell
Bcrtrand
Bocchino
Boegly, Jr.
Bogardus
Boliac
Bombaugh
Boswell
Bowerman
Brasowski
Breuer
Broker
Burchard
Burns
Canales
Carstea
Carter
Chen
Cheng
Christopher
Clausen
Cleary
Collins
Corbett
CotLer
Cowles
Cowser
Crawford
Cura
Curry
Dal Santo
Del]inger
Dennis
Denny
James K.
Roy C.
Bipin C.
Morris
Richard D.
James V.
Grant D.
John G.
Ed F.
Franklin A.
Alfred
Robert J.
Russell
Thomas W.
C. R.
Robert W.
Linda R.
Rene R.
Robert M.
William J.
Raymond B.
Charles E.
Karl J.
James T.
Herbert F.
Leon
C. Thomas
Gunter
John K.
Eugene A.
Manuel J.
Dan
Stephen R.
Hsiu-Luan
Daniel H.
Jay
John F.
Joseph G.
Robert V.
William E.
Jack
John 0.
K. E.
Kiium W.
Jerome J.
Lloyd
Dario J.
Ba r ry
Patrick
Dale A.
P. 0. Box E
P. 0. Box 1267
MS E-201, Germantown
401 M Street, S. W.
4704 Harlan St.
7655 Old Springhouse Road
1930 Bishop Lane
1716 Heath Parkway
7927 Jones Branch Drive
P. 0. Box 12194
1558 Washington Street, E.
25 Main Street
4th Floor, Capital Plaza Tower
5120 Belmont Road
620 Fifth Avenue
20030 Century Boulevard
Ridgeway St.
P. 0. Box 101
2400 Ardmore Boulevard
P. 0. Box X
6900 Wisconsin Avenue
248 401 Building
8500 Shoal Creek Boulevard
P. 0. Box 225621, MS-349
232 Valleton Lane
110 South Orange Avenue
Acorn Park
Wallneyerstrasse 6
IERL, MD-60
P. 0. Box 1620
Nyala Farm Road
7929 Westpark Drive
763 New Ballas Road, South
650 Winter Avenue
P. 0. Box 880
P. 0. Box 2521
Bldg. 01, Room 2020, 1 Space Park
1 Lethbridge Plaza
8500 Shoal Creek Boulevard
8500 Shoal Creek Boulevard
1 Space Park
8301 Greensboro Drive
P. 0. Box X
One Space Park Drive, K4/1136
151 Bear Hill Road
8500 Capital Drive
345 Courtland Street
Box 12313
2200 Churchill Road
IERL, MD-62
Oak Ridge
Ponca City
Washington
Washington
Denver
McLean
Louisville
Fort Collins
McLean
Research Triangle Park
Charleston
Belleville
Frankfort
Downers Grove
New York
Germantown
Norris
Florham Park
Pittsburgh
Oak Ridge
Chevy Chase
Chattanooga
Austin
Dallas
Walnut Creek
Livingston
Cambridge
4300 Essen
Research Triangle Park
La Jolla
Westport
McLean
St. Louis
Paramus
Morgantown
Houston
Redondo Beach
Mahwah
Austin
Austin
Redondo Beach
McLean
Oak Ridge
Redondo Beach
Waltham
Milwaukee
Atlanta
Research Triangle Park
Springfield
Research Triangle Park
TN
OK
DC
DC
CO
VA
KY
CO
VA
NC
WV
NJ
KY
IL
NY
MD
TN
NJ
PA
TN
MD
TN
TX
TX
CA
NJ
MA
GERMANY
NC
CA
CT
VA
MO
NJ
WV
TX
CA
NJ
TX
TX
CA
VA
TN
CA
MA
WI
GA
NC
IL
NC
37830
74601
20545
20460
80212
22102
40277
80522
22102
27709
25311
07109
40601
60515
10020
20767
37828
07932
15221
37830
20015
37401
78758
75265
94596
07039
02140
27711
92038
06680
22102
63141
07652
26505
77099
90278
07430
78758
78758
90278
22102
37830
90278
02154
53222
30342
27709
62706
27711
U.S. Department of Energy
Conoco Inc.
U.S. Department of Energy
U.S. EPA
Rocky Mountain Energy Company
International Research & Tech. Corp.
American Air Filter Co., Inc.
Environmental Research & Technology Inc
Radian Corporation
Research Triangle Institute
WV Air Pollution Control Commission
Pennwalt Corp., Wallace & Tiernan Div.
Dept. for Nat. Res. & Env. Prot.
Mittelhauser Corporation
BP North America Inc.
The Aerospace Corporation
Tennessee Valley Authority
Exxon Research & Eng. Co.
Energy Impact Associates
Oak Ridge National Laboratory
WAPORA, Inc.
Tennessee Valley Authority
Radian Corporation
Texas Instruments, Incorporated
Ind. Refiners of Calif.
Foster Wheeler Energy Corp.
Arthur D. Little, Inc.
Landesansalt fur Immissionsschutz
U.S. EPA
Systems, Science & Software
Stauffer Chemical Company
UOP/SDC
Environmental Science & Engineering
Burns & Roe Industrial Service Corp.
EG&G
Texas Eastern Corp.
TRW, Inc.
HydroQuaJ Inc.
Radian Corporation
Radian Corporation
TRW, Inc.
TRW Energy Systems Group
Union Carbide Nuclear Co.
TRW, Inc.
EG&G, Environmental Consultants
Camp, Dresser & McKee
U.S. EPA, Region IV
Northrop Services Inc.
Illinois EPA
U.S. EPA
-------
CTl
tn
Drummoud Charles J. P. 0. Box 10940
Duliamcl Paul EV-34, MS E-201, GTN
Dunn James E. 737 Executive Park
Durmington Frank M. 50 Stamford St.
Ellis Linda E. P. 0. Box 8405
Enoch Harry P. 0. Box 11888, Iron Works Pike
Erskine George 1820 Dolley Madison Boulevard
Evans Robert 3424 S. State Street
Evers Robert W. 1000 Chestnut Street Tower II
Evers Theo 4200 Linnean Avenue N. W.
Faist Michael B. 8500 Shoal Creek Boulevard
Felix W. Dale 329 Building, 300 Area
Ferrell James K. Dept. Chemical Engineering
Fischer William H. P. 0. Box 1498
Fox Robert D. 9041 Executive Park Drive
Freeman Philip G. Box 8213, University Station
Friedman Bernard S. 4800 S. Chicago Beach Dr., Rm.l616N
Friedman Max 1 Penn Plaza
Fritschen Herman A. P. 0. Box 300
Geyer Roseann 2970 Maria Avenue
Giddings Jeffrey P. 0. Box X
Gieck Joe 1500 Meadow Lake Parkway
Ginsbnrg Robert 59 East Van Buren
Grano, Jr. John R. P. 0. 7167 Ben Franklin Sta.
Gray W. Scott 50 Beale St., P. 0. Box 3965
Greene Jack H. IERL, MD-60
Greene Kevin 59 East Van Buren
Griffin Mike P. 0. Box 3809
Gryka George E. Nyala Farm Road
Guenther Fred H. 5265 Hohman Avenue
Gulledge William P. P.O. Box 10940
Hangebrauck Robert P. IERL, MD-61
Hanson Douglas M. 225 Wildwood Avenue
Headley Larry P. 0. Box 880
Heap Michael P. 8001 Irvine Boulevard
Heaton Richard P. 0. Box 1663
Hellman Karl H. 2565 Plymouth Road
Henschel D. Bruce IERL, MD-61
Herman Mark N. P. 0. Box 101
Holubowich Alexandra McGraw Hill, 1221 Ave. of Americas
Honefenger Ronald L. 2700 South Post Oak
Howard F. Sidney One Davis Drive
Huang F.draund T. 1126 South 70th Street
Huang Hann S. 9700 S. Cass
Hudson P. E. (Ted) 8500 Shoal Creek Boulevard
Hughes Larry W. P. 0. Box 391
Ireland Sydney J. 8400 Westpark Drive
Jackson James 0. P. 0. Box 1633, MS-486
Janes T. Kelly IERL, MD-61
Jennings Larry 4704 Harlan Street
Pittsburgh
Washington
Louisville
Boston
Kansas City
Lexington
McLean
Chicago
Chattanooga
Washington
Austin
Richland
Raleigh
Reading
Knoxville
Grand Forks
Chicago
New York
Tulsa
Northbrook
Oak Ridge
Kansas City
Chicago
Washington
San Francisco
Research Triangle
Chicago
Butte
Westport
Hammond
Pittsburgh
Research Triangle
Woburn
Morgantown
Santa Ana
Los Alamos
Ann Arbor
Research Triangle
Florham Park
New York
Houston
Belmont
Milwaukee
Argonne
Austin
Ashland
McLean
Los Alamos
Research Triangle
Denver
PA
DC
KY
HA
MO
KY
VA
IL
TN
DC
TX
WA
NC
PA
TN
ND
IL
NY
OK
IL
TN
MO
IL
DC
CA
Park NC
IL
MT
CT
IN
PA
Park NC
MA
WV
CA
NM
MI
Park NC
NJ
NY
TX
CA
WI
IL
TX
KY
VA
NM
Park NC
CO
15236 U. S. Dept. of Energy/PETC
20545 U.S. Department of Energy
40207 CoaLiquid
02114 Metcalf & Eddy
64114 Black & Veatch Consulting Engineers
40578 Kentucky Department of Energy
22102 The MITRE Corporation
60616 Institute of Gas Technology
37401 Tennessee Valley Authority
20008 Netherlands Embassy
78758 Radian Corporation
99352 Battelle-Northwest
27650 N. C. State University
19603 Gilbert/Commonwealth
37919 IT Enviroscience
58202 U.S. Department of Energy
60615 Consultant
10119 Chemico Air Pollution
74102 Cities Service Company
60062 Mcllvaine Co.
37830 Oak Ridge National Laboratory
64114 Black & Veatch, Cons. Engineers
60605 Citizens for a Better Environment
20044 Inside EPA Weekly Report
94119 Bechtel National, Inc.
27711 U.S. EPA
60605 Citizens for a Better Environment
59701 MT Energy and MHD Res. & Dev. Inst.
06680 Stauffer Chemical Company
46325 Norhtern Indiana Public Service Co.
15236 Pittsburgh Energy Technology Center
27711 U.S. EPA
01801 Bioassey Systems Corp.
26505 Department of Energy
92705 Energy & Environmental Research Corp.
87545 Los Alamos Scientific Laboratory
48105 U.S. EPA
27711 U.S. EPA
07932 Exxon Engineering
10016 SynFuels
77056 Transco Companies, Inc.
94002 Lurgi Corp.
53214 Allis-Chalmers Corporation
60439 Argonne National Lab.
78758 Radian Corporation
41101 Ashland Oil, Inc.
22102 Science Applications, Inc.
87545 Los Alamos Scientific Laboratory
27711 U.S. EPA
80212 Rocky Mountain Energy Company
-------
Jessup
Johnson
Johns ton
Jones
Jones
Jones
Josephsou
Jost
Junkin
Kalish
Kapsalopoulou
Kaufman
Kelly
Kendell
Kilgroe
Kim
Kingsbury
Kirchgessner
Klein
Knauss
Kniffin
Komai
Krishnan
Kuntz
Lagemann
Lessig
Lillian
Loran
Luthy
Mack
NacKenzie, Jr.
Maddox
Madenburg
Ma gee
Malki
Mansoor
McAllister
McMichael
McSorley
Michael
Miller
Mirchandani
Mixon
Moghissi
Mohn
Mohr, Jr.
Morgan
Mulder
Hulvihill
Murray
Deborah H.
Larry D.
Ross M.
Fred L.
Hershal T.
N. Stuart
Julian
Jack L.
Preston D.
Robert
Ariadni
Joseph W.
Robert M.
James
James D.
Jung I.
Garrie L.
David
Jerry A.
James
Troy
Ralph Y.
R.
Gail
Robert C.
Dennis C.
Daniel
Bruno
Richard G.
Karen L.
Kenneth W.
Emily L.
Richard S.
Robert A.
Kal
Yardena
Robert A.
William J.
Joseph A.
Don R.
M. Dean
Dilip M.
Forest 0.
A. Alan
Nancy C.
Donald H.
Dennis L.
Willem C.
James W.
Charles
1231-25th Street, N. W.
IERL, MD-62
300 W. Washington Street
One Woodward Avenue, 6th Floor
MS-E333
P. 0. Box 12194
1151-l6th Street, N. W.
800 N. Lindbergh
8301 Greensboro Drive (Rm. 657)
P. 0. Box 150, Building 2506
6621 Electronic Drive
10 Bl Phillips Building
Box 5035, Riddick Hal]
Washington University, Box 1226
IERL, MD-61
9190 Red Branch Road
P. 0. Box 12194
IERL, MD-61
P. 0. Box X
3399 Tates Creek Road
801 North Eleventh
P. 0. Box 10412
251 South Lake Avenue
32 South Ewing
IERL, MD-61
2223 Dodge
Oper. & Env. Safety Div., EV-133
100 West Walnut Street
Schenley Park
P. 0. Box 12194
6630 Harwin Drive
1007 Market St., Central Res. & Dev.
P. 0. Box 7808, II Plaza
8500 Shoal Creek Boulevard
31 Inverness Parkway
1725 K Street, N. W.
P. 0. Box 13000
P. 0. Box 12194
IERL, MD-63
245 Summer Street
Box 305, S. Illinois Avenue
Two World Trade Center
P. 0. Box 12194
RD-682
1000 Prospect Hill Road
Oilman Hall
P. 0. Box 32
Kiggelaerstraat 15
One Oliver Plaza
01/2060, 1 Space Park
Washington
Research Triangle
Chicago
Detroit
Washington
Research Triangle
Washington
St. Louis
McLean
Plaquemine
Springfield
Bartlesville
Raleigh
St. Louis
Research Triangle
Columbia
Research Triangle
Research Triangle
Oak Ridge
Lexington
St. Louis
Palo Alto
Pasadena
Helena
Research Triangle
Omaha
Washington
Pasadena
Pittsburgh
Research Triangle
Houston
Wilmington
Boise
Austin
Birmingham
Washington
Research Triangle
Research Triangle
Research Triangle
Boston
Belleville
New York
Research Triangle
Washington
Windsor
Berkeley
Miamisburg
2596 TL Den Haag
Pittsburgh
Redondo Beach
Park
Park
Park
Park
Park
Park
Park
Park
Park
Park
Park
DC
NC
IL
MI
DC
NC
DC
MO
VA
LA
VA
OK
NC
MO
NC
MD
NC
NC
TN
KY
MO
CA
CA
MT
NC
ME
DC
CA
PA
NC
TX
DE
ID
TX
AL
DC
NC
NC
NC
MA
IL
NY
NC
DC
CT
CA
OH
NETHERLANDS
PA
CA
20037
27711
60606
48226
20545
27709
20036
63166
22102
70764
22151
74004
27650
63130
27711
21045
27709
27711
37830
40502
63101
94303
91101
59601
27711
68102
20545
91124
15213
27709
77071
19898
83729
78758
35243
20006
27709
27709
27711
02107
62221
10048
27709
20460
06095
94720
45342
15102
90278
Bureau of National Affairs
U.S. EPA
Marblehead Lime Company
American Natural Service Company
U.S. Department of Energy
Research Triangle Institute
Environmental Science & Technology
Monsanto Company
TRW Energy Systems Planning Division
Dow Chemical
Versar Inc.
Phillips Petroleum Company
North Carolina State University
In These Times
U.S. EPA
Hittman Associates, Inc.
Research Triangle Institute
U.S. EPA
Oak Ridge National Laboratory
Dames & Moore
Sverdrup Corporation
Electric Power Research Institute
Jacobs Enginering Group Inc.
Montana Dept. of Natural Resources
U.S. EPA
InterNorth, Inc.
U.S. Department of Energy
Ralph M. Parsons Company
Carnegie-Mellon University
Research Triangle Institute
ERT, Inc.
E. I. DuPont de Nemours & Company
Morrison-Knudsen Co., Inc.
Radian Corporation
Combustion Engineering
Sobotka & Co., Inc.
TRW, Inc.
Research Triangle Institute
U.S. EPA
Stone & Webster Engineering Corp.
Rexnord
Envirosphere Co.
Research Triangle Institute
U.S. EPA
Combustion Engineering, Inc.
University of California
Monsanto Research Corporation
Ministry of Health & Env. Protection
Dravo Engineers & Constructors
TRW, Inc.
-------
Mustl Lee A. c/o Boeing Co., Box 3766
Neufeld Ronald D. Dept. of Civil Engineering
Newell Gordon W. 2101 Constitution Avenue
Nichols Duane G. Research Division
Noichl 0. T. P. 0. Box 538
Notch Mark 9190 Red Branch Road
Offen George R. 1901 Fort Myer Drive, Suite 1012
O'Shea Thomas P. P. 0. Box 10412, 3412 Hillview Ave.
Page Gordon C. 8500 Shoal Creek Boulevard
Panzer Jerome P. 0. Box 51
Parkhurst Ben P. 0. Box 546
Patkar Avi N. 11499 Chester Road
Patterson L. W. P. 0. Box 2511
Patterson Ronald K. ESRL, MD-57
Petrie Thomas W. Dept. of Thermal & Env. Engineering
Phillips James H. 230 South Dearborn St.
Phillips Joseph W. River Oaks Building
Pittman Steve 701 North Park Street
Place Barry G. P.O. Box 512
Potter Victoria 1725 K Street, N. W.
Price Donald A. 1 Space Park
Raden Dennis P. 12700 Park Central Place, Suite 1400
Ramachandran G. 555 17th Street
Rapean John C. 3333 Highway 6 South
Reinert Bruce D. 805 Goethals Drive
Reveal William S. 3412 Hillview Avenue
Rhodes William J. IERL, HD-61
Rice Beth L. P. 0. Box 1368, Air Program
Rickert Loutillie P. 0. Box X
Riegel Kurt W. 401 M Street, S. W. (RD-681)
Riggin Ralph M. 505 King Avenue
Rosenfeld Michael 245 Summer Street
Rosenfield Henry P. 0. Box 2521
Ruckriegel Michael J. P. 0. Box 101
Ruggeri Sam P. 0. Box 487
Salemme Robert Box 43
Samela Daniel 1 Penn Plaza
Samematsu Henry P. 0. Box 111
Sauter, Jr. Lawrence J. 825 N. Capitol St., N. E., Rm. 7102
Schlosberg John P. 0. Box 2752
Seeliger Jochen Friedrichstrabe 1
Shaughnessy Mary E. 1000 Independence Avenue
Sherman Sandra Box 13015, Iron Works Pike
Sievers Henry E. 6330 Highway 290 East
Singer Philip C. Dept. of Environ. Sci. Eng.
Sizemore Freddie A. 1558 Washington St., E.
Skrinde Rolf T. 1319 Dexter N., (P.O. Box 9445)
Sliger Glenn 16200 Park Row, Industrial Park Ten
Smith N. Dean IERL, MD-61
Smith Tracey L. P. 0. Box 2180
Seattle
Pittsburgh
Washington
Library
A] lentown
Columbia
Rosslyn
Palo Alto
Austin
Linden
La ramie
Cincinnati
Houston
Research Triangle" Park
Carbondale
Chicago
Muscle Shoals
Herrin
Milwaukee
Washington
Redondo Beach
Dallas
Denver
Houston
Richland
Palo Alto
Research Triangle Park
Jefferson City
Oak Ridge
Washington
Columbus
Boston
Houston
Florham Park
Canton
Schenectady
New York
Los Angeles
Washington
Allentown
4300 Essen
Washington
Lexington
Austin
Chapel Hill
Charleston
Seattle
Houston
Research Triangle Park
Houston
WA
PA
DC
PA
PA
MD
VA
CA
TX
NJ
WY
OH
TX
NC
IL
IL
AL
IL
WI
DC
CA
TX
CO
TX
WA
CA
NC
MO
TN
DC
OH
MA
TX
NJ
OH
NY
NY
CA
DC
PA
F.R.
DC
KY
TX
NC
WV
WA
TX
NC
TX
98124
15261
20418
15129
18105
21045
22209
94303
78758
07036
82070
45246
77001
27711
62901
60604
35660
62948
53201
20006
90274
75251
80123
77079
99352
95070
27711
65102
37830
20460
43201
02107
77001
07932
44701
12301
10019
90051
20426
18001
.GERMANY
20585
40511
78723
27514
25311
98109
77084
27711
77001
Resources Conservation Co.
University of Pittsburgh
National Academy of Sciences
Conoco Coal Development Center
Air Products & Chemicals, Inc.
Hittman Associates
Acurex Corp.
Electric Power Research Institute
Radian Corporation
Exxon Research
Western Aquatics, Inc.
PEDCo Environmental
Tenneco Inc.
U.S. EPA
Southern Illinois University
U.S. EPA
Tennessee Valley Authority
Illinois South Project, Inc.
Allis-Chalmers
Sobotka & Co., Inc.
TRW, Inc.
Phillips Coal Company
ARCO Coal Co.
Shell Development
Hanford Env. Htth. Foundation
Electric Power Research Institute
U.S. EPA
MO Dept. of Natural Resources
Oak Ridge National Laboratory
U.S. EPA
Battelle
Stone & Webster Engineering Corp.
Texas Eastern Transmission Corp.
Exxon Research & Engineering Co.
American Elect. Power Serv. Corp.
General Electric Co.
Stone & Webster Engineering Corp.
Los Angeles Dept. of Water & Power
Federal Energy Regulatory Commission
International Coal Refining Company
Gesamtverband des deutschen
U.S. Department of Energy
Institute for Mining & Minerals Research
Texas Air Control Board
University of North Carolina
WV Air Pollution Control Commission
Olympic Associates Co.
Pullman Kellogg Research Center
U.S. EPA
Exxon Company
-------
O1
en
CO
OUSGPO:
»-•
ID
00
10
I
1
Cfl
Ul
(D
O
ID
to
Smitlison, .Jr.
Soukup
Stern
Stevens
Szluha
Taylor
Tell
Thocm
Thomas
Thomas
Trabert
Tranquill
Tyndall
Vlahakis
Vyas
Wadden
Wallace
Wanimcl
Weinreich
Weitzel
West
Wj ener
Williamson
Willson
Wipperman
Witmer
Yen
Yu
Yu
Zaklan
Zoueshtiagh
G. Ray
Charles L.
Jay L.
Nicholas J.
Adam T.
Don
Barbara Lorin
Terry L.
Susan J.
William C.
Angela L.
Paul C.
Frank
John
Kaushik
Richard
Anna W.
A. W.
Gary N.
Richard L.
Chris
Richard
Phebus C.
Warrack G.
Robert C.
Fred E.
Alan F.
Kar Y.
Wen C.
Gerri
Hahid
505 King Avenue
Military Road
P. 0. Box 2744 Terminal Annex
P. 0. Box 1500
500 Boyd Building, Wyoming Office
Suite 1400, Park Central III
29623 W. Western Highway
Suite 900, I860 Lincoln St.
4 Research Place
8500 Shoal Creek Boulevard
816-16th Street
Midland Building
500 Eastowne Drive
12th and Pennsylvania Ave., N. W.
P. 0. Box 880
P. 0. Box 6998, School/Public Health
P. 0. Box 12194
P. 0. Box 509
One Woodward Avenue
509 Estes Avenue
MD-51
One Penn Plaza
Ammonia from Coal Projects
P. 0. Box 8213, University Station
100 Federal S.
Mailstop E-201, Env. Safety & Eng.
P. 0. Box 2752
One Space Park Drive, R4/1136
7929 Westpark Drive
1313 Dolley Madison Boulevard
50 Beale Street
Columbus
Rothschild
Los Angeles
Somerville
Cheyenne
Dallas
Southfield
Denver
Rockvilie
Austin
Wilmette
Cleveland
Chapel HilJ
Washington
Morgantown
Chicago
Research Triangle
Beacon
Detroit
Schaumburg
Research Triangle
New York
Muscle Shoals
Grand Forks
Boston
Washington
Allentown
Redondo Beach
McLean
McLean
San Francisco
OH
Wl
CA
NJ
WY
TX
MI
CO
MD
TX
IL
OH
NC
DC
WV
IL
Park NC
NY
HI
IL
Park NC
NY
AL
ND
MA
DC
PA
CA
VA
VA
CA
43201 Battelle Columbus Laboratories
54474 Zimpro Inc.
90051 Joy Mfg. Co.
08876 Research-Cottrell
82002 Industrial Siting Administration
75251 Phillips Coal Co.
48034 Swanson Environmental, Inc.
80295 U.S. EPA, Region VIII
20850 NUS Corporation
78758 Radian Corporation
60091
44115 The Standard Oil Co.
27514 GCA Corporation
20461 Department of Energy
26505 EG&G
60680 University of Illinois
27709 Research Triangle Institute
12508 Texaco Inc.
48226 American Natural Service Co.
60193 Ecological Analysts, Inc.
27711 U.S. EPA
10001 Envirotech/Chemico
35660 Tennessee Valley Authority
58202 Grand Forks Energy Technology Ceiiter
02110 Herrick & Smith
20545 Department of Energy
18001 International Coal Refining Company
90278 TRW, Inc.
22102 UOr/SDC
22101 Dynalectron
94119 Bechtel
------- |