f/EPA
,
EPA t-
Januai
Fuel Contaminants:
Volume 3.
Control of Coal-related
Pollutants
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development. U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of. and development of. control technologies for energy
systems; and integrated assessments of a wide-range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service. Springfield, Virginia 22161.
-------
EPA-600/7-79-025a
January 1979
Fuel Contaminants:
Volume 3.
Control of Coal-related Pollutants
by
E.J. Mezey, Seongwoo Min, B.R. Allen,
W.C. Baytos, and Surjit Singh
Battelle-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
Contract No. 68-02-2112
Program Element No. EHE623
EPA Project Officer: Lewis D. Tamny
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental
Research Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily reflect
the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendations for use.
ii
-------
ABSTRACT
Volume III of the series of reports on fuel contaminants is a
continuation of the removal technology evaluation studies reported in
Volume II. Specifically the objective of the study was to identify possible
future control strategies for removal of pollutants in coal and coal-derived
liquids based on the findings from reviews on contaminant chemistry (Volume
I) and the evaluation of removal technology used for various solid and liquid
fuels. Alternative approaches to isolating or removing contaminants were
also identified as were new approaches with potential for contaminant removal.
Rationale used to select and tentatively rank the various approaches are
described. Of the approaches considered, six were selected for preliminary
assessment by experimentation. These are:
• Biological action on coal-derived liquids—Although
bacteria can tolerate low levels of aromatic hydrocarbons,
the addition of aromatic sulfur compounds increases the
toxicity of these solvents to bacteria. Hence the pros-
pects of using non-mutated bacteria for sulfur and/or
nitrogen removal from coal liquids appears small.
• Enhancement of pyrite removal during immiscible fluid
agglomeration—Pyrite removal equivalent to that obtained
for float-sink analysis was obtained by pretreatment and
oil agglomeration. The same technique was found effec-
tive for the recovery of over 90 percent of coal from
coal cleaning plant fines.
• Extraction of clean fuels from coal liquids—Light hydro-
carbons can be used to extract 83 percent of coal liquid
iii
-------
at super critical conditions to yield a low sulfur and
nitrogen fuel.
• Concentration of organic sulfur and nitrogen and ash
from coal liquids—Up to 76 percent of the sulfur and
about 10 percent of the nitrogen can be removed by
passing coal liquids over various special porous media.
• Conversion of coal liquefaction residues to environmentally
acceptable fuels—Treatment of coal liquefaction residue
with hydrogen-carbon monoxide mixtures reduced nitrogen
content by as much as 14 percent.
• Improvements in pyrite liberation from coal—This part of
the study was not undertaken because of significant
advances by others
This report was submitted in fulfillment of Contract No. 68-02-2112
by Battelle's Columbus Laboratories under the sponsorship of the U.S.
Environmental Protection Agency. This report covers the period June 15, 1975
to March 31, 1978.
iv
-------
CONTENTS
Abstract iii
Figures vii
Tables ix
1. INTRODUCTION 1
Technical Objectives 2
2. CONCLUSIONS 3
3. RECOMMENDATIONS 5
4. BACKGROUND 6
Characteristics of Fuel Contaminants 6
Distribution of Contaminants in Fuels 9
Summary of Techniques Used for Removal of Contaminants
from Coal 13
Summary of Techniques Used for Removal of Contaminants
from Coal-Derived Liquids 17
Interrelational Aspects of Contaminant Removal 18
5. SELECTION OF REMOVAL TECHNOLOGIES TO BE CONSIDERED
FOR FUTURE STUDY 20
Selection Criteria Development 20
Results of Screening 22
Selection of Removal Technologies for Preliminary
Assessment 28
6. BIOLOGICAL ACTION ON COAL-DERIVED LIQUIDS 29
Literature Survey 29
Experimental Studies 48
Experimental Results 54
Conclusions 59
-------
CONTENTS - (Continued)
7. USE OF AGGLOMERATION TO RECOVER CLEAN COAL VALUES FROM
PREPARATION PLANT WASTES AND THE ENHANCEMENT OF
PYRITE REMOVAL 61
Introduction 61
Conclusions 62
8. EXTRACTION OF CLEAN FUELS FROM COAL-DERIVED LIQUIDS .... 65
Introduction 65
Background 67
Experimental Program 76
Optimum Extractions of Clean Fuel from a Coal Liquid ... 88
Process and Economic Considerations 90
Description of Battelle Extraction Process 93
Comparison of Battelle Extraction Process to "Conventional"
Solvent Refined Coal/Liquefaction Process 94
Conclusions 95
9. USE OF POROUS MEDIA TO CONCENTRATE ORGANIC SULFUR, NITROGEN
AND MINERAL MATTER CONTAMINANTS IN COAL DERIVED LIQUIDS . . 97
Introduction 97
Experimental Program 98
Discussion 102
Conclusions 105
10. CONVERSION OF COAL LIQUEFACTION RESIDUE TO ENVIRONMENTALLY
ACCEPTABLE FUEL 106
Introduction 106
Experimental Program 107
Discussion 112
Conclusions 112
11. IMPROVEMENTS IN PYRITE LIBERATION FROM COAL 113
Discussion 113
12. REFERENCES 115
APPENDIX - THEORY OF EXTRACTION WITH COMPRESSED GASES A-l
vi
-------
FIGURES
Number Page
1 Relationship of Suggested Contaminant Removal Methods
Within a Coal-Based Complex 23
2 Structure of Bituminous Coal Postulating Sulfur and
Nitrogen Occurrence in Coal 31
3 The Aerobic and Anaerobic Environment in the Earth's
Carbon Cycle 37
4 The Effect of Surface Area on Bacterial Oxidation 40
5 Catechol and Protocatechuic Acid as Key Intermediates in
the Oxidation of Benzene Derivatives 43
6 Solubility of High Molecular Weight Hydrocarbons in Ethylene
Gas as a Function of Temperature and Pressure 74
7 Effect of Critical Temperature of Gases on Extraction of a Coal
Tar Component (Phenanthrene) at 40 atm and 40 C (313 K) . . . . 75
8 Experimental Arrangement for Solvent Extraction at Elevated
Temperatures and Pressures 79
9 Extraction of SRC Process Dissolver Product by Gases as
a Function of Temperature (3400 psig) 81
10 Extraction of SRC Process Dissolver Product by Gases as
a Function of Temperature (Pressure = 1500 psig) 82
11 Extraction of SRC Process Dissolver Product by Butane
and Ethylene, at 300 C, as a Function of Pressure 83
12 Carbon in Coal Used for Fuel Oil Production (by Hydrogenation)
and the Requirement for Carbon in Same Coal for Hydrogen
Production 89
13 "Conventional" Solvent Refined Coal Liquefaction Process for
Boiler Fuel Production 91
14 Proposed Battelle Extraction Process
vii
92
-------
FIGURES - (Continued)
Number Page
15 Equipment Arrangement Used for Porous Material Adsorption
Studies 99
16 Arrangement of Equipment Used in Coal Liquefaction
Residue Treatment with Gases 109
A-l Vapor-Phase Solubility of Naphthalene in Ethylene,
Calculated and Experimental Values at 35 C A-4
viii
-------
TABLES
Number Page
1 Elemental Analysis of Typical Fuels 7
2 Distribution of S and N Contaminants in Fuels 10
3 Trace Elements Associated with Mineral Matter in Solid
and Liquid Fuels 12
4 Trace Elements Associated with Organic Matter of Solid
and Liquid Fuels 14
5 Areas Selected for Further Study 24
6 Nutrients for Aerobic and Anaerobic Bacteria 33
7 Some Natural Bacteria Useful to Form Mutants 35
8 Microbial Species and Biodegradation of Hydrocarbons
Found in Coal/Coal Liquids 42
9 Types of Microorganisms Found in Peat as Determined by
Differential Media Counts 46
10 Aerobic Sulfur-Utilizing Bacteria 50
11 Anaerobic Sulfur-Utilizing Bacteria 50
12 Culture Media 51
13 Summary of Sulfur-Utilizing Thiobaccilus Growth Experiments,
with Aeration and Shaking 55
14 Results with Unidentified Mixed Culture-Experiments in
Defined Mineral Media Plus Solvent Only (7 Days) 57
15 Results with Unidentified Mixed Culture-Growth Experiment in
Defined Mineral Media Containing Solvent and Either
Thiophene or 2-Methyl Thiophene (7 Days) 58
16 Results with Aerobic Mixed Culture-Growth Experiments in
Defined Mineral Media, with Thiophene in Toluene or
Benzene as Sole Sulfur Source (9 Day) 60
ix
-------
TABLES - (Continued)
Number Page
17 Constituents of Solvent Refined Coal 66
18 Solvent Types for Selective Solubilization of Coal 68
19 Filtration Solids from Solvent Fractionated Coal Liquids ... 70
20 Typical Data on Solvent Deasphalting of Residuum 72
21 Feed Material for Extraction of Coal Liquids 77
22 Summary of Reaction Conditions, Percent Extracted, and
Product Recovery for Gas Extractions 84
23 Analysis of Liquid Product and Residue After Extraction of
Coal Liquid 86
24 Sulfur and Nitrogen Balance for Selected Extraction
Experiments 87
25 Materials Used as Porous Media 100
26 Results of Experiments on the Use of Porous Media to Remove
Contaminants from Coal Liquids (Feed Rate 10 ml/min) .... 103
27 Analyses of Coal Liquefaction Residues, Percent 108
28 Effect of Treatment of Coal Liquefaction Residues with
Various Gases on Their Sulfur and Nitrogen Content Ill
A-l Computed Values of -0 A-3
-------
SECTION 1
INTRODUCTION
Major air pollutants have made their presence felt for many years
and were the obvious first targets when the Air Quality Act of 1970 was
enacted. Thus, a concerted effort is under way with Federal government
support to develop practical and effective methods for their control --
especially the elimination of sulfur emissions from the combustion of fuels.
Analogous major efforts are also under way on other visible or readily iden-
tifiable pollutants such as particulates, oxides of nitrogen, etc. The
impact of control of air pollutants on the pollution of water and land became
of equal concern and subsequent targets for control.
In-accordance with the responsibilities mandated by the Congress,
the Environmental Protection Agency is developing information on the control
of pollutants emitted to the air, water and land from stationary sources.
This includes not only the major pollutants mentioned above, but also other
pollutants such as trace metals which, although emitted only a very low con-
centrations, also pose serious problems because of their toxicity.
Treating fuel to remove contaminants prior to combustion is an
attractive technique because of the possibility of removing several contami-
nants in one processing step. The combustion of a clean fuel would obviate
the need for stack-gas treatment methods. It now appears that certain
pollutants, especially S02 and particulates, can be removed by an integrated
stack-gas treatment process, but the removal of other pollutants such as NOX
and volatile trace metals would probably be difficult to integrate into a
stack-gas treatment process. While some NOX is formed from the nitrogen in
the combustion air, a considerable portion is also formed from fuel nitrogen.
Accordingly, the primary objective of the current research program is to
identify and examine the feasibility of fuel contaminant removal prior to
combustion, with emphasis on sulfur and nitrogen. However consideration was
-------
given to trace metals, halides, phosphorus, and other potential air
pollutants.
The program has provided a survey of the literature on the
characteristics of the sulfur, nitrogen, and trace element contaminants in
coal, petroleum, shale oil, and tar sand oil, and has classified and
analyzed methods employed to remove the contaminants from these fuel
Q 2)
types. ' ' During this study, removal methods were identified that had not
been used for various reasons but that may be used today. In addition,
alternative approaches for isolating or removing contaminants were also iden-
tified, as were new approaches with potential for contaminant removal. As
part of the study, it was desirable to verify existing data or obtain new
data to sup'port the basis for possibly new and novel removal techniques of
the more promising approaches for contaminant removal. This report describes
the rationale involved in selecting such approaches based on the review find-
ings. Results of some preliminary assessments of the techniques selected for
study in the laboratory conclude the first and second phase effort on the
Fuel Contaminant study.
TECHNICAL OBJECTIVES
The objectives of this study, which is a continuation of the
(1 2)
removal technology evaluation studies ' are twofold: (1) to identify
possible future environmental control techniques for potential pollutants in
coal and coal liquids and (2) to make preliminary evaluation of some selected
contaminant removal techniques. This study was done to determine the feasi-
bility of using these techniques to produce a fuel low in environmental
pollutants prior to combustion of the coal or coal-derived liquids.
-------
SECTION 2
CONCLUSIONS
Some of the contaminant removal methods identified during the
review of the chemistry and removal technology of fuel contaminants are not
in use today for various reasons but may be used in the future. Alternative
approaches to contaminant removal were identified as were new approaches
with a potential for contaminant removal. Of the various approaches, ranked
according to rationale developed in the study, six were selected for pre-
liminary assessment by experimentation. The conclusions from each of the
areas studied and reported in the following sections are summarized below.
• Biological action on coal-derived liquid.
Both aerobic and anaerobic bacteria can tolerate low levels (0.5 percent
or less) of aromatic hydrocarbons but higher levels completely inhibit
their growth. The addition of aromatic sulfur compounds to the aromatic
hydrocarbons increases their toxicity to the bacteria. The prospects of
using non-mutated bacteria to remove organic sulfur and nitrogen com-
pounds from coal derived liquids appears small.
• Enhancement of pyrlte removal during immiscible fluid agglomeration.
Pyrite removal equivalent to that obtained by float-sink analysis
(42 percent removal) was obtained by chemical pretreatment of fine coal
before oil agglomeration. The oil agglomeration technique was also found
to be effective for the removal of 90 percent or more of the coal value
in sediments from slurry ponds and black water.
The process has potential for the control or treatment of effluents
from coal cleaning plants and the removal of hazards associated with
slurry ponds.
• Extraction of clean fuels from coal liquids .
Light hydrocarbons (C- to C,) can extract up to 83 percent of coal liquid
3
-------
at super critical conditions for these hydrocarbons. The extract
isolated is very low in sulfur and nitrogen and there is good indication
of the concentration of the sulfur and the nitrogen in the residue
fraction along with the ash. This was accomplished without any evidence
for thermal degradation of the feed material.
• Concentration of organic sulfur and nitrogen and ash forming minerals
from coal liquids.
Porous materials with low surface area but large pore volume have been
able to remove up to 76 percent of the sulfur (ash free) from coal
liquids when passed through a fixed bed of the material. A 10 percent
removal of nitrogen was also obtained. Both nitrogen and sulfur
removals appear to occur by separate mechanisms and both seem to be
independent of ash removal.
• Conversion of coal liquefaction residues to environmentally acceptable
fuels.
Treatment of coal liquefaction residues or still bottoms with Hj-CO mix-
ture removed as much as 14 percent of the nitrogen originally present.
Sulfur removal with hydrogen and ammonia was only minor. More severe
conditions than those used in this screening study should increase sulfur
removal with ammonia and increase the reactivity of the H^-CO mixture.
• Improvement in pyrite liberature from coal.
This part of the study was not undertaken because of significant advances
made by others.
-------
SECTION 3
RECOMMENDATIONS
Based on results of this study, it is recommended that the oil
agglomeration technique be further developed as a method to control effluents
from coal cleaning plants in order to reduce the hazards and environmental
impact of increased quantities of wastes that would result as the Nation
shifts its energy dependence to coal and strives for the production of clean
fuels in an environmentally sound manner. Further work is needed to demon-
strate the applicability of the process to various origins of coal wastes
and the effluents from coal cleaning plants with varying degrees of com-
plexity. The technique should also be extended to the reduction of mineral
matter being fed to coal liquefaction plants by pretreating the coal feed.
The development of a bench scale process suitable for the treatment of both
sediments and black water effluent should also be undertaken.
Also recommended is further study into the use of light hydro-
carbons as solvents for the extraction of clean fuels from coal liquids of
various origins, i.e., catalytic and noncatalytic processes without the need
for filtration of distillation.
Finally it is recommended that the concept of the use of porous
media for the physical removal of low levels of sulfur and nitrogen be
expanded to greater range of materials and for the treatment of coal liquids
of different origin.
-------
SECTION 4
BACKGROUND
To date the study has reviewed the information on the methods
employed to remove the types of contaminants identified from solid and liquid
fuels. It has categorized and analyzed these methods. Also, known but
unusual contaminant removal methods were identified for possible use in
today's world. In addition, new approaches to the liberation, isolation, or
removal of contaminants were identified. Criteria had to be developed to aid
in selecting approaches or techniques suitable for further study and tech-
niques for immediate evaluation as part of this task. Once techniques
were selected, laboratory support studies were needed to verify existing
data or obtain new data to support the basis for the potential removal
techniques.
To add continuity to the reports, the background information
presented here summarizes the findings from literature reviews on contaminant
characterization and contaminant removal techniques. This background infor-
mation was used to develop the selection criteria for possible novel
approaches to contaminant removal. Although petroleum, shale oils, and tar
sand oils were included in the first part of the study, this work emphasizes
the contaminants in coal and coal-derived liquids.
CHARACTERISTICS OF FUEL CONTAMINANTS
Understanding the characteristics and differences of coal, coal
liquids, petroleum, shale oil and tar sand oil, aid in characterization of
the sources of sulfur, nitrogen, and trace element contaminants in these
fuels. As given in Table 1, fuel types typical of those considered in this
review exhibit marked differences in the amount of ash-forming matter they
contain and in the elemental composition of their cmmbustible part. Coals
have a much higher ash content than do petroleum crudes. Coals also have
the lowest value for the H/C ratio and are considered in a state of
unsaturation with respect to hydrogen. For coal to approach petroleum crude
6
-------
TABLE 1. ELEMENTAL ANALYSIS OF TYPICAL FUELS
c
Weight Percent „ ,„
H/C
HO N S Ash (Atomic)
Coal (moisture-free)
Subbituainous 69.2
(Big Horn)
Bituminous 78.7
(Pittsburgh)
4.7 17.8 1.2 0.7 6.5
5.0 6.3 1.6 1.7 6.9
Coal Liquids
(Big Horn)
(Pittsburgh)
Shale Oil
Utah Asphalt
Petroleum Crude
(Pennsy Ivania)
89.2
89.1
80.3
82
85
8.9
8.2
10.4
11
14
1.03
1.5
5.9
3
1
0.4
0.8
2.3
2
1
0.04 >1
0.2 >1
1.1 %1
2 <1
1 <1
0.81
0.76
1.20
1.10
1.55
1.61
1.98
in character, hydrogen must be added to coals or coal liquids. The H/C
ratios of shale oil and tar sand oil fall between those for coal and
petroleum. Formation of coal liquids by hydrogenation of coal increases the
H/C ratio. Such processes also significantly reduce the amount of sulfur
and nitrogen remaining in the fuel.
The review of the characteristics of the sources of sulfur,
nitrogen, and trace element pollutants in coal, petroleum, tar sand oil, and
shale oil suggests that at least part of the sulfur and most of the nitrogen
originate from compounds common to all of these fuels. These compounds
are primarily organic sulfur and nitrogen compounds. The principal differ-
ence between the characteristics of the solid and liquid fuels is the way
in which these compounds are combined in the fuel. The noncombustible
matter associated with coal distinguishes it from the liquid fuels and
-------
is the source of inorganic sulfur and most of the trace elements. In liquid
fuels, the trace elements are present primarily as oil soluble compounds or
associated with the colloidal suspensions commonly present in them.
Contaminants in Solid Fuels
In coal, the nitrogen contaminants are present primarily as
organic compounds which are part of the three-dimensional carbon skeletal
structure that makes up the organic matter of coal. Indirect evidence
suggests that the nitrogen compounds exist as both alkyl and aryl
derivatives of amines, quinolines, and carbazoles. The sulfur in coal is
present both as organic and inorganic compounds. The organic sulfur
compounds, also part of the carbon skeleton of coal, consist of both
alkyl and aryl derivatives of thiols, thioether, disulfides, and thiophene.
The major sources of inorganic sulfur in coal are the pyrites and the
sulfate salts. Pyrites and other mineral matter exist in raw coal as dis-
crete phases ranging in size from gross inclusions in the coal seam to
micron-size particles disseminated throughout the organic matter of coal.
Trace elements that exist in raw coal can be categorized as those found
principally with the organic matter of coal and those present principally
in the mineral matter associated with coal. However, some trace elements
exhibit a dual role and are found in both components. Trace metals and
metalloids exist as complexes of the organic oxygen, nitrogen, and sulfur
compounds or as crystal-lattice impurities in the mineral matter associated
with coal seams. Nonmetallic trace elements are found primarily as part
of the inorganic mineral impurities. Some, such as phosphorus, are found
also in the organic part of coal.
It can be concluded that nonuniform distribution of impurities
exists not only from one coal seam to another but also within the coal seam.
Despite this limitation, intrinsic centers of sulfur and nitrogen contamina-
tion have been identified. The organic sulfur and nitrogen compounds are
uniformly distributed in the organic part of the coal, as are the trace
elements associated with it. Inorganic sulfur and the trace elements associ-
ated with the mineral matter in the coal vary with the source of coal.
8
-------
However, with the use of modern mining and coal-preparation techniques, a
more uniform product is available for use.
Contaminants in Liquid Fuels
In petroleum and tar sand oil, and to some extent in shale oil,
most organic sulfur and nitrogen compounds found in the asphaltene and resin
fractions are similar to those found in coal. In petroleum nearly 50 percent
of these fractions consist of molecules in the 4,000 to 8,000 molecular-
weight range. Between 10 and 50 percent of the nitrogen and 3 and 35 percent
of the sulfur in petroleum are found in these fractions. Although coal
liquids are known to contain resin and asphaltene fractions, the liquefaction
process by the addition of hydrogen to coal also reduces the sulfur content
(and at times the nitrogen content) to levels near 1 percent. The contami-
nants in coal liquids remaining after removal of the solids are present as
organic sulfur and nitrogen compounds that are difficult to remove and are
usually found in the higher molecular-weight asphaltene and resin fractions of
coal liquids.
It can be concluded that in the liquid fuels, the intrinsic centers
of sulfur and nitrogen contamination are found in the colloidal suspensions
(otherwise known as asphaltenes) and in the more soluble resins, both of
which make up anywhere from 1 to 43 percent (but more commonly less than
6 percent) of the petroleum, about 37 percent of tar sand oils that have
been characterized, 13 to 71 percent of the shale oil, and 10 to 30 percent
of the coal liquids. However, the molecular weights of the asphaltenes and
resins in petroleum crudes are greater than those in synthetic crudes.
DISTRIBUTION OF CONTAMINANTS IN FUELS
Sulfur and Nitrogen Contaminants
The distribution of sulfur and nitrogen contaminants in the fuel
types is shown in Table 2 . Large variations in the total sulfur and
nitrogen content exist between the fuels and within a fuel type. Coal is
unique with respect to the other fuels in that sulfur is also present in
the form of pyrites. The organic sources of sulfur in coal have been
determined indirectly primarily from studies on depolymerized coal. The
exact nature of organic sulfur in solid coal can only be surmised from
-------
TABLE 2. DISTRIBUTION OF S AND N CONTAMINANTS IN FUELS
Fuel
Contaminants
Parent
Type and Source Structure
Sulfur. Total
Inorganic
Pyrites
Organic
Mercaptans
Sulfldes
Thlophenes
Benzothlophenes
nitrogen. Total
Basic
Pyrldlnes
Qulnolinea
Acridities
Nonbaslc
Pyrroles
Indoles
Carbazoles
Benzamldes
(a) Colorado shale oil.
(b) Tar sands Including
(c) 48 percent of total
FeS2
R-SH(C>
R-S-R(e>
Q
CgQ
o
CO
ceo
w
cu
050
Q-COMHx
Coal
0.4-13X
x(c.d)
x(f)
x(8)
X
1-2.11
X
x(f)
x(f)
x(f)
x(f)
Coal
Liquids Shale.
Primary Oilw
<1% 1.1%
X X
X X
>1% 2.31
X X
X X
X X
X X
X
Tar
Sand.. Petroleum
01 1W Crude
0.2-6.3% 0.1-51
X
X
X
X
-------
these studies. There is strong evidence that mercaptans, sulfides, and
disulfides are present In the coal structure. The same type of compounds
are usually present In tar sand oils and petroleum but not in coal liquids
and shale oils. Thiophenes and benzothiophenes are known to be present in
coal liquids and shale oil as well as in tar sand oil and petroleum.
Total fuel nitrogen content is highest in coal and shale oil.
Little is known for certain about the types of nitrogen compounds contained
in coal since their characterization was determined from studies on coal tars
or depolymerized coal. There is good evidence that pyridine, quinoline,
acridine, indole, carbazole, and porphyrlns or their derivatives account for
the nitrogen in solid coal. Thus, it may be concluded that the nitrogen in
coal is present as aromatic compounds. Quinolines, pyrroles, indoles, carba-
zoles, acridines, and porphyrins are present in coal liquids, shale oil, tar
sand oil, and petroleum. Benzamides have been found in shale oil and petro-
leum. The porphyrin compounds may contain trace metals as coordination
compounds.
Values for the amount of each type of organic sulfur and nitrogen
compound are available for only a limited number of fuels and then for only
a few sources of the fuel type (e.g., petroleum). Others have not yet been
characterized or are being characterized (e.g., coal liquids, shale oils,
and tar sand oils). Until additional data are available, a quantitative
comparison would be difficult. But it has been determined that the total
nitrogen content of U.S. coals (moisture- and ash-free (maf)) varies In the
range of 1 to 1.9 weight percent for lignite and high volatile bituminous
coal, respectively. This amount of nitrogen is present as an integral part
of the chemical structure of coal.
Trace Elements
The trace elements in the fuels covered in this review can be
categorized as either being in fuel's organic matter or the mineral
matter. As shown in Table 3, coal has most of the trace elements closely
associated with these minerals. Petroleum and tar sand oils by
comparison contain much less mineral matter and hence less trace
elements are associated with them (e.g., those elements commonly associated
11
-------
TABLE 3. TRACE ELEMENTS ASSOCIATED WITH MINERAL MATTER
IN SOLID AND LIQUID FUELS
Mineral (percent of total)
Trace Elements
Coal
Shales, kaolins (52)
Sulfides, pyrites (25)
Carbonates (9)
Chlorides (0.1)
Quartz (1-20)
Apatite
Sulfates
Petroleum (tar sand oils)
Chloride (trace)
Quartz (trace)
Shales, kaolins (trace)
Shale Oils
Shale minerals
Coal Liquids (speculation)
Removed with mineral
matter
Si, Al, Ti, K (B, Cu, F, Hg, Sn, V)
(a)
Fe, Zn, Cd, Pb, Ni, Cu (Hg, B, As, Be,
Sb, Ge)
Ca, Mg, Mn (B, Cd, Mo, Se, V)(a)
Na, K, Cl (Br, I, Mg, Ca)(a)
Si (B, Cr, Mn, Cd, Mo, Ge, Se, V, Zn)
Ca, P, F (As, V, Cl, Mn, Ce)^
Fe, Ca (Mn)(a)
(a)
(a)
Na, K, Mg, Ca, Cl, Br, I (Mn)
Si
Si, Al, K
As, Ba, B, Cu, Cr, Pb, Li, Mo, Mn, P,
Se, Sr, Tl, Ti, V, Zn, Ag, Au
Eg, Zn, Zr, Cd, As, Pb, Mo, Mn (Co,
Ni, Cr, Se, Cu, Sb, V, Ti, Ga, P)(b)
(a) Elements in parentheses are known to have high correlation with
minerals but not necessarily detected with minerals.
(b) The elements in parentheses are more commonly found with mineral
matter but are also found in organic portion of coal.
12
-------
with brines and suspended sands and clays). Shale oils contain finely
divided shale minerals carried over during retorting and contain trace
elements found in the shale minerals. Upon coal liquefaction, the trace
elements associated with the mineral matter in coal remain insoluble and are
removed along with the mineral matter.
The trace elements more closely associated with the organic
matter in coal are shown in Table 4 . These trace elements in coal are
held in the organic matrix as complexes of organic compounds (containing
oxygen, sulfur, or nitrogen), porphyrins or amino acids, or as alkyl or
aryl derivatives. Most of the trace metals found in petroleum and tar
sand oils are found in the organic phase as oil soluble compounds. They
exist primarily as complexes, organic acid salts, and alkyl and aryl
derivatives. Shale oil also contains a large number of oil-soluble trace
elements which may have formed during retorting. Coal liquids would be
expected to contain in solution those elements most closely associated with
the organic part of coal or those that enter into solution during the
liquefaction process.
SUMMARY OF TECHNIQUES USED FOR REMOVAL
OF CONTAMINANTS FROM COAL
Raw coal, which contains large amounts of undesirable mineral
matter, undergoes considerable upgrading in modern coal-preparation opera-
tions."' Significant amounts of sulfur present as gross pyrite inclusion
and other ash mineral bodies present in the mined coal are readily removed
during coal-washing operations. When such processes are used, 15 to 30
percent of the pyritic sulfur in the run-of-the-mine coal is removed from
coal crushed to a top size of 1/4 inch (6 mm). Part of the finely dissemi-
nated pyrite can be removed by physical means only if the size of the coal
is further reduced. In separations using the dense media cyclone, a bottom
size of 32 mesh (0.5 mm) can be treated to remove up to 30 percent of the
pyrite. In froth flotation, pyrite removal can reach about 50 percent when
the coal is crushed to minus 28 mesh (0.6 mm). Staged froth flotation
employing pyrite depressants in the second stage reduces the pyritic sulfur
content of coals from 50 to 80 percent. Specialized methods for pyrite
13
-------
TABLE 4. TRACE ELEMENTS ASSOCIATED WITH ORGANIC MATTER
OF SOLID AND LIQUID FUELS
Derivative Trace Elements
Coal
Complexes of 0, N, S ligands Ge, Be, B, Ti, U (Ga, Y. La, Ni, Co,
and organic acid salts Zn, V, Ca, Al, Si, P) (a)
Porphyrin, amino acids V, Ni, Cu, Fe
Alkyl or aryl Ge (P, Se)(a)
Petroleum (tar sand oils)
Porphyrin Ni, V
Nonporphyrin Ni, V, Fe, Co, Zn, Hg, Cr, Cu
Tetradentate complexes V, Ni, Fe, Cu, Co, Cr (Cl)
(S, N, 0)
Alkyl or aryl Hg, Sb, As
Organic acid salts Na, As, Hg, Fe, Sb (Mo)^
Unknown Se, I
Shale Oils
Oil soluble As, Sb, Be, B, Cd, Ca, Cr, Co, Cu,
F, Ge, Pb, Mg, Mn, Hg, Mn, Mo,
K, Se, Na, Sr, Te, Ti, V, Y, Zn
Coal Liquids (speculation)
Liquid soluble (?) Ge, Ba, Ga, B, Ti (P, U, V, Sb,
>50 percent organic affinity Co, Ni, Cr, Se, Cu)
(a) Elements forming compounds to a lesser extent are in parentheses.
(b) Cl~ suggested as counter ion.
(c) Elements in parentheses are more closely associated with organic
part of coal but are also present in mineral matter.
14
-------
removal from coal that has been reduced in size to minus 200 mesh (0.074 mm)
and even minus 325 mesh (0.044 mm) have met varying success. Typically, 40
to 50 percent of the pyritic sulfur can be removed. In one case in which an
oil agglomeration technique was used, the amount of pyrite removal reached
90 percent. However, the minus 325 mesh (0.044 mm) coal was specially pre-
conditioned.
In chemical refining, essentially all of the pyritic sulfur in
coal is reported to be removed by treatment with aqueous solutions of sodium
hydroxide or ferric sulfate. Both processes require elevated temperature.
With sodium hydroxide, partial removal of organic sulfur occurs for selected
coals, while ferric sulfate treatment does not attack it. In both these
processes pyrite is more efficiently removed when finer sized coal is
treated.
Liquefaction or depolymerization of coal to produce a cleaner
solid fuel, as in the case of solvent-refined coal (SRC), is an alternative
to the extensive size reduction of coal needed to gain access to the finely
disseminated pyrite and mineral matter. During such a liquefaction,
noncatalytic hydrogenation of coal occurs mostly from the hydrogen-donor
type solvent that is mixed with the coal. After the liquefaction, the
mineral matter and the finely disseminated pyrite (now reduced to pyrrho-
tite or ferrous sulfide) originally in coal are released. They and the
unrcacted coal are removed prior to utilization. Typically, when all of
the ash minerals are removed, the sulfur is lowered to values equal to
or less than that attributable to organic sulfur in coal. Nitrogen values
are usually not lowered in such a process. Most of the cyclic and hetero-
eyclic organic sulfur and organic nitrogen originally present in the coal
remain as such in the liquefaction product (SRC). Further removal of this
sulfur and nitrogen must be by catalytic hydrotreatment of the SRC to release
most of the sulfur as f^S and part of the nitrogen as NH-j.
Desulfurization and denitrification of coal by carbonization or
pyrolysis are only partially effective since, during the processing, the
sulfur and nitrogen not removed overhead remain in the coke or char in
a form that is bound deeply in graphitic-type structures. Processes
employing reactive gases, alkalies, salts, and acids during carbonization
or pyrolysis are capable of increasing the amount of sulfur and nitrogen
15
-------
removed, but complete removal has not been attained. Unless the coal used
in these processes is low in ash and pyrite by virtue of their origin or
preparation, most of these components will remain in the coke or char.
The gasification of the carbon value in coal releases the sulfur
and nitrogen contaminants bound in the coal structure as well as those con-
taminants present as discrete phases. However, before the low-Btu gas can
be utilized, these released gaseous contaminants and the particulates must
be removed downstream from the gasifier. Although such an approach would
appear to be an effective way to remove the contaminants from coal, the
solid fuel is usually converted in the process to a low-grade gaseous fuel.
It may be concluded from these facts on the removal of contami-
nants from coal that:
• Release from coal of the finely disseminated pyrite
requires extensive size reduction of the coal to
enable even partial removal of the pyritic sulfur.
This is true whether the pyrite-removal method is
based on chemical refining or on differences in
specific gravity, surfacial behavior, or magnetic
properties.
• Only about one-half of the sulfur originally present
in coal as pyrite is removed during liquefaction or
depolymerization of coal by noncatalytic hydrogen-
ation. The pyrrhotite or FeS must be removed before
the product can be utilized as a low-sulfur fuel.
• The sulfur and nitrogen present as cyclic and hetero-
cyclic organic sulfur and organic nitrogen are relatively
unaltered by either the physical methods or chemical
refining and only a small amount of the organic sulfur
is released during the noncatalytic liquefaction process.
• Carbonization of coal is only partially effective for the
removal of the sulfur and nitrogen contaminants. Those
that remain in the product are tied up in the char structure.
• Gasification of coal releases all of the contaminants
contained in coal, but extensive posttreatment of the
16
-------
gasified coal Co remove gaseous and particulate pollutants
is required before the gas can be utilized.
SUMMARY OF TECHNIQUES USED FOR REMOVAL
OF CONTAMINANTS FROM COAL-DERIVED LIQUIDS
I
Coal liquids formed during the initial stages of the hydrogen-
ation of coal contain the noncombustible portion and unreacted coal as
slurry materials; and they also contain most of the contaminants originally
present in the coal.^ ' At this point in processing, several alternative
approaches for the removal of contaminants become available. As one alter-
native, the removal of the mineral matter, unreacted coal, and iron sulfide
as in the solvent-refined coal (SRC), will provide a product fuel which
is reduced in ash and total sulfur and is a solid at ambient temperatures.
This same product can be used as a feedstock for a catalytic hydrotreatment
process. Another alternative is to catalytically hydrogenate a coal-oil
slurry to produce a liquid fuel (liquid at ambient temperatures) and
then remove the suspended solids. During the catalytic hydrogenation, much
of the organic sulfur and part of the organic nitrogen is removed. Still
another alternative is to leave the solids in the liquid after hydrotreat-
ment, then distill the product fuel and leave the insoluble material in the
residue (as well as some of the sulfur and nitrogen that is more difficult
to remove). Other variations of the process exist, but these alternatives
appear to be most common.
Near complete removal of the organic sources of sulfur and nitrogen
requires exhaustive hydrogenation using amounts of hydrogen well in excess of
the stoichiometric equivalent of the contaminants being removed. The overall
hydrogen utilization is poor because the contaminant-removal reaction occurs
concurrent with hydrogenation of the coal, and produces less desirable hydro-
carbons and light fractions mixed with H_S and NIL.
Even though frequently used, the concentrations of sulfur,
nitrogen, and trace elements in the coal liquid products probably should not
be used as the only measure of the effectiveness of the overall contaminant-
removal method, since the final liquid products have different processing
histories. The fraction of the coal recovered as an environmentally accepta-
ble fuel should also be considered in the comparisons.
17
-------
From this summary on contaminant-removal from coal-derived liquids,
it may be concluded that:
• Simple organic sulfur contaminants can be removed by
chemical (hydrogen) treatment.
• More complex organic sulfur and organic nitrogen molecules
existing in coal derived liquids can be removed by
catalytic hydrotreatment to form I^S and NI^.
• Many of the metals in liquid fuels interfere with
catalytic hydrotreatment and must be absent or in very
low concentration before such treatment is undertaken.
• Conversion processes in which lighter liquids are
recovered from heavy liquid fuels by thermal cracking
are relatively ineffective for contaminant removal.
• Ilydrotreatment reactions change many of the properties
of the fuel as well as remove sulfur as i^S and nitrogen
as NH3.
INTERRELATIONAL ASPECTS OF CONTAMINANT REMOVAL
An obvious interrelationship exists between the commercial
coal-preparation processes used to remove contaminants from run-of-the-mine
coal and the need for quality coal feedstock used in other types of
contaminant-removal processes. The processes based on liquefaction, chemical
refining, pyrolysis, gasification, and some types of physical methods
attempt to remove contaminants that usually can be removed only partially or
are impossible to remove by the combined commercial preparation processes
(i.e., grinding, washing, dense-media separation, and froth flotation). The
limit to which the size of the coal can be ground to optimize processing
cost and minimize fuel losses during the coal preparation also influences
the extent of contaminant removal. However, when feed coal is to be pre-
pared in such a facility for utilization in, for example, chemical refining,
trade-offs would have to be made between coal losses and maximum removal of
reagent-consuming contaminants prior to chemical processing.
In any study on the contaminant removal from solid and liquid
fuels, it is necessary to consider how the removal of one class or type of
contaminant affects concurrent or subsequent removal of another contaminant.
18
-------
For example, liquefaction by catalytic hydrogenation removes the organic
sulfur, some organic nitrogen, and half the pyritic sulfur (FeS2 is converted
to FeS) as H2S. However, removing ash minerals from liquid fuels is
a costly and difficult step. If the coal were cleaned to remove pyrite
(physical separation or chemical refining) prior to catalytic hydrogenation,
it might not be necessary to separate the ash from the liquid fuel after
catalytic hydrogenation. Typical combined processes include:
• Pyrite removal by chemical refining followed by coal
liquefaction by hydrogenation.
• Iron sulfide removal by magnetic means after coal
liquefaction.
• Mineral and pyrite removal by magnetic means before
hydrogen treatment.
• Demetallization of coal liquid before catalytic
hydrotreatment for sulfur and nitrogen removal.
19
-------
SECTION 5
SELECTION OF REMOVAL TECHNOLOGIES
TO BE CONSIDERED FOR FUTURE STUDY
As part of the objective of the Phase I program, methods for
contaminant removal were reviewed and analyzed with regard to their potential
for further study. Potentially new and novel techniques based on the
chemistry of the fuel contaminants and potential removal mechanisms were con-
sidered. A systematic categorization of the methods of contaminant removal
according to the contaminants and the type of removal was included in the
Task 1 reports/1'2^
The selection of specific contaminant removal techniques was based
on various criteria discussed below. These same criteria were used to
finalize and rank a list of techniques suggested for further study.
SELECTION CRITERIA DEVELOPMENT
In choosing the specific areas for study in the initial screening
of fuel contaminant removal techniques, several criteria were used. The
criteria were developed from the Task 1 review of the information on the
chemical and physical characteristics of the contaminants in fuels and
(2)
the techniques employed to remove them from the fuels. The following
criteria, used most frequently to develop a list of techniques for contami-
nant removal, are related primarily to coal and coal-derived liquids.
• Criteria based on chemical and physical characteristics of the
contaminants.
• Pyritic Sulfur - About 40 to 60 percent of the pyrite
present in run-of-the-mine coal can be liberated and
removed easily from coal. The remaining pyrite is
micron-size particles distributed uniformly in the
organic matter. Therefore, release of this pyrite
will depend on the extent of size reduction or depoly-
merization the coal undergoes to expose the fine
20
-------
particles. The released pyrite might be separated or
removed by physical or selective chemical means.
• Organic Sulfur - In order to remove the organic sulfur
from the coal structure, either the molecule(s) must
be released from the coal polymer and then selectively
removed or the molecule must be disrupted to release
the sulfur atom. Selective removal would result in
substantial loss in fuel value. Disruption of the
molecules requires conditions of varying severity
depending on the stability of the molecule.
• Nitrogen - To remove the nitrogen present as organic
molecules locked into the coal structure, the
molecules must be released from the coal polymer.
The molecules can then be selectively removed or
disrupted to release the nitrogen atom. Removal
will result in substantial loss in fuel value. Dis-
ruption requires conditions more severe than those
for the organic sulfur molecules.
• Trace Elements - Separating the mineral matter dis-
tributed in coal will do much to remove trace elements.
The problem lies in the extent of release possible at
a practical limit of coal size reduction. Removal of
trace elements associated with the organic matter
requires disruption of coal structure.
Criteria based on techniques used to remove fuel contaminants.
• Fuel Value Recovery - Do the techniques used to remove
contaminants from fuels produce the maximum possible
amount of the desired environmentally acceptable fuel?
Is the transfer of fuel contaminants to a residual fuel
to provide clean fuel avoided or minimized? Is the
energy required in maximizing fuel recovery kept as low
as possible?
• Utilization of Interrelation! Aspects - Can two or
more contaminant-removal techniques when combined
produce a clean fuel with the in-tn-tmimi fuel losses?
21
-------
For example, can extensive removal of mineral matter
and pyrites provide a quality feedstock for use in
organic sulfur and nitrogen contaminant removal
processes?
• Optimize Hydrogen Utilization - Can the efficiency
with which hydrogen is used in removing organic sulfur
and nitrogen from a fuel be improved by treating
fractions isolated from coal-derived liquids enriched
in these contaminants?
• Transferable Technology - Can a technique of contami-
nant removal used for petroleum be utilized for coal-
derived liquids?
• Contaminant Disposal - Does the process minimize the
environmental threat posed by the disposal of the
removed contaminants allowing for their chemical nature and
form?
Consideration of these criteria provided the basis for ranking
several contaminant removal concepts. The ranking was done so that, in any
experimental evaluation program, higher ranked concepts that show no immedi-
ate success can be replaced by one of intermediate or low rank. Successful
approaches would be set aside for larger programs. Thus, as many of the
listed areas as possible might be evaluated in a limited time.
Each of the approaches considered could produce an environmentally
acceptable fuel. The points in the process where these might operate on a
coal-based complex is shown in Figure 1. As an example "Immiscible Fluid
Agglomeration of Coal" operates at the coal-cleaning facility, and "Physical
Adsorption for Contaminant Removal from Coal Liquids" operates in lieu of
filtration of coal-derived liquids.
RESULTS OF SCREENING
After preliminary assessment of the approaches, they were
critically reviewed during discussions with other EPA contractors active in
coal and coal liquid contaminant removal programs and the EPA Project
Officers. Their comments were weighed in the final rationalization which
reduced their number to fourteen (14) listed in Table 5. (The order of the
22
-------
10
u>
I
r
Raw 1
Liquefaction
k
*
^\Carbonizatior
\ r
Coal
Gasification
Coal*
fr
Cleanup pre_Hefl
Non-Catal
Coal*
Cleanup
*»
»___..
1I2
A
fr
ter \^_s
rtlr ... , . ... .. ....
SRC
Coking
j *Gaa
t> Char
H2
A
Product t — »>Ga80
— *Fuel
..,. — * Char
*Char*
\J FiTteration,
.. ,„..— ,..^Jl IVfpMlljiUnn
Catalytic
^ Gas
. . -> Tars
>• Coke
Oil
Cleanup
asific.
Gasification
* Potential points in process for contaminant removal method.
FIGURE 1 . RELATIONSHIP OF SUGGESTED CONTAMINANT REMOVAL METHODS
WITHIN A COAL-BASED COMPLEX
-------
TABLE 5. AREAS SELECTED FOR FURTHER STUDY
Ro.
l
Contaminant
Ranoval Study
Description
Comment*
Determine Influence of
mineral*, trace element*,
and metal complexes on
hydrotreatlng
N)
*»
Effect of reaction* of
NHj or HjO with organic
eulfur compound*
Selective ineolublll-
latloo of coal liquid
fraction*
To better undsrstsod the role and fate of ash Pro
mlnerale and trace elements during coel liquefaction
and aubsequent hydrotreatlng the following ereae
ahould be considered for further atudy.
The behevlor of model eulfur and/or nitrogen
compounde during hydrotreatlng In the presence of Con
unaltered coel minerele and trace elemente with and
without th* presence of liquid* analogoua to coal
liquids should be investigated. Such mineral* would
be recovered in en unaltered or only alightly altered
etete by low-tempersture sshlng from a coal known to
contain moat of th* trace element*. Variation* in
•elected ratio* of the trace elemente In coal* would
be Incorporated into the atudy In order to aacertaln
correlation of hydrotreatlng activity. I.e., aulfur
and nitrogen removal. Trace metals might be supplied
•• coordination complex** formed from nitrogen and
eulfur containing Uganda and/or porphyrin derivative*
known to be present In coil*.
An area which has received little1 attention le Pro (1)
the effect of the product* from hydrotroitment of
coal upon *ulfided catalyst* encountered during hydro-
tresting. Th* producte receiving eome attention are (2)
HHj and H2S. Howevar, conelderable water la alao
formed especially from lower rank coala and ahould ba
added to th* list. The back reectlon of these
product* with coal should elso be of concerni
The potentlel for the reaction of KH. with varlou*
organic aulfur compound* to produce the more etable
organic nitrogen compounde should be investigated. The
•tudy would determine If nitrogen readily removed from
more bade amlnea during catalytic hydrodenltrogenatlon
(HDN) aa ammonia 1* capable of forming organic nitrogen
compound from organic eulfur compounde or if the
ammonia decompoae* and "active" hydrogen 1* formed to
alter the hydrode*ulfurls*tlon (HDS) rat*.
Daasphaltlng of petroleum with low-molecular Pro (1)
weight lolvant* such aa propane ha* been used commer-
cially and la reasonably effective In rejecting metal*
and aulfur. The commercial use haa been limited,
particularly in recent year*, by th* problem of die-
posing of the rejected asphaltaae frectlon. The
proce** haa not been tested on coal liquid*, and th*
disposal of the reject fraction may not be aa much of (2)
a problem at a coal liquefaction facility aa at a
petroleum refinery. For example, recycle of thle
material to the liquefaction reectore may b* feaaible.
(1) Would provide fundamental data on mineral activity
on hydrodeaulfurlsatlon and hydronltrogenetlon
and auggest new approaches to bydrotreatlng
cetalysts.
(1) Study would be limited to a etudy of model eulfur
and nitrogen compound*.
(2) taaulta obtained from model compound* may not be
directly applicable to hydrotreatmant.
Would provide fundamental data related to the
mechanism of HDS and HDN and possible competing
reaction*.
Would determine If overlapping rataa exist for the
hydrodeeulfurlsatlon of the more etable eulfur
compound* and the lea* etable nitrogen compounde.
(1) Study would be done only on model compound* u*lng-
well known catalysts.
The hydrotreatmant of the isoleted fractions would
be expected to be more specific for eulfur and/or
nitrogen removal *lnc* less hydrocarbon would be
present to consume hydrogen through saturstlon of
the highly aromatic coal liquid before HDS end UDN
would take place (I.e., better hydrogen utilization
for HDN and HDS).
Ascertain Che utility and effectlveneaa of concen-
trating organic aulfur and nitrogen by solvent
precipitation as compared to distillation of coal
liquids [I.*., minimize residue formation
(polymerisation) and unwanted gaseous hydrocarbon
formation due to excessive exposure to high
temperature* during distillation].
-------
TABLE 5. (Continued)
Ho.
Contaminant
•••oval Study
Description
Commenta
Use of pyrtte depres-
ssnts In conjunction
with IsalsclbJe fluid
agglomeretlon of coal
to
Correlation of the
mechanise of chealcal
cocainution of coal to
tha Mchanisn ralatad
to coal dlaaolutlon/
dapolymerltatlon
Inaleclble fluid* such aa kerosene and fuel oil
are capable of separating the aah Minerals fro* finely
ground coal suspended In aqueoua slurry through a
process of selective agglomeration. This approach la
attractive because the coal does not have to be dried
after Mining, wet size-reduction and coal-washing
operationa. However, pyrlte removal la poor due to
the similarity of Its surfacUl properties to those
of coal. The use of reagente known to alter the
eurface properties of pyrlte or the use of coal waetea
known to contain Perrobaclllus Perrooxldans should be
Investigated as a Beans of improving the pyrlte
rcBoval performance of the agglomeration process. As
p-rt of the euggested study, substitution of coal
liquids for kerosene or fuel oil should be Inveetlgeted
frr those operations preparing coal for liquefaction
processes. The process has the added advantage of
allowing the Isolated coal fraction to be dewatered
readily. Por thla reason, Its applicability to brown
coals ahould be Investigated.
Both ammonia and methanol are known for their
ability to penetrate coal through naturally occurring
faults. These chemicals disrupt the bonding forcee
scross the phase boundaries between coal and the
Impurities it contslns, which results In forceless
breakage of the coal body. The mechenlam causing
the disruption la not well understood. A similar
action Is suggested when coal la dlssolved/depoly-
merlsed by certain solvents such as pyrldlne,
phenols, aromatic hydrocarbons, etc., to yield s
coal liquid. A study should be undertaken to geln
a better understanding of the comminution snd
dlssolutlon/ddpolynorlistlon mechanisms. This
would provide fundamental data useful in selecting
Improved solvent systems for the liquefaction of
coal.
Con (1) Would require selection of solvsnt system to yield
maximum rejection of eulfur and nitrogen compound*
and still provide economic advantagee for Its use
over total hydrotreetment of the coal liquid or
separation by dlstlllstlon.
(2) Solvent recovery and maintaining pure solvent
character may be a problem.
Pro (1) Can provide a low ash, low pyrlte fuel directly
or a feed for e coal conversion plant.
(2) Oil agglomeration provides s means of producing
low moisture feed for conversion proceesea with
minimum thermal drying.
O) Agglomerating oil may be derived from coal
liquids.
Con (1) Requires else reduction to liberate pyrlte and
aah minerals.
(2) Doss not reduce orgenlc eulfur OK nitrogen content
of coal.
Pro (1) Study would provide fundamental data on the
mechanisms of comminution, dissolution, and
depolymerlsstlon of coal.
(2) A model eolvent eyatem for enhanced coal
penetration may be developed.
(3) With a depolymerlsetloo process, an alternative
to liquefaction by hydrogen treatment at high
tempereturee might be developed.
Con (1) A review of reported rstss of coel tolublllsstlon
would have to precede eny laboratory work.
(2) Study would havs to bs limited to known reectlve
cosIs.
(3) Solvent recovery and stability are critical to
euccess of proceaa.
-------
TABLE 5. (Continued)
No.
Contaminant
Removal Study
Description
Coa»ente
Physical adsorption for
contaminant removal
from coal liquids
Treatment of coal with
00 or CO-H, mlxturea
to fora liquid*
Cleanup of co«l lique-
faction residues
ro
Acid treatment for
coal liquefaction
"residue" cleanup
10 Hydrotneroal action
to procaae oil ehale
The high boiling fraction of coal liquid*
contalna trace elements and sulfur and nitrogen
contaminants. These fractions would be treated In
a fluldlsed or moving bed of alumina (or alsillar
material) particles that have low surface areas .
(<100 a2 g) but large pore else dimensions (>200 A).
Such a material may promote rapid diffusion Into
the pores and enhance the removal of the contami-
nants by filtration. Since the adsorption material
has a low aurface area, it would be strong enough
physically to undergo repeated regeneration cyclea.
Aa an alternative to hydrotraatlng, the concept
of removing eulfur with 00 or Hj-CO mixtures should
be considered.
Coal liquefaction realduea would contain,
besides the contaminants, some catalyst particles.
During coal carbonisation it has been found that
when the process is carried out in the presence of
certain gases (e.g., H_, NU.) eulfur and nitrogen
contamlnanta are removed. Then coal liquefaction
residues are gasified, the effect of the above gases
on sulfur and nitrogen removal may be enhanced because
of the presence of the catalyst particles.
The reeldue obtained from coal liquefaction
la rich in coal aah and trace elements. This resi-
due msy be clesned up by acid washing (e.g., UC1,
UF) prior to gasification. The trace elemente would
be more amenable to acid leaching from the coal
liquefaction realdue than from the gasification char.
Pro
Con
(1) Would provide a means of Improving filtarablllty
of coal liquids to remove ash minerals ss well as
s means of concentrating eulfur and nitrogen
impurltlee In reeldue.
(1) Yield loaaee of liquid might be greater than
present methode.
Pro
Con
Pro
The concept of short-range aolubllltatlon and
repreclpltatlon of the mineral matter In oil shale
Inherent In a hydrothcrmal process should be
investigated. This would be done to reduce the
effect of trace element contamination of shsle oil
formed during oil shale retorting. It IB anticipated
that the release of kerogen and other organic consti-
tuents of oil shele may occur during the eolublllsatlon-
rapreclpltation sequence. Con
In HjS and NU.,
(1) Process would reduce hydrogen requirements for
liquefaction.
(1) If COS formation la substantial, disposal may be
difficult.
Pro .(1) Would provide a means of grsster hydrocarbon
recovery end gas streams richer
for more efficient removal.
(2) Would reduce amount of carbon in the char formed
in totel liquefaction procesa.
Con (1) Performance of new concept (energy and product
quality) would have to exceed performance of
exleting proceee of reeldue pyrolyele/gaelflcatlon.
Pro (1) Dissolution of ash minerals In residue would
liberate oil and unreacted carbon to Improve oil
recovery.
(2) Provide a meens of trace element recovery from
liquefaction residues.
(3) Produces a reduced aah feed for gaelflcetlon.
Con (1) Dispose! of spent sclds and unwanted salts
prssents a potential problem.
(1) Provldea alternatives to retorting at high
temperatures.
(2) Reduces contamination of ebale oil by trece
elements.
(3) Reduces losses of hydrocarbon values ss chsr.
(4) Provldea s means of recovery of elkell value
from oil shele.
(1) Water cleenup required before reuse or disposal.
(2) Water lossss may be critical.
-------
TABLE 5. (Continued)
Me.
Contaminant
ReBOvi1 Study
Description
Comments
11
Use of solid "getters"
In hydrotraatlng
12
10
13
Oxidatlva denltro-
genetlon of coal
liquids
Acid/ban* treatment
of coal liquids
Biological action on
coal liquids to remove
•ulfur and nitrogen
contaminants
Tha Motivation In this area la tha exploratory
study on tha thenal hydrotreatlng of tar aand oil in
tha presence of pulverised coal, which acts aa a
"setter" for s>etals In the oil end for coke fonwd In
the process. This study used Canadian tar aand oil
and Canadian coal. The properties (primarily porosity)
of U.S. western coel Indicate that It ahould work well
aa a "getter", and thla flta well with our western
reserves of shsle oil and tar aand oil. Other possible
"getters" include coke or asphalt from any sourcs and
chars produced In coal gasification of liquefaction
processes.
This represents an extension of the atudles on
petroleum of selective oxidation with nitrogen oxides
followed by extraction of the oxidised species with
Mthsnol. Removal of both aulfur and nitrogen were
demonstrated, although the nitrogen data were more
Halted. The potential for nitrogen renoval warrants
further study, particularly because of the poeelbility
that at least part of the NO required for the oxida-
tion Bight bs obtained from tha nitrogen removed froai
the fuel.
The primary motivation in thla area la the
limited data indicating very good denltrogenetlon
of a ahale-oll fraction by a apeclflc combination
of acid and bass trestments. Also, s base treatment
was reported aa removing at least one metal (araenic)
from ahale oil. No data are available on such treat-
ments of cosl liquids. Tha objective would be the
renoval of nitrogen or metels, and not sulfur.
The atudles on the biological action on amlno
sclds and fatty acids show that hydrogenatlon by
enueroblc organlama possessing hydrogensse enzyme
ay items is possible st room temperature and low pres-
sures of hydrogen. Such action may lie transferable
to coal liquids for the removal of organic nitrogen
anil sulfur contaminants by hydrogenatlon or by other
biological action that breeka the carbon-aulfur or
carbon-nitrogen bonde.
Pro
Con
(1) Utilises cosl chsr from liquefaction process to
adaorb trace element contaminsnts from coal
liquids.
(1) Concept would have to show advantage over
exietlng filtration techniques for removal of
ash mlnerala.
Pro
Con
(1) Would utilise nitrogen removed from coal.
(1) Reaction le not selective end coel quality la
degreded.
Pro
Con
Pro
Con
(1) Would provide e nonhydrogenatlon method for the
removal of the very atable organic nitrogen
compounds In coal liquids.
(2) Trsce element removel would be poesible.
(1) Acid treatment would produce unwented sludge
thst would be difficult to dispose of.
(2) Cerbon velue would be loet with nitrogen
compounds.
(1) Bacterial sctlon would provide selective removal
of eulfur end nitrogen contaminants.
(1) Bscterisl action Inherently slow. However. If
the eniyme systsm reeponeible for the action
could be Isolated end concentrated, a viable,
low preeeure/low temperature proceee might be
devleed.
-------
listing does not necessarily reflect the view of the EPA to the importance
of the proposed study areas, but it does reflect one of early prioritizations
set by Battelle Columbus Laboratories.) In Table 5 the contaminant removal
methods or concepts selected for further study are described along with
comments about the approach. The descriptions and comments critically dis-
cuss the salient features of the study areas which in some cases were select-
ed to answer fundamental questions on the chemical mechanisms and limitations
of a process.
SELECTION OF REMOVAL TECHNOLOGIES
FOR PRELIMINARY ASSESSMENT
Another of the objectives of this study is to assess rapidly,
through either a literature search and preliminary experiments, a
selected number of removal technologies. The following six areas were
selected for preliminary screening evaluations:
• The biological action on coal-derived liquids.
• Enhancement of pyrite removal during immiscible
fluid agglomeration.
• The extraction of clean fuel from coal liquids.
• Concentration of organic sulfur and nitrogen
and ash from coal liquids.
• Conversion of coal liquefaction residues to
environmentally acceptable fuels.
• Improvements in pyrite liberation from coal.
These six areas at the time of selection were given a higher
priority than others being considered. The intent of this prioritiza-
tion was done so as to permit replacement of the higher priority
approaches that show no immediate success with one of the remaining 8
cited in Table 5. Also those approaches that are successful would be
set aside for larger programs. In this manner an evaluation of as
many of the areas as possible could be made. This approach was not
initiated and only the results of the studies cited above comprise the
next six sections of the report.
28
-------
SECTION 6
BIOLOGICAL ACTION ON COAL-DERIVED LIQUIDS
LITERATURE SURVEY
The objective of the literature survey was to develop concepts on
the biological approach to removing sulfur and nitrogen from coal and coal-
derived liquids on the basis of information in the literature.
The role of microorganisms in forming and degrading petroleum and
coal has been the basis for various fundamental studies. ' ' Another study
has shown that asphalt, a by-product of the petroleum industry, can be
degraded by certain sulfate-reducing microorganisms. ' Thus, selective
degradation of sulfur and nitrogen compounds in coal and coal liquids may be
possible by using such microorganisms.
These microorganisms include bacteria, yeasts, and molds.
Since most microorganisms are active at or near ambient conditions, sulfur
and nitrogen contaminants might be removed from coal and coal liquids
under conditions much milder than the high temperatures and pressures
required in the conventional hydrodesulfurization and hydrodenitrification
processes.
In this survey, wherever possible, the ability of microorganisms
to selectively degrade hydrocarbons containing sulfur and nitrogen is
compared with the kind of sulfur and nitrogen compounds present in coal.
Sulfur and Nitrogen Contaminants
in Coal and Coal Liquids
The report on the characterization of the sulfur and nitrogen
contaminants in coal and coal liquids discussed earlier showed that, except
for pyrites, the contaminants are primarily organic and consist of
aromatic and heterocyclic compounds. The H/C ratio and the total
sulfur and nitrogen content for coals, liquids derived from these coals,
and liquids derived from a typical petroleum crude were compared in Table 1.
Coals as compared to coal liquids and petroleum crude have the lowest H/C
29
-------
ratio and are considered in a state of unsaturation with respect to hydrogen.
The postulated structure of coal "molecule" given in Figure 2 shows the
organic sulfur and nitrogen contaminants.
In coal liquids, free of solids, the cyclic and heterocyclic
sulfur and nitrogen compounds are found primarily in the high-molecular-
weight asphaltenes (benzene soluble and heptane insoluble) and resin (methanol
soluble) fractions of coal liquids. Some nitrogen in the coal liquids is
present in porphyrin-type structures. '
The cyclic and heterocyclic compounds containing sulfur and
nitrogen are very stable. When conjugation of a sulfur atom occurs in a
ring system, its stability appears to be greater than that in straight- chain
compounds. The sulfur atoms display not only electron-releasing but also
electron- accepting conjugative effects. Nitrogen, which has five valence
electrons, also forms very stable heterocyclic compounds. Two valence
electrons are involved in the formation of covalent bonds to the adjacent
carbon atoms in the ring; one valence electron is involved in either the
bond to a hydrogen atom, an alkyl group, or conjugation in the ring. The
two remaining electrons (the "lone pair") are usually not involved in bond
formation but account for the basic properties of some amines. These
inherent behaviors of electrons of sulfur and nitrogen compounds may be
important when considering interaction between these compounds and micro-
organisms .
Patents on Petroleum Desulfurization
Processes Using Microorganisms
The desulfurization of petroleum by using microorganisms is
described in many patents. (°-12) jj^ microorganisms used in these
patents include aerobic, anaerobic, or a combination of the two. The micro-
(12)
organisms are found in nature, but one patent uses a combination of
naturally occurring microorganisms with prepared mutants. In any case
the microorganisms are grown under controlled conditions of temperature, pH,
and in the presence of specific nutrients.
A brief discussion on the claims of these patents follows.
Strawinskiit ' claimed that crude oils and like substances are
desulfurized by bringing them into contact with mineral salt nutrient
medium in the proportion of approximately 5 percent oil to 95 percent
30
-------
FIGURE 2. STRUCTURE OF BITUMINOUS COAL POSTULATING SULFUR AND
NITROGEN OCCURRENCE IN COAL (from Wiser, W. H., EPRI
Conference on Coal Catalysis, Sept. 24-25, 1973)
-------
nutrient medium. The nutrient medium contains a diverter, usually a car-
bohydrate or protein, which the microorganism will consume in preference
to the petroleum hydrocarbons. A reduction of sulfur content of 12.5 percent
was noted when a crude was treated over a period of 4 days.
Also, Strawinskii^) proposed the use of aerobic bacteria to convert
organic sulfur to sulfate and then anaerobic bacteria to reduce the sulfate to
hydrogen sulfide. Aerobic bacteria such as certain species of Pseudomonas,
Alca}igenesT Bacillus or any microorganism which is capable of converting the
sulfur-bearing complexes in a petroleum hydrocarbon to sulfates are grown
in a mineral-salt nutrient. Small increments of the petroleum crude were
added to the nutrient. There was no inorganic sulfur in the nutrient-
petroleum mixture. The anaerobic bacteria consist of Vibrio desulfuricans.
Vibrio estuardi. Vibrio thermodesulfuricans or Desulfovibrio.
The nutrients for aerobic and anaerobic bacteria are shown in
Table 6. The medium may be further altered by the addition of pure
accessory growth substances, like vitamins and/or amino acids. The pH is
adjusted to 7. The addition of certain ions of copper, mercury, zinc
bismuth, and iodine to the medium may stimulate the activity of the micro-
organism only when present in trace concentrations. The patents do not
give any sulfur removal data. Most organisms were grown within a temperature
range of 24 to 30 C and with the pH in the range of 6.5 to 7.5.
ZoBell/^' in his patent claimed that sulfur is removed from
petroleum hydrocarbons containing relatively complex sulfur compounds by
bringing the hydrocarbons into contact with a substantial amount of hydrogen
in the presence of hydrogenase producing microorganisms selected from group
consisting of Desulfovebrio desulfuricans and Sporovibria and a nutrient
medium. In this way sulfur is split off in the form of gaseous products
which are removed. The hydrogen can be produced ^n situ by reaction of
Clostridium microorganisms on carbohydrates. The process can be carried out
either in batches or continuously and can be controlled by varying the partial
pressure of hydrogen, the temperature, pH, etc. The partial pressure of
hydrogen is maintained within 25-100 percent of total pressure which is kept
below 10 atm. The temperature is maintained between 25 and 75 C and the pH
between 6.5 and 7.5. Generally the reaction is exothermic; hence, cooling may
32
-------
TABLE 6. NUTRIENTS FOR AEROBIC. AND
ANAEROBIC BACTERIA^)
For Aerobic Bacteria
Na2S203«5H20
NaH2P04H20
K2HP04
NH4N03
MgCl26H20
MnCl24H20
CaC03
FeCl2
Water
Grams
10
1
2
2
0.25
0.01
0.01
0.01
1000 ml
For Anaerobic Bacteria
K2HP04 0.5
NH4C1 1.0
MgS04-7H20 2.0
CaS04«2H20 1.0
or
Na2S04 1.0
CaCl2-2H20 0.1
Mohr's Salt (ferrous ammonium sulfate) Trace
Water 1000 ml
33
-------
be required. When hydrogenase is used alone in the absence of other enzymes
or catalysts, a mineral salt solution buffer having a pH of 6-8 is used.
Catalyst for reducing sulfur-containing compounds can be isolated from marine
sediments, the cultures apparently grow best in an aqueous medium of the
composition of seawater, enriched by calcium lactate 0.02, ascorbic acid 0.02,
ferrous ammonium phosphate 0.02, sodium bicarbonate 0.05, and potassium
sulfate 0.5 weight percent.
A patent by Kirshenbaunr ' claimed to desulfurize petroleum crude
by using aerobic bacteria. The bacteria convert organic sulfur compounds
to inorganic sulfur compounds and these are removed by chemicals, e.g.,
lime treatment. The microorganisms, Thiophyso-volutans. Thiobacillus
thiooxidans and Thiobacillus thioparus. used in this patent are found in
nature at places of crude storage and spills. The conditions for desulfur-
ization are similar to those in previous patents. However, certain surface
active agents are recommended for increasing the contact between the oil
and water layer.
(12)
In another patentVA ' hydrocarbons are oxidized by using natural
and mutant microorganisms. Some of the microorganisms considered in this
patent are in Table 7. Adding a mutant to Desulfovibrio aesturii at 38 C
resulted in a large increase in the oxygen intake by the microorganism.
The evidence presented in these patents shows that biological
desulfurization of petroleum crudes is possible. What is required are con-
trolled process conditions and nutrients. Bacteria from the group,
Desulfovibrio. are particularly important. This bacteria dehydrogenates the
oil but stops short of complete oxidation. This may be due to hydrogen sul-
fide buildup which poisons the oxidizing system of Desulfobrio bacteria.
Reaction by Aerobic and
Anaerobic Microorganisms—
Some important chemical reactions brought about by microorganisms
are as follows.
• Oxidation of sulfur
2S + 21^0 + 302 > 2H2S04 + Energy
34
-------
TABLE 7. SOME NATURAL BACTERIA USEFUL
TO FORM MUTANTS (Reference 12)
Aehromobactrr aerophllum
Achromobacter citrophilum
Achromobacter pattinator
Achromobacttr tulfurtum
Achromobacttr thalattiut
Achromobacter iophagut
Achromobacter dclicatulut
Achromobacter aijnnmarinut
Achromobacter ciicloclatttt
Achromobacttr ttationit
Achromobactcr delmarvat
Achromobacttr agilt
Achromobacttr ccntropuncta-
tam
Agarbactcrium 611/9
Agarbactcrium reducant
Airtrbacttrium vitco'um
Alcaligenet metalcallgtnti
Alcolitjenti rtctl
BacillH* thtrmonmtilolytlotu
Jtaeilliu lattrotporat
Bacillu* brtvil
BaciUu* thtrmoUqurJacitn*
DaciUut tottm
Bacilt** hexacarbovorum
Bacilluf lactorubefacifnt
Bactllu* mycoiitrt corallinu*
Bnelllut bruntsll
Baclllui tolunUcum
Bacillut naphthnUnlc*t
Bneillnt jthennnthrenlciu
Bacillu* nubtillt
Itncillut flrnnt»
Btcillut macerant
Baclttu* oirculnn»
JJaciilui ethanicu*
Badltti* JtauttophituM
RacWui tlifrmoitiautatletu
RnrMu* cnlliloinr.ti*
Bncillu* mlcltatli*U
Barilla* thermoaltmtnto-
Jtacilltn
Ritftertttm
RncffHtim
fnrtfrinm
Bnffrritim
facitni
Jtnrtfrium
Bneteriam
Rirtrrlum
Partrrinm
Kncterinm
Jtartrrlum
Jtnrtrrluai
Jtnrtrr1nm
BnrtfriHitt
rnphttinllnlfn*
pArfinnfArenietu
*tutseri
fluoretcent
gloMfitrnti
rnhrfarinit
tntrrirfnm
lilonrnm
Corymbacterlum
phtheriticum
CorynebAlcttritun limpltx
Ut*ul/ovihrio dttulfuricant
Detulfovtbrio aeituaril
Deiulfovibrto rubentichlkil
Deiuljovibrio halehydrocar-
lionoelatticiu
Flaaobaettrium okeanoloitet
ftavobacteriun marine typ*.
cum,
Plauobacterium mannor
-------
• Oxidation- reduction (glucose, potassium nitrate)
C6H12°6 + 12KN03 - > 12KN02 + 6H20 + 6C02 + Energy
(Oxidizable) (Reducible) (Reduced) (Oxidized) (Oxidized)
• Oxidation of phenanthrene
C14H10 - ^ C11H8°2 + C02 + Ener8y
(Phenanthrene) (l-hydroxy-2-naphthoic acid)
• Reduction of carbon dioxide
C02 + 4H2 - > CH4 + 2H20 + Energy
• Reduction of sulfates
3CaS04 + 2(C3H503)Na - > 3CaC03 + Na2C03 + 2H20 + 2C02 + 3H2S + Energy
(sodium
lactate)
Some reactions mentioned above are included in the
earth's carbon cycle shown in Figure 3. The carbon cycle is maintained
by the combined activity of plants, animals, and microorganisms. Micro-
organisms active in the presence and absence of oxygen play an important role
in this growth and decay cycle. The normal decay of organic matter and its
conversion into carbon dioxide is an aerobic process. The aerobic process
uses free oxygen whereas the anaerobic process takes place in the absence of
oxygen. In the aerobic process hydrogen is transferred to oxygen to form
water while in the anaerobic processes hydrogen may either be transferred to
organic molecules or ions like nitrates or sulfates.
Along with the various aerobic and anaerobic degradations of organic
compounds, degradation of sulfur and nitrogen compounds also occurs. Aerobic
bacteria of the genus Thiobacillus oxidize elemental sulfur to sulfuric
acid, which reacts with a base to form sulfates. Thiobacillus and other
species also oxidize hydrogen sulfide and sulfides. In such reactions, the
intermediate and final products are elemental sulfur and sulfuric acid,
36
-------
AEROBIC ENVIRONMENT
Complex Organic
Matter
v
CH4
CnHxn
ANAEROBIC ENVIRONMENT
FIGURE 3. THE AEROBIC AND ANAEROBIC ENVIRONMENT IN
THE EARTH'S CARBON CYCLE (Reference 3)
respectively. Sulfuric acid is also produced by an anaerobic species of
(13)
Thiobacillus. ' This anaerobic species oxidizes sulfur to sulfuric acid
and the oxygen for this process is obtained from other salts like nitrates.
An anaerobic specie, Desulfovibrio desulfurican. is able to reduce elemental
sulfur and sulfates to hydrogen sulfide.
Aerobic and anaerobic microorganisms are also active in the
degradation of complex organic nitrogen compounds like proteins and amino
acids. The final products of aerobic degradation are ammonia, carbon
dioxide, sulfates, and water. The aerobic bacteria can further degrade many
of the products formed during an anaerobic degradation.
Several of the patents on desulfurization of petroleum by
aerobic microorganisms suggest that the desulfurization is achieved in the
presence of an enzyme, e.g., hydrogenase, produced by the microorganism and
hydrogen gas. The reaction scheme may be represented by
37
-------
_ .- _ . . Altered
Orgamc-Sulfur Bacteria
Compound "2 Enzyme compound ^
The enzyme "activates" the molecular hydrogen which then desulfurizes the
(8-10)
petroleum.
Thermodynamic Feasibility—
A reaction aided by aerobic or anaerobic microorganisms can be
analyzed like a conventional chemical reaction. Thus the more negative the
change in Gibbs free energy (AG) for a reaction the more thermodynamically
favored is the reaction as written. For certain reactions microorganisms
seem to act as a catalyst. Like in catalytic reactions, the presence of the
microorganisms may be neglected when calculating AG, as the concentration of
the microorganisms is low with respect to the reactants. In the reactions
discussed below, the magnitude of negative AG for an oxidation and an
oxidation-reduction reaction was calculated. The AG values show that the
aerobic reaction is thermodynamically more favored than the oxidation-
reduction reaction. However, the Gibbs free energy values give no indication
of the reaction rates.
In the reactions discussed below, the magnitude of the AG for
an oxidation and an oxidation-reduction reaction is compared to illustrate
the greater potential for aerobic reactions to proceed. The reactions of
xylene are considered instead of the classical glucose system, commonly dis-
cussed in microbiological degradation, because xylene-type hydrocarbons
dominate coal and coal liquids. The xylene-type hydrocarbons include cyclic
sulfur and nitrogen compounds found in coal and coal liquids.
• Aromatic (xylene) oxidation
by aerobic systems
C8H10 + ^2 °2 * 8C02 + 5H2° AG =
(Oxidized)
38
-------
Aromatic (xylene) oxidation-reduction
by anaerobic systems
^ KN02 + Y °2
(Reduced)
C8H10 H
(Oxidized)
C H + 21 KNO
C8H10 + 2 KN03
h 2 °2 ^ 8C02 + 5H2°
> Y ra°2 + 8C02 + 5H2°
AG - -723 kcal
In anaerobic reactions, no free oxygen is available to add directly to the
xylene or to act as hydrogen acceptor. With the help of enzymes, the anaero-
bic bacteria reduce the nitrate and oxidize the xylene. The free energy
change is -723 kcal. This value is less than that for the oxidation reac-
tion, which indicates that, although this reaction is favored, it is not as
favored as the direct oxidation reaction of the aerobic system.
Beerstecher has given examples where microorganisms are able to convert
xylene-type hydrocarbons to other products. One such example for benzene is
shown below:
(Benzene) HOOC
This degradation of benzene suggests direct rupture of the ring. Based on
such microbial degradation reactions, it can be postulated that organisms may
be found that can selectively degrade cyclic sulfur and nitrogen compounds.
Thus, when such selective degradation occurs in a coal or coal liquid only
the sulfur and nitrogen compounds would be altered by the microorganism.
Factors Influencing the
Growth of Microorganisms —
Microorganisms are cultured in various mixtures of mineral salts
(phosphates, nitrates, sulfates, chlorides) in water. An adequate source
of phosphate is necessary to obtain optimal levels of growth. Some micro-
organisms can use all types of sources of nitrogen; others require specific
39
-------
nitrogen compounds. Some hydrocarbon-utilizing microorganisms are known
to grow in extreme environments, for example, 25 percent salt solutions.' '
Other factors, like the pH, temperature, and dispersal of the hydro-
carbon in the media, influence growth of the bacteria and the subsequent
activity of the enzymes. The dispersal of the hydrocarbon in the liquid
media greatly enhances the activity of bacteria as in Figure 4. Various
techniques are used to subdivide the hydrocarbons; all involve
either the formation of emulsions or adsorption upon suitable materials.
As coal derived liquids have the inherent property of forming emulsions in
water, they may provide suitable environment for microorganism growth.
•o ,
1s
00
o
U
c
-------
Biological Action in Coal
Derived Liquids
Microbiological degradation of hydrocarbon types found in coal and
coal derived liquids by oxidation reactions is well documented. ' '
Several patents discussed earlier claim that microorganisms will remove
sulfur compounds from petroleum ~ and recently it was proposed that
microorganisms be used to clean up certain oil spills. In this clean up
of oil spills, it was not determined whether the degradation was selective
towards certain compounds found in an oil spill.
A majority of the microbial species that are known to consume
coal/coal liquid-type hydrocarbons are listed in Table 8. This table illus-
trates certain characteristics of the more important species and includes
fungi, a nonchlorophyll-containing plants. A large number of molds also
(3\
utilize hydrocarbons.v '
Oxidation of Hydrocarbons—
There are certain generalizations regarding the oxidizing ability of
various microorganisms. Anaerobic sulfate reducers do not oxidize hydro-
(14)
carbons smaller than decane. This is particularly interesting as the
type and size of hydrocarbons found in coal and coal liquids are comparable
in size to decane. Organisms such as Mycobacteria are generally restricted
in their growth to the hydrocarbon-water interface whereas Pseudomonas grow
in a dispersed fashion throughout the medium, Table 8.
The oxidation reactions of various straight-chain (paraffin)
hydrocarbons in the presence of aerobic bacteria are of limited importance
in this study since coal and coal liquids contain small amounts of
paraffinic hydrocarbons. The oxidation reactions of benzene and its
derivatives produce various acids prior to the cleavage of the benzene
ring as in Figure 5. Certain bacteria oxidize benzene at room temperature,
and the product is catechol. However, in the absence of bacteria, benzene
is oxidized with air at 600 C, and the product is again catechol.
When a polycyclic aromatic is attacked, a stepwise oxidation
involving cleavage of an end ring occurs in a manner analogous to cleavage
of the benzene ring. Oxidation of each ring proceeds in turn until the
41
-------
TABLE 8. MICROBIAL SPECIES AND BIODEGRADATION OF HYDROCARBONS
FOUND IN COAL/COAL LIQUIDS(a>
Family
Genus
Hydrocarbon,..
Biodegraded10'
Remarks
Fungi
N)
Nltrobacterlaceae
Pseudomonadaceae
Micrococcaceae
Achromobacterlaceaa
Bacterlaceae
Hycobaccerlaceae
(Actlnooycetalea)
filamentous bacteria
Streptomycetaceae
Thiobacillua
T. denitrificana
Paeudomonaa
Pa. aeruginosa
Pa. boreopolia
Pa. oleovorana
Desulfovibrio
D. desulfuricana
Sarcina
S. species
Achromobacter
A. centropunctatum
Bacterium
B. phenanthrenius
Mycobacterium
Micromonospora
Crude asphalt
Kerosene
Petroleum
Naphthalene
Naphthalene
Petroleum
Hydrocarbons
Phenanthrene
Petroleum
Kerosene
Petroleum
Phenanthrene
Moat hydrocarbons
Most hydrocarbons
Numerous species
Isolated from soil
Isolated from cutting
oils
Obligate anaerobe
isolated from oil
field waters
Isolated from oil-
soaked soil
Numerous species
Isolated from soil
Isolated from Wisconsin
lake bottoms
Yeast and Molds
Endomyces
Hanaenia
Petroleum
Petroleum
Isolated from soil
Isolated from soil
(a) From Tables 19 and 21 of Reference 3 and Table 7 of Reference 4.
(b) Blodegradation may not be complete in some cases.
-------
OH
p-hydroxybenzy la Icahol
CHO
DH OH
p*hydVoxybei3aldehyde
I
COOH
phenol
p-hydroxybenzofc
acid COOH
IHCOOH
oandalic acid
benzaic acid
OH
gL>hydroxyb*nzoic benzene
acid
COOH
COOH
OH
caltchel
/I-keteadipic acid
FIGURE 5. CATECHOL AND PROTOCATECHUIC ACID AS KEY
INTERMEDIATES IN THE OXIDATION OF BENZENE
DERIVATIVES (Reference 4)
43
-------
rings are broken. ' Thus, in the oxidative degradation of the polycyclic
aromatics, they will pass through the salicyclic acid and catechol deriva-
tives of benzene as shown in Figure 5.
Enzyme Activity—
The activity of the microorganisms in using hydrocarbons depends
on the enzymes the organisms produce. The ability of an enzyme to catalyze
/3 13)
a certain reaction and activity is determined by several factors:
• Enzymes are specific and will catalyze only certain
kinds of reactions; in addition, they act on but one
kind of substance. For example, enzyme deaminases
reacts with amino acids to produce ammonia and acids.
• The activity of an enzyme is directly related to the
operating temperature. Enzymes have an optimum operating
temperature which is usually below 60 C. At higher
temperatures, most enzymes are rapidly deactivated.
• Enzyme activity is sensitive to the pH of the solution
although specific enzymes are active in acid,
alkaline, or neutral solutions.
• Certain neutral salts like NaCl and KNC>3 enhance the
activity of some enzymes, but salts of heavy metals,
such as HgCl2 and CuSOA,, will, in time, inactivate
/3 g\
most enzymes. ' However, small amounts of heavy
metal salts were considered necessary nutrients in
the patents discussed earlier.
Microbiological Activity in
Coal and Petroleum Deposits
Microbiological activity probably played a significant part in
'3 A \
the formation and accumulation of coal and petroleum.v » ' In the carbon
cycle discussed earlier, the available carbon is constantly recycled, pro-
viding energy and building material for life. This cycle allows little
carbon to be accumulated. However, the various rates of microbiological
activity and unfavorable conditions of pH, oxygen, and nitrogen have probably
44
-------
led to the accumulation of carbon as coal and petroleum. For coal,
the first stage of this accumulation is peat formation.(3»^)
Microbial Activity in
Conversion of Peat to Coal—
Most microbial species found in peat are similar to the species
identified in the biodegradation of hydrocarbons, Table 8. The species
belongs to the family of Mycobacteria and Streptomycetes. The peat deposits
support two forms of microorganisms: aerobic microorganisms found near
the surface and anaerobic microorganisms found near the bottom of the peat
deposit as in Table 9. In this table, the sphagnum peat is acidic compared
to woody peat and thus supports different types of microorganisms. The
anaerobic microorganisms count in acidic peat is higher than that for woody
peat. The cellulose-decomposing and -nitrifying bacteria present in woody
peat are absent from the acidic peat. Fungi in both peats are active at the
surface of the deposit. This table also shows that the total number of
organisms increases with depth and suggests that the containers used in any
possible microbial treatment of coal to remove sulfur and nitrogen should
not be shallow.
One theory of coal formation postulates that the difference in
*
activity of aerobic and anaerobic microorganisms could influence the
nitrogen content of coals.^ ' According to this theory the anaerobic con-
ditions promote the decomposition of the nitrogenous complexes in the peat.
If this is so, then some anaerobic microorganisms present in peat may be used
to decompose selectively organic nitrogen compounds in coal and coal
derived liquids.
The activity of microorganisms that leads to variations in products
in peat and coal deposits is an important finding for this study. Such an
activity suggests that microorganisms may be found that may selectively
attack certain types of sulfur and nitrogen compounds in coal or coal
liquids. B
as follows:
liquids. Rogoff has summarized the microbial activity in peat deposits
Differential decomposition occurs as a result of
inherent resistance of the original plant matter to
decay. The waxy and resinous materials in coal con-
tain substances which inhibit the growth of micro-
organisms (biocides).
45
-------
TABLE 9. TYPES OF MICROORGANISMS FOUND IN PEAT AS DETERMINED
BY DIFFERENTIAL MEDIA COUNTS (Reference 4)
Microorganisms per g moist peat
Depth
(cm.)
0
26-40
50-62
110-120
2.5-10
22.5-30
90-120
150-180
240-220
270-330
pH
6.2
6.4
6.5
6.3
4.4
4.0
4.1
4.2
5.7
6.0
Capable of
developing
in air
^6, 200, 000
25,600,000
2,300,000
1,600,000
(c)100,000
260,000
650,000
750,000
1,250,000
2,000,000
Bacteria
Anaerobic^ Cellulose Nitrifying'^ Fungi
decomposing (a)
Woody Peat
6,000 12,000
12,000 12,000
12,000 12,000
18,000 6,000
Sphagnum Peat
200,000
100,000
200,000
300,000
200,000
300,000
18,000 26,000
12,000 2,000
12,000 0
6,000 0
_ .
-
-
-
-
-
Approximate.
^Includes facultative aerobes.
cIncludes facultative anaerobes,
• Different products are produced from the same plant
matter under different environmental conditions.
This may be demonstrated by the attack of micro-
organisms on lignin under aerobic conditions
giving carbon dioxide and water as products as
opposed to anaerobic attack in which the lignin
structure is merely altered.
Microbial Activity in Petroleum Reservoirs—
Living microorganisms have been isolated from oil field waters
originating several thousand meters below the earth's surface. At least
50 species of such microorganisms are known. These microorganisms can grow
under diverse conditions: (1) within a temperature range of 0 to 85 C,
(2) under hydrostatic pressures of up to 150,000 psi, and (3) in various
saline solutions.
(3)
The interest in the various microorganisms found in
46
-------
petroleum crudes is prompted by efforts to determine the origin of the
crudes. However, certain microorganisms, like the sulfate reducing bacteria,
play various roles in a petroleum reservoir: (1) they dissolve carbonates
from the rock thus releasing the oil, (2) they produce certain detergent-type
chemicals that help release oil from the rock, (3) they attach themselves to
the rock and prevent the movement of oil, and (4) they attack long-chain
paraffin compounds converting them to smaller chains/3'19^
Conclusions on Biological Actions on Coal-Derived
Liquids Based on Literature Survey
The literature reviewed for this study suggests that microorganisms
may be found that are effective in desulfurizing coal and coal derived liquids,
This desulfurizing activity of microorganisms is based on information found
la; (1) studies on the metabolic activity of certain aerobic and anaerobic
microorganisms and (2) data from the various patents on desulfurization of
petroleum.
The aerobic microorganisms that are active in desulfurizing
hydrocarbons function by converting the organic sulfur to a sulfate (804")
ion. This sulfate ion is then removed from the solution. The desulfuriza-
tion reactions of anaerobic microorganisms may be catalyzed by enzymes.
These enzymes convert molecular hydrogen to an 'active1 hydrogen that reacts
with sulfur compounds to form hydrogen sulfide. The microorganisms need
nutrients and controlled conditions for growth. The nutrients include
inorganic salts particularly phosphates. The conditions necessary for the
growth of these microorganisms are that (1) the temperature be in the range
of 0 to 85 C and (2) the pH be maintained in the range of 6.5 to 7.5; however,
many microorganisms will grow at other pH values. Besides the nutrients,
temperature, and pH of the solution, growth of the microorganisms is
influenced by certain organic compounds in the coal and coal liquids which
are preferred over the others by these microorganisms.
Besides the desulfurization of coal and coal derived liquids,
denitrification by microorganisms may also be possible as is observed
during the formation of coal.
47
-------
EXPERIMENTAL STUDIES
Outline of Experimental Program
Based upon the information obtained in the literature search,
a five-step experimental program was initiated with the intention of
demonstrating biological removal of organic sulfur and nitrogen-
containing compounds from coal-derived liquids.
The experimental program is outlined below:
Step 1. Grow microorganisms on defined media (salts media
containing inorganic sulfur salts) to determine
the growth rates of the organisms
Step 2. Add approximately 1 percent hydrocarbon character-
istic of coal-derived liquids (benzene or toluene)
to the salts media and determine the growth rates
on the modified media
Step 3. Add approximately 1 percent model sulfur compounds
(thiophene, thianaphthene, methylthiophene) to the
hydrocarbon(s) used in Step 2 and determine growth
rates in the presence of the organic sulfur compounds
Step 4. Repeat Step 3 but remove all inorganic sulfur salts
from the media
Step 5. Grow microorganisms on actual coal-derived liquids.
Steps 2 and 3 of the experimental program were needed to demon-
strate that aromatic hydrocarbons and organic sulfur compounds typical of
those found in coal-derived liquids were not toxic to the microorganisms.
No attempt was to be made in Step 3 to demonstrate sulfur removal — only
that the sulfur compounds were not toxic at concentrations typical of
those found in coal-derived liquids. If no growth could be achieved in
48
-------
Step 3 with 1 percent model compounds in the hydrocarbon phase, the con-
centration of the model compounds would be reduced to determine their
toxic level. Assuming Steps 1-3 were successful, analytical verification
of organic sulfur removal would be made in Steps 4 and 5. The same ex-
perimental program could then be repeated with model nitrogen compounds
to determine if microorganisms could be used to remove nitrogen-containing
compounds in coal-derived liquids.
Bacteria Cultures
Both aerobic and anaerobic sulfur-utilizing bacteria were
selected for the laboratory study based upon the frequency of their
referral in the literature and patent claims. The microorganisms and
their sources and cultivation media are given in Tables 10, 11, and 12.
The pure cultures, which were obtained from the American Type Culture
Collection (ATCC) and from Dr. Patrick J. Dugan (Ohio State University),
were maintained on the growth media recommended by ATCC and Dr. Dugan.
Stock cultures were maintained in 10 ml of their respective medium and
transferred to fresh medium on a biweekly schedule. The anaerobic bac-
teria were maintained in a Gas Pac 100 anaerobic jar, backflushed with
a. hydrogen-carbon dioxide gas mixture
The mixed culture, which was obtained from a settling pond at a
coal cleaning plant in Ohio, was maintained on ATCC medium No. 450 (under
aerobic conditions) and on commercially prepared, sulfate API broth (under
anaerobic conditions) which was obtained in sterile, 10 ml serus vials.
The Thiobacillus are characterized as being: gram-negative, strict
aerobes, short rods with a single polar flagella, and occurring singularly
or in pairs. They obtain their energy by oxidizing reduced sulfur compounds
to sulfuric acid. Their optimum temperature for growth ranges between 28 to
30 C and their optimum pH range is between 2 and 3.5.
Desulforibrio desulfuricans is characterized as a strict anaerobe,
and derives its energy for metabolic activity by chemical reduction of sulfate
to hydrogen sulfide. Its optimum temperature for growth is 25 to 30 C and its
optimum pH range is 6 to 7.5. I), desulfuricans is a gram-negative, mobile,
short curved rod, occurring either singly or in short chains.
49
-------
TABLE 10. AEROBIC SULFUR-UTILIZING BACTERIA
Organism Source Growth Media
Thiobacillus thiooxidans ATCC No. 19377 ATCC No. 238
Thiobacillus thiooxidans J. Dugan, Ohio ATCC No. 450
State University
Thiobacillus intermedius ATCC No. 15466 ATCC No. 152
Unidentified Mixed Coal Cleaning ATCC No. 450
Culture Plant Settling Pond
(a) ATCC — American Type Culture Collection, 12th Edition, 1976,
TABLE 11. ANAEROBIC SULFUR-UTLIZING BACTERIA
Organism Source Growth Media
Desulfovibrio desulfuricans ATC No. 13541 ATCC Medium No. 42
Sulfate APl(a>
broth
Unidentified Mixed Culture Coal Cleaning Sulfate API broth
Plant Settling
Pond
(a) API - American Petroleum Institute recommended practice 38,
3rd Edition, 1975, prepared by Difco Laboratories, Cat. No.
0500-86.
50
-------
TABLE 12. CULTURE MEDIA
ATCC 238
Thiobaccilua Medium B
HH4C1 0.1 g
KH2P04 3'° «
MgCl2 0.1 g
Na2S203.5H20 5.0 g
Distilled water 1.0 L
Adjust pH to 4.2 Sterilize by steaming for 30
minutes on three successive days.
ATCC 450
T2 Medium for Thiobaclllua
Solution A:
Na2S203.5H20 5.0 g
NH4C1 1.0 g
KNO,
2.0
'3 «•» 8
Distilled water 250.0 ml
Solution B:
KH.PO
2.0
4 '•« 8
Distilled water 250.0 ml
Solution D:
MgS04.7H20 0.8 g
FeS04.7H20 (2Z, w/v, In N HC1) . . 1.0 ml
Trace Hetals (see Med. 426) .... 1.0 ml
Distilled water 250.0 ml
The four solutions are sterilized separately
and combined aseptlcally for the completed
medium. The pH of the final medium la 7.0.
ATCC 125
Thiobaclllua Medium
(NH4)2S04 0.2 g
MgS04.7H20 0.5 g
CaCl
0.25 g
3.0 g
PeSO 0.005 g
Tap Water 1.0 I
KH2PO
1.0 g sulfur (precipitated) is placed In each dry
flask. The salt solution Is prepared, and 100
ml amounts are carefully poured down the side
of the flask without wetting the sulfur. Flasks
are then sterilized In flowing steam on three
consecutive days, 1/2 hour per day. Care must
be taken to ensure that the sulfur remains on
the surface throughout sterilization.
ATCC 426
Thlobaccillua Medium
Trace Metal Solution
EDTA 50.0 g
ZnSO, 22.0 g
CaCl, 5.54 g
5.00 g
4.99 g
•*
Ammonium molybdate 1.10 g
CuSO, 1.57 g
CoCl2 1.61 g
Distilled water 1.0 L
Adjust pH to 6.0 with KOH
MnCl,
-------
TABLE 12. (Continued)
01
ATCC 152
Thiobacciulltia Medium
Na,S,0..5H70
HH.Cl
MaCl«
K.HPO.
KH«POA
Fed.
0.08
10.0
1.0
0.5
0.6
0.4
0.02
1.0
1.0
g
g
g
g
g
g
g
g
L
ATCC 42
Deaulfovibrio Medium
Peptone ..
Beef extract
Yeast extract .
MaSO,
Na»SO,
24
Fe(NH.),,(SO.),
Tap water
Adjust pH to 7.0.
5 0
3 0
0 2
. 1.5
1.5
0 1
. 5.0
. 1.0
g
g
g
g
g
g
g
L
API SULFATB BROTH*
Sulfate-Reducing Medium
Sodium lacate, USP, millimeters . . 4.0
Yeast extract, grans 1.0
Ascorbic acid, grams 0.1
MgS04.7H20, grams 0.2
K.HPO. (anhydrous), grams 0.01
Fe(S04)2(NH4)2.6H40, grams .... 0.2
NaCl, grama 10.0
Distilled water, milliters . . 1,000.0
*Ten nl volumes are packaged in sterile
serum vlala by Difco Laboratories
-------
The mixed culture used in this study was not characterized except
that cell growth was obtained using the API broth and ATCC medium No. 450.
Model Organic Sulfur Compounds
Two representative organic sulfur-containing compounds were
selected for the laboratory tests: thiophene and 2-methyl thiophene. For
biological evaluation, these compounds were dissolved in reagent-grade toluene
and benzene appropriate to the test design. The thiophene and the 2-methyl
thiophene are analytical-grade reagents from Aldrich Chemical Company.
Experimental Technique
Standard microbiological equipment and technique were used through-
out the experimental work. In the experiments involving the Thiobacillus,
100 ml of the appropriate media were aseptically dispensed to sterile 500 ml
flasks fitted with a gas sparging tube. These flasks, after inoculation
and addition of the test reagent, were placed in a New Brunswick Gyrotory
incubator-shaker and continuously sparged with water-saturated compressed air.
Temperature was set at 30 C and oscillations at 80 rpms. With the aerobic
mixed culture, glass-stoppered flasks were selected so that the volatile
organic sulfur compounds would be retained in the medium and would not
evaporate during the test period. Oxygen was supplied to the actively growing
cultures by opening the flasks to the atmosphere for short periods of time
(10 minutes/day) during the experiment.
In the experiments involving the anaerobic-mixed culture, anaerobic
conditions were maintained by performing all experiments in the Difco-API-
sulfate-broth vials. Inoculations and additions of the organic sulfur
compounds were made with syringe injections. Caution was exercised to
exclude the introduction of air bubbles into the vials.
Detection and Estimation of Growth
Bacterial growth in the test media was determined by: (1)
observing the increase in turbidity, (2) measuring the change in pH,
(3) microscopic examination, and (4) observing the color change of pH
indicators.
53
-------
Characteristically, the sulfur utilizing bacteria are relatively
slow-growers — their doubling times are usually measured in days. Thus,
to positively establish that growth has occurred, a combination of the four
criteria mentioned above were used. Observing the increase in turbidity
alone is not a positive indicator since sulfur salts can either precipitate
or form colloids which can be misinterpreted as microbial growth. Therefore,
growth was confirmed by both measuring the medium's pH change and by phase
microscope examination. Several of the ATCC mediums for aerobic bacteria
include a pH indicator, chlorophenol red. This indicator responds to the
presence of sulfuric acid production and effects a color change within the
medium. The API sulfate broth includes a ferrous ion which precipitates as
black iron sulfide. The sulfide is produced when the bacteria reduce sulfate;
thus, the formation of the black precipitate indicates bacterial growth within
the medium.
EXPERIMENTAL RESULTS
The results from four shake-flask experiments performed with pure
cultures of Thiobacillus are summarized in Table 13. In Experiment 1, a series
of 10 flasks each containing 100 ml of ATCC Medium No. 238 was inoculated with
Thiobacculus thioxidans. After inoculation, toluene and benzene (at 0.25 and
0.50 percent) were added to duplicate flasks which were aerated and shaken at
30 C for ten days. No growth was observed throughout the test period in any
of the flasks, including the inoculated controls which did not contain any
test solvents.
In Experiment 2, 100 ml of ATCC Medium No. 152 were inoculated with
approximately 1 ml of Thiobacculus intermedus stock culture. The objective
of this experiment was to measure the growth rate for this organisms. Within
24 hours of incubation, growth was observed as indicated by the decrease in
the pH of the medium. Sulfuric acid is a metabolic by-product and its
presence was observed by the color change of the pH indicator, chlorophenol
red.
In Experiment 3, a series of 10 flasks each containing 100 ml of
ATCC Medium No. 152 was inoculated with Thiobaccilus intermedius at the 1
percent level. Toluene and benzene (0.25 and 0.50 percent) were added to the
54
-------
TABLE 13. SUMMARY OF SULFUR-UTILIZATING THIOBACCILUS GROWTH EXPERIMENTS, WITH AERATION AND SHAKING
Experiment
No.
Organism
ATCC Media
No.
Inoculum,
Experimental
Variables
Remarks and Results
T. thiooxidians
238
Ul
In
T. intermedlus
T. intermedius
152
152
T. thiooxidians
238
Toluene, 0.25 v/o
11 0.50 v/o
Benzene, 0.25 v/o
0.50 v/o
Controls (no solvent)
Controls (no solvent)
Toluene, 0.25 v/o
0.50 v/0
Benzene, 0.25 v/o
" 0.50 v/o
Controls (no solvent
Controls (no solvent)
No growth was observed in
either experimental or con-
trol flasks—no change in
pH or in optical density
after 10 days
Growth occurred within one
day after inoculation as
indicated by color change
of chlorophenol red
No growth occurred in any
of the flasks within 13
days
A series of 10 replicate
flasks were set up. The
objective was to establish
growth, then challenge the
active growing cultures
with toluene and benzene.
No growth occurred in any
of the flasks after 40 days
as measured by pH change
-------
test flasks and incubation at 30 C, with shaking, continued for 13 days.
No growth occurred in any of the flasks, including the controls.
In Experiment 4, a series of 10 flasks containing 100 ml of ATCC
Medium No. 238 was inoculated with Thiobaccilus thioxidans at the 4 percent
level. It was planned to obtain an active growing culture, then challenge
the test flasks with toluene and benzene. However, no growth occurred in
any of the flasks during the 40-day incubation period.
No experiments were conducted with the Desulforibrio desulfuri-
cons pure culture because a viable stock culture of the organism could
not be established.
Various attempts were made to stimulate the growth of the Thio-
baccillus stock cultures. Most were concerned with medium variations such
as: (1) substitute tap water for the double distilled water, (2) adjust
the pH to 3 with sulfuric acid rather than 4.2, as in ATCC medium No. 238,
(3) substitute cotton plugs for the urethane foams, (4) add sterile glu-
cose to the medium so that a supply of organic carbon is available, (5) oxy-
genate and carbonate the mediums with oxygen and carbon dioxide, (6) filter
sterilize, rather than steam sterilize, and (7) increase the volume of inocu-
lation to the fresh mediums. None of the variations tested improved the
viability of the stock cultures.
Since the stock cultures did not propagate reliably, the uniden-
tified mixed culture, isolated from a coal cleaning plant lagoon, was used
in all subsequent experiments.
The results of experiments in which the unidentified mixed cul-
ture was challenged with benzene or toluene and these same solvents con-
taining thiophene or 2-methyl thiophene are summarized in Tables 14 and 15.
The anaerobic bacteria in the culture tolerated a higher level of toluene
than benzene (Table 14), whereas the aerobic bacteria were more tolerant
of benzene. In no case was growth observed in cultures containing more
than 0.5 percent solvent.
Addition of the model sulfur compounds to toluene and benzene
increased the toxic effect of the solvents (Table 15). Toluene containing
as little as 5 percent thiophene completely inhibited growth of the bacteria.
2-methyl thiophene was slightly less toxic to the anaerobic bacteria.
56
-------
TABLE 14. RESULTS WITH UNIDENTIFIED MIXED CULTURE-EXPERIMENTS
IN DEFINED MINERAL MEDIA PLUS SOLVENT ONLY (7 Days)
Ul
Solvent
Concentration
(v/o)
Anaerobic
Aerobic
Toluene
Benzene
0.1
0.5
1.0
5.0
0.1
0.5
1.0
5.0
N.D. N.D.
N.D. N.D.
Medium
API Sulfate
broth
N.D. N.D. ATCC No. 450
•H-+, -H-, + indicates degree of growth (observed visually)
-, indicates no growth.
N.D. - Not determined.
-------
in
00
TABLE 15. RESULTS WITH UNIDENTIFIED MIXED CULTURE-GROWTH EXPERIMENT IN DEFINED MINERAL MEDIA
CONTAINING SOLVENT AND EITHER THIOPHENE OR 2-METHYL THIOPHENE (7 Day)
*
System
Anaerobic
Aerobic
Toluene (0.5 v/o)
( . Thiophene 2-Methyl Thiophene
System^' 5 v/o 5 v/o 10 v/o 20 v/o
Anaerobic - -H- - -
Aerobic - - N.D. N.D.
Benzene
(0.1 v/0) (0.5 v/o)
2-Methyl
Thiophene Thiophene Thiophene
5 v/o 10 v/o 20 v/o 5 v/o 10 v/o 5 v/o 10 v/o
N.D. N.D. + - -
+ + N.D.
2-Methyl
Thiophene
10 v/o
-
(a) Aqueous media given in Table 12.
-H-f, -H-, + indicates degree of growth.
-, indicates no growth.
ND - not determined.
-------
Solutions with two levels of benzene (0.1 and 0.5 percent) in which
bacterial growth occurred were tested. At the 0.1 percent level, addition of
1 percent 2-methyl thiophene to benzene inhibited growth of the bacteria.
Thiophene was less toxic to aerobic than the anaerobic bacteria, but growth
of bacteria cultures was inhibited by 10 percent thiophene present in benzene.
No growth was observed, at the concentrations of model sulfur compounds
tested, when the benzene level was increased to 0.5 percent.
The results of experiments in which the aerobic mixed culture
was challenged with benzene or toluene and these same solvents containing
thiophene but with no other source of sulfur are summarized in Table 16.
The aerobic organisms did not appear to grow in the medium that contained
no sodium thiosulfate despite the fact the sulfur was available in the
form of thiophene.
CONCLUSIONS
The following conclusions are based on the laboratory experiments
with naturally-occurring sulfur-utilizing bacteria:
(1) Both aerobic and anaerobic bacteria can tolerate
low levels (0.5 percent or less) of aromatic
hydrocarbons (toluene and benzene) but higher
levels completely inhibit growth of the bacteria.
(2) The addition of aromatic sulfur compounds (thio-
phene and 2-methyl thiophene) to toluene and
benzene at levels typical of those that might
be found in coal-derived liquids increases the
toxicity of these solvents.
(3) The naturally-occurring, aerobic bacteria used
in the experimental program could not utilize
thiophene as their sole source of sulfur.
(4) The prospects for using non-mutated bacteria to
remove organic-sulfur- and nitrogen-containing
compounds from coal-derived liquids appear to be
small because of the toxic nature of the aromatic
constituents of these liquids.
59
-------
TABLE 16. RESULTS WITH AEROBIC MIXED CULTURE-GROWTH
EXPERIMENTS IN DEFINED MINERAL MEDIA, WITH
THIOPHENE IN TOLUENE OR BENZENE AS SOLE
SULFUR SOURCE (9DAY)
__
Controls ,,*
(o)
Complete Incomplete Experimental
Medium with Medium, no Toluene 0.1 v/o"Benzene 0.1 v/o
Thiosulfate Thiosulfate 1 v/o Thiophene 5 v/o Thiophene
Growth -H-
PH (7 days) 5.7 7.5 7.9 7.9
(a) No solvents or organic model sulfur compounds added to salts medium
(ATCC No. 450).
(b) ATCC Medium No. 450 containing solvents and model sulfur compounds but
no thiosulfate.
(c) Initial pH 7.0.
+++, -H-, - indicates degree of growth.
-, indicates no growth.
60
-------
SECTION 7
USE OF AGGLOMERATION TO RECOVER CLEAN COAL VALUES
FROM PREPARATION PLANT WASTES AND THE
ENHANCEMENT OF PYRITE REMOVAL
INTRODUCTION
In the first phase of the program on fuel contaminants a review
was made of the utility and limitations of several technologies reported
in the literature for the removal, before combustion, of contaminants in
coal that produce pollutants when coal is utilized as a fuel. The oil
agglomeration technique was one of those that appeared to have the potential
for the removal of trace elements and under special conditions the physical
removal of pyritic sulfur liberated from coal during grinding. The results
of the review also suggested that the technique might be used to clean up
coal slurry pond sediments accumulated during coal cleaning plant operations
by removing the coal values they contain. The two areas are related in that
pyrite removal from old coal slurry pond sediments is enhanced by the auto-
trophic bacterial actions on the surface of pyrite that are supported in
these sediments. Understanding the cause of these effects could be
useful for developing a technique for removing pyrites from freshly ground
fine coal slurries by oil agglomeration. Conversely, any technique developed
to enhance pyrite removal from fresh coal could in effect be used to speed
up natures process and the recovery of low pyrite coal from coal waste
streams, both fresh and aged.
This part of the report summarizes the experimental results which
extend the state of the art of oil agglomeration to the recovery of coal from
coal cleaning plant wastes and the removal of pyrite from freshly ground coal
to the extent that they can be removed during float-sink analysis.* In this
* Detailed description of the experimental program are the subject of Volume 4
of the series on Fuel Contaminants. Only a summary of the findings are
presented here.
61
-------
program, both an Illinois No. 6 coal and an Ohio Pittsburgh Seam Number 8
coal were investigated.
Objective
The objectives of this study were to determine the feasibility of
coal recovery from coal cleaning plant wastes by immiscible fluid agglomera-
tion technique and to evaluate the effects of physical and chemical treat-
ments on enhancement of pyrite removal during agglomeration of these wastes
Background
Water immiscible liquids, usually hydrocarbons, have been used to
separate coal from its impurities. In principle it is an extension of
principles of froth flotation where the hydrocarbons wet the hydrophobic sur-
face of coal and the mineral impurities which are mostly hydrophilic remain
in aqueous suspension. Separation of the two phases takes place after
agglomeration or coalescence occurs and produces agglomerates of clean coal
containing the oil and an aqueous suspension of the mineral matter nearly
free of combustible material. Effective separation can be made with coal
with a size of minus 28 mesh (0.149 mm) and often with sizes too small for
any other recovery scheme. Hydrocarbon fluids such as kerosenes and fuel
oils have been found very effective for enhancing the separation of the mineral
matter from finely divided coal suspended in an aqueous slurry (i.e., reduce
ash). The selective agglomeration process is attractive because the coal
does not have to be dried after wet face mining, wet size reduction and
conventional coal preparation operations. In addition the agglomerated
coal can be readily dewatered by mechanical action providing energy trade-
offs between oil use and drying.
CONCLUSIONS
This study has demonstrated that coal recoveries of 90 percent or
greater are attainable from fine coal slurry wastes using the oil agglomera-
tion technique. These high levels of recovery are attainable from fresh
black water sediments generated during coal cleaning, aged sediments
62
-------
accumulated in slurry ponds and excavated, weathered, and partially dried
slurry pond sediments. The quality of the coal was good and had lower ash
and sulfur content than the coal shipped from the mine.
The residue from the agglomeration process contained between 2 to
5 percent of the oil used in agglomeration and very little coal, i.e.,
90 percent ash or greater. The residue suspension obtained after agglomera-
tion settled more rapidly than the original slurry. The residue material
appears to be well suited for land disposal.
The experimental results suggest the following environmental and
conservation advantages of the agglomeration process based on these current
results and on understanding of the relationships between various contami-
nants, mineral matter and organic matter.
• Recovered coal contains lower ash and reduced sulfur
(and trace heavy metals) than the parent coal.
• Recovered coal is easily dewatered and the product
remains dust free.
• Volumes of waste from coal cleaning facilities can be
significantly reduced resulting in less impoundment and
thus less land utilization.
• Waste management characteristics are improved since
residues are faster settling, more compatible with soil
since they are not altered, and less prone to support
bacterial activity that cause acid drainage.
• Rather than disposal of the concentrated mineral
matter it may be amenable as a raw material for
ceramics, cement and other construction purposes or
for the recovery of useful mineral values such as
alumina.
Coal derived liquids such as the distillate recovered from SRC
dissolver product is able to yield 90 percent or greater coal recoveries.
Its behavior is not any different than that of the petroleum derived liquids
such as No. 2 fuel oil, No. 6 fuel oil-kerosene mixture and kerosene alone.
With regard to the enhancement of pyrite removal during agglomer-
ation, use of sodium metaphosphate will remove 42 percent of the pyritic
63
-------
sulfur from freshly ground minus 48 mesh coal. This is the same as that
obtained by float-sink analysis suggesting removal of all the liberated
pyrite in the ground coal. Equally good results were obtained when a coal
slurry was treated with oxygen in the presence of sodium carbonate at 25 C.
These same treatments could be applied to black water and slurry pond
sediments to enhance pyrite removal during coal recovery.
An estimation of the product cost recovered from agglomeration
step as an add-on to an existing coal cleaning plant or as a portable
facility is about $14/ton.
Benefits that might be realized by the adaptation of the oil
agglomeration technique by the coal industry include the following:
• Reduces the hazards and environmental impact of exist-
ing coal slurry ponds.
• Recovers valuable resources from wastes.
• Allows environmental control of effluents from coal
preparation plants faced with increased throughput
to meet energy needs and environmental constraints.
• Applicable to coal preparation plants used to
prepare feed for coal conversion plants.
• Dewaters wet coal fines.
• Permits direct application of technology developed
for clay stabilization to residues.
• In principle, may reduce risks of catastrophic
dam failure when used as a replacement for raw wastes.
64
-------
SECTION 8
EXTRACTION OF CLEAN FUELS FROM
COAL-DERIVED LIQUIDS
The objective of this study is to develop a solvent extraction
process for treating coal-derived liquids to produce a fuel that is low in
contaminants (ash, sulfur, and nitrogen).
INTRODUCTION
Coal-derived liquids, such as those produced by the Solvent
Refined Coal (SRC), H-Coal, and Synthoil processes, are highly aromatic.
Their average molecular weight depends on the coal liquefaction process.
A primary coal liquid prepared in any of these processes consists mostly
of two- and three-ring aromatics and heterocyclic compounds. Only 60 per-
cent of one type of SRC can be vacuum distilled in the range of 90-300 C
as shown in Table 17. This limited recovery during distillation of coal
liquids is probably due to thermal degradation of the coal liquid, the by-
products being gases (e.g., methane and ethane) and a residue which is
highly graphitic in nature and which holds the sulfur and nitrogen deep in
its structure. This residue is not suitable for further liquefaction,
recycle, or hydrotreatment for sulfur and nitrogen removal. Furthermore,
separation by distillation alone depends solely upon the vapor pressure of
the constituents overhead. While distillation is well suited for petro-
leum upgrading, its use for upgrading coal-derived liquids is considered
marginal. This is especially true when one considers the separation of
non-coal solids, attrited catalysts, and unreacted coal from the liquefied
coal. The distillation temperatures are high enough to promote hydro-
cracking of the high boiling hydroaromatics and tend to promote reactions
leading to by-product hydrocarbon gas, char formation, and hydrogen loss.
Use of selective extractions with solvents conducted at rela-
tively low temperatures where degradation is minimized should permit
65
-------
TABLE 17. CONSTITUENTS OF SOLVENT REFINED COAL
(20)
Feed
Colonial Mine, Hopin County, Kentucky
Sulfur 3.33 percent, dry basis
Nitrogen 1.62 percent, dry basis
Product
Gas
Water
Volatiles
Cold trap oil
Cut I oil
(distilling to 90 C, 3 nmHg)
Cut II oil
(distilling 90 C to 300 C, 3 nmHg)
Heavy residue
Vacuum bottom
Dry minerals
Feed
5.29
2.29
0.19
1.00
6.44
59.58
100.77
Percent
Raw Coal
14.58
6.31
0.52
2.76
6.34
0.63
50.52
20.45
102.11
(a)Greater than 100 percent because of hydrogen addition.
66
-------
isolation of a hydrogenated, contaminat-free fraction (a clean fuel). Such
fractionation would enhance any subsequent upgrading by allowing the use of
hydrotreatment methods which take into consideration the chemical makeup of
the various fractions.
Ideally, extraction should lead to portioning the primary coal
liquid into streams containing:
1. Aliphatics, naphthenes, and light aromatics
2. Aromatics, including polynuclear compounds
3. Asphaltenes, preasphaltenes, and ash components.
To minimize hydrogen consumption and the production of gas and char, the
first group should not be subjected to the hydrogenation/hydrocracking
necessary to produce smaller molecules from condensed ring systems. Nor
should the third group be treated for heteroatom (S and N) removal by hydro-
Lreatment in the presence of the first two, since this would be costly in
hydrogen consumption due to hydrogenation of the nonheterocyclic compounds.
This study reviews the extraction of coal-derived liquids with
such solvents as benzene, tetralin, toluene, and acetone and summarizes the
background information leading up to the experiments that suggested extrac-
tion with light hydrocarbon solvents near or above their critical pressure
and can be used to isolate the fuel contaminants from clean fuel.
The development of some theoretical basis for the super critical
extraction technique is given in the Appendix.
BACKGROUND
«
Solvents
Various solvents have been used to extract components of coal and
/2)
coal liquids . The solvents can extract oils (Heptane soluble material),
resins (methanol soluble), and asphaltenes (benzene soluble) from coal
liquids.
Solvents are of various types, e.g., aliphatic, aromatic, hydro-
aromatic, as shown in Table 18. The aliphatic solvents are considered poor
solvents for coal liquids primarily because the coal and coal liquids con-
tain only small amounts of oils soluble in aliphatic solvents. But the
aromatic solvents (e.g., benzene) dissolve large fractions (60-70 percent)
of coal liquids. The other classes of solvents, the hydroaromatic (e.g.,
67
-------
TABLE 18. SOLVENT TYPES FOR SELECTIVE
SOLUBILIZATION OF COAL
Solvent Type
Function
Aliphatic
Heptane
Hexane
Pentane
Propane
Ethane/Ethylene
Aromatics
Benzene
Toluene
Xylene
Cresol, Phenols
Aromatics. Active 0. N, H Group
Pyridine
Tetrahydrofuran (THF)
Tetralin, Anthracene Oil
Other Solvents
Acetone
Trichloroethylene
Extract oil
Extract asphaltenes
Extract asphaltenes, solublize
more coal
Extract asphaltenes/paraffins
(^Extracts oil under conventional solvent extraction conditions.
68
-------
anthracene oil, tetralln), aromatic bases (e.g., pyridine), and aromatic
acids (e.g., phenols), can solubilize undissolved coal present in the coal
liquid. The hydroaromatic "hydrogen donor" solvents improve the quality of
the coal liquid by increasing the H/C (atomic) ratio of the coal liquid.
During such an improvement in the H/C ratio, the coal liquid is partially
desulfurized.
Aliphatic solvents appear to be solvents in the true sense for
coal and coal liquids, that is, they are chemically inert towards both the
extract and the residue (unreacted coal). The extract and residue ob-
tained by the action of aromatic solvents are more susceptible to oxida-
tion than the original feed. This indicates that aromatic solvents
probably extract coal and coal liquids by chemical action.
(21)
Rodgers evaluated various types of solvents for extracting
feeds of SRC and COED (Char Oil Energy Development) products for periods of
10-20 minutes at room temperature and found that aromatics containing a
nitrogen atom in the ring structure (pyridine, quinoline) and polar mole-
cules containing oxygen (acetone, tetrahydrofuran) were excellent solvents.
Solvents like cresol, pyridine, and quinoline extracted large amounts of
COED and SRC products. The solid residue, after the extraction, had an
ash content of 21 percent and 50 percent, respectively, as in Table 19.
The sulfur in the solid residue varied from 2.3 to 7.4 (weight percent)
and the ratio of sulfur in solids to sulfur in feed was in the range of
0.28 to 0.57. Some solvents, e.g., cresol and pyridine, that produce 4
to 9 weight percent solids compared to the 22 weight percent solids with
acetone and toluene did not selectively concentrate the sulfur in the
soluble or insoluble fraction of the feed. The iron content of the dry
solids, as in Table 18, is too low to account for sulfur as FeS or FeS_.
This indicates that sulfur in solid residue is organic in nature.
Solvent Deasphalting
Solvent deasphalting of petroleum fractions has been used exten-
sively in petroleum refining. In such a process, petroleum residuum is
contacted with a light hydrocarbon (the solvent), such as propane, normally
69
-------
TABLE 19. FILTRATION SOLIDS FROM SOLVENT
FRACTIONATED COAL LIQUIDS<21)
Solvent
Dry Solids
Collected,
wt percent
Feed
Ratio, (a)
[I]
Analysis of
Dry Solids, wt percent
Ash Sulfur
Iron
SRC (Solvent Refined Coal)
Toluene
Acetone
Cresol
Pyridine
Te trahyd ro furan
Quinoline
22.1
20.0
4.0
4.6
8.6
4.5
SRC (Feed)
0.55
0.67
0.25
0.26
0.38
0.28
1.0
14.1
12.8
50.1
55.0
33.4
57.7
3.2
COED (Char Oil Energy Development)
2.3
2.6
5.7
5.3
4.2
5.8
0.9
3.1
6.9
9.0
6.9
3.9
8.3
0.5
Toluene
Acetone
Cresol
Pyridine
Tetrahydrofuran
Quinoline
COED (Feed)
13.6
9.0
7.7
9.0
9.1
9.1
0.39
0.28
0.38
0.28
0.28
0.28
1.0
13.6
20.4
21.8
21.0
21.1
21.5
2.1
4.4
4.6
7.4
4.6
4.6
4.6
1.5
2.8
13.7
5.0
15.3
14.5
14.6
(a) A is the sulfur in solid product, B is sulfur in feed
(b) Example: Calculation of A/B ratio. Feed, SRC, 100 g; Solvent, Toluene.
Then A - (22.1 x 0.023) and B - 0.9 and the ratio A/B - 0.55.
70
-------
in a temperature range of 65-120 C and at pressures of up to 600 psi.
Deasphalting is a liquid-phase extraction process. The phases are sepa-
rated and the extract is flashed to recover the solvent, which is then
recompressed and recycled.
The amount of asphalt free oil produced during propane deasphalting
of petroleum residuum depends on the operating temperature. For a
residuum there is a temperature that will produce almost a deasphalted
residuum. Some typical data on solvent deasphalting are shown in
Table 20. The primary objective of deasphalting the residuum is to
remove metals, and it is found that sulfur and nitrogen compounds are
also removed.
Solvent Extraction Above Critical Temperature
One novel method that may produce a contaminant free product from
a coal liquid, e.g., SRC, extracts at temperatures above the critical tem-
perature of the solvent. The extraction thus occurs when the solvent is in
the gas phase.
This mode of solvent extraction is being used by the National Coal
Board, U.K., to study the structure of coal. In one of their recent studies,
up to 17 percent of a low rank coal was extracted at 350 C. The extract
consisted of aromatic hydrocarbons and the yield of gas was very small. This
indicated that during such an extraction the degradation of coal to gas was
f22,22a)
small.
In related study on the extraction of two- and three-ring aromatic
hydrocarbons that are present in coal liquids, it was found that ethylene
(23)
was a good solvent for removing these hydrocarbons from mixtures . The
degree of extraction was a function of temperature and pressure. In one
pressure range investigated, higher pressures increased the solubility of
the hydrocarbon to a value which was greater than the concentration at the
normal vapor pressure.
Diepen and Scheffer^ 'investigated the extraction of naphthalene
(2 ring aromatic) by ethylene (solvent). At extraction conditions of 12 C
and 100 atm, the naphthalene extracted was 25,600 times that obtained during
conventional extraction at 12 C and 1 atm pressure. Higher pressures increased
71
-------
TABLE 20. TYPICAL DATA ON SOLVENT DEASPHALTING OF RESIDUUM
(20)
Crude Oil Type
Residuum feedstock
Specific gravity at 16 C (60 F)
Sulfur (weight percent)
V (ppm)
Ni (ppm)
Cu + Fe (ppm)
Heptane insolubles (weight percent)
Deasphalted oil product
" Yield (volume percent of feed)
Sulfur (weight percent)
V (ppm)
Ni (ppm)
Cu + Fe (ppm)
Heptane insolubles (weight percent)
Asphaltene fraction
Sulfur (weight percent)
Heptane insolubles (weight percent)
Metals rejection to asphaltene fraction (percent)
Sulfur rejection to asphaltene fraction (percent)
Gach Saran
-
1.030
2.66
372
120
10
51.5(a)
1.89
8.0
7.6
< 0.006
3.25
20
98.5
66
Gach Saran
1.030
2.66
372
120
10
75(b)
2.25
89
40
< 0.006
3.8
48
81.0
41
West
Texas
0.986
27.6
16.0
14.8
66.0
1.3
1.0
0.8
96.6
California
1.027
136
139
94
52.8
2.3
8.1
3.5
98.2
(a) This yield value is low in V and Ni.
(b) This yield value is high in V and Ni.
-------
The solubility and a pressure versus composition plot for the ethylene-
naphthalene system, Figure 6, shows that at 35 C and 1660 psig, 6.1 weight
percent of naphthalene is solubilized in ethylene compared to about 2 weight
percent at 1500 psig.
(24)
Ellis extracted a mixture of hydrocarbons, dodecane and naph-
thalene by ethylene. In this extraction, at 75 C and 2100 psig, the vapor
contained 34 weight percent of the dodecane/naphthalene mixture, as is
shown in Figure 6. Also in this study, the concentration of the mixture
in the vapor was found to be a function of pressure. The hydrocarbons
extracted increased from 1.6 weight percent at 1175 psig to 34 weight per-
cent at 2100 psig.
Zhuze^25) used a propane-propylene mixture for deasphalting
petroleum. At extraction conditions of 100 C and 1500-1600 psig the yield
of deasphalted petroleum from petroleum asphalt and cracked residue was
30 and 50 weight percent; respectively. Also it was shown that diverse
products of petroleum refining, from residues to lighter fraction, could be
deasphalted with compressed gases.(25)
The above-mentioned solvent extraction studies utilized ethylene
and propane-propylene as the solvent. In similar extractions other gases
(e.g., methane, carbon dioxide, ethane, carbon tetrafluoride) were success-
fully used.
The extractions of 2 to 3 ring aromatics by various gases show
that the critical temperature of the gas may be an important criterion in
determining the degree of extraction by the gas. When phenanthrene (M.W. 178)
was extracted at 40 C (313 K) and 40 atm by certain gases, e.g., nitrogen,
methane, carbon tetrafluoride, ethylene, carbon dioxide and ethane, it was
observed that the extract (phenanthrene) increased markedly with increasing
critical temperature of the gases as shown in Figure 7. However, the
extraction with ethylene was greater compared to that by carbon dioxide
even though ethylene and carbon dioxide have similar critical tempera-
tures. This suggests that the physical and chemical nature of the gas
(solvent) molecule should be considered when selecting an effective solvent
system<26'27>.
73
-------
30
20
U
tt
10
I I
I 4 I I I I
Q Qulnoline-Dodecane Mixture
^ Cyclohexane-Beozeae
Q Quiaoline
^ Naphthalene
A Dodecane-Naphthalene Mixture
A I
3000
2000
Preuure, psla
1000
FIGURE 6.
SOLUBILITY OF HIGH MOLECULAR WEIGHT HYDROCARBONS IN ETHYLENE
GAS AS A FUNCTION OF TEMPERATURE AND PRESSURE (2 3,24)
74
-------
30
oo
0)
4J
U
2 20
AJ
4J
60
304 K
200 300
Critical Temperature, K
FIGURE 7. EFFECT OF CRITICAL TEMPERATURE OF GASES ON EXTRACTION OF A
COAL TAR COMPONENT (PHENANTHRENE) AT 40 ATM AND 40 C
(313 K)(27)
-------
The Effect of Pressure on Solubility—
In various binary and ternary systems, it is observed that the
solubility of high-molecular-weight components in the solvent, at tempera-
tures above the critical temperature of the solvent, increases rapidly with
pressure (Figure 6). It is observed that the increase in solubility is
(24)
greater for nonpolar components
The Effect of Temperature on Solubility—
The effect of temperature, as compared to pressure, on the solu-
bility of high-molecular-weight components is not well documented. Diepen
:ii
.(28)
(23)
and Scheffer show that in the pressure range of 60-80 atm the solubility
was increased as temperature was raised from 12 to 35 C. Todd and Elgin
observed similar results.
Thus, it appears that at relatively low pressures, increasing tem-
perature slightly above the critical temperature (of the solvent) decreases
the solubility, but at higher pressures, increasing the temperature will
increase the solubility of the hydrocarbon in the solvent (gas).
EXPERIMENTAL PROGRAM
The coal liquid selected for this study was obtained from the
Wilsonville, Alabama, SRC plant. The sulfur, nitrogen, and ash concentra-
tions in the feed coal are given in Table 21. This coal liquid, as it leaves
the dissolver at the SRC plant, is usually flashed to remove most of the
solvent. The solvent is then recycled in the process. The residue obtained
after the initial flashing operation is filtered at elevated temperatures
and pressures to remove the ash and the product is SRC (solvent refined
coal).
The experimental program included preliminary investigations to
determine the feasibility of producing a clean fuel (low in ash, sulfur,
and nitrogen) from a coal-derived liquid. The clean fuel was extracted
from the coal liquid by using the following hydrocarbon gases: methane,
ethane, ethylene, propane, propylene, and butane. The selection of the
above gases for the program was based on the following criteria:
76
-------
TABLE 21. FEED MATERIAL FOR EXTRACTION
OF COAL LIQUIDS
Source: Southern Services, SRC Technical Report 6,
Wilsonville, Alabama SRC Pilot Plant
Feed Coal
- W.
Ash
Sulfur
Sulfur
Sulfur
Kentucky
(pyrite)
(sulfate)
(organic)
Nitrogen
Dissolver
Ash
0.22
1.48
2.58
#14
8.
1.
0.
1.
1.
93%
12%
10%
85%
14%
Product (a)
S
0.23
0.35
0.49
N
0.74
0.76
0.76
Used
Remarks
on Runs
Used on Runs
Used
1-17
18-35
for Distillation
(a) Values for smaller samples taken from main
sample storage as percent.
77
-------
o These gases are readily available, e.g., a product
of coal liquefaction
• These gases are readily recoverable from the heavier
product
e This gas-liquid extraction will require conventional
extraction conditions, e.g., temperatures and pressures
similar to those being used in present refinery
extraction operations like propane deasphalting.
Nitrogen gas was used only for comparison purposes.
Extraction of Unfiltered Coal Liquid
The experimental arrangement for the extractions is shown in
Figure 8. The high pressure source was a small (20-25 scfm) compressed air-
operated high pressure pump.
A typical experiment consisted of charging the reactor with 50 g
of the coal liquid and then introducing the gas, e.g., ethane, into the
reactor. The reactor was then heated to the reaction temperature and
maintained at that temperature for 10 minutes (reaction time). At the end
of the reaction time a valve was opened and the gas and dissolved coal
liquid flashed into ice-cooled condensers. The clean fuel was then
recovered from the condensers.
The weight of the gas introduced into the reactor was calculated
from the following equation:
pv = ZnRI,
where
p - pressure, atm T • temperature, K
v • volume, 1 R » gas constant, 0.082
Z » compressibility factor.
The volume of gas in the reactor, in some cases, was verified by slowly
venting the contents of the reactor through a wet test meter.
The gas to feed (coal liquid) ratio was in the range of 0.8 to 8.
The extraction experiments and reaction conditions are summarized in Table
22. In this table, the column titled "percent extracted" gives the quantity
of clean fuel extracted form the coal liquid based on its weight loss. The
78
-------
Thermo -
couple XT
Stirrer
motor
VO
Bleed
valve
MS)
To high
pressure
source
o
Heater
1000-ml
stirred tank reactor
(6000 psi, 300 C)
Vent,
analysis
Condensers
FIGURE 8. EXPERIMENTAL ARRANGEMENT FOR SOLVENT EXTRACTION
AT ELEVATED TEMPERATURES AND PRESSURES
-------
quantities extracted when the gas was methane, nitrogen, ethane, or ethylene,
at pressures of 3400 psig, are shown in Figure 9. It is seen that ethaane
and ethylene were equally effective in extracting a clean fuel comprising as
much as 76 percent of the feed at 300 C. The yield of clean fuel is a func-
tion of temperature. The extraction capacity of ethane/ethylene increased
from 12 percent at 100 C to 76 percent at 300 C. Nitrogen and methane showed
poor extraction results with only 12 percent and 32 percent clean fuel
extracted respectively at 300 C.
The results of extraction of clean fuel from the coal liquid with
such as propane, propylene, and butane are compared with the results for
ethane and ethylene in Figure 10. Butane was effective in extracting 83 per-
cent of the feed at a temperature of 300 C. The effect of pressure on the
degree of extraction for butane and ethylene is compared in Figure 11. This
comparison shows that at a pressure of 3400 psig ethylene extracted 76 per-
cent of the feed, and butane gave similar results but at a pressure of 1500
psig.
Atmospheric pressure distillation experiments with 120 to 160 g of
feed (sp. g. 1.075) at a maximum pot temperature of 315 C, yielded about 50
g of light liquids. This yield is only about half that obtained by extrac-
tion with gases. Hydrogen, methane, and ethane were detected in measurable
amounts in the overhead vapors, thus indicating some degradation during
the recovery.
Quality of Liquid Product (Clean Fuel) and Residue
Samples of the condensed liquid product and residue (material
remaining in the reactor) selected from experiments with the highest extrac-
tion yields were analyzed to determine their quality. The analyses of the
products and residue after extraction of the coal liquid with ethane and
ethylene are given in Table 22. The sulfur in the product is 20 percent
of that in feed, the nitrogen varies between 40 to 80 percent of that in
the feed, and the product is essentially ash free. The residue had 5 to 20
times as much ash as feed material. The analyses of products and residue
after extraction of feed with butane are also given in Table 22. The ash
content of the liquid product is essentially zero whereas the residue now
80
-------
e
0)
01
o.
01
4J
o
2
4J
o
•o
o
M
PL.
S!
iH
O
n
oa
n
01
0)
o
o
Wi
PU
en
80
70
60
50
30
20
10
Nitrogen
W Methane
r~j Ethane
Ethylene
100 200
Extraction Temperature,
300
FIGURE 9. EXTRACTION OF SRC PROCESS DISSOLVER PRODUCT BY GASES AS
A FUNCTION OF TEMPERATURE (3400 psig)
81
-------
80
70
c
0)
u
b
0)
a
0)
4-1
U
a
Ui
4J
x
u
u
•o
o
0)
iH
O
a
09
a
a>
a>
u
2
Qu
CO
60
50
40
30
20
10
D
I
I
O Butane
£3 Propylene
Q Propane
^ Ethylene/
Ethane
200 300
Extraction Temperature, degrees C
FIGUKE 10. EXTRACTION OF SRC PROCESS DISSOLVER PRODUCT BY GASES AS
A FUNCTION OF TEMPERATURE (PRESSURE te 1500 psig)
82
-------
c
0)
o
h
0)
o.
-------
TABLE 22. SUMMARY OF REACTION CONDITIONS, PERCENT EXTRACTED, AND
PRODUCT RECOVERY FOR GAS EXTRACTIONS
Product Recovery
Seawla
No.
32727-15
3:727-16
32727-17
32727-18
32727-19
32727-23
32727-24
32727-25
32727-27
32727-28
32727-29
32727-31
32727-32
32727-33
32727-34
32727-35
32727-36
32727-37
32727-38
32727-39
32727-40
32727-42
32727-43
32727-45
32727-46
32727-48
32727-49
32727-50
32727-51
32727-52
32727-53
32727-54
32727-56
32727-57
Run
No.
1
2
3
4
5
6
7
8
9
10
11
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Weight
Loee
0
3
1.1
0.9
0.6
5.6
36.1
38.3
4.4
6.2
21.0
0
11.1
38
5.6
7.5
22.1
11.0
19.9
36.6
39.7
42.0
28.7
19.2
24.4
5.4
32.5
19.3
3.0
26.8
26.7
18.2
16.4
27.3
Percent
Extracted
0
6
2.2
2.0
1.2
11.2
72.2
76.6
8.8
12.4
42.0
0
22.2
76
11.2
15
44.2
22.0
39.8
73.2
79.4
84.0
57.4
38.4
48.8
10.8
65.0
38.6
6.0
53.6
53.4
36.4
32.8
54.6
Ut of Gee
Ut of SRC
3
4.6
3
4.6
3.0
7.8
4.5
4.5
2.14**'
7.8
1.2(b)
0.98
1.6
4.5
1.6
0.77
4.5
4.5
4.5
4.5
4.5
5.53
5.53
4.00
4.00
4.19
4.19
2.1
5.3
5.3
5.3
7.3
2.8
4.5
Te«p.
C
193
114
18
18
100
69
184
280
177
100
300
100
300
300
300
200
300
200
300
300
300
300
200
200
300
200
300
300
200
200
200
200
300
200
Preeeure,
Initial
2000
3000
2000
3000
2000
1500
700
700
700
1500
700
1500
1000
700 '
1000
500
700
700
700
700
700
f k
4
4Cc)
4(0
/ 0
49nO
6:iO
3600
Gee
»2
H2
»2
»2
H2
C2H6
C2H6
C2H6
C2H4
C2H4
C2»4
CHA
CH4
C2H4
N2
H2
C2H4
C2H4
C2H4
C2H4
C2H6
C4H10
C4H10
C3H6
C3H6
C3H8
C3H8
CH4
C2H4
C2H4
C2H6
C4»10
C4»10
C2H6
Final
Weight,
B
ND
ND
ND
ND
ND
222.6
229.2
235.5
223.6
229.1
232.9
157.5
164.5
171.6
162.8
158.9
167.5
160.6
168.2
171.9
162.1
157.7
162.1
154.8
156.8
152.5
169.4
163.6
—
161.3
158.5
153.2
152.5
158.5
Initial
Height,
8
ND
ND
ND
ND
ND
221.5
225.1
225.5
227.5
225.1
222
157.0-
157.1
157.9
157.9
158.0
158.3
157.8
158.8
157.8
150.2
151.0
150.2
150.9
148.6
150.7
150.8
150.6
—
151.0
149.0
150.1
148.4
148.0
Prod.
B
ND
ND
ND
ND
ND
1.1
4.1
10.0
ND
4.0
10.9
0.5
7.4
13.7
4.9
0.9
9.2
2.8
9.4
14.1
11.9
6.7
11.9
3.9
8.2
1.8
18.6
13.0
—
10.3
9.5
3.1
5.1
10.5
Autoclave
Uaeb,
B
ND
ND
ND
2.35
ND
ND
ND
ND
—
3.1
5.9
2.5
2.5
4.3
4.6
4.4
8.5
9.8
7.5
0.9
0.9
1.2
5.8
9.8
7.8
5.8
7.8
6.1
—
5.3
4.5
4.6
6.4
5.8
Total
Recovery,
B
—
—
—
—
—
1.1
4.1
10.0
—
7.1
16.8
3.0
9.9
18.0
9.5
5.8
17.7
12.6
16.9
15.0
12.8
7.9
17.7
13.7
16.0
7.6
• 26.4
19.1
—
15.6
14.0
7.7
11.5
16.3
00
(a) Calculated from 1700 pal preesure, 177 C.
(b) Calculated ttom 1500 pai preaeure, 300 C.
(c) Cu ft.
-------
has 7.7 weight percent ash compared to 1.5 weight percent ash in the feed.
The sulfur and nitrogen concentrations in the residue are 15 and 4 times,
respectively, those in the product. Thus, by using butane extraction yields
as high as 84 percent of the feed are feasible and the product is low in
ash, sulfur, and nitrogen.
The results in Table 23 may only be representative of the minimum
quality of products obtained because the amounts of product recovered in
Experiments 32727-25 and -42 represent only about 20 percent of the material
extracted, i.e., the heavy oils. The remaining part of the extract (i.e.,
light oils) were not captured under the experimental conditions employed.
Despite this experimental limitation, an estimate of the sulfur and nitrogen
contaminants the light oils contained was made by performing a sulfur and
nitrogen balance. Since the quantities of feed material, residue from
extraction the product captured in the traps were measured for Experi-
ments 32727-25, -29, and -42 and they had been analyzed for sulfur and
nitrogen, it was a simple matter to estimate the quantity of sulfur and
nitrogen they contained and their percentage. The results of these calcu-
lations are given in Table 24. The sulfur was estimated to be 0.14 percent
in the volatilized material and 0.23 percent in the feed (see Table 21),
while the nitrogen decreased to 0.65 percent from 0.74 percent. In Experi-
ment -42, the sulfur was estimated to be 0.28 percent, while the nitrogen
was about 0.71 percent (compared to 0.35 percent and 0.76 percent in the
feed). If these estimates are valid, then sulfur values appear to remain
with the residue while the concentration of nitrogen is only slightly less
than it was in the feed material. Further characterization of these frac-
tions is needed before this can be shown conclusively. Furthermore, the con-
ditions during extraction may not have been optimum for maximum rejection of
sulfur and nitrogen compounds.
Cracking of Light Oils
Gas samples, taken after the pressure was released, were analyzed
by gas chromatography. The analysis showed that no apparent chemical degrada-
tion occurred as measured by the presence of light gases, e.g., hydrogen,
methane, and ethane. This is in contrast to the gas analysis of the products
during atmospheric pressure distillation of the coal liquid at 300 C, when
85
-------
TABLE 23. ANALYSIS OF LIQUID PRODUCT AND RESIDUE AFTER EXTRACTION OF COAL LIQUID
Experiment
No. /Gas
32727-257
or! Ethane
32727-297
Ethylene
32727-427
Butane
Percent of
Coal Liquid
Extracted
76
36
83
Analysis ,
Product, weight percent Residue, weight percent
AshC HNS Ash C HNS
0 85.7 7.4 0.3 0.07 6.9 87.6 6.1 1.3 0.58
0 89.0 7.7 0.6 0.04 0.1 89.4 6.7 0.8 0.24
0 88.7 7.3 0.3 0.11 7.7 75.7 4.9 1.3 1.64
-------
TABLE 24. SULFUR AND NITROGEN BALANCE FOR SELECTED EXTRACTION EXPERIMENTS
oo
Experiment
No. (Run)
32727-25(8)
(Percent of
32727-29(11)
(Percent of
32727-42(23)
(Percent of
Feed
Sulfur
0.121
Feed)
0.125
Feed)
0.269
Feed)
(~\
a)
•
Total ,.v
Volatilized Products v '
g
Nitrogen
0
0
0
.388
.403
.416
Residue, g
Sulfur
0.073
(60)
0.084
(67)
0.153
(57)
Nitrogen
0.163
(42)
0.279
(69)
0.121
(29)
Product
Sulfur
0.007
6
0.004
(3)
0.007
(3)
Caught, g
Nitrogen
0
0
0
.030
8
.065
(16)
.020
(5)
Sulfur Nitrogen
g
0.041
(34)
0.037
(30)
0.109
(40
% 8
0.14 0.195
(50)
0.43 0.059
(15)
0.28 0.275
(66)
%
0.65
0.68
0.71
(a) Estimated from analysis of single sample.
(b) Estimated based on sulfur and nitrogen differences and weight of material not captured.
-------
hydrogen, methane, and ethane were detected in measurable amounts. The
presence of such gaseous products suggests that distillation at 300 C will
produce gases by cracking light hydrocarbons in the feed, thereby reducing
the yield of total liquids. This does not appear to be the case when
gaseous extraction of coal-derived liquids is used.
OPTIMUM EXTRACTIONS OF CLEAN FUEL
FROM A COAL LIQUID
Compared to a liquid fuel, the feed coal is deficient in hydrogen.
A typical high-volatile bituminous coal has a hydrogen to carbon (H/C) atom
ratio of 0.8, whereas a liquid fuel, e.g., fuel oil No. 1, has a hydrogen
to carbon (H/C) ratio of 1.88. This indicates that hydrogen must be added
to coal to produce liquid products. For a self-contained process, this
hydrogen must be produced from the coal; this is possible by reacting coal
with steam as:
C + H20 -f CO -I- H2.
Although this reaction is endothermic and thus requires an external source
of energy, the stoichiometry of this reaction still may be used to estimate
the amount of carbon in coal that is needed for hydrogen production. The
percent of carbon in coal required for hydrogen production when 40, 50, and
60 percent of the coal is converted to a fuel oil with various H/C ratios
can be determined from Figure 12. When 40 percent of a high volatile
bituminous coal (H/C - 0.797) is converted to a liquid fuel (H/C - 1.6),
the rest is a residue with a H/C - 0.8, then 14 percent of the carbon in
the coal must be converted to hydrogen. Similarly when 60 percent of the
coal is converted to a liquid (H/C « 1.9) then 33 percent of the coal must
be converted to hydrogen to supply the extra hydrogen required by the fuel
oil. These quantities of coal required for hydrogen production will be
higher when due consideration is given to the heat requirements of the
carbon (coal) - steam reaction.
With consideration given to these classical requirements for a
self-contained process which is in hydrogen balance overall, it is generally
agreed that a coal liquefaction/liquid recovery step that converts greater than
50 percent of the feed coal to the fuel oil type liquids is not practical.
88
-------
01
§
•O
O
c
0)
00
o
30
20
o>
(0
a)
O
O
g 10
1
o
I I I
High Volatile Bit. Coal
H/C - 0.797
Fuel Oil # 1
Fuel Oil # 4
Fuel Oil # 6
I
60%
(a)
(a)
Carbon in coal used
for fuel oil produc-
tion, percent feed
carbon.
1
1
1
0.5
1.0 1.5 2.0
H/C Atomic Ratio
2.5
FIGURE 12. CARBON IN COAL USED FOR FUEL OIL PRODUCTION (BY
HYDROGENATION) AND THE REQUIREMENT FOR CARBON IN
SAME COAL FOR HYDROGEN PRODUCTION
89
-------
However, given the opportunity to recover over 80 percent of the converted
feed as a clean fuel as is the case in the liquefaction/extraction process
reported here, shortages in the coal-hydrogen balance might be made up by
adding coal to the residue gasifier. This is especially true when the
liquefied coal-is recovered with minimal thermal decomposition and most of
the hydrogen in the converted coal is retained as the clean heavy fuel
formed in the process. Recycle oil requirements of a liquefaction process
could be met by the recovered extract.
PROCESS AND ECONOMIC CONSIDERATIONS
The experimental data have shown that a clean fuel can be
extracted from a coal liquid. In this evaluation, design data for a clean
(29)
fuel demonstration plant were taken and a comparison is made between a
conventional process that uses filtration and distillation and a new process
that uses extraction by gases as means for separating a clean fuel from the
bulk feed. The essential components of a conventional plant flew sheet,
with feed and product quantities, are shown in Figure 13. The coal to recycle
oil ratio is 0.5 and the coal dissolution occurs at 850 F (490 C) and 1000
psig (68 atm).
A hypothetical flow sheet for the new plant that would use extrac-
tion by gases is shown in Figure 14. A gasifier is common to both processes
since part of the coal liquid residue from the extraction must be gasified
to produce hydrogen. The liquefaction vessel remains unchanged, but two options
are considered as alternates to filtration and distillation: (1) extraction
with a light gas, e.g., ethane or ethylene, and (2) extraction with a heavier
gas, e.g., butane. The major cost savings for this type of plant results from
the elimination of filtration and distillation.
The extraction with light gases has shown that at temperatures of
300 C and pressures of 3500 psig (238 atm) the weight ratio of the gas to
coal liquid feed should be high, e.g., 5, and for a plant with 10,000 t/d
(29)
coal feed the corresponding quantity of gas to be handled would require
too many compressors. (The required output of the compressor was estimated
to be 1 x 10 cu ft/min or more of ethane/ethylene.)
90
-------
Optional
Section
Naphtha
Product
FIGURE 13. "CONVENTIONAL" SOLVENT REFINED COAL LIQUEFACTION PROCESS
FOR BOILER FUEL PRODUCTION*29'
-------
VO
to
Hydrogen, carbon monoxide
High purity hydrogen
Bay be required {or
hydrotreataent of
heavy oil product
Light
Gas
Makeup
FIGURE 14. PROPOSED BATTELLE EXTRACTION PROCESS
-------
When a heavy gas, e.g., butane, is used for extraction instead of
a light gas, this material actually exists at standard conditions as a
liquid, so that a conventional pump may be used instead of a compressor.
The pumps are inherently lower cost units than are compressors and
considerable savings could be realized if extractions were done using
readily liquefiable hydrocarbons. The capacity of the pumps will be dictated
by the extent of extraction desired in the process, i.e., from 50 to -100
percent of the feed coal liquid. Arguments relating to the magnitude of the
extraction relative to an overall coal to hydrogen balance were discussed
earlier. The issue was purposely left unresolved because the extraction
of coal liquid with gases may be more efficient and faster than conven-
tional filtration and distillation for the separation of ash from the coal
liquid and the separation of light and heavy fuel fractions from the
residue. The overall impact of such a process change on clean fuel recovery
and the energy efficiency of the total process would have to be assessed.
DESCRIPTION OF BATTELLE EXTRACTION PROCESS (see Figure 14)
1) Raw coal is slurried with residual liquids and solids
from the contactor and heavy oil product.
2) Slurry is pumped to the dissolver which is under hydrogen
pressure. Operating conditions normally used in the SRC process, i.e.,
1000 psig, 490 C (850 F).
3) Materials exit the dissolver and are sent to the contactor
where hydrocarbon gases (butane, ethylene or C3-C4 mixtures) are used to
extract the clean fuels from the coal liquids.
4) The liquids from the contactor are pumped to devolatilizers
(flash tanks) where the pressure is rapidly reduced in a 2-stage system.
Heavy oils are collected from the first stage, part of which is recycled
as solvent to the coal slurrying tank. The remaining heavy oil product may
require hydrotreatment for additional sulfur and nitrogen removal. Light
oils are recovered in the second stage and may not require further treatment.
(As yet the quality of the light fuel fractions, i.e., %S, %N, MWt, etc., are
93
-------
not known). The heavy fuel is typical of the values forund in Table 23, i.e.,
£0.1 percent S and 0.3 percent N.*
5) Gas which is released during the devolatilization process
is compressed to the operating pressure and returned to the contactor.
The €2 to C, hydrocarbon gases produced in the gasifier are also used.
Additional light gas may have to be added to the system. If butane or
€3-04 mixtures are used as the extraction medium, a pump could be used
instead of compressors.
6) Part of the residual fluids from the contactor are gasified
to produce hydrogen for the dissolver. Some gas treatment is required to
purge H«S from the system. If high purity hydrogen is required for
hydrotreatment of the heavy oils, shift conversion, and CO. removal will
be required as secondary gas treatment. Such a system is common to both
processes.
COMPARISON OF BATTELLE EXTRACTION PROCESS
TO "CONVENTIONAL" SOLVENT REFINED COAL/
LIQUEFACTION PROCESS
The front end of the Battelle process is similar to a conventional
system (see Figure 14) in that coal is slurried with a recycled liquid and
sent to a dissolver. However, material leaving the dissolver in a conven-
tional system is filtered. The filtration step is very time-consuming as
well as labor intensive. The ash load to the gasifier is also reduced by
the elimination of the diatomaceous earth in the filter cake. In the
conventional process, part of the filtrate is recycled with the raw coal in
the coal slurrying tank. Filtering the stream which is recycled puts an
added burden on the filters (i.e., the product and recycle must be filtered).
This is eliminated by the use of the extraction process. Following filtration
in the "conventional" process, part of the filtrate is distilled to produce
fuel gas, fuel oils, and naphtha. The distillation is known to degrade the
light and medium liquids to hydrogen, fuel gases, and heavy liquid fuels.
*Such values of S and N are typical of the Wilsonville SRC recycle solvent.
94
-------
The extraction process eliminates the distillation and related thermal
degradation. The Battelle process appears to produce better grades of fuels
with good recovery of desirable products.
The Battelle extraction process suggests the following benefits:
o Cleaner fuel with greater recovery
e Simpler process
• Eliminates filtration and distillation steps
e Better efficiency may be realized. (Heat and
material balances are needed to help decide.)
e Possibly eliminates need for hydrotreatment of
the oils to produce environmentally acceptable
fuel
• Fart of residue after extraction is recyclable to
coal dissolver or can be feed for further hydrotreatment.
It is enriched in sulfur and nitrogen contaminants and
amenable to more severe hydrotreatment for their removal.
CONCLUSIONS
It has been shown that light hydrocarbon gases such as ethane,
ethylene, propane, propylene, and butane may be used to extract a clean
fuel from a coal liquid. The clean fuel product has the following
characteristics:
• Ash content is low (almost zero)
a Sulfur content is 10-14 percent of total sulfur in feed
• Nitrogen content is 19-50 percent of total nitrogen
in feed.
When a heavy gas, e.g., butane, is used up to 83 percent of the
feed is extracted at pressures of 1500 psig (102 atm) and at a temperature
of 300 C.
Lighter gases, e.g., ethane and ethylene, require pressures of
3500 psig (238 atm) to achieve 76 percent extraction at a temperature of
300 C.
There does not appear to be any thermal degradation of the coal
liquid to hydrogen and light hydrocarbons during extraction of the clean
fuel as is the case with distillation.
95
-------
A process that uses butane instead of ethane/ethylene for the
production of clean fuel may be more economical with respect to both
capital and operating costs.
96
-------
SECTION 9
USE OF POROUS MEDIA TO CONCENTRATE ORGANIC SULFUR,
NITROGEN AND MINERAL MATTER CONTAMINANTS
IN COAL DERIVED LIQUIDS
The objective of this part of the program was to determine if a
major fraction of sulfur, nitrogen and mineral components of a coal-derived
liquid stream can be adsorbed and retained on a porous material.
INTRODUCTION
The role of catalysts in the mechanism of hydrodesulfurization
(HDS) and hydrodenitrification (HDN) of both petroleum and nonpetroleum feed-
stocks have been comprehensively reviewed in the literature as reported
(2)
earlier in this study. The mechanistic and kinetic models developed
for petroleum HDS have been postulated for the HDS reaction occurring during
coal hydrogenation and liquefaction. However since both HDS and HDS reactions
are interdependent, especially on sulfided metal catalysts, both processes
must be included in the model. These dual-function catalysts contain both
metallic and acidic sites at which special interactions occur to promote HDS
and HDN reactions. For sulfur and nitrogen contaminant removal, a critical
step in the reaction is the adsorption of sulfur or nitrogen compounds on the
active catalyst sites (usually acidic).
The support for many catalyst systems used in HDS and HDN treatment
are alumina, Fuller's earth, iron oxide and similar materials which possess in
their own right high surface areas and significant pore volume. By exposing
2
coal liquids to porous materials with low surface areas (< 100 m /gm) but
large pore dimensions (> 200 A) such as alumina particles with various calcin-
ing histories, the ash minerals and sulfur and nitrogen contaminants might be
removed. Such a material may promote rapid diffusion into the pores and
enhance removal of these contaminants by preferentially adsorbing sulfur and
nitrogen compounds much as they are in the initial stages of catalyzed HDS and
HDN reactions. The mineral matter would be occluded by the bed of particles
97
-------
of adsorbent and retained. If these processes occur, then the spent substrate
would be expected to contain a high concentration of sulfur and nitrogen con-
taminants suitable for further processing under conditions more amenable to
conversion to clean fuels by hydrotreatment. Since the porous adsorption
material was selected for its low surface area, the spent adsorbent would be
expected to be strong enough physically to undergo repeated gasification
regeneration cycles in a fluidized or moving bed.
Such a separation could provide a clean liquid from high boiling
fractions of coal-derived liquids and serve as a means of concentrating
organic sulfur and nitrogen contaminants into a fraction for more severe HDS
and HDN treatment (i.e., higher temperature and pressure reaction conditions).
However, losses of liquids would have to be held to levels below that now
being lost in filtration and/or distillation of coal liquids.
EXPERIMENTAL PROGRAM
The coal liquid selected for this study was obtained from the
Wilsonville, Alabama, SRC Plant and is the same material used in the extrac-
tion experiments described in Section 8, i.e., it is the SRC dissolver product
containing solvent.
The equipment used in the study is shown schematically in Figure 15.
The pump was operated at a flow rate of 10 ml/min. The SRC liquid flowed up
through the reactor containing the fixed bed or porous material. The porous
material could be maintained in the temperature range of 25 to 200 C.
The materials selected for this study are listed in Table 25. The
choice of the materials was based on the activity of the alumina, the silica-
alumina content, porosity and availability. Material of a minus 6 plus 10
U.S. Standard Sieve size was used in all experiments. Some of the materials
had to be crushed to obtain this size range. Volumes of 22 and 10 ml were
used to charge the reactor. When a volume of 10 ml was used, the remaining
reactor volume was filled with nonreactive ceramic fish spines.
Procedure
A typical run consisted of packing the reactor with a known volume
of the porous material. Then the reactor was heated to the predetermined
temperature. When this temperature was reached the SRC-liquid was pumped
98
-------
VO
Feed
Reservoir
Pump
Thermocouple
N
-*-
n
n
Produce Receivers
Heater
Tube Reactor
A = Adsorbent Bed
FIGURE 15. EQUIPMENT ARRANGEMENT USED FOR POROUS MATERIAL ADSORPTION STUDIES
-------
TABLE 25. MATERIALS USED AS POROUS MEDIA
o
o
Manufacturer
Norton
Uarshaw
Alcoa
BCL
U.R. Grace
Number
SH-5102
SH-5123
H-Zeolon
Al-X-649
-65-5-S
F-l
A_A>
N.A.
N.A.
N.A.
(c)
Bulk
Density
g/cc
1.7-2
2.2-2.4
0.6
1.5
1.5
0.61
0.2
Surface
Area
m2/g
0.7-1.3
0.02
450
1.0
5
1.69
(c)
Chemical
Nature
Inactive
Inactive
Active
Inactive
Active
Inactive
Inactive
(a) N.A. = not available
(b) 25 percent Fe203 - Bauxite Mixture Sintered
(c) 80-90 percent void space
-------
through the bed at a predetermined rate. A total of 120 ml of the SRC liquid
was treated in each experiment. Three liquid samples were collected and
analyzed for sulfur, nitrogen and ash. The first liquid sample (Cut 1) was
the first 10 ml of liquid which passed through the bed, the second liquid
sample (Cut 2) was the 100 ml of liquid following and the third liquid sample
(Cut 3) is the last 10 ml of liquid that passed through the bed. The porous
material is recovered from the reactor and stored. The spent bed was regen-
erated by heating to 400-450 C in a controlled stream of air. The performance
of the regenerated materials were evaluated under similar conditions to that
used for fresh material.
The first and third cuts were all analyzed for their ash content
first. From these values the amount of ash removal was calculated. Initially
only those samples which had a 50 percent reduction in ash content were to be
analyzed for sulfur and nitrogen. This criteria was later modified in order
to determine if sulfur and/or nitrogen removal was independent of ash
removal.
Results
A total of 33 experiments were completed at 25, 100, and 200 C
using fresh and regenerated adsorbents with the objective of reducing the
ash content by at least 50 percent. The results of the ash determinations
for selected samples of the first and third 10 ml cut of the 120 mis passed
through the adsorbent bed are compared to ash values for the feed materials
in Table 17. Also included in the table are the ash-free sulfur and nitrogen
values determined for samples selected for analysis and the feed materials.
Feed materials with the ash content of 2.58 and 3.25 percent had higher sulfur
content (0.50 and 0.44 percent S, respectively) than the feed material with
an ash content of 0.360 percent (0.23 percent S). This suggests that the
sulfur content is attributable to both organic and inorganic forms.
It should be noted that increases in ash content were observed
when adsorbents SA-5102, F-l, inactive AljO-j and bauxite were used. This
was assumed to be due to the presence of fines formed during crushing, screen-
ing and/or loading of these materials. After regeneration of the absorbent,
the tendency to increase the ash content was not as severe or absent (see
SA-5102). However, beds of regenerated bauxite, F-l, and H-Zeolon become
plugged when used again for treatment.
101
-------
DISCUSSION
Ash Removal
A comparison of the ash removals indicates that a reduction of about
50 percent of the ash content occurs only when the ash content of the feed
material is low, in this case 0.360 percent. This occurs for a known chemi-
cally active adsorbent (H-Zeolon) and for an inactive form (SH-5123). It also
occurs at 25 C and 100 C.
In the case of SH-5123 the smaller bed of 10 ml seemed to be the
best giving a 55.5 percent removal for the first 10 ml and even a greater
removal by the end of the run of 61.9 percent.
For the filled reactor (22 ml of adsorbent) the removal remained
the same while for H-Zeolon it decreased. The reason for these variations
is not clear. One may speculate that the voids between adsorbent particles
are filled by the end of the run and the filtration is more effective. This
would happen more quickly in the smaller bed treating equal volumes. (Ceramic
fish spines alone were found to be ineffective for ash-forming mineral
removal.) The ash removal for the other materials studied, even those con-
tributing to the ash in the first cut* was improved by the end of the run in
9 of the 13 experiments reported in Table 26. Of the four that showed an
increase in the ash removal two were for regenerated adsorbents.
When coal liquids with ash contents of 2.58 and 3.25 percent were
fed through the adsorbents, removal never approached 50 percent. For
H-Zeolon removal reached 14.2 percent, while SH-5123 reached 22.2 percent.
These results suggest that the fresh porous materials are not as effective
on coal liquids with higher ash content. The removal did improve slightly
at the end of the run in most experiments using fresh porous material. Regen-
erated materials were not as effective at the end of the run as at the
beginning or became plugged. With the material SH-5102, no significant differ-
ence was observed while operating at 25 C rather than 100 C. With SH-5123,
however, the ash removal of 42 percent at 25 C fell to 2.1 percent at 100 C
and 2.6 percent at 200 C. The material F-l was found to be more effective
for ash removal at 100 C than at 200 C. (No experiments were done at 25 C
with F-l.)
102
-------
TABLE 26. RESULTS OF EXPERIMENTS ON THE USE OF POROUS MEDIA TO REMOVE
CONTAMINANTS FROM COAL LIQUIDS (Feed Rate 10 ml/rain)
o
10
Active Bed
Material
U-Zeolon<«>
H-Zeolon
B-Zeolon (Regen)O')
SH-5123
SH-5123
SH-5123
SH-5123
SII-5123
SA-5102<">
SA-5102
SA-5102 (Regen)
P-lW>
F-l
Inactive A^OjW)
Bauxite
Bauxite (Regen)(b>
Peril te
Peril te (Regen)(b*
Volime.
•1
22
22
22
22
10
22
22
22
22
22
22
22
22
22
22
22
22
22
Te»p.,
C
100
100
100
25
25
100
200
100
100
25
100
100
200
100
100
100
100
100
Aah Values. Percent
Product
Feed
0.360
3.25
3.25
0.360
0.360
2.58
2.58
3.25
2.58
2.58
3.25
3.25
2.58
2.58
3.25
3.25
3.25
3.25
Cut 1
0.178
3.05
2.41
0.210
0.16
2.56
J.08
2.53
2.71
2.94
2.80
3.26
2.73
2.69
3.31
Plugged
2.90
2.75
Cut 3
0.262
2.79
-.(c)
0.210
0.137
2.52
2.51
2.95
2.74
2.69
2.95
2.80
2.64
2.51
2.58
2.83
3.07
Aah Reduction
(Increaae). percent
Cut 1 Cut 3
50.5
6.2
25.8
41.6
55.5
0.8
(1.9)
22.2
(5.2)
(14.1)
13.2
(0.3)
(6.1)
(4.5)
(1.7)
10.8
15.4
27.2
14.2
41.7
61.9
2.1
2.6
9.2
(6.4)
(4.5)
9.2
14.0
(2.8)
2.5
20.6
12.9
5.5
Sulfur/NltroRen. percent. Aah Free
Product
Feed
0.23/0.74
0.44/0.75
0.44/0.75
0.23/0.74
0.23/0.74
0.50/0.7S
0.50/0.78
_—
0.50/0.78
0.44/0.75
0.50/0.78
0.50/0.78
0.44/0.75
0.44/0.75
Cut 1
0.13/0.74
0.48/0.78
0.52/0.75
0.20/0.75
0.20/0.75
0.46/0.88
0.43/0.76
0.14/0.79
0.52/0.75
0.13/0.80
0.47/0.79
0.49/0.78
0.49/0.73
Cut 3
0.13/0.73
0.39/0.72
' (0.47/0.74)(<>
0.20/0.74
0.20/0.74
0.45/0.79
0.34/0.74
0.12/0.83
0.45/0.72
0.12/0.80
0.47/0.71
0.47/0.79
0.46/0.68
(a) Supplied by Norton
(b) Regenerated at 400 C with flow of exceae air
(c) Sample loac
(d) Supplied by Alcoa
(e) With aah
-------
Sulfur and Nitrogen Removal
Sulfur in the coal liquid was reduced to the 0.12 to 0.14 percent
level (ash free) with the use of H-Zeolon, SA-5102 and F-l but the treatment
conditions and relationship to ash content differ for each material. This
was the lowest level attained in these experiments. Nitrogen removal was
essentially zero except that in the third cuts using inactive A1203 or perlite
about a 10 percent reduction was measured.
H-Zeolon reduced the ash-free sulfur of 0.23 percent to 0.13 percent
for the feed with 0.36 percent ash but did not prove too effective for the
feed with 3.25 percent ash where the values appeared essentially the same
(feed = 0.44 percent; Cut 1 - 0.48 percent; Cut 3 » 0.39 percent). After
regeneration the H-Zeolon appeared to contribute sulfur to the coal liquid.
By contrast the material SA-5102 (which is not very effective in removing
a.sh forming minerals) was able to reduce the sulfur to 0.12 to 0.14 percent
from a feed with 0.50 percent sulfur (2.58 percent ash) at 100 C. However
porous material, F-l, could reduce the sulfur to 0.12 to 0.13 percent from a
feed with 50 percent sulfur only at 200 C. At 100 C, using a feed with 0.44
percent sulfur (3.25 percent ash), no removal was measured. Bauxite and
; I
perlite were found ineffective with the other porous media studied exhibiting
only a small tendency to remove sulfur.
The ability of inactive Al_0_ and expanded perlite to reduce
nitrogen from two high-ash-containing coal liquids by 10 percent is signifi-
cant since they did not affect the sulfur content. This suggests that the
removal mechanism for organic nitrogen may be different than for some of the
sulfur (organic or inorganic) and is not directly related to ash removal.
It should be noted that in those cases where high ash coal liquids
were fed and ash removals of 14 to 20 percent were obtained, sulfur reduction
did not occur. Even in the case of SH-5123 where a low-ash feed was used
and a large percentage of ash was removed, the sulfur only decreased from
0.23 to 0.20 percent. Therefore one could speculate that sulfur removal does
not necessarily follow effective ash removal if pyrrohtite were the source of
sulfur. In fact, the opposite seems to be true (see SA-5102 and F-l results).
104
-------
CONCLUSIONS
Some porous materials with low surface areas but large pore dimen-
sions have been found to remove up to 76 percent of the sulfur (ash free)
from coal liquids passed through them in a fixed bed. These same materials
can also remove about 20 percent of the ash-forming minerals from feeds con-
taining about 3 percent. From feeds containing about 0.4 percent ash,
50 percent or better can be removed. A 10 percent removal of nitrogen can
be accomplished using expanded perlite. Both nitrogen and sulfur removals
appear to occur by separate mechanisms and both seem to be independent of
ash removal.
105
-------
SECTION 10
CONVERSION OF COAL LIQUEFACTION RESIDUE
TO ENVIRONMENTALLY ACCEPTABLE FUEL
The objective of this study was to ascertain whether techniques
used by the coal carbonization industry to produce clean coke can be used
for the cleanup of coal liquefaction residues.
INTRODUCTION
As reported earlier in this study, the thermal decomposition reac-
tions of coal during carbonization for coke formation are favorable to the
(2)
enrichment of most contaminants in the solids. Up to 65 percent of the
sulfur and 60 percent of the nitrogen originally present in the coal end up
in the coke. This fraction was reduced by special treatment with gases
during carbonization. Removal of 80 percent of the sulfur was realized by
exposure of the coal during carbonization at 800 to 1000 C with either ammonia,
hydrogen, water gas or steam. Changes in nitrogen content were not reported.
The same principals are believed to be applicable for the removal of sulfur
and nitrogen from coal liquefaction residues which are usually enriched in
sulfur and nitrogen and trace elements originally in the feed coal. The sul-
fur and nitrogen are considered to be refractory or too difficult to remove
at reasonable HDS and HDN treatment conditions but still not fixed in the
5
carbon matrix as in coke.
Coal liquefaction residues are defined as the materials obtained
after removal of liquids during solids separation and the heavy oils left
after distilling off light cuts (i.e., bottoms). Characteristics of con-
taminants usually present in residues are as follows:
• Sulfur - present as FeS and as organic sulfur compounds
(2-4 member aromatic rings); usually difficult to convert
by hydrogenation.
106
-------
• Nitrogen - present as organic nitrogen compounds; usually
difficult to remove by hydrogenation; enriched in resi-
dues, especially bottoms.
• Trace elements - normally those present are those con-
tained in .mineral matter of raw coal.
• Catalyst - eroded or attrited catalyst if used in the
process.
Prevalent methods proposed or anticipated for utilization of these residues
are as follows:
• Typically insoluble solids would be retorted to obtain
added liquids and char. The char would be used for fuel
or reacted with water to form hydrogen. Such solids
typically contain 5 to 30 percent fixed carbon.
• Still-bottoms would be coked to provide additional
recoverable liquids and solid coke for fuel and/or
hydrogen production (would be enriched in sulfur and
nitrogen and undesirable as a fuel).
If the coal liquefaction residue can be converted into liquid hydrocarbons
(rather than gas) and a char which is made low in sulfur and nitrogen by
treating the residues with active gases, greater recovery of clean fuel
value would be possible Removal of the sulfur and nitrogen during the con-
version of residues or still bottoms to char will reduce the amount of sulfur
and nitrogen eventually fixed into the coke-like matrix during carbonization
(coking) . The cleaner char would be used to produce hydrogen stream requir-
ing less cleanup down stream. In addition, such an approach could provide
for greater yield of carbon as hydrocarbons and could provide gas streams
richer in H-S and NH, for more efficient cleanup during the treatment. No
energy penalty for cooling the gas stream would have to be considered. How-
ever, such a process would have to exceed the performance of existing pro-
cesses using pyrolysis and gasification with respect to overall environmental
impact, energy requirement, and the quality of the fuel products.
EXPERIMENTAL PROGRAM
Residues of coal liquids used in this study were from two sources.
One was prepared from SRC dissolver product described in Section 8 by topping
107
-------
the liquid at a 400 C bottom temperature. (One batch, prepared under vacuum,
gave a residue which was 18.4 percent of the original weight charged; the
other batch was prepared at atmospheric pressure and was 21 percent of the
original weight charged. Both batches were crushed and blended thoroughly
before analysis and use.)
The other residue was supplied by Hydrocarbon Research, Inc. (HRI)
at Lawrenceville, New Jersey (Sample No. LO-1052 from their syncrude run
130-79-168 which processed Illinois No. 6 coal). The HRI residue is from the
vacuum bottoms. The analysis of both materials is shown in Table 27.
The equipment used in the study is shown in Figure 16. The system
could be pressurized to 80 to 100 psi from the cylinder of gas being used for
the treatment.
Flow of the gases at pressure were controlled by double-valving
and the flow rates of 10 and 100 ml/minute were measured at the vent of the
reactor train by measuring the rate of rise of coap bubbles in a burette
at atmospheric pressure. The cold trap was cooled to -78.5 C with a solid
carbon dioxide-acetone slurry to trap condensable material released during
the run. An automatic sampling gas chromotograph was used to detect the
release of noncondensable (at -78.5 C) products formed during the reaction.
The temperature in the reactor was controlled at either 400 or 500 C in a
resistance heated furnace.
TABLE 27. ANALYSES OF COAL LIQUEFACTION RESIDUES, PERCENT
Material Moisture Ash Total S Nitrogen
SRC Residue 0.20 19.5 2.29 1.76
HRI LO-1052 0.08 25.8 2.41 1.28
108
-------
Cylinder
Gas
Source
o
VO
Furnace
Automatic
Sampler and
GC Analyzer
Flow
Measure
Cold Trap
A = Residue packed column
FIGURE 16. ARRANGEMENT OF EQUIPMENT USED IN COAL LIQUEFACTION
RESIDUE TREATMENT WITH GASES
-------
Procedure
In a typical run, about 22 ml of crushed residue was weighed and
packed into the reactor. Then the reactor was fastened into place and
checked for leaks at the operating pressure. The furnace and cold traps and a
H--CO mixture was used was at best sporadic and could not be correlated to
changes in the sulfur and nitrogen values in the residue.
When the percentage of nitrogen and sulfur found in the treated
residue increased from that originally present in the feed and N/S weight
ratio remained essentially the same (0.768 versus 0.884), it was assumed that
the increase was due to the loss of volatile hydrocarbons from the residue.
This behavior was found especially characteristic for hydrogen. When a
mixture of hydrogen and carbon monoxide was used (H. to CO volume ratio = 1),
a measurable reduction in the amount of nitrogen in the residue was observed
despite an apparent increase in the percentage of sulfur present due to
110
-------
TABLE 28. EFFECT OF TREATMENT OF COAL LIQUEFACTION RESIDUES WITH
VARIOUS GASES ON THEIR SULFUR AND NITROGEN CONTENT
Percent,
Moisture and Ash Free
Treatment
Original Char (SRC Residue)
H2 at 400 C, high flow(a>
H2 at 500 C, low flow(a)
H2 at 400 C, low flow
H2 at 500 C, high flow
NH3 at 500 C, high flow
low flow
H2-CO(b) at 500 C, low flow
high flow
low flow
BH. at 500 C with active alumina
low flow
high flow
Original Still Bottoms ^
H2-CO(b) at 500 C, low flow
H.-CO at 400 C, high flow
NH3 at 400 C, low flow
HE. at 500 C with active alumina
high flow
H. at 500 C, low flow
H2 at 400 C, high flow
N
2.19
2.19
2.30
2.29
2.44
3.49
2.62
1.89
2.17
1.89
3.19
2.85
1.73
1.61
1.58
2.00
2.24
1.75
1.71
S
2.85
2.83
2.98
2.96
3.18
2.72
3.04
3.04
3.03
3.01
3.01 *
2.95
3.25
3.18
3.30
3.32
3.44
3.09
3.22
Weight
Ratio
N/S
0.768
0.774
0.772
0.774
0.767
1.283
0.862
0.622
0.716
0.628
1.060
0.966
0.532
0.506
0.479
0.602
0.651
0.566
0.531
(a) High flow - 100 ml/min; low flow - 10 ml/mln.
(b) H. to CO volume ratio - 1.
(c) Hydrocarbon Research Sample No. LD-1052.
Ill
-------
hydrocarbon loss. The lower N/S weight ratio was obtained for both coal
liquefaction residues treated with the H2-CO mixture. Treatment with
ammonia gave a slight reduction of sulfur, but since the treated residue
was cooled in ammonia, there was an increase in the nitrogen present in both
residues. The presence of active alumina appeared to enhance the loss of
hydrocarbons since the sulfur values increased significantly. However,
nitrogen was retained by the residue in these experiments also.
DISCUSSION
The experimental results suggest that when ammonia is passed
through the SRC residue at 100 psi and 500 C, small reductions in sulfur
occur. At 400 C no sulfur removal was obtained. Hydrogen and a mixture of
hydrogen and carbon monoxide were not effective for the removal of sulfur.
Treatment with the mixture of hydrogen-carbon monoxide at 400 and
500 C was found to remove nitrogen from both the SRC and the HRI residues (it
was not possible to determine the effect of carbon monoxide alone). The
10 ml/min flow rate gave better removal. Reasonable agreement between dupli-
cate runs were obtained for the SRC residue at 500 C. The longer residence
time appeared to favor the denitrification reaction and produced a 4 percent
reduction. In the case of the HRI residue, treatment at the high flow rate
but at 400 C provided the greater nitrogen reduction—about 9 percent. Both
higher temperatures and pressures may enhance nitrogen removal.
Treatment with ammonia gas in the presence of an activated alumina
which is known to promote the decomposition of NH, at 500 C was more effec-
tive for removing hydrocarbons than just ammonia alone, but it did not show
the activity reported in the literature during coal carbonization. Higher
temperatures might have helped.
CONCLUSIONS
Nitrogen in coal liquefaction residues can be reduced by treatment
with a mixture of hydrogen and carbon monoxide at 100 psi and at 400 C.
Treatment of residues with ammonia at even 500 C did not remove sulfur to the
extent that the literature reports suggest during the carbonization of coal
at 1000 C.
Higher temperatures and pressures may help both sulfur and nitrogen
removal and promote the conversion of a greater faction of the carbon value
to hydrocarbons. Further work in this area would be desirable.
112
-------
SECTION 11
IMPROVEMENTS IN PYRITE LIBERATION FROM COAL
The objective of this study was to enhance pyrite liberation by
chemical comminution compared to mechanical grinding.
DISCUSSION
Quite early in the program, it became apparent that work in this
area should not be initiated because the progress made in chemical comminu-
tion by others would only be duplicated by the planned effort. Although
there was no activity in the area, it seemed appropriate to summarize the
finding that prompted our action.
Bench-scale studies with ammonia-treated coals at Syracuse Research
Corporation indicate that chemical comminution is capable of liberating more
pyrite and a comparable amount of ash-forming minerals than conventional
(30 31)
mechanical crushing to the same size consistency. '
The chemical comminution process is basically an improved method
for pyrite liberation. Although the mechanism is not well defined, it seems
that chemical comminution involves rapid migration of the molecules of the
comminuting agent throughout the naturally occurring system of faults in
coal. This weakens and disrupts the interlayer forces. The net result is
fracture of the coal, the breakage being induced selectively along the
bedding planes and mineral boundaries. The fragmented coal and mineral
matter can then be separated by some conventional cleaning processes.
The chemical comminution of coal with ammonia has been studied by
the Syracuse Research Corporation since 1971. Comparative washability data
for an Upper Freeport coal showed that 96.6 percent of the pyrite was removed
at 1.3 specific gravity from the chemically comminuted coal, while 90 percent
of the pyrite was removed from the mechanically crushed coal (-14 mesh) at
the same specific gravity. It was also shown that the chemically comminuted
113
-------
sample contained only 5.5 percent fines (-100 mesh) compared to 21 percent
fines (-100 mesh) for the mechanically crushed sample.
The researchers estimated the process requirements for a chemical
comminution plant producing 1000 tons per hour as:
Ammonia (1 percent of recirculated NH-)—0.5 ton/hr
Electricity—7830 kW
Steam—70 ton/hr
Manpower—3 men/shift.
The estimated capital cost for a plant capacity of 1000 ton/hr was
$12,500,000. The operational cost was $1.06 per ton of cleaned coal. This
covered only the chemical comminution units added to an existing coal clean-
ing plant.
Syracuse Research Corporation reported that the nitrogen content
of coal increased by 5 to 7 percent after the chemical comminution, depending
upon the type and extent of ammonia recovery treatment. However, it is not
known whether the nitrogen is in a form that would result in NO emissions
2C
during combustion.
114
-------
SECTION 12
REFERENCES
(1) Mezey, E. J., S. Singh, and D. W. Hissong, Fuel Contaminants
Volume I, Chemistry, EPA 600/2-76-177a, U.S. Environmental Pro-
tection Agency, Research Triangle Park, N.C., 1976, 177 pp.
(2) Mezey, E. J., S. Singh, and D. W. Hissong, Fuel Contaminants
Volume II, Removal Technology Evaluation, EPA 600/2-76-177b,
U.S. Environmental Protection Agency, Research Triangle Park,
N.C., 1976, 316 pp.
(3) Beerstecher, E., Petroleum Microbiology, Elsevier Press, Inc.,
New York, 1954, 375 pp.
(4) Rogoff, M. H. et al, Microbiology of Coal, Bureau of Mines 1C
8075, 1962, 85 pp.
(5) Kim, K. E., et al, Sulfur Recovery by Desulfovibrio in a Biochemi-
cal Method of Oil Shale Production, Science and Technology of Oil
Shale, edited, T. F. Yen, Ann Arbor Science, Mich. 1976, 252 pp.
(6) Findley, J.,etal, Degradation of Oil Shale by Sulfur-Oxidizing
Bacteria, Applied Microbiology, 28: 460-466, 1974.
(7) Davies, A. J. and T. F. Yen, Development of Desulfurization Pro-
cedure for Fuels, Preprint Am. Chem. Soc. Div. Fuel Chem., 19 (2):
218-223, 1974.
(8) Strawinski, R. J., Method of Desulfurizing Crude, U.S. Pat.
2,521,761 (1950).
(9) Strawinski, R. J., Purification of Substances by Microbial Action,
U.S. Pat. 2,574,070 (1951).
(10) ZoBell, C. E. Hydrogenation, Desulfurizing and Like Processes,
Canada Pat. 503,218 (1954).
(11) Kirshenbaum, I., Bacteriological Desulfurization of Petroleum, U.S.
Pat. 2,975,103 (1961).
(12) Harrison, W. M., Bacterial Treatment of Media Containing Hydro-
carbons and Sulfides, U.S. Pat. 3,105,014 (1963).
-------
(13) Sarles, W. B., et al, Microbiology, Harper and Bros., New York,
1951, 493 pp.
(14) ZoBell, C. E., Action of Microorganisms on Hydrocarbons, Bacteri-
ological Review., 9: 10(1): 1-49, 1946.
(15) Novelli, G. D. and C. E. ZoBell, Assimilation of Petroleum Hydro-
carbons by Sulfate-Reducing Bacteria, J. Bacteriology., 47: 447-448,
1944.
(16) Fonken, G. S., R. A. Johnson, Chemical Oxidation with Microorganisms,
Marcel Deliker, New York, 1972, 272 pp.
(17) Westlake, D.W.S., et al, Microbial Utilization of Raw and Hydro-
genated Shale Oils, Can. J. Microbiology 22: 221-226, 1976.
(18) Dryden, I.G.C., How Was Coal Formed, Coke and Gas., 18: 123-127,
181-184, 1956.
(19) Prevot, A. R., Manual for the Classification and Determination of
the Anaerobic Bacteria, Lea and Febiger, Philadelphia, 1966, 402 pp.
(20) Anderson, R. P., and C.H. Wright, Coal Desulfurization in the P&M
Solvent Refining Process, Amer. Chem. Soc. Div. of Fuel Chem. Pre-
print., (1) 20: 2-25, 1975.
(21) Rodgers, B. R., "Use Solvents to Separate Micron Sized Particles
from Liquid Streams'1, Hydrocarbon Processing., 5: 191-194 (1976).
(22) Bartle, K. D., et al, Chemical Nature of a Supercritical-Gas Extract
of Coal at 350°C. Fuel., 54: 226-230, 1975.
(22a) Harrison, J.S., Coal Liqeufaction in the UK Coal Processing Technology,
Vol. 2, Am. Inst. Chem. Eng., N.Y., 1975.
(23) Diepen, G.A.M. and F.E.C. Scheffer, On Critical Phenomena of
Saturated Solutions in Binary Systems, J. Am. Chem. Soc., 70:
4085-4089, 1948.
(24) Ellis, S.R.M., Vapor Phase Extraction Processes, British Chem. Engr.,
16 (4/5): 358-361, 1971.
(25) Zhuze, T. P., Compressed Hydrocarbon Gases as a Solvent, Petroleum.,
23: 298-300, 1960.
(26) Ewald, A. H., et al, The Solubility of Solids in Gases, Disc.
Faraday Soc., 15: 238-245, 1953.
(27) Paul, P.F.M., and W.S. Wise, The Principles of Gas Extraction,
Mills & Boon LTD., London, 1971, 85 pp.
116
-------
(28) Todd, D.B. and J.C. Elgin,, Phase Equilibria in Systems with
Ethylene Above its Critical Temperature, AIChE J., 1 (1): 20-27, 1955.
(29) Demonstration Plant. Clean Boiler Fuels from Coal. Office of Coal
Research, R&D Report No. 82, Ralph M. Parson Company, Los Angeles,
California, Contract No. 14-32-0001-1234 (September, 1973).
(30) Datta, R.S., P.H. Howard, and A. Hanchett, Feasibility Study of
Pre-Combustion Coal Cleaning Using Chemical Comminution, Final Report,
Fe-1777-4, Energy Research and Development Administration, Syracuse
Research Corporation, Syracuse, New York (November 1976).
(31) Howard, P.H., and R.S. Datta, Desulfurization of Coal by Use of
Chemical Comminution, Science, Vol. 197, No. 4304, pp 668-669
(August 1977).
117
-------
APPENDIX
-------
APPENDIX
THEORY OF EXTRACTION WITH COMPRESSED GASES
THERMODYNAMICS
It would be useful to know of the extraction capacity for a
compressed gas. But compressed hydrocarbon gases, when used for extraction,
form non-ideal vapor phase mixtures. Prausnitz discusses the implications
of non-ideality of the vapor. The fugacity, f, is not generally equal to the
partial pressure, but rather it is necessary to apply corrections for the
non-ideality of mixing as well as for the non-ideality of the pure vapor.
The fugacity coefficient, 0, is used to correct for the non-ideality of
the system.
(1 2)
Various methods ' for predicting the properties of gaseous
mixtures have been proposed, but for the most part these have been strictly
empirical. It will be shown later that the solubility of a hydrocarbon
(MW 128) in a hydrocarbon gas (MW 30) can be calculated by various methods,
e.g., Lewis fugacities, Redlich and Kwong equation and virial equation of
state. Only the solubility values calculated by the virial equation agree
with the experimental values. The virial equation of state is a series in
the reciprocal volume
PV . . B . C
RT M l + V + V2
where -2 is the compressibility factor.
When Equation (1) is applied to a mixture, the virial coefficients
are functions of composition as well as temperature. The composition
dependence of the virial coefficient is given by
A-l
-------
In the case of a simple binary mixture, Equations (2) and (3) become
+ 2y1y2B12 + y§B22 . (4)
°m " yicm + 3yiy2cii2 + 3yiyici22 + yfc222 etc-
where coefficients, B.. , C..., B._, C ._, depend on the forces acting between
like molecules. The remaining coefficients, called cross-coefficients,
depend on the forces acting between unlike molecules.
It has been shown that when there is ideal mixing of the gases,
the cross-coefficients are zero. Such a simplification implies that the
compressibility factor of a mixture at constant temperature and pressure is
a straight line function of the composition. But even for a simple mixture
of methane and ethane at 50 C and 60 atm, the plot (compressibility factor
versus mole fraction) is not a straight line but a parabola. Thus, even
this system does not involve ideal mixing.
A method has been developed by Prausnitz ' for calculating the
virial coefficients. To calculate the cross-coefficient, for a binary
mixture, the virial coefficient is represented by the following equation:
Bi1 T
v^- - e fc4— . w ) (6)
cij cij 1J
where
V .., characteristic initial volume, = 1/2(V . + V .).
Cij Cl Cj
6, a generalized function, computed values available as in Table A-l.
1/2 '
T , characteristic critical temperature, • k..(T .T .) . k. . values
C1J lj ci cj ij
are calculated from Statistical Mechanics. T is temperature.
W , accentric factor, - 1/2 (W± + W ), values tabulated in Reference (4).
By using the critical values of temperature, volume, and accentric
factor, the terms T/Tcl1» v c±4* and ^4 are calculated, and then from Table A-l
it is possible to determine the value of G. Thus, from Equation (6) the
cross-coefficient is calculated.
A-2
-------
The third virial coefficient (e.g., C_,,) may be neglected when
(3}
the density of the vapor is similar to that of the compressed pure gas.
There is good agreement, as shown in Figure A-l, between calculated
and experimental solubility values of ethylene and naphthalene mixtures.
It needs to be determined whether the virial equation of state can be used
to predict extraction of coal liquids by compressed gases.
TABLE A-l. COMPUTED VALUES OF -0
Partially Reproduced from Reference (3)
Reduced
Temp,
T/Tc
0.5
1.0
1.5
Solubility
0.0
4.008
1.155
0.483
Isotherms
Accentric
0.1 0.2
5.412 6.896
1.213 1.274
0.451 0.418
(Empirical)
Factor, W
0.3 0.4 0.5
8.473 10.144 11.923
1.339 1.408 1.481
0.382 0.345 0.304
Studies in the deasphalting of petroleum crude by a compressed
gas shows that a typical solubility isotherm of the petroleum crude in a
gas is qualitatively described by the equation:
T| * 0
where
n. * mole fraction.
p « saturation vapor pressure.
V° • molar volume of hydrocarbon in condensed phase, the greater
V° the lower the pressure corresponding to minimum solubility
V • partial molar volume of hydrocarbon in gaseous phase.
R - gas constant and
T - temperature.
A- 3
-------
O
I
•g,
§
u
I
W 1
£
(1) Ideal Gas
(2) Lewis Rule
(3) Redlich-Kwong
(4) Virial
10 * —
40 60 80
Total Pressure, atm
100
120
FIGURE A-l.
VAPOR-PHASE SOLUBILITY OF NAPHTHALENE IN ETHYLENE,
CALCULATED AND EXPERIMENTAL VALUES AT 35 C (Ref. 5)
A-4
-------
The isotherm for the solubility can be described by the observa-
tions that when p is equal to the saturated vapor pressure of the substance
at the given temperature then the mole fraction of the hydrocarbon is
unity. As pressure is increased, the solubility curve shows a downward
trend until a minimum value in n is reached. Further increase in pressure
leads to an increase in mole fraction of the hydrocarbon in the vapor.
The Effect of Pressure on Solubility
The limited data on binary and ternary systems shows that solu-
bility of high-molecular weight components in the solvent, at temperatures
above the critical temperature of the solvent, increases rapidly with
pressure.
The Effect of Temperature on Solubility
It appears that at relatively low pressures, increasing tempera-
ture slightly above the critical temperature (of the solvent) decreases the
solubility, but at higher pressures, increasing the temperature will increase
the solubility of the hydrocarbon in the solvent (gas).
REFERENCES
1. Prausnitz, J. M., Molecular Thermodynamics of Fluid Phase Equilibria,
Prentice-Hall, N.J., 1963, 523 pp.
2. Reed, R. C. and T. K. Sherwood, The Properties of Gases and Liquids,
McGraw-Hill, N.Y., 1966, 646 pp.
3. Prausnitz, J. M., Fugacities in High Pressure Equilibria and in Rate
Processes, AIChE Journal, .5 (1), 3-9, 1959.
4. Zhuze, T. P., Compressed Hydrocarbon Gases as a Solvent, Petroleum, 23,
298-300, 1960.
5. Ellis, S. R. M., Vapor Phase Extraction Processes, British Chem. Engr.,
16 (4/5), 358-361, 1971).
A-5
-------
TECHNICAL REPORT DATA
(Please rtad Inunctions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-02 5a
2.
3. RECIPIENT'S ACCESSION'NO.
4. TITLE AND SUBTITLE
Fuel Contaminants: Volume 3. Control of
Coal-related Pollutants
S. REPORT DATE
Januarv 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
E.J.Mezey, Seongwoo Min, B.R.Allen. W.C.Baytos
and Surlit Sineh
9. PERFORMING ORGANIZATION NAME AND AOORESS
Battelle-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
10. PROGRAM ELEMENT NO.
EHE623
11. CONTRACT/GRANT NO.
68-02-2U2
12. SPONSORING AGENCY NAME AND AOORESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final: 7/75 - 7/76
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES lERL-RTP project officer is Lewis D. Tamny, Mail Drop 61. 919/
541-2709. EPA-600/2-76-177a and -177b are earlier reports in this series.
i6. ABSTRACT
rep0rt gives results of a. study to identify strategies for removing
pollutants from coal and coal-derived liquids. Of the approaches considered, five
were selected for preliminary assessment by experimentation (a sixth, improve-
ments in pyrite liberation from coal, was not studied because of significant advan-
ces by others). Study findings include: (1) biological action on coal-derived liquids —
prospects of using nonmutated bacteria for sulfur and/or nitrogen removal from coal
liquids appear small; (2) enhancement of pyrite removal during immiscible fluid
agglomeration — removal equivalent to that obtained for float-sink analysis was ob-
tained by pretreatment and oil agglomeration, the same technique found to be effec-
tive for recovering > 90% of coal from coal cleaning plant fines: (3) extraction of
clean fuels from coal liquids — light hydrocarbons can be used to extract 83% of coal
liquid at supercritical conditions to yield a low sulfur .and nitrogen fuel: (4) concen-
tration of organic sulfur and nitrogen and ash from coal liquids—up to 76% of the
sulfur and about 10% of the nitrogen can be removed by passing coal liquids over
various special porous media: and (5) conversion of coal liquefaction residues to
environmentally acceptable fuels --treatment of coal liquefaction residue with H2/CO
mixtures reduces nitrogen content by as much as 14%.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS
c. COSATi Field/Group
Pollution
Coal
Coal Preparation
Liquefaction
Desulfurization
Nitrogen
Biodeterioration
Bacteria
Pyrite
Agglomeration
Hydrocarbons
Hydrogen
Carbon Monoxide
Pollution Control
Stationary Sources
Denitrogenation
Bacterial Action
Coal Cleaning
13 B
21D,08G
081
07D
07A
07B
06A
06M
07C
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (Thlt Report)
Unclassified
21. NO. Of PAGES
135
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Perm 2220-1 (»-73)
124
------- |