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

-------