United States      Industrial Environmental Research EPA-600/7-79-201
Environmental Protection  Laboratory         August 1979
Agency        Research Triangle Park NC 27711
Pollutants from Synthetic
Fuels Production:
Sampling and Analysis
Methods for Coal
Gasification

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-600/7-79-201

                                                    August 1979
Pollutants from Synthetic Fuels Production;
          Sampling and Analysis Methods
                  for Coal Gasification
                                by

                   S. K. Gangwal, P. M. Grohse, D. E. Wagoner,
                 D. J. Minick, C. M. Sparacino, and R. A. Zweidinger

                         Research Triangle Institute
                            P. 0. Box 12194
                   Research Triangle Park, North Carolina 27709
                           Grant No. R804979
                       Program Element No. EHE623A
                      EPA Project Officer: N. Dean Smith
                   Industrial Environmental Research Laboratory
                     Office of Energy, Minerals, and Industry
                      Research TriangleL.Park, NC 27711
                             Prepared for

                   U.S. ENVIRONMENTAL PROTECTION AGENCY
                      Office of Research and Development
                          Washington, DC 20460

-------
                                 ABSTRACT

     A laboratory-scale coal gasification facility has been developed and
is in use to study the generation, sampling, chemical analysis, process
evaluation, and environmental assessment of pollutants from coal gasifi-
cation operations.  This report describes sampling and analysis methods
pertaining to this coal gasification environmental study.
     Collection and sampling techniques are described for partlculates,
organic condensibles and vapors or gases in the raw product stream of the
gasifier as well as for solid residues.
     Gas chromatography (GC) procedures are described for the measurement
of fixed gases, C-|-C5 hydrocarbons, sulfur gases and Cg-Cg aromatics.
Atomic absorption (AA) procedures for measurement of toxic trace elements
include those for arsenic, selenium, lead, cadmium, chromium, and mercury.
     Volatile orgam'cs are collected from the gas stream via polymeric
sorbents (Tenax GC and XAD-2), and analyzed by glass capillary GC/mass
spectrometry.  The major nonvolatile byproduct (tar) is prefractionated by
solvent partitioning into acid, base, and neutral fractions.  Each fraction
is analyzed by capillary GC/MS or high performance liquid chromatography
(HPLC).  Typical results are given to illustrate the nature of the compounds
studied, the methodologies and the sensitivities of the methods.
                                     ii

-------
                     TABLE OF CONTENTS
Section
Abstract 	
List of Figures 	
List of Tables 	
Acknowledgements 	 ....
1.0 INTRODUCTION 	
2.0 CONCLUSIONS, PROBLEM AREAS AND PLANS 	
3.0 SAMPLING METHODS 	
3.1 SOLIDS AND PARTICULATES 	
3.2 LIQUIDS 	
3.3 GASES AND VAPORS 	
4.0 ANALYSIS OF GASES AND VAPORS 	
4.1 GAS CHROMATOGRAPHY (GC) 	
4.2 EVALUATION OF POLYMERIC SORBENTS AND GC/MS ANALYSIS
5.0 ANALYSIS OF LIQUIDS 	
5.1 SEPARATION OF TAR AND CONDENSATE 	
5.2 TAR PARTITIONING 	
5.3 GC/MS ANALYSIS 	
6.0 ATOMIC ABSORPTION METHODS 	
6.1 SAMPLE PREPARATION AND ANALYSIS 	
6.1.1 Coals and Residues 	
6.1.2 Tars 	
6.1.3 Aqueous Condensate and Scrubber Solutions .
6.1.4 Sorbent Materials for Gaseous Effluents . .
6.2 QUALITY CONTROL PROCEDURES 	
6.3 TYPICAL RESULTS 	
7.0 FOURIER TRANSFORM INFRARED SPECTROMETRY 	
7.1 HIGH RESOLUTION ANALYSIS OF GAS SAMPLES 	
7.2 GC/FTIR ANALYSIS OF SEMIVOLATILES IN XAD-2 EXTRACT
Page




1
3
5
5
5
7
12
12
21
37
37
38
40
47
47
47
49
. 49
. 49
49
49
54
54
. 57
     7.3  LOW RESOLUTION FTIR OF COAL, ASH, AND TAR	    59

8.0  SPECIAL TECHNIQUES FOR ANALYSIS OF TARS AND AQUEOUS CON-
     DENSATES	    65

     8.1  PNA ANALYSIS USING GLASS CAPILLARY GAS CHROMATO-
          GRAPHY (GC)2	'	    65
     8.2  SIMULATED  GLC  DISTILLATION OF TAR	    69
     8.3  CLASS  CHARACTERIZATION USING ELEMENT SPECIFIC
          DETECTORS	    69
     8.4  ION  CHROMATOGRAPHIC  ANALYSIS OF AQUEOUS  SAMPLES  ...    71
                                111

-------
TABLE OF CONTENTS (continued).

Section                                                                 Page
References	    72
Appendices
     I.   SELECTION AND PREPARATION OF POLYMERIC SORBENTS  	    1-1
    II.   TENAX and XAD-2 ANALYSIS WITH GC/MS COMPUTER 	   II-l
   III.   QUALITATIVE GC/MS RESULTS FOR ORGANIC VAPORS AND TAR
          FRACTIONS	  III-l
                                     iv

-------
                            LIST OF FIGURES

Number                                                                Page
  1      Samples collected during gasification  of coal  	     6
  2     Glass sampling system 	     8
  3     Continuous analysis system  for H2»  02> C02, CO, and ChL  •  .     10
  4     Schematic of GC analysis system  	     13
  5     Hydrocarbon analysis system 	     16
  6     Comparison of dual  and single flames modes  for a  coal
        gasifier product gas	     19
  7     Performance of Tenax GC cartridges  at  high  hydrocarbon
        levels	     24
  8     Tar partitioning scheme 	     39
  9     Range of selected trace elements in coals 	     51
 10     Percent loss of effluents (Pittsburgh  No.8  coal)   	     53
 11      Chemigram of infrared bands (cm~ )  versus time during GC/
        FTIR analysis of extract of XAD-2 resin  sampled during  surge
        phase of run 41	     58
 12     Comparison of coal  spectra.  Top =  Illinois No.6; Bottom =
        Zap lignite	     61
 13     Comparison of tar spectra.   Top  = Illinois  No.6;  Bottom =
        Zap lignite	     62
 14     Comparison of ash spectra.   Top  = Illinois  No.6;  Bottom =
        Zap lignite	     63
 15     PNA fraction of coal gasification tar  by capillary GC-FID:
        Determination of selected PNA's  	     67
 16     Analytical system for organic vapors in  gas samples ....    II-2
 17     Gas liquid chromatograph-mass spectrometer  computer (GC/MS/
        COMP) layout	    II-3

-------
                            LIST OF TABLES

Number                                                                Page
  1      Conditions for GC Analysis	    14
  2      Cg-Hydrocarbons,  Cg-Cg Aromatics  and  Sulfur Species  in  Pro-
        duct Gas Stream from Semi batch Air-Steam Gasification of
        4x8 mesh Illinois No.6 at 200 psia:  ppm versus  onstream
        time	    18
  3      Integrated Concentrations for Selected Compounds  in  the
        Gasifier Product  Gas	    20
  4      Sample Stability  Check   	    22
  5      Performance Evaluation of XAD-2 Traps   	    25
  6      Estimated Total Hydrocarbons Collected on XAD-2 Traps   ...    26
  7      Breakthrough on XAD Cartridges 	    27
  8      Components of Standard 	    28
  9      Evaluation of Recoveries from XAD-2 Cartridges  	    30
 10      Recovery of Compounds from XAD-2  Resin with Evaporation of
        Solvent from 5 ml  to 0.5 ml	    31
 11      Recovery of Compounds from XAD-2  Resin with Evaporation of
        Solvent from 195  ml to 5 ml   	    32
 12      Average Percent Recoveries from XAD-2   	    33
 13      Percent Deviation of Response Ratios  from Ideality as a
        Function of Peak  Width	    33
 14      Average of Deviations (%) of Response	    34
 15      Quantitative GC/MS Results for Selected Compounds on XAD-2
        Resins	    36
 16      Elemental Analyses of Crude Tar	    43
 17      Typical Preliminary and Partition Results for Tar and
        Condensate	    44
 18      Quantitative Results for Selected Compounds in Tar
        Fractions	    45
 19      Quantitative Results for Phenols, Cresols, and  Xylenols in
        Condensate by Reverse Phase HPLC  and  GC/MS 	    46
 20      Average Material  Balance for Three  Illinois No.6  Coal
        Gasification Runs	    50
 21      High Resolution FTIR Analysis of  Selected Gas Samples   ...    56
                                     VI

-------
LIST OF TABLES (continued).

Number                                                                  Page
 24     Selected Compounds in PNA Fraction as Weight % of Coal
        Gasification Tar (MS vs.  capillary-FID)	     68
 25     Conditions for Simulated  GLC Distillation 	     70
 26     Error in Boiling Points of PNA's as Predicted by the
        Procedure .  .  .  .	     70
 27     Corrected Boiling Point Distribution for Tar Produced from
        Gasification of Illinois  No.6 Coal (Run  52)	     71
 28     Operating Parameters for GLC/MS/Comp System	    11-4
 29     Operating Parameters for LKB 2091 GLC/MS/COMP System  ....    II-7
 30     GC/MS Analysis of Polynuclear Aromatics  from Run 21 	   III-2
 31     GC/MS Analysis of Nonpolar Neutrals from Run 21	111-3
 32     GC/MS Analysis of Tar Bases from Run 21	III-4
 33     GC/MS Analysis of Steady-State XAD-2 Extract from Run 21  .  .   111-5
 34     GC/MS Analysis of Surge XAD-2 Extract from Run 21	III-6
 35     GC/MS Analysis of Glass Fiber Filter Extract from Run 21  .  .   III-7
 36     GC/MS Analysis of Tenax Sample Upstream  of Steady-State
        XAD-2 Trap for Run 21	II1-8
 37     GC/MS Analysis of Tenax Sample Downstream of Steady-State
        XAD-2 Trap for Run 21	III-9
                                   vii

-------
                             ACKNOWLEDGMENTS

     The sampling and analysis procedures reported herein were developed and
implemented for the project entitled "Pollutants from Synthetic Fuels Pro-
duction" under sponsorship of the Fuel Process Branch, Industrial Environ-
mental Research Laboratory, U.S. Environmental Protection Agency, Research
Triangle Park, N. C.  The guidance of Dr. N. Dean Smith, Project Officer, Mr.
William J. Rhodes, Program Manager, and Mr.  T. Kelly Janes, Branch Chief, are
acknowledged.
     The collaboration of David Green, John Cleland, Duane Nichols, and
Forest Mixon of the Process Engineering Department at RTI have been central
to the conduct of this project.  Substantial contributions to analytical
method development and execution have been provided by Jesse McDaniel, Steven
Frazier, Kenneth Tomer, and Mitchell Erickson.  Finally, the authors express
thanks to Dr. Harold McNair, Professor of Chemistry at Virginia Polytechnic
Institute and State University for his valuable suggestions regarding specific
analytical techniques.
                                     viii

-------
                             1.0  INTRODUCTION

     The Research Triangle Institute (RTI), under sponsorship of the U.S.
Environmental Protection Agency (EPA), has undertaken a research project to
study pollution problems associated with synthetic fuels production processes.
The project is currently in the third year of a five-year program.   So far,
the major thrust of the program has been the gasification of coal.
     Nine U.S. coals have been gasified in a laboratory reactor and some 57
gasification runs have been completed.  The reactor and the associated
sampling system were designed, fabricated and put into operation during the
first year of the program.   A versatile and flexible sampling system has
resulted from a series of modifications which were made through experience
gained from the earlier runs.  This report describes the sampling and
analysis methods for characterization of the gaseous, solid and liquid
effluents from the laboratory coal gasification reactor.
     The coal gasification reactor has been operated in a semi batch mode.  A
1.5 kg batch of coal is dropped into the hot reactor through which steam and
air are flowing at high temperature (~1000°C) and pressure (~200 psia).  A
surge condition prevails for a few minutes following the coal drop until
conditions stabilize as volatiles are driven off due to pyrolysis.   Char
gasification then ensues.  Provision has been made in the sampling approach
to characterize the two periods as distinct from each other.  Emphasis is
placed on the analysis of trace organics, sulfur species, tars and aqueous
effluents with the use of modern instrumental techniques.
     Gas samples are collected in glass bulbs and analyzed by gas chromato-
graphy (GC).  A variety of detectors including thermal conductivity (TC),
flame ionization (FID) and flame photometric (FPD) are used for measurement
of permanent gases, C-,-C5 hydrocarbons, Cg-CQ aromatics and sulfur gases.
Volatile and semivolatile organics are
-------
XAD-2 resin is extracted using methylene chloride and the extract is analyzed
by GC/MS.  Standard atomic absorption (AA) procedures are used for measurement
of toxic trace metals including As, Sb and Hg in coals and residue and
additional AA procedures have been developed for analysis of Se, Cr, Pb, Ni,
Cd and Cu in coals, residues and tars.
     The major nonvolatile product is condensed tar which is separated from
the aqueous condensate and solvent partitioned into acidic, basic and three
neutral fractions.  Each fraction is then analyzed by capillary GC/MS or high
performance liquid chromatography (HPLC).
     Several special techniques are under development for further characteri-
zation of the various effluents.  These include Fourier transform infra-red
                                                                    2
spectroscopy (FTIR), GC-FTIR, glass capillary gas chromatography (GC ),
simulated GC distillation and ion chromatography.  Details of the above
procedures and typical results obtained are described in this report.
     Related reports are simultaneously being prepared on other aspects of the
screening test runs of the RTI nonisothermal pollutant generation facility.  A
report on the technical  aspects of the runs presents operating conditions and
test results; and, a second report deals with the health and environmental
significance of the results.   Separate reports are also planned on related
topics.  One is intended on the transient behavior of the gasification test
runs relative to both reactor operation and the concentration of effluents.
Another is to be a comparison of the results of this project with the available
comparative data on pilot plant and commercial gasification operations.
     Future reports will also present information relative to parametric
studies, which examine the generation and control of potential pollutants in
coal  gasification under various selected operating conditions.  The parameters
under consideration are coal  type, coal  particle size, reactant flow rates,
chemical additives, and other factors.  Information being generated in this
project is intended to provide a basis for the assessment of the potential
health and environmental severity of the effluents from coal  gasification
processes and is intended for use in the reduction of such.

-------
                 2.0  CONCLUSIONS, PROBLEM AREAS AND PLANS

     A sampling and analysis protocol  has been developed and applied for
collection and measurement of solid, liquid and gaseous effluents from a
laboratory coal gasifier.  The type of samples collected include solid
residues and particulates, tars and aqueous condensates, semivolatile and
volatile organics and gaseous species.  Analytical  chemical  measurements are
made using a variety of techniques.  These include  ultimate  and proximate
analysis of coals, ash and tars; continuous analysis of primary gases; gas
chromatographic analysis of gaseous pollutants including C,-Cc hydrocarbons,
Cg-Cg aromatics and sulfur species; atomic absorption analysis of solid and
liquid samples for trace metals; sorption and analysis of volatile trace
organics by GC/MS/computer techniques; and the collection, partition and
analysis of aqueous condensate and semivolatile and nonvolatile tars using
GC/MS and HPLC methods.  Several special techniques have been developed or are
under development.  These include analysis of polynuclear aromatic species by
glass capillary gas chromatography; Fourier transform infra-red analysis of
gas, tar, coal and residue; simulated GC distillation of coal tars for measure-
ment of boiling point distributions; and ion chromatographic analysis of
cations and anions in aqueous condensates.
     Certain problem areas which need further developmental  work or alter-
native instrumental techniques have been identified.  Of major concern is the
characterization of sulfur, nitrogen and oxygen heterocyclic compounds in the
tar.  A study has been initiated using element specific detectors to char-
acterize nitrogen and sulfur compounds by boiling points.  Another area of
importance is the characterization and speciation of trace metals in the gas
phase.  To date, only XAD-2 resins have been analyzed for this purpose.
However, species like mercury vapor, methylmercury, arsine and hydrogen
selenide are probably not being trapped by XAD-2.  An effort is presently
underway to confirm this by installing a backup activated charcoal trap

-------
downstream of the XAD-2.  Charcoal is known to trap species like arsine very
            2
efficiently.   Polynuclear aromatics analysis is another area which requires
further work.  This is especially true for five-to-seven ring compounds which
do not yield to easy measurement by gas chromatographic techniques because of
their high boiling points (>500°C).  Yet they are known to be highly toxic and
mutagenic.  An alternative approach involving high performance liquid chromato-
graphy coupled to a spectrofluorometer will be evaluated.  This technique is
especially useful for ultra-trace level PNA's of the order of picograms.
Another area of concern is the spillover of PNA's onto the other tar fractions.
                                                                           3
Tar bases have been shown to be even more mutagenic than the PNA fractions.
It is felt that PNA spillover may be partly responsible for this.  This is
being further investigated.  A relatively simple procedure is to inject a
dilute tar sample directly onto a capillary GC-FID and determine selected
PNA's based on retention times.  Then by subtraction of the PNA's in the PNA
fraction the degree of spillover can be estimated.
     Future plans include continuation of the application of the sampling and
analysis protocol to coal gasification and other synthetic fuel-related
samples.  Modifications and improvements will be made as needed to improve
analytical efficiency and reduce analysis uncertainties.  Quality control
procedures have been in use all through the project.  During parametric tests
and modeling studies, however, there will be a need for making rigorous
quality control checks on the data.  Also degrees of accuracy and precision
need to be better defined so that the confidence levels of results can be
ascertained for future test runs.

-------
                          3.0  SAMPLING METHODS

     A versatile and flexible system is in use to sample various effluents
from the coal  gasification reactor.    This sampling system removes  parti -
culates, tars  and aqueous condensates, semivolatile organic material,  volatile
organics and fixed gases.  A simplified schematic of the various samples  that
are collected  is shown in Figure 1.   All liquid and solid samples,  unless
otherwise stated, are stored in a freezer in glass bottles with tight  teflon
                 D
caps and Parafilm  seals for further analysis.
3.1  SOLIDS AND PARTICIPATES
     As shown  in Figure 1, these samples are of three types:  feed  coal,
reactor residue, and particulates.  Approximately 50-100 g of a representative
sample of the feed coal and a like amount of the residue following  the run are
collected.  Portions of these samples are sent to commercial laboratories for
ultimate and proximate analyses by standard ASTM procedures.  The product
gases that emerge from the reactor immediately enter in a particulate trap.
This trap contains a stainless steel braided cup which functions as a flow
impinger.  The trap is further packed with glass wool as a medium to facilitate
the removal of solid particles from the hot gas stream and is insulated and
heated to prevent tar condensation.   Any finer particles which are not captured
in this trap are further removed in the condenser and through the use of glass
fiber filters that will be described in Sections 3.2 and 3.3.  Following the
run, the glass wool is carefully removed and refrigerated, pending  solvent
extraction and analysis.
3.2  LIQUIDS
                                                             1
     The gases leaving the particulate  trap enter a condenser  with an outer
jacket through which liquid freon is circulated for removal of tar and
aqueous condensate.  This trap may be tapped periodically during a run for

-------
                                                                     9
   Gasifier
1.
2.
3.
4.
5.
                   Particulate
                      Trap
       Condenser
                      Gas Sampling
                         System
Coal
Residue
Participates
Tar, Water and Insolubles
Entrained Tar and Water
6.
7.
8.
9.
Gas Bulbs
Polymeric Sorbents
Glass Fiber Filter
Scrubber Solutions
                Figure 1.  Samples collected during gasification of coal

-------
removal of the accumulated material  or at the end of the run  for total
collected material.   The tar and condensate are separated and the tar is
subdivided into several  fractions.   An elaborate sample preparation  technique,
described in Section 5.1, is used to facilitate partitioning  into different
sample classes including aqueous condensate, tar acids, tar bases, nonpolar
neutrals, polar neutrals, and polynuclear aromatics.  Solids  and particulates
which are insoluble in the neutrals  partitioning solvent (cyclohexane)  are
also collected.  Required quantities of each sample are stored in the freezer
and the rest are disposed of using adequate waste disposal  methods.
3.3  GASES AND VAPORS
     The gases leave the condenser and at this point the reactor pressure is
reduced through a backpressure regulator.  The gases then enter a glass
sampling system shown in Figure 2.   Entrained moisture in the gases  is further
removed at this point so that its interference with the subsequent sampling  of
gases and vapors is reduced.  Special emphasis is placed on the sampling
strategy to account for the surge and steady-state conditions characteristic
of the semi batch operation.  When a  batch of coal is fed to the hot  reactor, a
surge condition prevails during which most of the pyrolytic products are
released.  Typically the surge period lasts for about 15-30 minutes, during
which a dramatic change in concentration of the pollutant occurs, which could
be as much as three orders of magnitude.  Gaseous grab samples are taken  at  a
high frequency (every 1  to 2 minutes) during the surge condition to  efficiently
characterize the concentration changes.  The steady-state condition  is marked
by a drop in the methane concentration to a relatively stable value  of about
2 percent.  At this point, gas samples are collected with a time interval of
about 10 minutes or greater.  The glass containers used to collect these gas
samples are 500 ml in volume and are equipped with two high vacuum teflon
stopcocks.  Approximately 3-5 liters of gas is passed through them prior to
closing the stopcocks.  Samples are kept stored in a heated box at about 50°C.
Stability of gas samples and the GC analysis of these samples are described  in
Section 4.1.
     In addition to direct collection of gases in glass containers,  organic
vapors are adsorbed on polymeric sorbents.  Tenax-GC and Amberlite XAD-2 are
used for the gas analyses.  The sorbents are packed into two types of containers:

-------
                                                        All  Class
    From
Gasifier
00
                                                                                             CT:  Corrugated Teflon

                                                                                             T1-T2:  Toggle Valves
                                                                                             S1~S6:
Ports with
teflon stopcocks for
sample collection in
small Tenax cartridges
and glass bulbs.
y







Continuous
Gas
Monitors


Meter
                                            Figure 2.  Glass  sampling  system.

-------
large in-line traps, and small  cartridges.   The in-line traps  are  cylindrical
tubes with an internal  volume of approximately 1  liter.   One end contains  a
fritted glass disc to retain the sorbent.   Approximately 150-200 g of cleaned
XAD-2 (see Appendix I)  is packed in each trap.   This  is followed by a layer  of
glass wool, a stainless steel screen and a  glass  fiber filter  (Gelman type A).
End adapters are clamped to each end of each cylinder and sealed with Viton  o-
rings.  In order to distinguish between the surge and the steady-state con-
ditions the two traps are used in parallel, connected by a three-way teflon
stopcock.  This affects the switchover from one condition to another.
Approximately 1000 and 3000 standard liters of gas are produced during the
surge and steady-state conditions respectively.  In addition to the in-line
traps, samples are collected in small cartridges  (1.5 cm I.D.  x 10 cm long).
These are filled with sorbent (either XAD-2 or Tenax-GC) to a  length of 6  cm
with glass wool at both ends.  They are interfaced to the sampling system  both
upstream and downstream of the large XAD-2  traps by means of teflon stopcocks
                                                  3
and fittings.  When Tenax is used, only 100-150 cm  of sample  is  necessary
                                       2                       2
because of its lower surface area (30 m /g).  When XAD-2 (300  m /g) is used  in
these cartridges, 5-15 liters of sample are required.  Efficiency  of XAD-2 has
been evaluated using Tenax cartridges upstream and downstream  of  the large in-
line traps.  In addition for selected runs, two XAD-2 or Tenax cartridges  have
been used in tandem both during the surge and steady-state periods to determine
breakthrough and displacement.
     In addition to the collection of gases in bulbs and sorbents, a IN HgSO^
impinger is used in series with the glass sampling system for  measurement  of
ammonia.  Evaluation of the accuracy of the GC/MS analysis of samples col-
lected on the sorbents are described in Section 4.2.
     To ensure successful operation of the gasifier and to detect gross
system malfunctions, it is necessary to monitor the major product gases (H2>
COp, CO, and ChL) continuously.  In addition, it is necessary  to  monitor
oxygen in the product gas which should be absent under normal  operation.
However as the coal batch is gasified, 'a stage is reached when oxygen break-
through occurs indicating the completion of the run.   An elaborate system  has
been connected in parallel to the glass sampling system for continuous analysis
of l^j C02> CO, CH,, and 02  (Figure 3).  A sample conditioner removes traces

-------
              BPR
From
Gasifier
^Glass
Sampling
 System
                   SV
BP?.:  Back Pressure Regulator
TV:  Toggle Valve
SV:  Solenoid Valve
NO:  Normally Open
P:   Pump
REF:  Refrigeration System
NV, R:  Needle Valve
 A :  NDIR Analyzers
HU:  Hi oh Level  Alarm
FI:  Flow Meter
I
	To Dry Test Meter
  iout            ^  02
               rr
             Drain
                                                                                                      0-5 V DC
                           Figure 3.  Continuous  analysis  system for H2, 02, CO, C02 and

-------
of condensibles from the gases via refrigeration (~1°C) prior to their
entering the continuous analyzers.  ChL, CO and COp are measured using non-
dispersive infrared analyzers, H2 using a thermal conductivity analyzer and
02 using a paramagnetic analyzer.  Pressure and flow in the system are con-
trolled using a backpressure regulator and several rotameters.  The analyzers
are zeroed and spanned prior to the run with known concentrations using a
series of solenoid switch valves.  Following analysis, the gases are routed
back to the glass sampling system and mixed with the bulk gases for total
volumetric flow measurement using a dry test meter.
                                     11

-------
                     4.0  ANALYSIS OF GASES AND VAPORS

     Collection of gaseous and vapor effluents from the coal gasifier was
described in Section 3.3.  Further preparation of the samples and analysis of
selected species is carried out using gas chromatography (GC) and gas chromato-
graphy coupled to mass spectrometry (GC/MS).  GC is mainly used for measure-
ment of major gases and volatile pollutants which are directly analyzable from
the glass bulbs.  On the other hand, GC/MS is used for the analysis of trace
organics collected on polymeric sorbents and glass fiber filters.
4.1  GAS CHROMATOGRAPHY (GC)
     A simple schematic of the GC analysis system is shown in Figure 4.  As
stated earlier, the glass bulbs are stored in a heated box at 50°C.  The
analysis is carried out without removal of the bulbs from the box by means of
a vacuum sample removal system.  The system hardware is essentially similar to
                                        A
the one described by Gangwal and Wagoner  for measurement of sulfur compounds
using a dual flame photometric detector (FPD) coupled to a GC.  It consists of
a Sargent-Welch vacuum pump for evacuating the sample loops (1 ml) to less
than 1 mm of Hg absolute pressure, a Heise vacuum-pressure gauge (0-3450 mm)
with 2 mm graduations for measurement of sample loop pressure prior to in-
jection and three heated zero volume valves for injection of the sample onto
the GC columns.  Table 1 shows the various compounds that are analyzed and the
type of detector, column and conditions that are used.
     The sample injection and analysis procedure for each detector consists of
the following.  The sample loop (1 ml) and associated valving and tubings are
first evacuated using the vacuum pump (Figure 4).   A small amount of sample is
then used to flush the loop and the system is reevaluated.  The sample is then
introduced into the loop and the absolute pressure is carefully measured using
the vacuum-gauge.   Approximately 300-400 mmHg of sample pressure is used.
Using the zero volume valve, the sample is then injected onto the GC.
                                     12

-------
to
        To GC-TC


















-.

{





t





(
_y




t

^



\
\





r^
\
S


\
/

I
I




J


)
^


JU

c





r



5
^

>
r^i


•
To GC-FID
^
/•


1 HoatoH R
1 . nca LcO D
Bulbs
2. Sample I
(Zero Vo
3. Carrier
4. Heise Va
_, 5. Vacuum P
_j _£.. Swagel
Valves




^ • To GC-FPD
>S)2
. 	 / /• — -
tr f 1
^L

                                                                                                       with Glass

                                                                                                   Injection Valve
                                                                                                    lume)
                                                                                                    Gas Cylinder
                                                                                                  Vacuum-Pressure Gauge
                                                                                                   Pump
                                       Figure 4.  Schematic of GC analysis  system.

-------
TABLE 1.   CONDITIONS FOR GC ANALYSIS
Detector
Compounds Analyzed
Column and Oven
Temperature
Carrier Gas Flow
(ml/min)
Detector
Temperature (°C)
Sample Loop
Material and Size
Injection Valve
Material
TC
H2, C02, CO, N2, 02
CH^, C2Hg, CgH^
H2S, COS (>100 ppm)
6' x 1/8" SS Porapak N
7' x 1/8" SS Molecular
Sieve 13X
70°C
30
200
SS 1 ml
SS
FID
(a) C-,-Cc hydrocarbons
(b) C6H6, C7Hg, C8H10
(a) 6' x 1/8" Durapak
phenylisocyanate
(18°C)
(b) 8' x 1/8" TRIS-1,2,3
cyano ethoxypropane
(90°C)
20
150
SS 1 ml
SS
FPD
COS, CH3SH, C2H5SH
(CH3)2S, CS2, thiophene
6' x 1/8" Teflon Carbopack
B/XE60/H3P04
50°C for 2 min to 130°C at
32°C/min
30
150
Teflon 1 ml
Hastalloy-C

-------
Standards of known concentration are also analyzed in a similar fashion.   Peak
areas of the various components in the sample and standard are measured  using
a A/D converter coupled to a computer with a GC-software system.   Concentration
of the components of interest are determined in the sampling using equation
(1).
                          u     As  Pu                                     '

where:   subscript u refers to the sample and S to the standard and C,  A,
         and P are concentrations, peak areas and loop pressures, respectively.

Equation (1) is only applicable to the TC and FID.  The FPD is nonlinear and
a log-log calibration curve of area versus concentration for each sulfur
                                                        4
species is prepared by varying the sample loop pressure.   These curves are
then used for quantisation of the sulfur compounds.  Periodically, calibration
curves are also prepared for the TC and the FID using varying sample loop
pressures in order to confirm their linear range.
     The analysis of fixed gases, C-j-C^ hydrocarbons and H2S using the thermal
conductivity detector has been described in detail by Gangwal , et al .
Recently, a further improvement in the system allows analysis of COS in
addition to the above gases.   This is especially helpful when H^S and COS are
present in relatively large concentrations (>200 ppm).  At these concentrations,
the sulfur specific FPD is saturated because the amount of sulfur injected
falls outside the dynamic range for 1 ml samples.
     The flame ionization detector can be used either for the analysis of Cg-
CQ aromatics or C-,-Cg hydrocarbons using two columns in parallel.  Backflush
valves are incorporated into the system to reverse the carrier gas flow
through the column.  The schematic is shown in Figure 5.  This is necessary to
prevent the interference of the heavier materials with analysis of subsequent
samples.
     The flame photometric detector (FPD) is used  for the analysis of COS
(<200 ppm), CH3SH, C2H5SH, (CH3)2S, CS2 and thiophene.  The FPD uses two
flames instead of one used in the conventional Brodey and Chaney  design.
                                     15

-------
Sample to^^
Heise Gauge
                                    FID
                                                                                  Zero Volume Gas Sampling
                                                                                  Valve
                                                                                  Backflush Valve
                                                                                  Switch Valves
                                                                                  8' x 1/8" S.S. TRIS (1.2,3)
                                                                                  Cyano Ethoxy Propane Column
                                                                                  6' x 1/8" S.S. Durapak-
                                                                                  Phenyl Isocyanate Column
                   Carrier
                     Gas
                                Figure 5.  Hydrocarbon analysis system.

-------
             "                                                 o
Details of the design have been described by Patterson, et al.  The quenching
of the response to sulfur compounds in the presence of a hydrocarbon matrix
eluting simultaneously is a problem common to all  FPD designs.  This results
in inaccurate measurements.  The dual  flame design minimizes this problem by
breaking down the complex matrix to simpler compounds in the first flame and
measuring the optical emissions in the second flame.  This characteristic of
the dual flame FPD is demonstrated in Figure 6.  In the single flame mode, no
C$2 response is seen even though a larger thiophene response is  obtained for
the same amount of sample.  The FID under the same conditions indicates a
larger background of hydrocarbons at the point of C$2 elution and a smaller
background at the point of thiophene elution.  In addition, the dual flame
FPD has been found to be much more reproducible than the single flame FPD.
As much as a 75 percent different response has been observed for the same
sample analyzed consecutively on the single flame FPD.  This, in our opinion,
is due to hydrocarbons, still eluting from the previous sample and interfering
with the response of the next sample.
     Typical data obtained for various species during gasification run 52 are
shown in Table 2.  The pyrolytic products, e.g., hydrocarbons, mercaptans and
thiophene are reduced to very low levels after only 15-30 minutes into the
run.  In contrast, the gasification products, H2$, COS, prevail until the
very end of the run  (i.e., until oxygen breakthrough).  The results in Table
2 become more meaningful when the data are integrated over the duration of
                                                           3
the run.  Pollutants produced can then be expressed in pg/m  of gas.  For
three runs carried out with Wyoming, North Dakota and  Illinois No.6 coals,
integrated values are presented in Table 3.
     A few words are needed here as to the quality control procedures used  to
ensure the integrity of the data generated.  Since various types of analyses
are carried out on the same bulb, a partial vacuum  is created in the bulb
which increases following  each analysis.  For  this reason, the bulbs are
equipped with special high-vacuum glass teflon stopcocks capable of holding
the vacuum down to 1 mm Hg for several days.   In addition, since it is not
possible to complete the analysis of all bulbs immediately following
                                     17

-------
00
            TABLE  2.   C2-HYDROCARBONS,  Cg-Cg AROMATICS  AND SULFUR SPECIES  IN  PRODUCT  GAS  FROM  SEMIBATCH
                AIR-STEAM GASIFICATION  OF 4 x 8  MESH  ILLINOIS  No.6 AT 200  psia:   ppm  versus  onstream  time
Tine
(minutes)
C2H6
C2H4
C2H2
C6H6
C7H8
Ethyl benzene
p + m-Xylene
o-Xylene
H2S
COS
CH3SH
C2H5SH
cs2
Thlophene
1
13300
17000
110
2160
360
9
40
5
11700
280
70
20
90
130
3
25500
20400
130
2230
480
20
70
10
23000
470
140
40
150
150
4
21800
19000
120
2750
750
30
130
30
20300
440
100
30
210
170
6
18300
20100
180
2590
610
20
90
20
19600
430
90
40
240
140
7
16600
18000
160
3060
770
30
140
30
19500
420
60
30
190
160
9
14900
16900
170
3560
820
30
150
30
19300
370
50
20
210
160
11
13800
14100
130
3300
620
20
110
20
19000
380
90
20
180
160
13
12200
12200
150
3570
580
20
110
20
18100
360
80
10
180
160
18 23
9970 4130
8670 2350
100 <50
3520 1770
480 320
20 10
90 70
20 10
15900 9960
330 230
50 20
10 <1
150 70
140 60
33
230
<50

500
160
6
30
6
3400
130
2

2
5
43 54 63 73 107 137 177 226 287 322
<50


260 160 90 50 10 20 69 45
100 70 50 40 10 10 64 34
4 4 3 3 2 1 1 1 0.3 0.2
20 20 20 20 20 8 64 33
44443221 11
2830 2550 2690 3840 2500 4320 1680 270 2900 210
160 150 140 180 160 160 110 170 600 29
0.2

0.7
<5

-------
         Column:   6'X 1/8" Teflon
                   Carbopack B/H3PO4/XE 60
         Temperature:   50°C for 2 minutes
                   programmed to 130°C at
                   30°C/min
         28 ml/min helium carrier gas
             1. COS
             2. Methyl Mercaptan
             3. Thiophene
IN
                                           IN
                                                 CS2
                                            IN
                                           -4
     LAI
                Single Flame Response
No CS2 Response
                                                       8 Minutes

                                                       3
                                                           Dual Flame Response
                                                             FID Response
Fiqure 6.  Comparison of dual  and  single flames modes for a coal gasifier  product  gas.
                                          19

-------
             TABLE 3.  INTEGRATED CONCENTRATIONS* FOR SELECTED
                       COMPOUNDS IN THE GASIFIER PRODUCT GAS
(ug/m3)


C2H6
C2H4
C6H6
C7H8
Ethyl benzene
p,m-Xylene
o-Xylene
H2S
COS
CH3SH
^*O ^* C
cs2
Thiophene
Run 51
N. Dakota
NA
NA
NA
9.6E5
6.7E5
9.3E4
2.0E5
8.5E4
1.0E6
1.0E5
3.5E4
5.5E3
1.5E3
1.7E4
Run 47
Wyoming
2.4E6
1.6E6
1.6E4
2.3E6
1.3E6
5.8E4
3.9E5
7.6E4
1.6E6
1.2E5
2.3E4
1.1E4
3.9E3
3.7E3
Run 52
Illinois #6
1.8E6
2.7E6
3.4E4
1.4E6
3.6E5
1.8E4
9.8E4
2.0E4
6.1E6
6.1E5
1.4E4
1.7E4
5.7E4
6.0E4

*Analyzed by gas chromatography.
NA = Not analyzed.
                                      20

-------
sample collection, it is essential  that the samples  be  stable,  at  least  for
a few days in the bulbs.  A stability study was made on three bulbs  over a
period of seven days.  This is shown in Table 4.   Only  sulfur compounds  were
used in this study.   As seen from the table, only a  five percent (or lower)
change was seen in the concentration of the various  sulfur species.   This
was considered to be acceptable.   Each compound is quantitated by  means  of
calibration with a standard obtained commercially and certified within ± 2
percent accuracy or better.  Periodically, the standards are verified against
a permeation tube calibration system or against NBS  standards.   Accuracy of
peak integration using the computer system is periodically checked by comparing
the results against  alternative methods i.e., peak  height, triangulation,
cut-and-weigh or other independent integration systems  such as a small
electronic integrator.  Finally, since the major gases  (Hg. C02» CO and CH^)
are analyzed continuously, only a few samples are periodically verified b5f
the GC and compared with the data from the continuous analyzers.  A difference
of less than 10 percent between the two procedures is considered acceptable.
4.2  EVALUATION OF POLYMERIC SORBENTS AND GC/MS ANALYSIS
     As described in Section 3.3, organic vapors are collected on two polymeric
sorbents, Tenax-GC and XAD-2.  Criteria for selection of these materials are
described in Appendix I.  Therein further details are given regarding the
nature of these materials and how they are prepared  for vapor collection.
Conditions and procedures used for GC/MS analysis using high resolution
glass capillary columns with a thermal desorption interface for Tenax and
methylene chloride extraction for XAD-2 are described in Appendix II.   In
this section, the polymeric sorbents are evaluated for their adsorption
efficiency and typical quantitative GC/MS results for selected organics are
presented.
     Tenax-GC was evaluated using a sampling manifold which permitted the
flow through two Tenax cartridges (see Section 3.3).  An injection port was
installed upstream of the cartridge and* benzene-doped nitrogen flowed through
both cartridges at 2 1/min.  Radiolabelled toluene was injected into  the
port and the second  cartridge was replaced periodically with a fresh  cartridge.
Each cartridge was desorbed thermally onto a scintillation counting fluid and
                                     21

-------
                                           TABLE 4.   SAMPLE STABILITY CHECK
ro
ro
Sample No.*
SI
(7 minutes
from coal
drop)
S2
(16 minutes)
S3
(86 minutes)
Time
(hours)
3
28
49
74
98
123
147
3
28
49
74
98
123
147
3
28
49
74
98
123
147
COS
(ppm vol)
550
540
530
530
540
530
510
440
420
420
420
430
410
410
150
140
140
130
140
130
130
CH,SH
(ppm vol)
39
37
37
36
38
36
35
17
16
15
16
16
15
16
<1
C2HgS
(ppm vol)
20
19
19
19
20
19
18
9
9
8
9
7
7
6
<1
CS2
(ppm vol)
230
220
220
220
220
220
210
140
140
130
140
140
130
130
<1
Thiophene
(ppm vol)
250
250
250
240
250
240
230
150
150
150
150
150
140
140
<1
         Samples obtained  from low-Btu semibatch  gasification  of W.  Kentucky  #9  coal.

-------
radioactivity was determined.   Three different conditions were examined:   (1)
low toluene (0.8 mg), (2)  high toluene (22 nig), and (3)  low toluene with  3000
ug/1 benzene in the purge gas.  These experiments graphically illustrate
(Figure 7), the problem of extrapolating from low to high concentrations.
The amounts used are realistic in terms of actual gasifier process streams.
The change in breakthrough volume for toluene with toluene is large.   The
displacement of toluene by benzene is also of importance, since a large
background of hydrocarbons exists in the real sample and only a limited
number of them are sought at low levels.  This behavior is not unique to
toluene but the threshold at which the above effects become significant
changes with each sorbate.  This experiment indicated that the amount of  the
real sample through the Tenax cartridge should be low and 100-150 ml  was
found to be appropriate for the gasifier sampling program (see also Section
3.3).
     XAD-2 unlike Tenax-GC has a higher sample capacity.  This material was
evaluated under gasification conditions by taking small  discrete Tenax
samples upstream and downstream of the XAD-2 traps.  Further performance
checks were made on XAD-2 by collecting discrete (5-15 liters) samples on two
XAD-2 cartridges in tandem upstream of the large XAD-2 traps.  Results for
three gasification runs during which the above approaches were adopted are
shown in Table 5.
     It is apparent that breakthrough is occurring at the high loading
present in runs 26 and 32.  To compare this with the results from the Tenax
study, a summation of all the compounds quantitated on the XAD traps on a per
gram basis was made.  This is shown in Table 6.  In the Tenax experiments the
benzene mass was 8 mg/g at 20 percent toluene  breakthrough and 16 mg/g at 65
percent toluene breakthrough.
     This would indicate a greater capacity  for  collection of hydrocarbons by
XAD-2 since only 10 percent of such compounds  as toluene, C2-benzene, methyl
thiophene, C^-thiophenes and  naphthalenes were found downstream relative to
upstream on runs 26 and 33.   This is predictable on the basis of greater
surface area of XAD-2 relative to Tenax.  Capacity problems are apparent,
however, in the analysis of phenol and cresols.  The performance is good in
                                     23

-------
                     2000 \ig/i benzene  in  purge
                     0.08 mg toluene
                                                            22 mg toluene
                                                              .87 mg toluene
                             100       150

                           Purge Volume (J.)

Figure 7.  Performance of Tenax GC cartridges at high hydrocarbon levels.

-------
                                           TABLE  5.   PERFORMANCE  EVALUATION  OF XAD-2 TRAPS
Compound
Benzene
Thiophene
Toluene
Methyl thiophene
C2-benzene
C~-thiophene
Benzaldehyde
Phenol
Benzofuran
Cresols
Naphthalene
Biphenyl
Dibenzofuran
Acetic acid
Run 26 Surge Phase -
733 1 of gas in 34 min
Total mass on
XAD trap (ng)
—
—
—
Ill
285
37
Not found
650
225
290
32
11
12
Not found
Concentration found
on Tenax (pg/1)
Upstream Downstream
645 1088
101 116
324 37
67 3.5
243 1.1
16 2.3
12 9
1500 907
57 ND
265 143
510 42
25 4
7.6 ND
ND 39
Run 32 Surge Phase -
1500 1 of qas in 10 min
Total mass on Concentration found
XAD trao fmnl on Tenax l^/U
XAU trap (mgj Upstrean Downstream
124 405
12 133
72 44
6.4 3
39 1.5
2.2 ND
NOT 1.4 2.3
DETERMINED 360 250
18 1.8
22 21
160 45
4.1 4.8
1.6 2.4
3.4 6.2
Run 33 -
Mass on XAD
trap (mg)*
—
--
--
1.9
22
0.8
Not found
83
ND
40
175
0.8
2.8
Not found
Steady State Phase
Concentration found
on Tenax ipg/1)
Upstream Downstream
22 0.2
1.0 ND
15 0.4
0.6 ND
11.5 0.08
0.6 ND
ND 0.45
148 5.0
10 0.05
8.8 0.4
76 2.2
0.8 0.07
0.19 ND
ND 0.85
ro
en
            35% of the values found after 100 min on  line.

-------
     TABLE 6.  ESTIMATED TOTAL HYDROCARBONS COLLECTED ON XAD-2 TRAPS

          Run No.                            Hydrocarbons/g XAD (mg)

            26                                           18
            33                                            6
run 33 at 6 mg/g hydrocarbons but in run 26 at 18 mg/g hydrocarbons severe
breakthrough is observed.  Run 32 showed the exaggeration of the effects
seen in run 26 due to twice the gas volume and presumably even higher hydro-
carbon accumulations.
     The tandem XAD-2 cartridges were analyzed for runs 32 and 33 with the
results given in Table 7.  The higher breakthrough of naphthalene on the
cartridge was somewhat surprising until the residence of the sample in the
cartridge (0.75 sec) was compared with that of the trap (1.8 sec).  This
residence time is especially important for the less volatile compounds which
may form aerosols.
     Recovery studies were performed with 11 to 13 commonly observed organics
ranging from the most volatile, methyl thiophene, to the least volatile,
anthracene.  In these experiments a mixture of these compounds in methanol
was applied by direct injection into the bed of three prepared cartridges
(1.5 cm x 6 cm) and permitted to equilibrate.  The mixture contained d1Q-
anthracene as an internal standard.   After equilibration each cartridge was
extracted separately with methylene chloride for 22 hours in a Soxhlet
extractor.  The extracts were evaporated in K-D evaporators to 1  ml and
analyzed by GC/MS/COMPUTER (see Appendix II).
     The recovery experiments were repeated using a mixture of compounds
shown in Table 8 with a 20 g portion of XAD-2 to simulate the sample from
the XAD-2 traps.   In these recovery experiments three internal standards
were used, including d^-anthracene at the start of the evaporation and
p-chlorotoluene at the final  step before analysis.   Recoveries were deter-
mined based on each internal  standard.   In this evaluation only 4 ml of
~150 ml  was evaporated to 1  ml.
                                    26

-------
                                       TABLE 7.  BREAKTHROUGH ON XAD CARTRIDGES'
ro

Compound
Methyl thiophenes
C2-thiophenes
Cp-benzenes
Benzofuran
Phenol
Cresol
Naphthalene
Biphenyl
Dibenzofuran

Cartridge
1 (yg/D 2
0.8
<0.1
10.7
0.9
25
17
105
1.6
0.5
Run 32
No.
(ng/i)
ND
ND
0.2
ND
ND
ND
24
0.9
ND

Breakthrough
0
—
1.9
0
0
0
19
36
0

Cartridge
1 (yg/1) 2
31
5.6
<236
ND
239
121
263
2.2
0.8
Run 33
No.
(pg/i)
ND
ND
0.4
ND
ND
ND
32
1.2
ND

Breakthrough
0
0
<0.2
0
0
11
35
0

          Volume of process stream sampled was 5 liters.

-------
TABLE 8.  COMPONENTS OF STANDARD
ng/1
Anthracene-d-jQ
2-Chl oronaphthalene
p-Chlorotoluene
Indole
Phenol
3, 4-Dimethyl phenol
3,5-Dimethylphenol
2,3-Benzofuran
Toluene
1 , 2, 4-Trimethyl benzene
o-Cresol
p-Cresol
Indan
Indene
no
107
36
124
220
114
132
174.6
138.7
140.1
164.4
162.8
154.2
159.4
ng/i
Naphthalene
Anthracene
Phenanthrene
Benzofuran
Biphenyl
Methyl indole
Pyridine
Methyl quinoline
Ethyl benzene
Methyl thiophene
Ethyl thiophene
Quinoline
Methyl pyri dine

126
96
118
170
94
192.6
157.1
173.8
138.7
163.1
158.9
174.9


               28

-------
     The average and standard deviation of the response from each compound
ratioed against each internal standard response for the same injection were
computed.  From these the percent deviation relative to the average was used
as a measure of precision for each compound relative to each internal  standard.
     The recovery of the compounds selected for quantitation under the para-
metric variation phase of the study were evaluated.  The results shown in
Table 9 are for the discrete XAD-2 cartridge sampling procedure.
     One set of triplicates had two internal standards including p-chloro-
toluene, added just before the GC/MS/COMPUTER analysis.  Overall average
recoveries were 83 to 87 percent.  Anomalous results were obtained for indan
on one set of cartridges and for naphthalene on the other.
     Similar experiments were conducted for the XAD-2 trap samples.  Two sets
of conditions were examined; the usual 10-fold evaporation from 5 ml to 0.5 ml
and a larger 40-fold evaporation from 195 ml to 5 ml.  The latter was used in
some earlier work and occasionally when a larger sample mass was required.
The results are given in Tables 10 and 11.  Three internal standards were used
for the evaluation.
     The analysis of quinolines is very erratic (see below), and when this
class is eliminated, average recoveries are those shown in Table 12.  Inter-
pretation of these results is complex since there are most likely competing
effects.  There is the possible loss of d-|Q-A or d^-P during the sample
handling process.  Possible  loss of the sample compounds  during handling would
tend to compensate for this.  Add to these effects of varying instrumental
precision with compound  and  internal standard, and an accurate  assessment of
the limiting parameters  becomes more difficult.  Substantial amounts  of d^Q-A
are lost in the 40-fold  concentration step  relative  to  dg-P and almost no d&-P
was lost relative to p-CT.   In the 10-fold  concentration  of solvent other
factors  seem to be  involved, perhaps  instrumental, and  the changes are not
large relative to the variability of  the  recoveries.
                                     29

-------
          TABLE 9.   EVALUATION  OF RECOVERIES FROM XAD-2 CARTRIDGES
Compounds
Methylthiophene
Ethyl benzene
Ethyl thiophene
p-Ethyl toluene
Benzofuran
Indan
Indene
Phenol
o-Cresol
Dimethyl phenols
(2 isomers)
Naphthalene
Biphenyl
Anthracene
Mass Added
to Cartridge
(vg)
102
86
100
86
109
96
100
106
103
20
181
152
45
Percent Recovered
d Aa'b
Q10A
55 ± 8
63 ± 12
64 ± 10
91 ± 9
—
189 ± 22
42 ± 11
60 ± 2
66 ± 25
ND
127 ± 48
70 ± 42
135 ± 72
p-CTC
60 ± 25
67 ± 25
68 ± 27
83 ± 31
--
155 ± 49
42 ± 10
64 ± 25
63 ± 9
ND
122 ± 18
63 ± 4
124 ± 2.5
Replicate
p-CTC, d
72 ± 13
90 ± 10
73 ± 30
81 ± 28
26 ± 8
35 ± 8
33 ± 4
86 ± 6
100 ± 17
57 ± 7
230 ± 42
107 ± 9
102 ± 3
aAverage and standard deviation of triplicate analysis.
 Internal standard was d1Q-anthracene added at the beginning of the extraction.
clnternal standard was p-chlorotoluene added after the evaporation.
 Extraction repeated on another time.
                                    30

-------
  TABLE  10.   RECOVERY  OF COMPOUNDS FROM XAD-2 RESIN WITH EVAPORATION
             OF  SOLVENT FROM 5 ml  TO 0.5 ml

Spectrum
Nrv
liU .
—
—
--
--
48
50
80
180
185
188
225
230
256
280
300
360
360
409
422
450
504
510
592
736
741

741
Compound

Pyridine
Methyl pyridine
Toluene
Methyl thiophene
Ethyl benzene
Ethyl thiophene
p_-Xylene
Benzofuran
1, 2, 4-Tri methyl benzene
Phenol
Indan
Indene
o-Cresol
jD-Cresol
2, 6-Dimethyl phenol
3, 5- Dimethyl phenol
Naphthalene
Quinoline
Methylindole
Indole
Biphenyl
Methylquinoline
Dibenzofuran
Phenanthrene
Anthracene
p-Chlorotoluene
d5-Phenol
dig-Anthracene
Percent
a
V
__
—
.
—
88
51
55
22
67
72
44
59
69
65
74
75
57
236
50
80
133
159
77
136
118
61
72
recovery
h
d5pb
—
—
--
—
99
66
72
37
88
93
60
79
116
110
80
127
57
531
60
107
159
190
91
139
142
80
130
relative to
p
p-CTc
—
--
--
--
33
83
90
37
110
116
75
99
116
110
72
122
87
360
74
135
196
234
113
202
175
--
126
162

ad,g- Anthracene
bd,-Phenol
b




Cp-Chlorotoluene
                                    31

-------
   TABLE  11.   RECOVERY OF COMPOUNDS FROM XAD-2 RESIN WITH EVAPORATION
              OF SOLVENT FROM 195 ml to 5
Spectrum
No
1 1 \j •
—
--
—
—
5
8
20
98
101
105
154
160
185
214
240
320
320
280
290
422
480
500
570
712
719

100
719
Compound

Pyridine
Methyl pyridine
Toluene
Methyl thiophene
Ethyl benzene
Ethyl thiophene
o-Xylene
Benzofuran
1 , 2 ,4-Tri methyl benzene
Phenol
Indan
Indene
o-Cresol
f-Cresol
,6-Dimethyl phenol
3,5-Dimethylphenol
Naphthalene
Quinoline
Methyl indole
Indole
Biphenyl
Methylquinoline
Dibenzofuran
Phenanthrene
Anthracene
p- Chi oro toluene
d5-Phenol
dig-Anthracene
Percent recovery relative to
i .a
dioA
_-
—
--
--
323
263
336
345
236
259
178
166
203
219
130
236
131
346
90
146
212
442
86
139
137
263
275
• w
K
d5Pb
__
—
--
—
114
90
111
115
84
89
62
59
72
78
68
101
78
204
26
52
60
125
24
39
39
90
_ _
34

p-CTc
_ _
.—
— _
_-
90
100
124
124
93
99
70
65
80
87
53
88
47
123
32
57
75
155
30
49
48

112
38
^g-Anthracene
bd5-Phenol




Cp-Chlorotoluene
                                  32

-------
             TABLE  12.   AVERAGE  PERCENT  RECOVERIES  FROM  XAD-2
Degree of
Concentration
10-fold
40-fold

d]0-Aa d5-Pb p-CTc
78 ±29 94 ± 32 107 ± 46
201 ±79 72 ± 28 74 ± 28
 d,Q-anthracene
Dd5-phenol
"p-chlorotoluene
     Based on an idealized chromatographic peak and using  a 1.7  sec  scan
cycle, one can calculate the limits of error introduced solely by not
acquiring intensity data for an ion at the peak maximum.   These  data are
shown in Table 13.
       TABLE 13.  PERCENT DEVIATION OF RESPONSE RATIOS FROM IDEALITY
                           AS A FUNCTION OF PEAK WIDTH

Peak Width
10 sec
8 sec
6 sec
4 sec
Deviation
1 .7 sec cycle*
18.5
25
33.5
57
1.0 sec cycle
9.5
13.4
18.7
30
*Normal operating mode.
Actual measurements were made and calculated worst case results were not
realized.  Results are given in Table 14.  Relative retention times for
the compounds are given since, in general, peak widths increase with retention
time.  An unexpected trend was observed in these data.  The relative deviation
                                      33

-------
                                    TABLE  14.   AVERAGE OF DEVIATIONS (%) OF RESPONSE
c*>
Compound
Toluene
Ethylthiophene
Pyrldine
Ethyl benzene
Methylpyridine
Phenol
p-Chlorotoluene
2,3-Benzofuran
1 , 2, 4-Tri methyl benzene
In dan
Indene
p_-Cresol
p_-Cresol
Naphthalene
3,5-Dimethylphenol
3, 4-Dimethyl phenol
Quinoline
Methylindole
Indole
2-Chloronaphthalene
Biphenyl
4-Methylquinoline
Benzofuran
Phenanthrene
d,Q-Anthracene
Anthracene
Relative
Retention
0.16
0.16
0.2
0.44
0.56
0.96

1.28
1.32
1.68
1.74
1.96
2.2
2.88
2.92
3.08
3.28
3.4
3.6
4.0
4.1
4.2
4.8
5.88
5.90
5.92
d,Q-Anthracene
9
11
29
32
39
30

27
31
25
23
29
22
31
38
12
19
10
1

13
6
3
6

11
2-Chloronaphthalene
19
17
14
14
21
11

13
12
16
4
12
7
11
7
5
3
6
7

9
16
12
10

27
p-Chlorotoluene
10
6
8
7
19
9

7
4
5
6
2
14
7
11
17
7
16
20

13
25
24
10

5

-------
was lowest for the internal  standard with the retention time nearest to that
of the compound.  It is unclear at this time whether this observation is due
to the chromatography or some aspect of the mass spectroscopy since the mass
of the ions used also correlate with retention times and the deviations.
     Typical results from GC/MS analysis of the coal gasification process
streams collected on XAD-2 cartridges from three runs are presented in Table
15.  In addition qualitative identification of a large matrix of compounds is
presented in Appendix III.
                                     35

-------
                   TABLE 15.  QUANTITATIVE GC/MS RESULTS FOR SELECTED COMPOUNDS ON XAD-2 RESINS
                                                 (yg of compound/g  of coal)

Benzofuran
Indan
Indene
Phenol
Biphenyl
Naphthalene
Anthracene
Phenanthrene
Di benzof uran
Al kyl benzenes
Cresols
Xylenols
Thiophenes
Pyridines
Qu inclines
Indoles
Run 47
Wyoming
Surge
43
20
89
190
45
94
3
ND
0.7
>300
>110
56
74
ND
59
ND
Steady-State
7
3
36
48
1.5
56
0.4
ND
ND
112
36
21
14
ND
ND
ND
North
Surge
35
9
91
98
2
no
1
2
1
>460
77
160
78
ND
36
ND
Run 51
Dakota Lignite
Steady-State
2
0.5
7
11
0.5
19
0.5
1
0.5
100
5
1
2
ND
ND
ND
Run
Illinoi
Surge
3
1
10
25
1.0
52
0.6
1.0
ND
54
ND
ND
1
ND
ND
ND
52
s No. 6 Coal
Steady-State
2
1
5.5
15
2
100
1
2
0.2
62
ND
ND
3
ND
ND
ND
CO
       ND = Not detected.

-------
                          5.0  ANALYSIS  OF LIQUIDS

     Comprehensive analysis of liquids generated  during  coal  gasification  can
be separated into three categories.   Initial  activities  involve separation of
crude tar and aqueous condensate,  determination of  pH and total volume of
condensate and total  weight and density  of the crude tar.  The second step in
the analysis involves fractionation of a crude tar  sample into chemically
distinct classes of organic compounds by application of  a wet-chemical parti-
tion scheme.  In the final step, qualitative and  quantitative analyses are
performed on specific partition fractions and the amount of phenol, cresols,
and xylenols in the aqueous condensate  are ascertained.
5.1  SEPARATION OF TAR AND CONDENSATE
     Crude tar and aqueous condensate generated by gasification are contained
in a trap designed to collect liquid byproducts.   Prior to comprehensive
analysis, the tar and condensate must be isolated from each other.  Separation
is performed by allowing the mixture to drain through a filtration system into
a separatory funnel.  While a large portion of the crude tar is adsorbed
during filtering, some residue is carried onto the separatory funnel with the
condensate.  This is removed by swirling the funnel gently, allowing the  crude
tar to settle, then draining it off.  The tar free condensate  is poured into
a graduate cylinder where  the total volume, pH, and color are  recorded.   The
sample is then stored pending further analysis.
     Isolation of the crude tar begins  by dissolving the residue, which was
separated from the condensate,  in an organic solvent and transferring  this
solution to  the  separatory funnel.  An  aliquot of water  is added to  facilitate
identification of the  interface between the organic phase and  residual aqueous
condensate.  The organic  phase  is allowed to drain  into  a roundbottom  flask,
and the  solvent  removed.   The walls of  the condensate trap and  filter  system
                                       37

-------
 containing  the  adsorbed  tar  are  then  exhaustively  extracted,  the extracts
 combined, the solvent  removed, and  the  total weight  of  the tar determined.
 The  tar  is  redissolved in  solvent with  aliquots  removed for bioassay, ele-
 mental analysis, density determination, and partitioning.  Specific details
 and  results are presented  in upcoming sections.
     The filtration system used  to  adsorb the tar  consists of a powder funnel
 containing densely packed  glasswool,  followed by a funnel containing a fluted
 cellulose-acetate filter,  (50 cm, grade no.512).  A  500 ml or 1000 ml separa-
 tory funnel is  used to collect the  condensate.   In general, methylene chloride
 (Burdick and Jackson,  or Fisher  HPLC  grade) is used  to  isolate the crude tar;
 although ethyl  acetate (Burdick  and Jackson) was also used during a mass
 balance study for chlorine.  Solvent  evaporation is  performed by a Buchler
 rotary evaporator.  The  total weight  of crude tar and the tar density is
 determined and  the pH of the condensate is measured.
 5.2  TAR PARTITIONING
                                        g
     Tar samples are exceedingly complex  and require fractionation before
 direct analysis can be undertaken.  Other investigators have utilized either
 of two procedures for this process:   column chromatography or solvent partition.
 Chromatographic methods  separate the crude material  into fractions of like
 polarity and can function as a useful means of reducing a complex sample into
 one of more manageable proportions.     Solvent partition schemes have been
 devised, most notably by researchers from the tobacco industry,    in which
 group separations are accomplished on the basis of similar chemical pro-
 perties, e.g., acids, bases, etc.  The latter approach is more practical,
 particularly if fractions are to be chromatographed further.   A detailed
 schematic of the partitioning procedure used is shown in Figure 8 and is a
                                              12
modification of the method utilized by Novotny   for air particulate
extracts.
     Five fractions are produced by application of the scheme:  acids,  bases
and three neutral  fractions.   These three fractions are designated nonpolar
 (aliphatics and 1-2 ring  aromatics), medium-polar (polynuclear aromatic
 hydrocarbons-PNA's) and polar (oxygenated material).   Each group is then
 either analyzed directly  by gas chromatography/mass spectrometry (GC/MS), or
 is chromatographed using  high performance liquid chromatographic (HPLC)
 techniques.
                                    38

-------
                                [Cruj> Tor (ig.)
                                          WnhJXwHhlNNaOH
                         CH2CI2 byw	,    NtOH teywt

WahZXmth
H20
1
1
il
CH^Iiyw
WHh«rithCH2Cl2
NiOt
hy*r
                   r
               CH2CI2liytr	H20 hyw
                        WishZX
                        with 10%
                                   (pH-10)
                                                                 Adjust to pH 2 with 6NHC1
                                                                 Extract «ritkCH2Cl2
                        HjSOiind  I
                        IX with 2
                         20X|,
                   H2SO,
                                  Wash with
                                  cyclohflXMM
|   OntoicAridi
                                                                      AQ
                              CydohMim layw     H20 toytr
                                 (1) Evtpont* to
                                   dfyntss.
                                 (2) Dimln in CH^
                                               Mi«tt*»H2«rith1NHC1
                                               Extract 3X with CH2CI2
              With 2X with
              H20
                                          OrfwieAeidi
                               Wish2X
                                                                I
                                                               AQ
                        CH2CI2 liyir
                                     H2804lty.r
                                             AdJ8ifto»H12wit(i10NNiOH
                                             Extract 3X with CHi-C^
CH2CI2hy« ---- .  H2OUv«
                 I  fph-W
                                    OtftnitBtttt
                                                        I
                                                       AQ
                          »"*BXwithCH2CI2
                                      H20toy«
         (3)WMb3Xwith
           4:1CH3OH/H20
                                     |AdjuittopH12with1NNiOH
                                     Extract 3X with CH2CI2
    I
Cydohoxtm Ityir

      (DCoMMtnti
                        I
                    CH3OH/H20 toyor

                          With 4X with cyclohtxtM
                                             CH3OH/H20 (tyor


                                                  Fnondry
          Figure 8.   Tar partitioning  schematic.
                                      39

-------
 5.3   GC/MS ANALYSIS
      As with vapor phase samples, qualitative analysis of fractions subjected
 to GC/MS  is achieved by comparison of the mass cracking pattern of the unknown
 mass  spectra to an eight major peak  index of mass spectra.    Quantitative
 results are obtained via internal standard procedures using the detector
 response  of the peak intensity as provided by the mass spectrometer.  Internal
 standards consist of deuterated materials, phenol -d^, quinoline-dy and
 anthracene-d-|0.  This process can be applied to single species as well as,
 in certain cases, to classes of compounds.
      For  single species, in order to eliminate the need to construct complete
 calibration curves for each compound to be quantitated, the method of relative
 molar responses (RMR) was used.  The calculations are as follows:

                           A /moles
                     RMR -                                            0)
where A = peak area or other suitable response parameter (u and s designating
          unknown and internal standard).

Equation (1) is equivalent to:
                     RMR •
where g = grams and mw = molecular weight.

Rearranging,
                        _ Au • mwu • g$
                     9u ~ AS • mws • RMR

The value of the RMR is determined from at least three independent analyses.
                                     40

-------
     The basis of class quantitation by GC/MS lies in the selection of a  mass
spectral fragment ion which is common and unique to all  members of a class.
If, in addition, that ion is of approximately the same relative intensity for
all, or nearly all, class members, then all  compounds can be quantitated  on
the basis of that single ion.   In practice,  no ion is unique to one class,
but for samples which are highly enriched in single compound types, the
presence of a characteristic ion for one type is not unusual.  Research of a
mass spectral catalog   of fragmentation patterns of alkylated indoles,
pyridines, quinolines, and naphthalenes, revealed an ion that meets the
criteria outlined above for each of these classes.  Using the mass spectro-
meter to generate single ion plots (using the characteristic fragment ion)
peaks, are obtained for every class member.   Intensity data are then avail-
able for use to quantitate every compound.  Summation of the results for
every compound provides a reasonably accurate assessment of the amount of
material associated with individual classes.  The acid fraction is analyzed
by either glass capillary GC/MS or HPLC.  The latter method has recently  been
developed, and owing to its greater speed and reliability, will be routinely
utilized for future analyses.
     For GC/MS analysis, an aliquot of the acid fraction (spiked with phenol-
dr- as internal standard) is introduced into the GC/MS system.  The intensities
 b
of those peaks corresponding to phenol, cresols, xylenols and higher alkylated
phenols are  recorded and compared to the intensity of the internal standard.
Using response factors (previously calculated from reference standards),
quantitation is carried out on each compound.
     For analysis of tar acids by HPLC the method used  is the same as  is
described below for gasifier condensate.  Aside  from speed and ease of
analysis, the HPLC appproach utilizes the aqueous solution of acids instead
of the methylene chloride extracts which  are  used for the GC/MS procedure.
This  is  important since the difficulties  associated with the extraction  of
phenolics from water probably  results  in  an  inaccuracy  of the GC/MS method.
     The tar bases are analyzed  directly  by  glass capillary  GC/MS.  The
internal standard used here is quinoline-dy,  and quantitation  is  performed  as
described for  tar acids.  Compounds  routinely analyzed  are:   alkyl  pyridines,
quinoline,  alkylquinolines, alkylbenzoquinolines and  acridines.
                                    41

-------
      The subfractlonation  of the neutral  fraction  consists  of  solvent parti-
 tion  of a cyclohexane  solution  of the  neutrals  with  methanol/water  (1:4) and
 nitromethane.   The  polar neutrals are  recovered in the methanol/water fraction
 and the medium-polar neutrals are associated with  the nitromethane  fraction.
 Those materials remaining  in solution  in  cyclohexane comprise  the nonpolar
 neutrals,  and  are analyzed  directly  by GC/MS using wall coated open tubular
 (WCOT)  glass capillary columns.
      The medium-polar fraction  containing PNA hydrocarbons  is obtained by
 preferential partition from cyclohexane into nitromethane.  Analysis is
 carried out by GC/MS using  the  same  column as described earlier for nonpoTar
 neutrals.
      Those neutral materials that are  washed into methanol/water (1:4) from
 cyclohexane are termed polar neutrals.  The nature of the compounds comprising
 this  function  is such that  direct analysis by GC/MS  is not  practical.  Various
 modes of high  performance liquid  chromatography  (HPLC) have been carried out
 with  limited success.
      Because of the intimate admixture of the coal tar and  condensate in the
 tar/water  trap,  a large quantity  of  organic matter is associated in the
 condensate.  The bulk of this material is comprised of the more water soluble
 phenols.  Although analysis  for various hydroxylated aromatic compounds were
 initially carried out by GC/MS, due  to the well   known difficulty of extracting
 phenolics into  organic phases and  the  slower turn-around associated with
 GC/MS samples,  a more direct analytical method was sought.
     The use of HPLC for the analysis  of phenolics is relatively straight-
 forward, and a  procedure utilizing this technique, was developed.   Because of
 their overwhelming abundance in the  condensate sample compared to all  other
 hydroxyaromatics, only phenols, cresols (3 possible isomers) and xylenols (6
 possible isomers) were analyzed.  Chromatographic conditions were such that
 single peaks were obtained for phenol, all cresol isomers, all  xylenol  isomers
 and the  internal standard, isopropyl  phenol,  in  that  order.  The  HPLC  system
 uses a reverse phase column with methanol/water as mobile phase.   The UV
 transparency of this solvent system permits detection at 215 nm,  a wavelength
where all cresols and xylenols have about the same extinction  coefficient.
 Thus,  the quantisation of coexisting  species is  valid, and only three
                                    42

-------
calibration curves need be prepared instead of one for every isomer.   Using
reference standards, calibration curves for the three compound groups are
prepared and the regression equations (least squares) determined.   In all
cases the square of the correlation coefficients was greater than  0.99.  The
curves were validated through the analysis of known samples.
     Condensate samples are pH adjusted (~pH 4.5) to ensure that phenolics are
present in their acid form, a known amount of internal standard is added and
equilibrated and an aliquot is injected onto the hplc.  Peak areas are deter-
mined by electronic integration and amounts of phenol, cresols and xylenols
are calculated from the appropriate regression analysis.
     Typical results for the analyses of selected compounds in tar and aqueous
condensate from three runs using Wyoming, North Dakota and Illinois No.6 coals
are presented in Tables 16 through 19.  In addition, qualitative identifications
of a large matrix of compounds in various tar fractions are reported in Appendix
III.
             TABLE 16.  ELEMENTAL ANALYSES* OF CRUDE TAR
                              (expressed in wt %)

Run No.
47
51
52
C
81.4
82.9
85.5
H
7.8
6.9
6.5
N
1.1
1.0
1.0
s
1.5
2.5
1.9
0
7.8
6.5
4.9

*Analyzed by Galbraith Laboratories, Knoxville, TN.
                                    43

-------
                  TABLE 17.   TYPICAL  PRELIMINARY AND  PARTITION RESULTS FOR TAR AND CONDENSATE


Run
No.
47
51

52




Coal
Wyomi ng
N, Dakota
Lignite
Illinois
No. 6

Total
Tar
(g)
28.8
17.7

49.3

Tar

Density
g/ml
1.14
1.33

1.39


Acids
wt %
31.0
23.5

19.8


Base
wt %
2.6
4.5

2.7


Ins.
wt %
2.0
11.0

9.0


NPN
wt %
27.8
17.1

16.8


PN
wt %
18.1
7.7

7.2


PNA
wt %
18.5
136.5

44.6

Condensate

Volume
pH ml
11 500
9 700

8-9 640

Organic
Extractables
9
4.2
4.2

NA

Ins:   Insoluble in cyclohexane.
NPN:   Nonpolar neutrals.
PN:   Polar neutrals.
PNA:   Polynuclear aromatics.

-------
         TABLE 18.  QUANTITATIVE RESULTS FOR SELECTED COMPOUNDS
                                  IN TAR FRACTIONS
                                (expressed in grams)
Fraction
PNA








Organic
Bases



Organic
Bases




Naphthalene
Alkylnaphthalene
Fluorene
Dibenzofuran
Anthracene +
Phenanthrene
Fluoranthene
Pyrene
Chrysene
5-Ring Compounds
Alkylpyri dines
Quinoline
Alkylquinolines
Alkylbenzoquinolines
Acridine
Phenol
Cresols
Xylenols
o-isopropyl phenol
trimethyl phenol
Run 47
0.06
0.06
0.03
0.02
0.05
0.02
0.02
0.01
Trace
0.007


<0.0005
0.62
1.13
1.36
0.02
0.12
Run 51
0.57
0.20
0.09
0.08
0.22
0.05
0.04
0.04
0.01
0.04
0.04
0.01
0.01
0.003
0.39
1.19
0.17
0.25
0.05
Run 52
2.44
0.38
0.27
0.31
1.31
0.45
0.31
0.23
0.18
0.03
0.01
0.01
0.18
0.003
0.02
0.05
0.02
<0.0005
0.001
GC/MS Conditions:   LKB-2091  GC/MS, 15-20 m WCOT capillary column,  1% Se-30/
                   BaC03, 100°/2 min/8° min/265°.
                                    45

-------
 TABLE 19.  QUANTITATIVE RESULTS FOR PHENOLS, CRESOLS, AND XYLENOLS
                IN CONDENSATE BY REVERSE PHASE HPLC AND GC/MS
                              (expressed in grams)


Run No.         Phenol           Cresols          Xylenols          Method
47
51
52
0.58
2.03
1.93
0.28
1.04
1.02
0.10
0.16
0.12
GC/MS
HPLC
HPLC

GC/MS:
HPLC:
Finnigan 3300 GC/MS
1000/8°min/250°.
Reverse phase HPLC
600 psi; SF-770 VV
, 24 m SCOT 1% SE-30
analysis, 1:1 MeOH/H-
variable wavelength a
glass capillary column,
0, Water's
etector, 2
Associates, pump
ml/min.
                                    46

-------
                      6.0  ATOMIC  ABSORPTION  METHODS

     Selected trace elements  are analyzed  by  atomic absorption  (AA)  following
various digestion procedures.   The following  sample preparation  and  analysis
scheme is used.
6.1  SAMPLE PREPARATION AND ANALYSIS
     The various trace elements that have  been  analyzed  include  Hg,  As,  Se,
Sb, Cd, Pb, Ni,  Cu, Cr, and Be. They can  be  subdivided  into  three groups
based on the sample preparation and analysis  procedure.   Methods for Hg  are
unique and it can be classified as Group  1; As, Se, and  Sb  are  in Group  2, and
Cd, Pb, Ni, Cu,  and Be are in Group 3.  A  further subdivision is possible
based on the sample matrix.  During the course  of this project,  five different
types of samples have been analyzed.  These  include parent  coals, reactor
residues, tars,  aqueous solutions  (condensate or scrubber)  and  sorbent
materials for gaseous effluents.   A mass  balance has  been attempted  from the
results for various tests and is discussed further in Section 6.3.   Sample
preparation and  analysis procedures for the  three groups and  the various
matrices are given in the following subsections.
6.1.1  Coals and Residues
       For Hg analysis a 1-g  sample of coal  prepared  for gasification (approxi-
mately 6-14 mesh) is placed in a platinum cup.   The sample  is then  placed in a
quartz-lined Parr oxygen bomb calorimeter containing  10 ml  of 10 percent
nitric acid.  The vessel is pressurized to 24 atm of Q£ and the sample is
combusted.  After cooling, the resulting solution is  then diluted to 25 ml
with 10 percent nitric acid and submitted for analysis by the cold vapor
           1415
technique.   '    For Hg  in the residue, 1 g of residue material  is mixed with
0.5 g  benzoic acid  (after  being ground to approximately 200 mesh), subjected
to Parr 02 bombing  and analyzed as  for the coal  procedure.    For  Group 2 and
Group  3 elements,  the  residue  is  heated and analyzed in exactly  the  same way
as coals which  is  described below.
                                     47

-------
      For Group  2  elements,  the  coal  (or  residue)  is  crushed to 200 mesh and
 subjected  to  one  of  two procedures.  The first  procedure  is based on wet acid
 ashing; 0.1 g of  coal  sample  (or  0.03  g  residue sample) are treated to a
 mixture of 3  ml concentrated  HLSO. and 5 ml concentrated  HNO-, for 1/2 hour on
 an aluminum block heated to 200°C.   After 1/2 hour,  the mixture is cooled and
 treated with  an additional  2  ml of concentrated perchloric acid and digested
 overnight  for removal  of the  nitric  and  perchloric acids.  The mixture is then
 cooled and 7  ml distilled water,  2 ml  20 percent  ascorbic acid and 1 ml 0.06 M
 Nal are added,  respectively,  mixing  after each  reagent addition.  Finally, the
 mixture is extracted with 2 ml  of toluene.  The toluene phase is then analyzed
 for antimony  using electrothermal atomic  absorption.  One ml of 6.0 M Nal is
 then  added to each mixture and  reextracted.  The  toluene phase is then analyzed
 for As and Se by  electrothermal AA.  Sixty pi of  0.1 percent Ni (N03)2 is
 first added to the furnace and  dried.  Twenty yl  of  sample is then added and
 analyzed.  The entire  procedure takes,  two to three days.
     A second procedure may be  used  for coal (or  residue) analysis.     The
 coal  (1.0  g)  or residue (0.2  g) is subjected to a dry ashing at 750°C after
 mixing with 1 g of Eschka mixture (MgO and Na2CO.J.  After four hours the
 mixture is cooled, dissolved  in concentrated HC1 and diluted to 100 ml  with
 distilled  water.  The  resulting solution  is analyzed for As, Se, and Sb by
 hydride generation AA.  Here  the analyte  is reduced  to the respective hydride
 with sodium borohydride and the resulting gas is  swept into an argon-hydrogen
 flame for  subsequent analysis.
     For Group 3 elements, the coal  (or residue)  is  dry-ashed at 450°C for
 eight hours.   After the ash content  is determined, the sample is then sub-
jected to  acid digestion and  analyses  by AA; 0.1 g of ash is placed in a Parr
 acid bomb  teflon cup and wetted with 1 ml aqua  regia.  Three ml of concen-
 trated HF  is added to  the mixture.  The resulting mixture in the teflon vessel
 is placed  in the stainless steel bomb  and digested in an oven for 150°C for 30
minutes.   The vessel  is cooled, and the mixture is treated with 2.8 g boric
acid.   The resulting solution is diluted to 50 ml  with distilled water and
                                                                    18
analyzed for Cd, Pb,  Ni, Cu, Cr, Be and others by electrothermal  AA.
                                    48

-------
6.1.2  Tars
       The tar material  is extracted out of the condensate traps with
methylene chloride.  The extract is evaporated at room temperature under a
stream of N2-  For mercury, 1  g of tar is Parr 02 bombed and analyzed by the
procedure given in Section 6.1.1.  The resulting residue is dissolved in 1 ml
of aqua regia and subjected to electrothermal  AA for analysis of lead and
arsenic.
6.1.3  Aqueous Condensate and Scrubber Solutions
       The water condensate is extracted with  methylene chloride to remove
the organics.  The methylene chloride phase is combined with the tar con-
densate materials in Section 6.1.2.  The pH of the aqueous phase is adjusted
to 4.0 and extracted with methylene chloride to remove the phenolic material.
The aqueous phase is then analyzed directly by the cold vapor technique for
Hg and electrothermal techniques for the other elements as described in
Section 6.1.1.  Acidic scrubbers are analyzed  by direct aqueous injection
onto the AA.
6.1.4  Sorbent Materials for Gaseous Effluents
       These materials include XAD-2 resins and charcoal traps.  Again the
sorbents are subjected to Parr 02 bombing and analyzed for Hg, As, and Pb by
the appropriate AA procedures mentioned in Section 6.1.1.
6.2  QUALITY CONTROL PROCEDURES
     All samples are analyzed in duplicate.  Wherever possible, the NBS
(National Bureau of Standards) Standard Reference Materials are used in con-
struction of calibration curves.  These include SRM 1632a bituminous coal,
SRM 1633 flyash and SRM fuel oil which are analogous in matrix type to the
coal, residue, and tar materials encountered  in this study.  Wherever NBS
SRM's were unavailable, the method of standard additions was employed to
compensate for matrix mismatching between sample and standard.
                                     49

-------
  6.3  TYPICAL RESULTS
       Figure 9 shows the AA analysis for seven trace elments  in  six U.  S.
                                                        19
  coals.   For comparison, results from Gluskoter,  et al.    are presented.   Good
  qualitative agreement is found.  Figure 10 shows the percentages  of trace
  elements lost to gaseous and liquid effluents for Pittsburgh No.8 coal.
  These are obtained by difference between the amounts in the  coal  and the
                                                   20
  residue.  Again for comparison, results of Attari   are presented.   Finally,
  Table 20 shows overall  material balance and the  fate of trace elements from
  gasification of Illinois No.6 coal.  Mercury is  almost  totally  evolved and is
  expected to end up in the gas.   Gas phase analyses are  as yet not available
  but can be crudely estimated by difference.   Almost twice as much chromium is
  recovered than fed.   This may be a contribution  of the  stainless  steel reactor.
     TABLE 20.  AVERAGE MATERIAL BALANCE FOR THREE ILLINOIS NO. 6 COAL
                                    GASIFICATION RUNS

As
Be
Cd
Cr
Hg
Pb
Se
Coal in
5050
1396
183
27. 5K*
188
3716
2407
yg (Average)
Char Tar HgO
3056 176
1265
139 2
52K*
7.1
1639 41
576 27
633
—
7
0.90K*
--
65
668
Vapor
1768+
NA
NA
NA
NA
NA
NA
Material
Balance
76.5% M
dm)
90.6%
80.9%
192.0%
3.8%
45.0%
53.0%
*K = 1000.
**Accounting for vapor.
^Estimated from results obtained using a 100 mg charcoal trap.
NA: Not analyzed.
                                     50

-------
  .30 -
2 .20
  .10
       Hg
                                             Pb
              1  •  T
I
i   ill    t   i
I   P   K  M  W  N
                        9.0 -
                      •S?
                      t  6.0
                         3.0
                                                   .
                                     I   P   K  M  W  N
                            Se
                3.6
                2A
                 1.2
                        11
                      I   P  K  M W  N
                                        Legend:
                                     I. Illinois #6
                                     P. Pittsburgh #8
                                     K. Kentucky //9
                                     M. Montana Rosebud
                                     W. Wyoming Sub-bit.
                                     N. North Dakota lignite
                                     Note:  RTI analysis results
                                            shown relative to
                                            range of Gluskoter,
                                            et.al.
                     Raf. Gluskoter et a).. 6-77
      Figure 9.  Range of selected trace elements  in coals.

                              51

-------
                 As
      18
      12
       6
              i    t
           I   P  K  M  W  N
   1.8
I 1'2
   0.6
                Be
                                          1    II
                                                         I	1
        i   P   K  M   W  N
.30

f -20
.10


LrO
t
-•

•
•
i
i
i

1







*

i
i

<
>

i

30

f 20
10




1

*

, |
•

1 I
ur

>
«


i
i .


A*
\ m

IPKMWN IPKMWN
         Reference:  H. J. Gluakoter. et al. "Trace Elements in
                    Coal:  Occurence and Distribution"
                    Illinois State Geological Survey
                    report to IERL-EPA 6-77
Figure 9.  Range of  selected trace elements  in coals,  (continued)
                                  52

-------
100
 80
 60
 40
  20
Ref: A. Attari
    Institute of
    Gas Technology
                                                 8-73
         A. Attari
         RTl
         As    Be    Cd    Hg   Pb   Se
   Figure 10.  Percent  loss of effluents (Pittsburgh No. 8 coal).
                              53

-------
               7.0  FOURIER TRANSFORM  INFRARED SPECTROMETRY
     Application of developmental work has been initiated for the Fourier
Transform  Infrared Spectrometer  (FTIR) to the analysis of coal gasification
samples  in three distinct areas:  (1) high resolution FTIR of gas sample, (2)
gas chromatography/FTIR of semivolatiles, and (3) low resolution FTIR of coal,
tar and  ash.
     Results from the FTIR analysis at coal gas were confirmed for 13 com-
pounds and five additional compounds were identified.  Two compounds were
quantitated.  High resolution FTIR, a nearly universal detector, is especially
useful for determining the presence or absence of specific pollutants of
interest which are not easily determined by routine gas chromatographic
methods.  The technique also serves as an independent check on the quantita-
tion obtained by GC.
     Over 24 individual compounds were identified in an XAD-2 extract from the
GC/FTIR spectra, including benzene, phenol, naphthalene, and some of their
alkyl derivatives.  In many cases, positional isomers were distinguished
(e.g., 2,4-xylenol and 2,6-xylenol).  Compounds were identified by comparison
with gas phase spectra of standards.  Selected compounds were also quantitated.
This technique is superior to GC/MS in its ability to distinguish positional
isomers.  The relative utility of class quantitation by GC/FTIR versus GC/MS
is still under evaluation.
     The analysis of coal, tar, and ash samples by FTIR indicates that general
information regarding types of inorganics and organics may be obtained.  The
qualitative and quantitative capabilities of the technique for these samples
are under further investigation.
7.1  HIGH RESOLUTION ANALYSIS OF GAS SAMPLES
     High resolution FTIR is particularly suited for the analysis of low
molecular weight polyatomic gases produced during coal gasification.   At high
resolution (0.125 cm"1),  the sharp lines produced by rotational  energy level
                                    54

-------
transitions are easily distinguished and measured for qualitative and  quanti-
tative work.  Quantities of gases are determined by comparing the absorption
of a preselected sample spectral  line with that of a standard.   Thus  it is
possible to simultaneously identify and quantitate every gas species  present
which absorbs IR radiation and which has had a standard curve generated (H2,
N2, and 02 do not absorb; also quantities of H20 and C02 cannot be determined
due to interference from these species within the optical path of the  spectro-
meter).
     Gases were received in 500 ml glass gas bulbs fitted with teflon  stop-
cocks.  A sample was attached to an evacuated (<2 mmHg) Perkin-Elmer Model
186-035 10 m Variable Path Gas Cell and expanded into the cell to equilibrium.
The cell volume was 4.4 liters and the optical path set at 8.76 m.  The cell
was fitted with NaCl windows.  The pressure was brought to 1 atm with N2 gas
at ambient temperature.  The sample spectrum was then collected (100 scans),
Fourier transformed, and ratioed to a previously collected background spectrum
of N2-  Identification was made by comparison with standards or reference to
published literature.
     Standards of COS, NH3 and HCN were prepared by making appropriate
dilutions of pure gas standards with No-  The standards were introduced into
the cell in the same manner as the samples.  A series of dilutions for each
gas was made,  the spectra collected, and a plot of absorbance versus wave
number for  a characteristic absorption  line prepared.  The absorbance for a
corresponding  wave  number  in the  sample gas was then compared to  the standard
and the concentration for  that species  determined.
      Four  gas  samples were analyzed  by  High Resolution  FTIR.  Quantisation
was performed  on COS  and estimated on HCN.  Table  21 shows  the  concentrations
found  in the four gas  samples.   Estimates  were  made  on  percent  composition
for gases  qualitatively  identified based  on absorbance  relative to known
species.   It was notable  that  surge samples  showed significantly  higher
amounts of COS and  C2  hydrocarbon gases than  did  the steady-state samples.
There  was  little correlation  among steady-state samples,  however.  Principal
components of  all  samples  were CH4,  CO, C02,  H^O and COS.   A total of  13
confirmed  species  and 5  suspected components  have been  identified in  the four
samples.
                                      55

-------
      Table 21.   HIGH RESOLUTION FTIR ANALYSIS  OF  SELECTED GAS SAMPLES

 Species          Run 49          Run 49           Run 50           Run  50
	Steady State	Surge	Steady State	Surge
 ru               i i •               J.J.A              J.J.             j-j.
 UMyt              TTT              TTT              TT             TT
 C2Hg              -                -H-              TR             -H-
 C2H4              "                ++              -              ++
 C2H2              -                -               -               +
 C6H6
 m                4-4.               4-4-              4-4-             XX
 LU                TT               TT              TT             TT
 Pfi               i | I               i i i              XXJ.            .l~i~l.
 \j\Jn              II T               TTT             TTT            TTT
 COS               TR              120              96            370
 C2H40             -                                 TR?
 CH3SH             -                 TR
 (CH3)2S           -                 TR?
 CH3C1             ....
 CH3NH2            ....
 HC1                                -
 HCN            >148                -                -            <148
 HCHO             -                                                TR
 NH3              ?                 -
 AsH3              -
 Ni(CO)4          ....
 (CN),            .?.?
 CH3OH            -                                 T              TR
 CH2CHCHO         -                TR
 NOX                               -
 coci2            -
 HN03             -                                 TR?
 Legend:
         >  10%  Gas Composition
    ++   1-10%  Gas Composition
     +   10 ppm - 1% Gas Composition
    TR   Lowest detectable confirmation of presence     <10 ppm
    —   None detected
    ?    Assignment uncertain; interfering absorbance  or no reference
           standard
 Concentrations expressed in PPM for quantitatively analyzed species
                                   56

-------
7.2  GC/FTIR ANALYSIS OF SEMIVOLATILES IN XAD-2 EXTRACT
     Because of its speed and sensitivity, FTIR can be used as a detector for
gas chromatographic effluents (GC/FTIR).   The technique has been applied to
the analysis of the XAD-2 extracts which  have heretofore been analyzed by gas
chromatography/mass spectrometry (GC/MS).  Infrared is much better at certain
class distinctions than mass spectrometry, due to the characteristic fre-
quencies for all compounds containing a given functional group.  Thus, GC/FTIR
may prove superior to GC/MS for routine quantisation of such classes as "total
aromatics," "total phenolics," etc.
     The goal of this study was to demonstrate the feasibility of GC/FTIR for
the analysis of semivolatile coal gas effluents and determine whether it is
suitable as merely a confirmatory technique, or whether it offers advantages
over GC/MS which would make it a primary analytical technique.
     A sample of extract from XAD-2 resin of gasifier run 41 was qualitatively
analyzed on a 2 mm ID x 180 cm glass column packed with 3 percent OV-17 on
Chromasorb G (AW) 100/120 mesh.  The column was installed in a Varian 3700 GC
equipped with a thermal conductivity detector and interfaced to the Nicolet
7199 FTIR system through the Nicolet GC-7000 GC/FTIR Accessory.
     An injection of 1.0 yl of the XAD-2 extract was chromatographed under the
following conditions:  temperature program (50°C/3 min to 240°C/10 min at
8°/min); He flow (30 ml/min); T... (222°C); TC detector temperature (280°C).
Infrared spectra consisting of four low  resolution scans were  collected when
the eluting species exhibited an absorbance greater than 0.005 over any of the
five preselected windows corresponding to specific chemical  group absorptions.
     The resulting chromatogram of selected infrared bands,  called a  "chemi-
gram"  is shown  in Figure 11.  The  band from 3600-3700  cm    is  selective for
compounds containing the hydroxyl  group  (the  three large peaks are phenol  and
the cresol  isomers).  The  3025-3100 cm"   band  is characteristic for aromatics
and/or alkenes.  The  1630-1730  cm~  band is  characteristic  for carbonyl-
containing  compounds.   Unfortunately  some noncarbonyl  infrared absorbances
 (notably from  overtones  in the  naphthalene  spectrum)  interfered;  no carbonyls
were detected  except  for an artifact.  The  1560-1630  cm"   band is  charac-
teristic for aliphatic  compounds.  The 1475-1520 cm    band is characteristic
 for aromatics  and  is  therefore  a  cross-check  with  the 3025-3100 cm   band.
                                      57

-------
96DO -  37OO
1"|75 -  ISZO
                                                          1O90  1200  1320
                                                          TIME. SECONDS
       Figure  11.   Chemigram of  infrared bands (cm"1)  versus time during GC/FTIR analysis of
                    extract of XAD-2  resin sampled during  surge phase of run  41.

-------
     Interpretation of the chemigram indicates  that most  (but  not  all)  of  the
compounds are alkyl aromatics.   There are a few phenolic  compounds.   Table 22
gives the qualitative identifications.   Aliphatic compounds  do not appear  to
be significant.
7.3  LOW RESOLUTION FTIR OF COAL, ASH,  AND TAR
     The coal, ash, tar and other complex liquids or  solids  may be analyzed
with minimal sample preparation to give general characteristics of the  sample.
The spectra contain information about both inorganic  (sulfate, carbonate
silicate) and organic constituents (aromatics,  aliphatics, and carboxylates).
Alternatively, the spectra may be used simply as a pattern characteristic of
a  specific  sample class (e.g., coal rank).
     Two coal samples have been thoroughly examined—Illinois No.6 and North
Dakota  lignite.  The coal and resultant ash were finely milled and pressed
into KBr pellets before their low resolution (8 cm" ) spectra were collected.
The tar was dissolved in CHgCl,, and placed in a liquid solution cell to
obtain  its  IR spectrum.
     The spectra of the coal, tar, and ash from Zap lignite and Illinois No.6
coal are presented in Figures 12 through  14, respectively.  A comparison of
the spectra for the two coals is presented in  Figure 12.  The weak band at
3058 cm   in  the Illinois No.6  spectra indicates greater aromaticity.  The
intensity  of  the shoulder at 1725 cm"  on  the major peak at   1610 cm"  is
indicative of the  carbonyl content.  Zap  lignite appears to contain more
carbonyl species.  The  peaks at 1095,  1035, 540, and 473 cm"   in  Illinois  No.6
have been  attributed  to inorganics  (vide  infra).  This would  indicate  that
 Illinois No.6 contains  more  inorganic  species.
     A comparison  of  the  tar spectra for the two coals  (Figure  13)  indicates
that the tars are  remarkably similar.  The major difference is  the much
higher aromatic  content of the  Illinois  No.6 tar  (3058 cm"  ).
     The two  ash  spectra  (Figure 14) are completely  different.  In  the Zap
lignite ash,  the  peaks  at 1451  (strong),  880 (sharp  peak at low end  of broad
                    _1
"hump"), and  710  cm    (weak, sharp)  have been  tentatively assigned  to  a
carbonate  species.   The peaks  at 1190  (shoulder),  1109  (strong),  and 622  cm"
 (medium, sharp)  have  been tentatively  assigned to  a  sulfate species.   In  the
                                     59

-------
      TABLE  22.   COMPOUNDS  IDENTIFIED  IN  XAD-2 SURGE  SAMPLE  FROM  RUN 41

        File No.*                      Compound

          11                  Methyl  chloride
          15                 Benzene
          29                 Toluene
          42                 m-Xylene (tent.)
          49                 o-Xylene (tent.)
          56                 Ethyl benzene  (tent.)
          62                 1,3,5-Trimethylbenzene  (tent.)
          66                 1,2,4-Trimethylbenzene  (tent.)
          71                  Alkylbenzene
          79                 Phenol
          89                 Monosubstituted alkylbenzene  (tent.)
          94                 Indan (tent.)
         100                 Indene
         106                  o-Cresol and p-cresol
         113                  m-Cresol
         124                  Alkylbenzene
         129                  2,6-Xylenol
         131                  2,4-Xylenol
         142                  Naphthalene
         151                  1-Methylnapthalene (tent.)
         156                  2-Methylnaphthalene (tent.)
         161                  Alkyl aromatic
         165                  Alkyl artomatic
         170                  Ester (background)
         177                  Dimethyl naphthalene isomer (tent.)
         183                  1,4-Dimethylnaphthalene (tent.)
         201                  d-jQ-Anthracene (std.)

*File numbers are located along  x-axis of chemigram.~~
                                   60

-------
     o
     o
      9

     o
     o
     o
     o

     in .
   u

   in
     co


     o
     o
3t
         OD   Z970
2110    1680    1250

WflVENUMBERS
820
3SD
Figure 12.  Comparison of coal  spectra.  Top = Illinois No.6; Bottom = Zap

          lignite.
                                 61

-------
  o
  o
   •
  o
  o
  «— «

  o
  in .
CJ
~
Jin
in
zo
  o
  o
  atbo   ssVo   es¥o   2110   isso   izso    sea    390
                          WflVENUMBERS

 Figure 13.  Comparison of tar spectra.  Top = Illinois No.6; Bottom = Zap
           lignite.
                               62

-------
  3100   2970   25\0
2110   1680
WflVENUMBERS
1250     820
330
Figure 14.  Comparison of ash spectra.
          lignite.
        Top = Illinois No.6; Bottom = Zap
                               63

-------
Illinois No.6 ash, the broad hump at 1082 cm"  and the small doublet at 790
cm   have been tentatively assigned to quartz.
     In summary, FTIR appears to be a useful technique for solid, liquid and
gaseous samples related to coal gasification.  GC/FTIR is currently usable as
a confirmatory technique to GC/MS.  Further work is needed to compare GC/MS
and GC/FTIR for class quantitation capability.  High resolution FTIR of coal  gas
is useful for validation of routine GC analyses and also for analyses of certain
compounds which cannot be determined by conventional GC detectors.   Low resolution
FTIR analyses of coal, tar and ash is under further investigation and its utility
for such analyses needs to be fully assessed.
                                     64

-------
               8.0  SPECIAL TECHNIQUES FOR ANALYSIS OF TARS
                           AND AQUEOUS CONDENSATES
     Of special importance is the characterization of the complex tar pro-
duced during coal gasification.  In Section 5.0, various procedures for
partitioning and analysis of tar and condensate were described.   In this
section, additional studies are described which were undertaken  for further
characterization of the liquid effluents.  These include polynuclear aro-
matics (PNA) analysis using glass capillary gas chromatography,  boiling range
measurement using simulated GLC-distillation, class characterization using
element specific detection and ion chromatographic analysis of the aqueous
samples.
8.1  PNA ANALYSIS USING GLASS CAPILLARY GAS CHROMATOGRAPHY (GC)2
         o
     (GC)   has been applied for quantisation of PNA materials in tar.  A
chemically  bonded temperature stable (300°C) methyl-silicone capillary
column was  used.  The conditions for analysis are shown in Table 23.  The
                21
'GROB' split!ess   method of sample injection was used and approximately 5-
15 yg were  injected for detection of the heavier PNA's, e.g.,  benzo(g,h,i)
perylene.   The splitless technique consists of  injecting 2 to 3 yl of the
sample and  then 30 seconds later, opening the splitter to remove- the excess
solvent.  This prevents a long solvent tail as  illustrated in the accompanying
chromatogram (Figure 15) in which 21 PNA's  have been  identified based on
retention times of standards.  Table 24  shows the analysis of several com-
pounds  in the  PNA fraction of  the coal tar  for  three  runs.  A comparison is
presented for  a few compounds  for which  GC/MS data  are available.  At present,
it  is  planned  to use this temperature  stable column  in a GC/MS for confirmation
of  these  compounds.  In  the  future, the  capillary-GC-FID technique will  be
routinely applied  to PNA analyses of coal  conversion  tars  and condensates.
                                     65

-------
               TABLE 23.  CONDITIONS FOR (GCr ANALYSIS
Instrument:
Column:

Injector:

Detector:
Oven:
Carrier Gas:
Makeup Gas:
Hydrogen:
Air:
Varian 3700 with all-glass capillary system.
High temperature OV-101, 25 m x 0.25 mm ID,
WCOT, glass (obtained from Quadrex Corporation,
New Haven, Connecticut).
Splitless (GROB), solvent vented 30s after in-
jection, 250°C
FID 280°C
40°C for 1 min, 4°C/min to 270°C and hold 40 min.
Helium 1.25 ml/min
Helium 29 ml/min
30 ml/min
300 ml/min
                                   66

-------
CTV
                       I   i
  1.  Naphthalene
  2.  2-Meth. Naph.
  3.  1-Meth. Naph.
  4.  Acenaphtylene
  5.  Acenaphthene
  6.  Fluorene
  7.  Phenanthrene
  8.  Anthracene
  9.  Fluoranthene
 10.  Pyrene
 11.  Benzo(a)fluorene
 12.  Benzo(b)fluorene
 13.  Benzo(a)anthracene
 14.  Chrysene &
       Triphenylene
 15.  Benzo(b)fluoran-
       thene
 16.  Benzo(k)fluoran
       thane
 17.  Benzo(e)pyrene
 18.  Benzo(a)pyrene
 19.  Perylene
20.  Dibenzo(a,  h)an-
      thracene
21. Benzo(g. h,  Opery-
      lene
                                                                                                        4
7"TTT


1
---

1
[j •
!
I '
• f i '
r i
1
\jf
I



; '


II
,i
•-
1
,

§--•
OB
O
o""

s
** 111
O III
J

If
III
I
r
'V




<
/


LJ
O


3


3





(
INS





1
TRU
- t





MEN!
1 i '
1MII
I'
:

*!
.i
f:
M
3X

18,
•
1
V
CARRIER GAS: H
MAKE UP: h
DETECTOR: F
HYDROGEN: 3
AIR: 3
INJECTION
MODE: '(
V
COLUMN: 2
TEMPERATURE: 4
SOLVENT: n
SAMPLE SIZE
1

1


'!

                                                                                                                              AMP FULL SCALE
                                                                                                                         19-
                                                                                                                                         i   20,
                                                                                                                       VARIAN 3700
                                                                                                                       HELIUM 1.25 ml/min
                                                                                                                       HELIUM 29 ml/min
                                                                                                                       FID 1280° C)
                                                                                                                       30 ml/min
                                                                                                                       300 ml/min

                                                                                                                       GROB' SPLITLESS (250° C),
                                                                                                                       VENT IN 30 SECONDS
                                                                                                                       25 m X 0.25 mm I.D. WCOT.
                                                                                                                       HIGH TEMP. OV101 (280° +)
                                                                                                                          C FOR 1 min. TO 270° C
                                                                                                                       AT 4° C/min, HOLD 40 min
                                                                                                                       n-C6&CH2C!2(1:1)
                                                                                                                                                 '    I
                                                                                                                                                 94 MIN
             Figure  15.    PNA  fraction  of coal  gasification tar  by  capillary GC-FID:   Determination  of  Selected  PNA's.

-------
        TABLE 24.  SELECTED COMPOUNDS IN PNA FRACTION AS WEIGHT % OF
                     COAL GASIFICATION TAR (MS vs. capillary-FID)
Compound
Naphthalene*
Benzothiophene
2-Methyl naphthal ene
1 -Methyl naphthalene
Biphenyl
C2-Naphthalenes
Acenaphthylene*
Acenaphthene*
Dibenzofuran
Fluorene*
Phenanthrene*
Anthracene*
Benzidine*
Carbazole
Fluoranthene*
Pyrene*
Benzo a)fluorene
Benzo bjfluorene
Benzo a) anthracene*
Chrysene* +
Triphenylene
Benzo(b)fluorenthene*
Benzo( k) f 1 uoranthene*
Benzo (e)pyrene
Benzo(a)pyrene*
Perylene
Indeno(l ,2,3-cd)pyrene*
Dibenzo(a,h)anthracene*
Benzo(g,h,i)perylene
Run 21
MS
5.73
--
1.00
—
--
—
--
--
0.74
0.80
2.30
--
--
—
1.30
0.9
—
—
—

0.80
—
—
—
--
--
—
--

Capillary-
FID
2.91
0.53+
0.38
0.33
0.26+
1.07
1.15
0.38
0.60+
0.63
1.31
0.47
0.05+
0.12+
0.91
0.67
0.26
0.17
0.47

0.46
0.31
0.16
0.21
0.35
0.09,
0.14
0.28
0.18
Run 23
MS
2.70
—
0.93
—
--
—
—
--
0.54
0.48
2.37
0.76
—
—
1.40
1.10
—
—
—

0.87
—
—
—
--
0.38
—
—

Capillary-
FID
1.73
0.36+
0.27
0.25
0.22+
0.74
1.09
0.27
0.54+
0.56
1.42
0.49
0.06+
0.12+
0.79
0.59
0.18
0.09
0.28

0.32
0.18
0.09
0.12
0.20
0.05.
0.08
0.16
0.11
Run 25
MS
1.30
—
1.10
—
--
--
--
—
0.57
0.78
1.90
0.66
--
--
1.10
1.10
--
—
—

0.78
—
--
--
--
0.29
--
—

Capillary-
FID
2.51
0.17+
0.74
0.79
0.29
1.57
1.80
0.41
0.78+
1.22
1.44
0.55
0.02+
0.26+
0.74
0.64
0.41
0.24
0.70

0.62
0.25
0.12
0.19
0.27
0.07.
0.1 7+
0.34
0.27
+Subject to corrections.
*Priority pollutant/consent decree compound.
                                    68

-------
8.2  SIMULATED GLC DISTILLATION OF TAR
     The boiling point curves of tars produced from various screening and
parametric tests were measured by a procedure based on ASTM D2887.22  Briefly,
the procedure consists of injecting a tar sample onto a column which separates
the components on the basis of their boiling points.   Retention times of
standard hydrocarbons of known boiling points are used to divide the chromato-
gram into boiling point fractions and areas under the chromatogram for each
fraction are measured.  Two modifications of the procedure were carried out.
Direct on-column injection was employed so that a low injector temperature
could be used to prevent septum bleed.  A problem with the ASTM procedure is
column bleed during temperature programming, resulting in excessive baseline
drift.  To alleviate this, two identical columns were used on the sample and
reference sides, respectively, of the thermal conductivity detector.  Con-
ditions for the simulated distillation are shown in Table 25.
     Another drawback of the ASTM procedure is that the PNA's tend to elute
faster from the column than normal hydrocarbons of similar boiling points.
This tends to give lower results for boiling points of samples with substantial
amounts of PNA materials.  To correct for these deviations,  it was necessary
to determine deviations for several pure PNA's in the boiling range from 200-
525°C.  These are shown in Table 26.  Using the data in Table 26, the results
were corrected to give the actual boiling point distribution.  Typical results
for a tar from gasification of Illinois No.6 coal are shown  in Table 27.
8.3  CLASS CHARACTERIZATION USING ELEMENT SPECIFIC DETECTORS
     This study was only recently initiated and work  is continuing.  It  is
planned to characterize the tar and  its various fractions  using a nitrogen
specific detector  (NPD) and a  sulfur  specific detector  (FPD).  The  objective
is to determine if substantial amounts  of  heavier  (3  ring  or greater) nitrogen
and sulfur compounds are present.  The  procedure will consist of using  identi-
cal packed OV-101 columns  (all glass  system) with  a  FID,  FPD, and an NPD with
identical column conditions and carrier gas  flow rate.  The  FID will  indicate
a  response to all compounds  present.  With  the  use of standards, the chromato-
gram  can be  separated  into two, three,  four,  five  and six ring  compounds.  Then
                                      69

-------
    TABLE 25.  CONDITIONS FOR SIMULATED GLC DISTILLATION
Instrument:
Detector:
Injector:
Column:

Flow Rate:

Temperature:
Sample:
Standard:
Varian 3700
TC (350°C)
On-column 230°C
Two 4' x 1/8" S.S. (3% OV-101 on 80/100 Chromosorb
HP)
30 ml/mm through each column (adjusted to produce
a straight baseline during programming)
30°C to 310°C at 5°C/min
Coal tar from gasification
C-C   normal hydrocarbons
  TABLE 26.  ERROR IN BOILING POINTS OF PNA'S AS PREDICTED
             BY THE PROCEDURE
Compound
Naphthalene
2-Methyl naphthalene
1 -Methyl naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Chrysene
Benz(a)pyrene
Perylene
Coronene
Observed Boiling Point
(°C)
207
227
229
251
257
273
300
302
336
342
382
419
422
447
Deviation from Actual
Boiling Point (°C)
11
14
16
19
22
22
40
38
39
51
66
84
78
78
                                70

-------
      TABLE  27.   CORRECTED BOILING POINT DISTRIBUTION FOR TAR PRODUCED
                 FROM GASIFICATION OF  ILLINOIS NO.6 COAL (RUN 52)
    Observed                         Corrected
  Boiling  Range                    Boiling Range            % Recovered
     (°C)                               (°C)                 (Cumulative)
0-100
100-150
150-200
200-250
250-300
300-350
350-^00
400-450
450-500
500-550
550-600
4-97
97-150
150-207
207-272
272-338
338-404
404-469
469-555
555-599
599-659
659-715
0
1.9
18.9
35.9 (56.7)
15.5 (72.2)
19.6 (91.8)
5.4 (97.2)
1.6 (98.8)
1.0
0.2
0

the same sample will be injected onto the NPD and the FPD systems to give the
amounts of nitrogen- and sulfur-containing compounds, respectively.   Preliminary
results for tars with the FPD have shown that greater than 90 percent of the
sulfur compounds contain less than three rings.   The information obtained from
such studies is considered extremely useful in the characterization of complex
materials such as tar.
8.4  ION CHROMATOGRAPHIC ANALYSIS OF AQUEOUS SAMPLES
     The principle of an ion chromatograph has been discussed in detail
          0-5
elsewhere.    Using a Dionex ion chromatograph (1C) preliminary work has been
carried out for characterization of various cations and anions in the aqueous
condensate and scrubber samples.  It has been found that the 1C will be most
suitable for the analysis of such ions as NH. , Na  , K , Cl~, F", and SOT
because of its cost effectiveness and accuracy.  Additionally, various modifi-
cations in the standard eluent-column procedure will allow measurement of CN~,
SCN", and  S".  Typical results from 1C have shown  about 4 g/1 of NH-, 2 g/1
of Cl" and 0.02 g/1 of F" for gasification run 23.
                                    71

-------
                                REFERENCES


1.   del and, J. G., et al.  Pollutants from Synthetic Fuels Production:
     Facility Construction and Preliminary Tests.  EPA-600/7-78-171,  U.  S.
     Environmental Protection Agency, Research Triangle Park, N.  C.,  1978.

2.   Grohse, P. M., R. B. Denyszyn, and D. E. Wagoner.  Sampling  and  Atomic
     Absorption Spectrometric Determination of Arsine at the 2 ug/nr  Level.
     Anal. Chem.. 50 (8): 1094-96, 1979.

3.   Nichols, D. 6., Pollutants from Synthetic Fuels Production.   Quarterly
     Report, EPA Grant No. R804979-03, RTI/1700/00-01Q, 1979.

4.   Gangwal, S. K., and D. E. Wagoner.  Response Correlation of  Low  Molecular
     Weight Sulfur Compounds Using a Novel Flame Photometric Detector.   0^
     Chromatogr. Sci., 17 (4): 196-201, 1979.

5.   Gangwal, S. K., et al.  Analysis of a Semibatch Coal  Gasifier Product
     Gas Using an Automated Gas Chromatograph.  J.  Chromatogr. Sci.,  16  (8):
     368-71, 1978.

6.   Gangwal, S. K., et al.  A Sampling and Analysis Procedure for Gaseous
     Sulfur Compounds from Fossil  Fuel Conversion.   Paper  presented at the
     EPA Oil Shale Sampling, Analysis and Quality Assurance Symposium,
     Denver, CO, March 26-28, 1979.  Proceedings to appear.

7.   Brodey, S. S., and J. E. Chaney.  Flame Photometric Detector: The
     Application of a Specific Detector to Phosphorus and  Sulfur  Compound
     Sensitive to Subnanogram Quantities.  J. Gas Chrom..  4: 42,  1960.

8.   Patterson, P. L., R. L. Howe, and A. Abu-Shumays.  Dual-Flame Photo-
     metric Detector for Sulfur and Phosphorus Compounds in Gas Chromato-
     graphic Effluents.  Anal. Chem.. 50: 339, 1978.

9.   Sharkey, A. G.   Carcinogenesis, Vol. 1, Polynuclear Aromatic Hydrocarbon
     Chemistry, Metabolism and Carcinogenesis.  R.  I. Freudent and P.  W.
     Jones, ed., Raven Press, 1976.

10.  Schiller, J. E., and D. Mathiason.  Separation Methods for Coal  Derived
     Solids and Heavy Liquids.  Anal. Chem., 49: 1225, 1977.

11.  Snook, M. E., W. Chamblain, R. F. Severson, and 0. Chortyk.   Chromato-
     graphic Concentration of Polynuclear Aromatic  Hydrocarbons of Tobacco
     Smoke.  Anal. Chem.. 47: 1155, 1975.
                                    72

-------
12.   Novotny, M.,  M.  L.  Lee,  and K.  Bartle.   Methods  of  Fractionation,
     Analytical  Separation  and Identification of  Polynuclear Aromatic Hydro-
     carbons in  Complex  Mixtures.   J.  Chromatogr.  Sci..  12: 606,  1974.

13.   Eight Peak  Index of Mass Spectra. Vol.1  (Tables  1 & 2), Vol.11  (Table  3).
     Mass Spectrometry Data Center, AWRE,  Aldermaston, Reading,  U.  K.,  1970.

14.   Parr Bomb Combustion of Sample with High Organic Content, Chapter
     7, EPA-600/2-76-160, U.  S. Environmental Protection Agency,  1976,
     86 pp.

15.   Parker, C.  R.  Water Analysis by Atomic  Absorption, Varian  Techtron
     Pty Ltd., Springvale,  Australia, 1972.

16.   Aruscauage, P.  Determination of Arsenic, Antimony  and Selenium in Coal
     by Atomic Absorption Spectrometry with a Graphite Tube Atomizer.   Journal
     Research U.S. Geological Survey, 5 (4):  405, 1977.

17.   Bosshart, R.  E.   Determination of Arsenic, Selenium and Antimony in
     Coal, Coal  Ash,  and Coal Refuse Materials by Hydride Generation-Atomic
     Absorption.  Bituminous Coal  Research,  Inc., December 1977.

18.   Pradhan, N. K.  Coal and Petroleum Analysis  by Atomic Absorption.
     Varian Techntron Pty Ltd., Springvale, Australia, March  1976.

19.   Gluskoter, H. J., et'al.  Trace Elements in Coal:  Occurrence and
     Distribution.  EPA-600/6-77-064, U. S.  Environmental Protection Agency,
     Washington, D. C., 1976.

20.   Attari, A.   Fate of Trace Constituents  of Coal during Gasification.
     EPA-650/2-73-004, U. S. Environmental Protection Agency,  Washington,
     D. C., 1973.

21.   Grob, K., and G. Grob.  Splitless Injection in Capillary Columns.
     J. Chromatogr. Sci.. 7: 584-590, 1969.

22.   Standard Test Method for Boiling Range Distribution of Petroleum
     Fractions by Gas Chromatography, ASTM D 2887, 1973.

23.   Small, H., T. S. Stevens, and W. C. Bauman.   Novel  Ion Exchange Chromato-
     graphic Method Using a Conductometric Determination.  Anal. Chem., 17
     (11): 1801, 1975.

24.   Pellizzari, E. D., et a]I.  Collection and Analysis of Trace Organic
     Vapor Pollutants in Ambient Atmospheres.  I. A Technique for Evaluating
     the Concentration of Vapors by Sorbent  Media.  J.  Environ. Sci. Tech..
     9: 552, 1975.
                                     73

-------
25.  Pellizzari, E. D., et al.  Collection and Analysis of Organic Vapor
     Pollutants in Ambient Atmospheres.  II.  Studies on Thermal  Desorption
     of Organic Vapor from Sorbent Media.  J. Environ. Sci. Tech.. 9:
     556, 1975.

26.  Pellizzari, E. D.., et al.  Collection and Analysis of Trace Organic
     Vapor Pollutants in Ambient Atmospheres.  The Performance of a
     Tenax-GC Cartridge Sample for Hazardous Vapors.   Anal. Letters,  9:
     45, 1976.

27.  Pellizzari, E. D.  Analysis of Trace Hazardous Organic Vapor Pollutants
     in Ambient Atmospheres.   Anal. Chem., 48: 803, 1976.
                                    74

-------
                                 APPENDIX I
               SELECTION AND PREPARATION OF  POLYMERIC  SORBENTS

     At the initiation of this effort various  alternative sampling  methods  for
vapors were considered and eliminated using  two major  criteria  for  evaluation:
(1) the broadest possible range of organic vapors be collected, and (2)  a
minimum of artifacts be produced in the sample.   Other criteria which were
considered included level of sample recovery and ease of interfacing with the
instrument system.  Cryogenic trapping was eliminated on several criteria,
especially on artifact production.  If the mechanics of cooling the product
gases were not difficult in themselves, the sample would be subject to extreme
vulnerability to artifact formation—collected as a liquid and/or solid  sample
with large quantities of water.  The condensed water would serve to collect
both acidic and basic inorganic gases (NHj, HC1, HF, HCN, NOX>..) which might
result in artifact formation.  Similar problems occur with impingers and the
range of volatility of the sample collected would be limited.  (Among the
sorbents used for collection of organics, polymeric sorbents cause the fewest
artifacts and cover the  broadest volatility range.)  Two polymeric sorbents,
Tenax-GC and Amberlite XAD-2, were used  in this program in a sampling strategy
which allowed them to complement each other.
     At the initiation of this study a  significant body of information on the
performance of Tenax-GC  for trace organic analysis especially  in ambient air
existed.       It was clear from these  data that artifacts were at a minimum
for  trace analysis but that artifacts could occur at higher concentrations.
The  collection efficiency and  breakthrough volume needed to be verified at
high concentrations and  the implications of displacement chromatography needed
to be  defined  in  order to transfer the  sampling  technology from trace analysis
to process stream analysis.
                                      1-1

-------
     The complementary nature of the two sorbents is a function of the basic
 physical properties of the two sorbents.  Tenax-GC is a highly temperature
 stable polymer of poly(2,6-diphenyl-p-phenylene oxide) with a surface area of
 30 m2/g (60-80 mesh).  The thermal stability of Tenax allows the collected
 sample to  be recovered thermally without the introduction of any solvent.  In
 addition,  Tenax has a very low affinity for water and most inorganic gases in
 comparison to organic vapors.  This results in fewer artifacts being created
 as a result of the sample method.  The major limitations to the use of Tenax
 as a sorbent for sampling process streams is its low surface area which trans-
 lates into low capacity and high cost.  XAD-2 on the other hand, has a surface
 area of 300 m /g and costs approximately 1/50 as much as Tenax.  The performance
 of XAD-2 is somewhat inferior in two areas, (1) its lack of thermal stability
 precludes  thermal desorption, and (2) it has a significant affinity for water
 and possibly inorganic gases.
     Given the properties of the sorbents, the sampling approach taken was to
 place XAD-2 in the larger in-line traps with valving to place each on-line
 separately.  Sampling ports were installed upstream and downstream of these
 traps to monitor their performance with small Tenax samples.
     Tenax GC is prepared by Soxhlet extraction with methanol and then pentane,
 for 22 hours each.  The gross amounts of solvent are removed under vacuum at
 elevated temperatures.  The sorbent is then packed into the glass cartridges
with glass wool plugs in both ends.  Within two weeks of sampling the cart-
ridges are thermally desorbed at ca. 275°C under helium purge to remove any
trace organics.  The cartridges are sealed in screw cap centrifuge tubes with
teflon cap liners.
     XAD-2 is prepared by a multisolvent extraction sequence of water, methanol,
and methylene chloride.   First it is soaked in deionized water, the excess
water decanted, and the resulting slurry poured into a 90 x 200 mm cellulose
thimble.  A plug of glass wool is then placed on top of the thimble and then
two thimbles stacked in the Soxhlet.  The XAD-2 is then extracted with each
solvent in the series for 22 hours and the resin dried in a vacuum oven with
mild heat for 48 hours.
                                     1-2

-------
                                 APPENDIX II
                TENAX AND XAD-2 ANALYSIS WITH GC/MS/COMPUTER

     Gas chromatography/mass spectrometry/computer (GC/MS/COMP)  systems using
high resolution capillary columns is used for the analysis of the Tenax
samples, the XAD-2 extracts and the glass fiber filter extracts.   This approach
permits both qualitative identification of species collected and their quan-
titative estimation.  Since all of the data acquisition and handling is done
by the computer and the data stored as a permanent file, retrospective searches
are possible.  In the early stages of this work complete qualitative interpre-
tations of each sample were performed before selecting specific  compounds for
quantisation.  As data accumulated, certain compounds and classes of compounds
were selected for routine quantitation.  Described below are procedures used
for analysis of Tenax samples and XAD-2 extracts.
     J_enax--The organics adsorbed on Tenax were recovered by thermal desorp-
tion which permitted the entire sample to be introduced into a gas chromato-
graphy column.  Two internal standards (perfluorobenzene and perfluorotoluene)
were added prior to the desorption step.  The inlet-manifold (Figure 16) for
introducing the sample into the analytical instruments consisted of four main
components:  a desorption chamber; a six-port, two-position high temperature,
low-volume valve; a gold-plated nickel capillary trap; and a temperature
controller.
     The sampling system inlet-manifold  is used for recovering vapors trapped
on Tenax-GC cartridges and  is  interfaced to  the GC/MS/COMP system (Figure 17).
The desorbed vapors are subsequently resolved by gas/liquid chromatography  and
mass cracking patterns automatically and continuously  obtained through  the  GC
run with a Varian CH-7 mass spectrometer.  Operating parameters  are those
depicted in Table 28 for the GC/MS/COMP  system for  the analysis  of  volatile
organic fractions.
                                      II-l

-------
                                                                         .PURGE
                                                                          GAS
                                                                            i
ro
                 ION
               CURRENT
               RECORDER
                             GLASS
                              JET
                           SEPARATOR
                          HUH
                               r
                            CARRIER
                             GAS
                   TWO
                 POSITION
                  VALVE
 CAPILLARY
    GAS
CHfiCMATCGHAPH
                                                                     CAPILLARY
                                                                       TRAP
                                      ANALYTICAL  SYSTEM
                                                                                THERMAL
                                                                                OcSOSPTJCN
                                                                                CHAMecS
HEATED
BLOCKS
                                         EXHAUST
                      Figure 16.  Analytical systems  for organic vapors in gas samples.

-------
lalct-
Maalfold-


CLC
                 Separator
Figure 17.   Gas  liquid chromatoqraph-mass spectrometer
            computer (GC/MS/COMP)  layout.
                             II-3

-------
          TABLE 28. OPERATING PARAMETERS FOR 6LC/MS/COMP SYSTEM
        Parameter
   Setting
Inlet-manifold
      desorption chamber
      valve
      capillary trap-minimum
                     maximum
      thermal desorption time
 GC
MS
      95M SE30 SCOT
           capillary columns
      carrier (He) flow
      transfer line to ms
      scan range
      scan rate, automatic-cycle
      filament current
      multiplier
      ion source vacuum
     260°C
     180°C
    -195°C
    +175°C
    10 min
20°C, 4/C°/min
 3 ml/min
210°C
m/e_ 20 - 300
1 sec/decade
300 A
6.0    6
4 x 10   torr
                                    II-4

-------
     The mass spectrometer is first set to operate in the repetitive scanning
mode.  In this mode the magnet is automatically scanned exponentially upward
from low mass to high mass values.  Although the scan range could be varied
depending on the particular sample, values used are from m/z 20 to m/z 300.
The scan is completed in approximately 5 seconds.   At this time the instrument
automatically resets itself to the low mass and the run is suitable for
further processing, since it gives some idea of the number of unknowns in the
sample and the resolution obtained using the particular GC conditions.
     The next stage of the processing involves the conversion of the spectral
peak times to peak masses.  To accomplish this, the spectra are read suc-
cessively into the Varia computer core and then sent at high speed to the
PDP/8 disk by means of the intercomputer interface.  Depending on the number
of peaks in each spectrum, the disk capacity allows the storage of 200-500
spectra.  Spectra containing time information are stored on the disk, and the
data are returned to an off-line digital computer (IBM 620-i) for mass conversion
by use of the calibration table obtained previously.  Normally one set of
calibration data is sufficient for an entire day's data processing since the
characteristics of the Hall probe are such that the variation in calibration
is less than 0.2 atomic mass units per day.  Very long runs may require several
cycles of mass conversion using the PDP/8 disk to complete the data processing.
     After spectra are obtained in mass-conversion form, full spectra of scans
from the GC run were recorded on  the Statos plotter.  The total ion current
(TIC) information available at this time are most useful for deciding which
spectra are to be analyzed.  At the beginning of the run where peaks  are very
sharp, nearly every spectrum  is inspected  individually.  Later in  the run  when
the  peaks are broader, only selected  scans are analyzed.
     Identification of resolved components is  achieved by comparing the mass
cracking patterns of the  unknown  mass  spectra  to an  eight major peak  index of
mass spectra.  Particular note is made  of  the  boiling  point  of the identified
compounds as a function of the elution  temperature  and the  order  of elution of
constituents in homologous series,  since  the SE-30 capillary column separates
primarily  on the basis of boiling point.
                                       II-5

-------
     XAD-2 -- The organics sorbed on the XAD-2 resin and glass fiber filter
are recovered by solvent extraction with methylene chloride.  The extract is
evaported from ca. 150 ml to 0.5 to 1.0 using a Kaderna-Danish (K-D) evaporator
and a micro K-D evaporator.  The internal standards, d-jQ-anthracene, dg-
phenol and p-chlorotoluene, are added and the extract is ready for GC/MS/COMP
analysis.  Weighed portions of the XAD from the integrated samples are placed
in a thimble and extracted as above.  In this case, only a portion (4 ml of
150 ml) of the extract is evaporated to 1 ml and the internal standard added.
     The XAD extracts are analyzed on the 1KB 2091 GC/MS/COMP system.  The
operating parameters are listed in Table 29.
     The mass spectrometer is set to operate in the repetitive scanning mode.
In this mode the magnet is automatically scanned exponentially upward from low
mass to high mass values.  Although the scan range can be varied depending on
the particular samples, values used are from approximately m/z 40 to m/z 500.
The scan is completed in approximately 1 second.  At this time the instrument
automatically reset itself to the low mass position in preparation for the
next scan, and the information is accumulated by an on-line PDP-11/04 computer
onto disk.  The reset period requires approximately 1.4 seconds.   Thus, a
continuous cycle of 2.4 seconds is maintained.
     The sample is injected and automatic data acquisition is initiated just
before elution of the first component.   As each spectrum is acquired by the
computer, each peak that exceeds a preset threshold is recognized and recorded
by mass and intensity.  This information is stored on the disk as the scan is
in progress.  In addition, TIC values are stored as they are acquired.   This
procedure continues until the entire GC run is completed.  By this time, there
are from 1500-2000 spectra on disk, which are then used for subsequent or
simultaneous data processing.  The data are also recorded on 9-track magnetic
tape for archiving or for subsequent data processing.
     The mass spectral data are processed in the following manner.  First, the
original spectra are scanned, and TIC information extracted.  This is taken
directly from the disk after data acquisition.   Then the TIC intensities are
plotted against the spectrum number of a Versatec electrostatic printer/plotter.
The information thus gained generally indicates whether the run is suitable
for further processing,  since it gives  some idea of the number of unknowns in
                                     II-6

-------
  TABLE 29.   OPERATING PARAMETERS  FOR  1KB  2091  GLC/MS/COMP  SYSTEM
      Parameter
 Setting
Inlet
   Split ratio
   Injector Temp
   Injection Size
GC
   15 M SE 30 WCOT
   glass capillary columns

   carrier (He) flow
   programming conditions
   separator temperature
MS
   scan range
   scan rate, automatic-cycle
   trap current
   multiplier
   ion source vacuum
   10:1
   260°C
   U
 2.5 ml/min
70°C for 6 min,
8°/min to 250°C

200°C
m/z 40 - 500
2.4 sec/cycle
50 A
350
10~6 torr
                                    II-7

-------
the sample and the resolution obtained using the particular GC conditions.
Full spectra of scans from the GC run are recorded on the Versatec plotter.
The TIC information available at that time is most useful for deciding which
spectra are to be analyzed.  At the beginning of  the  analysis  where  peaks  are
very sharp, nearly every spectrum is inspected individually.  Later in the run
when the peaks are broader, only selected scans are analyzed.
     Identification of resolved components is achieved by comparing the
fragmentation patterns of the unknown mass spectra to an eight major peak
index of mass spectra.  Particular note is made of the boiling point of the
identified compounds with the elution temperature and the order of constituents
in homologous series since the SE-30 WCOT capillary separates primarily on the
basis of boiling point.
                                     II-8

-------
                                APPENDIX III
       QUALITATIVE GC/MS RESULTS FOR ORGANIC  VAPORS AND TAR FRACTIONS

     Sample results are included in this Appendix for the chemical  compounds
which were identified in the XAD cartridges and the tar condensate.  These
results are presented for screening test run  21.  The tar condensate was
solvent-partitioned prior to analysis, the procedure for which is described in
detail in this report (see Section 5.2).  Substantial progress has been made
over the course of this project in the quantitative determination of the
specific compounds which are potentially hazardous, including consent decree
compounds.
                                      III-l

-------
     TABLE 30.    GC/MS  ANALYSIS  OF  POLYNUCLEAR AROMATICS  FROM  RUN  21
                                     -QUALITATIVE  RESULTS-
Chromato-
grephic
Peak No.
Elution
 Te»p.
 CC)  _
Coapound
Chromato-  Elution
graphic    Temp.
Peak Ho.    CC)
                                                        Coapound
    1       123       MChyl lnd«M lionets
    2       127       naphthalene
    2a      129       2,3-benzothlophene
    3       132       2-ethyl benzlaldazole (tentative)
    4       144       Mthyl naphthalene leoMr
    4a      145       5-Mthyl-2,3-benzothiophene
    5       146       Mthyl naphthalene leoMr
    6       156       blphenyl
    7       158       1- or  2-ethyl naphthalene
    8       160       dlaethyl naphthalene lamer
    9       162       dlaethyl naphthalene iaoner
   10       165       1- or  2-athyl naphthalene
   11       166       blphenyltne
   12       171       acenaphthene
   13       17!       dlbenzofuran
   13a      178       1.3-dihydro-4,6-dlMthylthl»no
                     (3,4-c)thiopehe (tentative)
   14       182       Mthyl acanaphthylene laoMr
   15       185       fluorene
   16       187       methyl aeenaphthylene iaoMr
   17       190       hydroxy fluorene leaner
   18       192       hydroxy fluorene iao«*r
   19       200       1-Mthyl fluorene
   20        200       9-fluor«none
   21       205       2-aethoxyfluorene
   22        206       dlbenzothlophene leoMr
   23        210       anthracene
   23a       212       phananthrane
                                                 ' 24       213       beazidine
                                                  24«      214       dlbenzothlophene leoner
                                                  25       219       l-nethylbenzo(l,2-b:4,3-b)dithi-
                                                                    ophene
                                                  26       222       3-nethyldibencothlophene
                                                  27       223       carbatolc
                                                  28       224       Mthyl phenenthrene lioner
                                                  29       227       4,5-»«thylene phenanthrene
                                                  29a      228       Mthyl pnenanthrene leoner
                                                  »       232       2,2'-41cyanoblphenyl
                                                  31    238*242      diMthyl phenanthrme laoMre
                                                  32    243-246      fluoreatnene end/or
                                                                   1,2.3,4-tetrahydrofluoianthane
                                                  33       248       2-nitro dlaethyl tarephthalate
                                                  34    249-250      pyrent and/or
                                                                   1,2,3,4-tetrahydrofluoranthene
                                                  35    259-265      benzofluorenone laoewra and/or
                                                                   •ethyl pyrene laoawrs
                                                  36       265     naphtho(l,2-b)thianaphthene
                                                  37       265      2.3-b*nzanthracene
                                                  38       265      benzphenaathrene leonere and
                                                                   trlphenylen*
                                                  38a      265      •ethylbenr(a)anthracene laoeurs
                                                  38b      265      e»thyl chryeene leoewr
                                                  3Bc      265      •ethylbenaophananthrene laoaere
                                                  38d      265      dlhydroxy anthraqulnone laonere
                                                  39       26S      p«ryl»ne
                                                  39a      265      benzpyrene
                                                     III-2

-------
TABLE 31.  GC/MS ANALYSIS OF NONPOLAR NEUTRALS FROM RUN 21
                     -QUALITATIVE RESULTS-
Chromato-
graphlc
Peak No.
4
5
6
7
8

9
10
11
12
13
14

16
17
18
19
20
21
22
23
24
25
26

27
28


29



30
31
32
34
35
36
37
39
40
41
Elution
Temp.
(°C)
66
72
73
74
76

76
78
81
82
83
83

86
88
89
91
94
96
99
101
103
105
108

108
110


113



117
119
120
121
122
126
127
131
132
133
Compound
Indene
saturated hydroc«rbons
saturated hydrocarbons
saturated hydrocarbons
C.-cyclohexane or methylcyclohexyl-
cyclohexane
C -thiophene (tentative)
nor-undecane
saturated hydrocarbon
dimethylstyrene or methyllndan
•aturated hydrocarbon
dicyclohexylbutane
n-pentylcyclohexane (tentative)
C12H21 i80""
naphthalene
thlanaphthene, dlnethyllndan
C.-benzene
n-dodecane
cyclohexylbenzene
dimathy linden
dlnethyllndan
dimethyllndan
mathylnaphthalene
nethylthianaphthene
nethylnaphthalene
trims thylindan
trine thy lindan, tetranethy linden or
5,8-dlmethyl-l-n-octyl-l,2,3,4-
tetrahydronaphthalene
l-methyl-4-n-heptyl-l,2,3,4-tetra-
hydronaphthalene or
dlmethyltetrahydronaphthalene or
trinethylindan
blphenyl
dlnethylthlalndene
athylnaphthalena
dime thy Inaphthalene
dlmethylnaphthalane
dime thy Inaphthalena
blphenylene , dlnethy Inaphthalena
acenaphthene
mathylbiphanyl
trimethylthiaindene
Chroma to-
graphic
Peak No.
42
43
44
47
48

50
51
55
58


59
60
61
62
64
66
67
68
69
70
71
72

73
74
75
76
77
78
79
80
81
82
83
84
85
86
* 87
88


Elution
Temp.
CO
135
136
137
144
147

149
152
160
163


165
166
169
170
173
178
179
182
184
185
187
190

195
195
199
205
206
214
216
219
228
234
237
253
255
257
261
265


Compound
C.-naphthalene, dlbenzofuran
pentamethyldihydroindene
ar-phenylphenol
fluorene
b iphenylae thane
C.-naphthalene (tentative)
hydroxy fluorene
hydroxy fluorene
1 , 1-diphenylathane
2-nethoxy fluorene
acrldone or phenanthrldone
l,l-dl-(4-aethylphenxl)dodecane
dibenzothiophene
phenanthracane
dj --anthracene
anthracene
dlhydromethylphenylbenzofuran
dibenzoheptafulvene
methyldlbenzothiophene
methylphenanthrene
4 , 5-methy lenephenanthracene
saturated hydrocarbon
phthalata
C16H12 Uooer or
2,2' -dicyanobiphenyl
dimathy Iphenanthrene
saturated hydrocarbon
fluoran thane
pyrene
saturated hydrocarbon
C17H12 i90a*T
C17H12 i80Ber
saturated hydrocarbon
phthalate
saturated hydrocarbon
unaaturatad hydrocarbon (tentative)
saturated hydrocarbons
C18H12 laol"r
C18H12 l80ner (tentative)
C18H12 i80~r
phthalates


                                III-3

-------
                         TABLE  32.   GC/MS ANALYSIS  OF  TAR BASES  FROM  RUN  21
                                                 -  QUALITATIVE  RESULTS-
Chromato- Elation
graphic    Temp.
Peak No.   (°C)
Compound
Chromato—
graphic
Peak. No.
Elation
 Temp.
                                                          Compound
   1      100.0         toluene
   3      100.0         dimethyl pyridine isomer
   4      100.0         dimethyl pyridine isomer
   5      103.6         trimetbyl pyridine isomer
   6      110.4         2,4-dimethyl-6-ethylpyridine
   7      112.4         toluidiae isoaer and/or
                        benzyl mine
   7a     113.2         C2-alkyl pyridine isomer
   7b     124.8         methyl ethyl pyridine isomer
   7c     125.6         n-methyl-o-toluidine
   8   133.6-135.6      quinoline
   9   138.0-141.2      isoquinoliae
  10      144.4         methyl quinoline isomer
  10a     146.0         methyl quinoline isoaer
  lOb     148.0         methyl quinoline isomer
  11      150.8         methyl quinoline isoaer
  12      154.4         methyl quinoline isoaer
  12a     155.2         ethyl quinoliae isomer
  13      158.4         ethyl quinoline isomer
  14      161.2         dimethyl quinoline isomer
                             15      164.0         diaethylquinoline isomer
                             16      171.2         dimethyl quinoline isomer
                             17      174.4         diphenyl amine
                             18      184.4         alpha naphthyl amine
                             19      190.4         carbazole
                             19a     200.0         xanthene
                             19b     200.0         9-methyl carbazole
                             19c     206.8         methyl  carbazole isomer
                             19d     207.6         methyl  carbazole isoaer
                             20      210.8         7,8-benzoquinoline
                             21      215.6         benzoquinoline isomer
                             22      218.0         3-aethylbenzoquinoline
                             23      219.6         acridine
                             23a     223.2         methyl  benzoquinoline isomer
                             23b     224.4         methyl  benzoquinoline isomer
                             24      226.0         phenyl  indole isomer
                             25      230.0         dimethylacridine isoaer
                             25«     236.8         aethyl  phenyl indole isomer
                             25b     240.8         aethyl  phynyl indole isoaer
                                                         III-4

-------
TABLE 33.  GC/MS ANALYSIS OF STEADY-STATE XAD-2 EXTRACT FROM RUN 21
                       -QUALITATIVE RESULTS-
Chromato-
graphic
Peak No.
la
2
2a
2b
2c
3
4
5
5a
5b
5c
5d
5e
5f
5g
6
6a
6b
6c
7
8
Elation
Temp.
60°
60°
60°
60°
60°
60°
60°
60°
60°
62°
64°
71°
72°
81°
85°
87°
90°
92°
96°
100°
102°
Compound
unknown
toluene
methylthiopheae isomer
hydrocarbon
C.-benzene isomer ethyl
xylene isomer
CgHg isomer styrene
xylene isomer
hydrocarbon
C, -benzene isomer
C.-benzene isomer
benzofuran
C.-benzene isomer
C.-benzene isomer
indan
indene
diethylbenzene isomer
diethylbenzene isomer
phenol (tent)
C^H.-benzene isomer
C,H, -benzene isomer
Chromato-
graphic
Peak No.
8a
8b
8c
8d
8e
9
9a
9b
10
lOa
lOb
IQc
lOd
lOe
11
lla
lib
12
13
14
14a
Elution
Temp.
105°
108°
111°
116°
118°
122°
124°
127°
145°
147°
148° '
160°
164°
171°
177°
182°
190°
200°
204°
218°
224°
Compound
methylbenzofuran isomer
C.H7 -benzene
cresol isomer
methyl indene isomer
cresol isomer
naphthalene
benzothiophene isomer
C_-phenol
8-methy Inaphtha lene
methyl benzothiophene isomer
a-methylnaphthalene
biphenyl
dimethylnaphthalene isomer
C12H18 isOBler
C^-biphenyl isomer
dibenzofuran (tent)
C.-H.^ isomer
Cj-H.g isomer (tent)
CjgH.g isomer (tent)
D._ anthracene i
C.gH.- isomer
                                 III-5

-------
TABLE 34.  GC/MS ANALYSIS OF SURGE XAD-2 EXTRACT FROM RUN 21
                       -QUALITATIVE RESULTS-
Chromato-
graphic
Peak No.
1
la
2
2a
3
3a
3b
3c
3d
4
4a
4b
4c
4d
4e
4£
4g

4h
4i

4j
4k
41
4m
5
5a
6
6a
6b
7
8
Sa
8b
8c
8d
8e
8f
8g
Sh
8i
Sj
8k
81
8m
8n
So
8p
Elutlon
Temp.
(*C)
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°

60°
60°

60°
60°
60"
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60°
60.1°
62.7°
63.5°
63.9°
65.9°
67.1°
68.3°
70.3°
Compound
aethylene chloride (bkg)
1,1, 1-trichloroethane
benzene
thiophene
cyclohexene
C?H14 isomer (tent)
CyH12 isomer
C7Hj0 isomer
C7HJ2 isomer
toluene
methylthiophene isomer
CglL, isomer
C8H16 isomer
CgHj^ isomer
CgH.g isomer
C-Hj, isomer
C.H., isomer
8 14
CgH^ isomer
CgH14 isomer

C8H16 i80IBer
CgH. , isomer
CgHlg isomer (tent)
CgH18 and CgH^ isomers
ethylbenzene
ethylthiophene isomer
dimethylbenzene isomer
dimethylthiophene isomer
C,- thiophene isomer
CgH- isomer styrene
dimethylbenzene isomer
CgH.Q isomer
CgHlg isomer (tent)
C9H16 isomer (tent)
C. -benzene isomer
CgHj, isomer
CgHjg isomer
saturated hydrocarbon
n-propylbenzene
C. -benzene isomer
C. -thiophene isomer
C, -benzene isomer
C3- thiophene isomer
C, -benzene isomer
C-j-thiophene isomer
saturated hydrocarbon
benzofuran
Chromato-
graphic
Peak No.
9
9a
9b
9c
9d
9e
9f
10
lOa
lOb
lOc
lOd
11
lla
lib
lie
lid

He
llf

Hg
llh
iH
12
12a
12b
lie
12d
12e
12f
12g
12h
12i
12j
12k
121
12m
12n
12o
12p
12q
12r
12s
13
13a
13b
13c
Elution
Temp.
71.5°
72.3°
75.1°
78.7°
80.3°
81.1°
83.5°
85.5°
89.1°
91.1°
93.1°
97.5°
98.3°
100.7°
102.7°
103.1°
107.1°

108.3°
111.5°

113.1°
113.5°
114.3°
119.9°
121.9°
123.9°
127.9°
128.7°
121.9°
141.9°
144.7°
146.7°
156.3°
167.5°
171.9°
172.7°
177.5°
186.7°
189.1°
198.7°
199.1°
209.1°
212.3°
213.1°
218.7°
229.9°
245.1°
Compound
C. -benzene isomer
•ethylstyrene isomer
C10H20 isOBer
C.QH_2 isomer
C, -benzene isomer
C^-benzene isomer
indan
indene
diethylbenzene isomer
diethylbenzene isomer
phenol
C . -benzene isomer
4
C/H7 -benzene isomer
C,H7-benzene isomer
methylbenzofuran isomer
methylbenzofuran isomer
C.H.. isomer
4 24
cresol isomer
methyl-(2 ,3-dihydroindene)
isomer
cresol isoner
methyllndene isomer
cresol and methyl indene isomers
naphthalene
benzothiophene
C2 -phenol isomer
2-ethylbenzimidazole (tent)
C12H26 i801D"
saturated hydrocarbon
P-methyl-naphthalene
a -methyl-naphtha lene
C13H2» isomer
biphenyl
C.-H- isomer
acenapbthene
diphenylethane
dibenzofuran
fluorene
C,,H,, isomer (tent)
ID lo
C.,H,0 isomer (tent)
C16H18 isomer
dibenzothiophene (tent)
phenanthrene
DJO anthracene (i)
C18H22 isOBer
methylphenanthrene isomer
C.-H.. isomer
                            III-6

-------
TABLE 35.  GC/MS ANALYSIS OF GLASS FIBER FILTER EXTRACT FROM RUN 21
                           -QUALITATIVE RESULTS-
Chromato-
graphic
Peak No.
5
Sa
6
7
8
8a
8b
8c
9
10
lOa
11
lla
lib
12
13

13a
13b
13c
13d
13e
13f
13g
Elution
Temp.
134°
135°
151°
153°
163°
167°
169°
172°
173°
177°
178°
183°
185°
189°
191°
193°

197°
199°
• 200°
204°
207°
212°
213°
Compound
naphthalene
benzothiophene
B-raethy Inaphtha lene
a-methylnaphthalene
biphenyl
dimethylnaphthalene isomer
dimethylnaphthalene isomer
dimethylnaphthalene isomer
acenaphthylene
acenaphthene
alkyl-biphenyl isomer
dibenzofuran
trimethylnaphthalene isomer
isopropylnaphthaleae
fluorene
2,2,4-trimethyl penta 1,3-diol
di-isobutyrate (bkg)
C,,H,nO isomer (tent)
lj lu
C,,H,nO isomer (tent)
1J 1U
C16H18 isomer
C16H18 isomer
methyl-fluorene isomer
C.CH., isomer (tent)
15 lo
dibenzothiophene
Chromato-
graphic
Peak No.
13h
14
I4a
14b
14c
14d
14e
14f
15
15a
15b
16
16a
16b
16c
16d
17
17a
18
20
21
21a
21b

Elution
Temp.
CO
217°
218°
219°
229°
231°
233°
234°
239°
250°
250°
250°
250°
250°
250°
250°
250°
250°
250°
250°
250°
250°
250°
250°

Compound
phenanthrene
DIO anthracene (I)
anthracene
unknown





methyl-phenanthrene isomer
methylene-phenanthrene isomer
C15^12 isomer
unknowns
C16H10 isoner
C16H10 i80mer
2-benzyl benzimidazole
C16H10 isoioer






diphenyl disulfide or unknown
methyl pyrene isomer
methyl pyrene isomer
unknown
phthalate compound (bkg)
C18H12 isOBer
C18^12 isoiner
phthalate compound (bkg)
phthalate compound (bkg)
C20H12 i80aer (tent)« (5
C-..H. , isomer (tent), (5
20 12









rings)
rings)


                                 III-7

-------
TABLE 36.  GC/MS ANALYSIS OF TENAX SAMPLE UPSTREAM OF STEADY-STATE
                      XAD-2 TRAP FOR RUN 21
                      -QUALITATIVE RESULTS-
Chromato-
graphic
Peak No.
1
2

3
4
4a
5
6
6a
7
8
8a
9
10
lOa
lOb
11
12
12a
12b
15
16
16a
17
17a
18
18a
19
20
20a
21
21a
22
22a
23
23a
23b
23c
24
24a
Elution
Temp.
CO
57°
58°

60°
70°
72°
74°
79°
82°
85°
86°
87°
90°
96°
97°
102°
103°
116°
118°
122°
134°
136°
137°
140°
140°
140°
143°
144°
146°
151°
152°
154°
154°
155°
156°
156°
157°
158°
159°
164°
Compound
co2
carbonyl sulfide + sulfur
dioxide
hydrocarbon
acetone
CC.H.- isomer
methylene chloride (bkg)
carbon disulfide
C^HgO isomer
perfluorobenzene (I)
n-hexane
t r i chlo rome thane
perfluorotoluene (I)
benzene
thiophene + carbon tetrachloride
trichloroethylene
acetic acid
toluene
methylthiophene isomer
CgHjg isomer
xylene isomer
xylene + C.-thiophene isomers
C.-thiophene isomer
styrene
hydrocarbon
xylene isomer
C.-thiophene isomer
C.H20 isomer
C10H22 i80Der
saturated hydrocarbon
benzaldehyde
C.-benzene isomer
phenol (tent)
C.-benzene isomer
saturated hydrocarbon
C.-benzene isomer
saturated hydrocarbon
benzofuran
C.-benzene isomer
C^-benzene isomer
Chroma to-
graphic
Peak No.
25
25a
25b
26
27
27a

27b
28
28a
28b
29

30
30a
31
31a
31b
32

32a
33
33a
34
34a
35
35a
36
38
39
39a
39b
40
40a
41
42
42a
42b
42c
43
Elution
Temp.
CO
164°
166°
167°
167°
168°
169°

169°
170°
173°
174°
176°

177°
178°
180°
181°
182°
184°

185°
187°
188°
192°
193°
208°
210°
211°
216°
219°
221°
223°
224°
225°
226°
230°
233°
240°
240°
240°
Compound
C.-benzene isomer
saturated hydrocarbon
cresol isomer
C.H.-benzene isomer
indene
saturated hydrocarbon +
C,-benzene isomers
acetophenone
cresol isomer
C.-benzene isomer
C, -benzene isomer
methylbenzofuran + C.-phenol
isomers
methylbenzofuran isomer
saturated hydrocarbon
C.-phenol + C^-benzene isomers
C.-phenol isomer
C,H^ -benzene isomer
C.-phenol + methylindene
isomers
C^H. -benzene isomer
C.-phenol isomer
benzole acid (tent)
naphthalene
benzothiophene
p-methylnaphthalene
methylbenzothiopheue isomer
or-methylnaphthalene
biphenyl
saturated hydrocarbon
C^-naphthalene isomer
C.-naphthalene isomer
unknown
C.-naphthalene isomer
C..H. isomer + hydrocarbon
if, O
C..,Hj_ isomer
dibenzofuran
C..H.- isomer
tributyl phosphate (tent)
diphenyl oxazole isomer
                               III-8

-------
          TABLE 37. GC/MS  ANALYSIS OF TENAX SAMPLE DOWNSTREAM OF STEADY-STATE
                                  XAD-2 TRAP FOR RUN 21
                                   -QUALITATIVE RESULTS-

Chromato-
graphic
Peak No.
1
2
3
3a
3b
4
5
6
6a
6b
6c
7
7a
7b
8
9
9a
10
lOa
lOb
11
lla

12
12a
12b
12c
12d
13
13a
14
15
15a
15b
Elution
Temp.
CO
56°
57°
58°
66°
67°
68°
70°
71°
73°
74°
76°
76°
78°
78°
87°
92°
93°
95°
97°
100°
113°
116°

125°
130°
136°
138°
146°
148°
151°
152°
153°
156°
159°
Compound
co2
sulfur dioxide
carbonyl sulfide
butene isomer + freon 11
acetone
n-pentane
methylene chloride (bkg)
carbon disulfide
C^H. isomer
C5HJO isomer
C/HgO isomer
hexafluorobenzene (I)
n-hexane
chloroform (tent)
perfluorotoluene (I)
benzene
carbon tetrachloride
acetic acid
C,H.0 isomer
trichloroethylene
toluene
4-methyl-2-ethyl-l,3-dioxolane
(tent)
silane compound (bkg)
xylene isomer
hydrocarbon
hydrocarbon
hydrocarbon
benzaldehyde
phenol
C,H,N isomer
hydrocarbon
hydrocarbon
hydrocarbon
Chroma to-
graphic
Peak No.
17
18
18a
19
19a
20
20a
20b
20c
21
21a
22
24
25
*
it
*

27
*
28
*
31
32
33
34
35
36
37
*




Elution
Temp.
CO
164°
166°
167° *
168°
173°
174°
176°
178°
179°
180°
182°
183°
186°
189°
190°
197°
204°

206°
207°
208°
210°
217°
220°
228°
229°
232°
236°
240°
240°




Compound
cresol isomer
indene
acetophenone
cresol isomer
C, -benzene isomer
C. -phenol isomer







methyl benzofuran isomer
C. -phenol isomer
C, -benzene isomer
C. -phenol isomer
C,H, -benzene isomer
C.-phenol isomer
benzoic acid
naphthalene
benzothiophene
C-H.-benzene
oethylbenzothiophene
(tent)
B-methy Inaphtha 1 ene
methyl benzothiophene
cr-methylnaphthalene









isomer


isomer

thiobenzoic acid or unknown
biphenyl
hydrocarbon
dimethyl naphthalene
C.-naphthalene isomer
acenaphthylene
acenaphthene
dibenzofuran (tent)
C13H10 isOBier






isomer










*Minor unnumbered peak.
                                             III-9

-------
                                TECHNICAL REPORT DATA
                         (Please read Instructions on the reverse before completing)
 1. REPORT NO.
 EPA-600/7-79-201
                                                     3. RECIPIENT'S ACCESSION NO.
 4. TITLE AND SUBTITLE
 Pollutants from Synthetic Fuels Production: Sampling
  and Analysis Methods for Coal Gasification
                               5. REPORT DATE
                                August 1979
                               6. PERFORMING ORGANIZATION CODE
  AUTHORIS) S.K.Gangwal, P.M.Grohse, D.E.Wagoner,
 D. J. Minick, C. M. Sparacino, and R. A. Zweidinger
                                                     8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Research Triangle Institute
 P.O. Box 12194
 Research Triangle Park, North Carolina 27709
                               10. PROGRAM ELEMENT NO.
                                EHE623A
                               11. CONTRACT/GRANT NO.

                                Grant No. R804979
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC  27711
                               13. TYPE OF REPORT AND PERIOD COVERED
                               Task Final; 8/78 - 7/79
                               14. SPONSORING AGENCY CODE
                                 EPA/600/13
 is. SUPPLEMENTARY NOTES ffiRL-RTP project officer is N. Dean Smith, Mail Drop 61, 919/
 541-2708. Grant-related reports include EPA-600/7-78-171, EPA-600/7-79-200, and
 EPA-600/7-79-202.     	
 16. ABSTRACT
          The report describes sampling and analysis methods involving a laboratory-
 scale coal gasification facility used to study the generation, sampling, chemical
 analysis, process evaluation, and environmental assessment of pollutants from coal
 gasification. It describes methods for particulates,  organic condensibles, and va-
 pors or gases in the raw product stream of the gasifier as well as for solid residues.
 It describes gas chromatography (GC) procedures for measuring fixed gases, C1-C5
 hydrocarbons, sulfur gases, and C6-C8 aromatics.  Atomic adsorption (AA) proce-
 dures for measuring toxic trace elements include those for arsenic, selenium, lead,
 cadmium, chromium,  and mercury. Volatile organics are collected from the gas
 stream using polymeric sorbents (Tenax GC and XAD-2), and analyzed by glass ca-
 pillary GC/mass spectrometry (MS). The major nonvolatile byproduct (tar) is pre-
 fractionated by solvent partitioning into acid, base,  and neutral fractions. Each
 fraction is analyzed by capillary GC/MS or high-performance liquid chromatography
 (HPLC).  Typical results are given to illustrate  the nature of the compounds stu-
 died, the methodologies,  and their sensitivities.
 7.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                   b. IDENTIFIERS/OPEN ENDED TERMS
                        c. COSATI Field/Group
 Pollution
 Coal Gasification
 Sampling
 Analyzing
 Dust
 Aerosols
Organic Compounds
Vapors
Gases
Residues
Pollution Control
Stationary Sources
Synthetic Fuels
Particulates
136
13H
14B

11G
07D
07C
 8. DISTRIBUTION STATEMENT

 Release to Public
                   19. SECURITY CLASS {This Report)
                    Unclassified
                        21. NO. OF PAGES
                            102
                   20. SECURITY CLASS (TMspagc)
                    Unclassified
                        22. PRICE
EPA Form 2220-1 (t-73)
                ra-io

-------