EPA-600/2-77-147
September 1977
Environmental Protection Technology Series
                            SYNTHETIC FUEL  PRODUCTION
                                       FROM  SOLID WASTES

                                      Municipal Environmental Research Laboratory
                                            Office of Research and Development
                                           U.S. Environmental Protection Agency
                                                    Cincinnati, Ohio 45268

-------
                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 ENVIRONMENTAL PROTECTION TECH-
NOLOGY  series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia  22161.

-------
                                          EPA-600/2-77-147
                                          September 1977
SYNTHETIC FUEL PRODUCTION FROM SOLID WASTES
                     by

                Roy C. Feber
      Los Alamos Scientific Laboratory
         The University of California
        Los Alamos, New Mexico  87545

                    and

              Michael  J.  Antal
Aerospace and Mechanical  Sciences Department
            Princeton, New Jersey  98540
         Interagency Agreement No.
              EPA-IAG-D5-0646
              Project Officer

               Albert J. Klee
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
           Cincinnati, Ohio  45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
           CINCINNATI, OHIO  45268

-------
                               DISCLAIMER
       This report has been reviewed by the Municipal Environmental
Research Laboratory, U.S. Environmental Protection Agency, and approved
for publication.  Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.

-------
                             FOREWORD
     The Environmental Protection Agency was created because of public
and government concern about the dangers of pollution to the health and
welfare of the American people.  Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environ-
ment.  The complexity of that environment and the interplay between its
components require a concentrated and integrated attack on the problem.

     Research and development is that necessary first step in problem
solution and it involves defining the problem, measuring its impact, and
searching for solutions.  The Municipal Environmental Research Laboratory
develops new and improved technology and systems for the prevention,
treatment, and management of wastewater and solid and hazardous waste
pollutant discharges from municipal and community sources, for the
preservation and treatment of public drinking water supplies, and to
minimize the adverse economic, social, health, and aesthetic effects of
pollution.  This publication is one of the products of that research; a
most vital communications link between the researcher and the user
community.

     In particular, this study examines the potential and evaluates the
use of char produced from the pyrolysis of solid wastes as a source
of synthetic fuel.  It reflects our encouragement of greater interest
in the use of integrated schemes to meet our future energy requirements
in an environmentally acceptable way.
                                 Francis T. Mayo, Director
                                 Municipal Environmental Research
                                 Laboratory

-------
                             ABSTRACT
     The work described in this report has two objectives:  first, to
evaluate potential catalysts for the commercial practice of the gasifi-
cation of chars produced by the pyrolysis of municipal or industrial
wastes; second, to determine the potential for synthetic fuel production
from solid wastes produced in this country, and to explore the feasibil-
ity of providing the heat required for the gasification reactions by
coupling a chemical reactor to a solar collector.

     To meet the first objective, a small scale, fixed bed, flow-through
reactor was assembled, and a number of potential catalysts were tested
on chars from a number of sources.  The alkali metal carbonates are
superior to any other catalysts tested for gasification with both steam
and carbon dioxide at 650 C.  With these catalysts, rates of gasification
by steam are increased by factors of two to three, and rates of gasifi-
cation by carbon dioxide, by factors up to ten.  The rates are compar-
able with those observed elsewhere for other carbonaceous materials.

     To meet the second objective, several possible schemes for coupling
a solar collector and a gasification reactor are suggested, and economic
analyses of the systems are attempted.  It is concluded that a feasible,
economically attractive system is possible.

     This report was submitted in fulfillment of Interagency Agreement
EPA-IAG-D5-0646 by the Los Alamos Scientific Laboratory under the
sponsorship of the Environmental Protection Agency.  Work was completed
as of December 1975.
                                   IV

-------
                             CONTENTS

Foreword   	i i i
Abstract   	i v
Figures    	vi
Tables     	vi i
Acknowledgment  	viii

     I.  Introduction   	1
    II.  Conclusions    	4
   III.  Recommendations  	6
    IV.  Sources and Characterization of Chars 	7
     V.  Experimental Method 	23
    VI.  Experimental Results   	29
   VII.  Systems Study   	44

References	69
Appendices

     A.  Optimization of the reactor's volume  	72
     B.  Radiant heat transfer in gaseous HoO and CO? + H?0 mixture..
                                                                  75

-------
                                   FIGURES

Number                                                                   Page

  1      Monsanto char  carbon of probable biological  origin  	 9
  2      Garrett char  carbon of probable biological  origin   	 9
  3      Monsanto char  mineral  speroid  	 10
  4      Garrett char  mi seel 1aneous  fine carbon   	 10
  5      Monsanto char  metallic inclusions  	 11
  6      Garrett char  bright metallic inclusions   	11
  7,8    Monsanto char  probable pyrolyzed wood   	 13
  9,10   Monsanto char  principally inorganic matter  	 14
  11     Monsanto char  principally carbonaceous matter  	 15
  12     Garrett char  inorganic matter  	 16
  13     Garrett char  calcite crystal  	 16
  14,15  Sugar char  	 17
  16     Schematic of gasification reactor tube   	 25
  17     Char reactor equipment  	 26
  18     Reacted uncatalyzed sugar char  	 35
  19,20  Reacted catalyzed  sugar char  	 37
  21     Sugar char  K2COs  catalyst  	 38
  22     Sugar char  Li2COs-K2C03 catalyst  	 	 38
  23     Sugar char  KgCOs-DMSO catalyst  	 40
  24     First reactor design  	 54
  25     Second reactor geometry  	 55
  26     Schematic of window assembly  	 57
  27     Schematic of a tubular absorber  	 61
  28     Schematic of a nested annular fluidized-bed  reactor   	•	 62
  29     Light reflection between two mirror surfaces  	 73

-------
                                    TABLES


Number                                                                    Page

  1.   Thermodynamics of Char Gasification Reactions 	3
  2.   Equilibrium Between Char and Steam at 650 C   	3
  3.   Source and History of PERC Chars 	7
  4.   Fluorescence Analysis of Inorganic Material in Char  	12
  5.   Density and Surface Area of Chars  	18
  6.   Chemical Analyses of Monsanto and Garrett Chars  	20
  7.   Chemical Analyses of PERC Chars   	22
  8.   Gasification of Chars to CO by COg  	31
  9.   Gasification of Chars by Steam   	-.	33
 10.   Removal of Inorganic Constituents from Char by Leaching  	41
 11.   Recovery of Catalyst Fromars   	41
 12.   Composition of a Synthetic Solid Waste  	42
 13.   Analysis of Gas Evolved from Steam-Pyrolyzed Synthetic
        Solid Waste 	43
 14.   Average Analysis of Raw Municipal Wastes  	46
 15.   Hydrogen Production Potential of Wastes Produced in the  	46
        U. S. A. (1971)
 16.   Methanol Production Potential of Wastes Produced in the
        U. S. A. (1971)  	47
 17.   Mass and Heat Balance Calculations for the Adiabatic
        System  	49
 18.   Heliostat Economies of Scale   	51
 19.   Effect of Temperature on Radiant Heat Loss by Reactor 	58
 20.   Dimensions of Annular Reactors  	67
 21.   Economic Analysis of a Municipal  Synthetic Fuel Plant 	67
                                     vn

-------
                         ACKNOWLEDGMENTS


     The experimental part of this work was done at the Los Alamos
Scientific Laboratory, and the systems study was done by Prof.
Michael J. Antal.

     The authors are indebted to M. C. Tinkle for implementing the
experimental  part with the assistance of E. Virgil and W. W. Washichek.
They also wish to thank Prof. E. F. Thode, New Mexico State University,
for his interest and advice.
                                 vm

-------
                                SECTION I

                               INTRODUCTION
     The pyrolysis of municipal and industrial solid wastes has received
considerable attention as an environmentally acceptable alternative to the
disposal of such wastes by incineration, open dumping, or as landfill.  If,
in addition, the useful energy and materials contents of the wastes are
regarded as a valuable and renewable national resource, disposal by open
dumping or as landfill is clearly undesirable.  Also, of course, no volume
reduction has been accomplished.  Controlled incineration is capable in
principle of providing a significant volume reduction and of recovering a
significant fraction of the energy content as the heat of combustion.
However, additional equipment necessary to make the process thermally effic-
ient and environmentally acceptable is stated to have a relatively high
capital cost.

     A number of processes have been developed for the pyrolysis, or
destructive distillation, of solid wastes, some of which may be regarded
as commercial.  Much has been written on the subject, and one particulary
informative review has been prepared by Huang and Dalton.  Most of the
processes provide the heat necessary for pyrolysis by partially combust-
ing a portion of the wastes with a sub-stoichiometric amount of oxygen
(from air or pure oxygen).  Potentially useful liquids or a low Btu gas
are produced.  In addition a char which may contain 40 to 60% ash remains.
It is the utilization of this char to produce a synthetic fuel  which is the
subject of the present work.  If successful, an additional reduction of
the ultimate volume of solids to be disposed would also result.

     Synthetic fuel may be produced from the carbon contained in char by
gasification of the carbon with steam and/or carbon dioxide to produce
a gas mixture which also contains hydrogen, carbon monoxide, and methane.
The following set of reactions describes the process:

                                                       (1)

                                                       (2)

                                                       (3)

                                                       (4)
H20(g) -
C02(g) -
C0(g) +
C(s) +
«• c(s) ;
«• c(s) 5
H20(g);
2H2(g);
t H2(g) +
H 2CO(g)
^ C02(g)
^ CH4(g)
C0(g)

+ H2(g)


-------
Combining Reaction 1 with the water gas shift reaction, Reaction 3, gives:

                2H20(g) + C(s) * C02(g) + 2H2(g)         (5)

In a  flowing system the proportions of the desirable products hydrogen,
carbon monoxide and methane will depend on the temperature and the equi-
librium constants and relative rates of the above reactions at that
temperature.

      In Table 1 are summarized equilibrium constants and the standard
enthalpies of Reactions 1-5 from 600 to HOOK.  Thermodynamic data for the
tables are from the JANAF Thermochemical Tables.2  The data are based on
graphite as the standard state for carbon.  Therefore, the numbers will
be slightly in error for those reactions involving chars, as the carbon
in chars is non-graphitic.

      With the exception of Reaction 4, the gasification reactions are endo-
thermic and therefore require an external source of heat.  Although the
heat  might be provided by burning a portion of the char, the premise of
this  work is that it is preferable to extract a maximum amount of syn-
thetic fuel from the char and that the required heat will be provided
by solar process heat.  This premise in turn implies that gasification will
be done at about 650 C (923 K), a temperature which is regarded as
attainable by a reasonable extrapolation of current solar process heat
technology.  At that temperature the gas compositions shown in Table 2
would be obtained at the indicated pressures if an excess of char were
at equilibrium with steam (assuming ideal gas behavior).

      These calculations show a very favorable distribution among reaction
products at equilibrium at 650 C.  In fact, of course, equilibrium with
respect to the carbon will not be reached because gasification reactions
1, 2, and 4 are known to be slow, as are reactions in the gas phase
involving methane.

      One primary objective of this study is therefore to evaluate potential
catalysts for the commercial practice of Reactions 1 and 2 when applied
to chars resulting from the pyrolysis of municipal or industrial solid
wastes.  A second objective is to determine the potential for synthetic
fuel   production from solid wastes produced in this country and to explore
the feasibility of providing the heat required for the gasification
reactions by coupling a chemical reactor with a solar furnace.

-------
TABLE 1.  THERMODYNAMICS OF CHAR GASIFICATION REACTIONS

Equilibrium Constant (pressure in atm.)
Reaction*
T,OK 1 234 5
600
700
800
900
1000
1100
Reaction

TjOK
600
700
800
900
1000
1100
4.85 x 10-a i.7i x 10-o 28.4 100. 1.38 x
2.31 x 10-3 2.43 x 10-4 9.47 8.93 2.18 x
4.21 x TO'2 9.93 x 1Q-3 4.24 1.40 0.179
0.407 0.176 2.31 0.322 „ 0.941
2.50 1.73 1.44 9.75x10-2 3,60
11.0 11.1 0.993 3.63x10-2 11.0
Enthalpy, kcal
Reaction*
1 234 5
32.168 41.460 -9.292 -19.916 22.876
32.301 41.351 -9.050 -20.429 23.251
32.391 41.190 -8.799 -20.857 23.592
32.447 40.996 -8.549 -21.207 23.898
32.475 40.779 -8.304 -21.482 24.171
32.477 40.543 -8.066 -21.696 24.411
ID'2




*Reaction
1 H20(g) + C(s) + H2(g) + C0(g)
2 CD2(g) + C(s) £ 2CO(g)
3 COIg) + H20(g)* C02(g) + H2(g)
4 C(s) + 2H2(g) t CH4(g)
5 2H20(g) + C(s)£ C02(g) + 2H2(g)
TABLE 2. EQUILIBRIUM BETWEEN CHAR AND STEAM AT 650 C

Pressure
atm
0.5
1.0
2.0
5.0
10.0
, Moles C consumed/ Partial pressure of products,
mole added steam H20 H2 CO C02
0.682 0.051 0.235 0.139 0.062
0.619 0.149 0.435 0.218 0.153
0.574 0.400 0.768 0.333 0.358
0.541 1.341 1.518 0.565 1.026
0.530 3.139 2.425 0.828 2.203
atm
CH4
0.013
0.045
0.141
0.550
1.405

-------
                                SECTION II
                               CONCLUSIONS

     The  alkali metal carbonates are superior to any of the other catalysts
 tested  for  gasification by both steam and carbon dioxide,  other cata-
 lysts tested  included metallic Fe, V205, CoMoO*, NiMoO., a zeolite, and
 fly  ash from  a coal-fired plant.  Because ease of catalyst recovery must
 be considered for a practicable gasification process, or that catalyst.
 must be inexpensive enough and environmentally acceptable enough to dis-
 card, the number of candidates is probably limited, given the present
 state of  the  art.

     When 10  wt% of the alkali metal carbonates is added to Monsanto or PERC
 chars,  the  gasification rates with steam are increased by factors of two
 to three  and  those  with carbon dioxide by factors to ten.  Because un-
 catalyzed sugar char is less reactive than uncatalyzed Monsanto or PERC
 chars,  its  gasification rates are increased by the greatest amount.
 Acid leashing of ash-containing chars increased slightly reactivity with
 respect to  carbon dioxide, but decreased slightly steam gasification
 rates.

     A  comparison of all steam gasification rates suggests that inorganic
 material  in the chars, particularly in Monsanto char for which more data
 exist,  is itself acting as a catalyst for the reaction with steam.
 Attempts  to  improve contact of alkali metal carbonate catalysts with
 chars and thereby enhance gasification rates (by such techniques, for
 example,  as lowering the melting point of the catalyst below the gasifi-
 cation  temperature)  did, in general, increase the percent conversion of
 carbon  dioxide to carbon monoxide, but were not demonstrably effective
 in increasing the rate of steam gasification.  The general conclusion is
 drawn that  intimate  contact between catalyst and char is not critical.

     The  rates of catalyzed and uncatalyzed reactions tend to decrease with
 time of gasification.  Therefore, to judge if a catalyzed gasification
 of char can gasify  some acceptably large fraction of the available car-
 bon and be  an economically competitive source of synthesis gas, some
 kinetic questions need to be answered.

     The  major product of steam gasification is a mixture of hydrogen and
 carbon  dioxide at a  ratio of about 2:1 with catalyzed and uncatalyzed
Monsanto  and  PERC chars.  The amount of carbon monoxide is usually
*2.5%.   In the present apparatus the lowest practicable rate of water
addition and  the rate of steam gasification correspond to a partial

-------
pressure of steam throughout the reaction zone high enough to drive the
water gas shift reaction far to the right.  Indeed, the amount of carbon
monoxide found in the product gas is close to that which would be calcu-
lated if the shift reaction were at equilibrium.

     The gasification of raw solid wastes produced in this country can make
a significant contribution to national hydrogen or methanol requirements.

     Several possible schemes for coupling solar collectors to a gasifi-
cation reactor have been explored for their technical and economic feasibil-
ity.  Although many engineering features of such systems cannot be specified
at present, it is concluded that the development of a commercial solar-
steam pyrolysis systems for the production of synthetic fuel from solid
wastes deserves further effort.

-------
                               SECTION  III

                             RECOMMENDATIONS

     It has been concluded that alkali metal carbonates have proven their
effectiveness as catalysts for the gasification of char by carbon cioxide
or steam.  However, to determine if the system is a practical one for the
production of a synthesis gas and can make an impact on the char disposal
problem, further work should be done.

     The major thrust of work to be done next should be on engineering
kinetic studies.  There are three related aspects to the kinetics problem
which should be addressed.  First, the preponderance of evidence is that
char reactivity decreases with reaction time.  Therefore it is desirable
to define feasible limits to the extent to which carbon may be removed
from the chars.  Second, in the present apparatus the lowest practicable
rate of water addition corresponds to a partial pressure of water high
enough to drive the water gas shift reaction far to the right.  As a
result we produce H£ and C0? at a ratio of near 2:1.  A significant con-
centration of CO would be desirable for a synthesis gas.  Assuming that
a large scale system would be a flow system, one need to know, for
example, the effect of the relative rates of the carbon-steam reaction
and the forward and reverse rates of the water gas shift reaction as a
function of water partial pressure.  These relative rates will determine
the H2/CO/C02 ratio in the product gas.  Third, kinetic data are required
for engineering design of a large scale system.  A bench scale, stirred-
bed, continuous reactor should be capable of providing the necessary
data for these purposes.

     In addition, it would be appropriate to continue the search for cata-
lysts better than the alkali metal carbonates and for methods of con-
tacting char and catalyst which simplify catalyst recovery procedures.

-------
                                SECTION IV

                  SOURCES AND CHARACTERIZATION OF CHARS
SOURCES
     Chars from the pyrolysis of solid wastes were obtained from the
Monsanto Corporation, the Garrett Corporation, and the Pittsburgh Energy
Research Center (PERC).  The processes with which these organizations are
identified are described in some detail in Volume 2 of Reference 1.

     The Monsanto char derived from a pilot plant for the Landgard process,
in which pyrolysis takes place at 650 to 980 C.  The source wastes are
believed to have been municipal wastes.  Most of our experimental work
was done with this char.  The source for the char from the Garrett process
was Douglas fir tree bark which had been flash pyrolyzed to about 480 C.
Four samples of PERC char were obtained, each of which had a different
source and history (Table 3).

                TABLE 3.   SOURCE AND HISTORY OF PERC CHARS
    Code #	Source	Treatment	

    1221-5      Raw municipal refuse from        Carbonization tempera-
                F.A.M. plant in Altoona,         tures between 500° and
                Pa.                              9009:.

    1222-6      Same as above, but con-          Carbonized at 750 Sc.
                tains plastic film re-
                moved from 1221-5

    1223-1      Heil mill industrial ref-        Carbonized between 500°
                use.                             and 900<€.

    1224-3      Gondard mill industrial          Carbonized at 900°C.
                refuse.
     All of the PERC chars were chemically analyzed, but only #1221-5 was
used in gasification studies.  In contrast with the Monsanto and Garrett

-------
chars, the PERC chars were very heterogeneous and contained large chunks
of pyrolyzed material apparently originating from metal, paper, wood,
and fabric present in the wastes.

     As a reference, ash-free char, a sugar char of high surface area was
prepared following directions given in the literature. 3  This procedure
required a final temperature of 800 C.  Finally, a limited amount of
work was done with a char sample produced directly from the steam
pyrolysis of a synthetic waste.

CHARACTERIZATION

Microscopic Analysis

     A microscopic examination of grab samples of the Monsanto and
Garrett chars showed that they are largely composed of amorphous carbon,
some of which had a cellular structure indicating carbon of biological
origin.  There were also some particles that gave metallic reflections
and some particles that appeared to be crystalline.  Three photomicro-
graphs each of Monsanto and Garrett chars are shown in Figures 1 to 6,
Figures 1, 3, and 5 show metallographic sections of the Monsanto char.
Figures 2, 4, and 6 are sections of Garrett char.  All of the metallo-
graphic sections were etched with hydrogen ion bonbardment (cathodically
vacuum etched with hydrogen) at 3 kV for 15 minutes.  Figures 1 and 2
show carbon of probable biological origin in both Monsanto and Garrett
chars.  In Figure 4 some interesting cubes of mineral matter may be
seen in the Garrett char.  The mineral matter has a grey intensity in
these photomicrographs.  In Figures 5 and 6 metallic inclusions are shown
in both of the chars.  The metallic material shows up as the lightest in-
tensity in the photomicrographs.


Scanning Electron Microscopy

     Fugure 7 is a photomicrograph of organic constituents in Monsanto
char at a magnification of 1000X, and Figure 8 shows the same organic
constituents at 3000X.  Figure 9 shows a piece of inorganic matter in
Monsanto char at 3000X.  An electron fluorescence analysis of this
piece of material gave as the major components:  Mg, Ca, Si, and Al
with a trace of Ag.  The same piece of material at 10.000X is shown
in Figure 10.  Figure 11 is of a piece of primarily carbonaceous
material with a bright spot of inorganic material in the center.  A
normalized electron fluorescence analysis of the bright spot is
as follows:
                                    8

-------
                                              «;v       '•^» ^ «,
Figure 1.  Monsanto char.  Carbon of probable biological origin.   100X
                  •
 Figure 2.   Garrett char.  Carbon of probable biological origin  100X

-------
             »

Figure 3.  Monsanto char.  Mineral spheroid upper left.   100X


          %
                  »
                                      4*
                                       *
                                      -
                                                      «



Figure 4.   Garrett char.   Miscellaneous  fine  carbon.
                 Mineral  cubes  lower left.
100X
                              10

-------
                                               l\ *
                                                  ! < ^
     Figure 5,  Monsanto char.  Metallic inclusion upper center.    100X

                      *

     Fugure 6.  Garrett char.    100X
Bright miscellaneous inclusions.  Miscellaneous minerals (grey)
                                 11

-------
      TABLE 4.   FLUORESCENCE ANALYSIS OF  INORGANIC MATERIAL IN CHAR

      Elementwt. %  as uxlde
Na
Mg
Al
Si
Ca
Ti
Fe
Cr and Zr
P, S, Cl, K,
Mn, Ni, Cu, Zn
3.11
14.08
37.74
12.06
6.51
8.32
18.18
Present in sample but not in standard

Not detected

      ngures  \L  and  id  snow  scanning electron pnotomicrograpns at
 1000X and  3000X  of Garrett char.  The inorganic material in Figure 12
 was  composed  principally of  K, Ca, and Fe with traces of Al.  A crystal
 of what  appears  to be calcite appears in Figure 13.  Electron fluorescence
 analysis indicated that Ca was the principal material in this crystal.

      Figures  14  and  15  are photomicrographs of sugar char at 100X and
 3000X, respectively.

 X-Ray Analysis

      Char  from the Monsanto  pyrolysis contained two major crystalline
 components:   a-quartz and §  hexagonal phase with lattice constants
 a  =  4.33  A and  c =  17.08 A.  In addition, there were a few very weak
 unidentified  reflections and a pronounced background modulation having
 a  broad  peak  near d  = 3.5 A, perhaps indicative of the presence of
 non-graphitic carbon.   The structure and parameters-of the hexagonal
 phase were judged to be very similar to those of CaCOo (calcite), and
 probably represent a somewhat impure form of this carbonate.

      Char  produced by the Garrett process also contained a-quartz and
 impure CaCO,.  However, there was significantly less of the carbonate,
 and  it was not as crystalline as in the Monsanto sample.  There also a
 appeared to be a larger fraction of non-crystalline material, resulting in
 a  higher background but with a less broadened peak near d = 3.8 JL  It is
 unlikely that this peak is due to non-graphitic carbon alone, but it
 probably originates from a mixture of nearly amorphous components.

      Because  the samples contained particles which had a pronounced
magnetic behavior, a crude magnetic separation was effected, and the
resulting fractions were examined separately.  The magnetic fraction
still contained appreciable a-quartz, but none of the carbonate could
be detected.   Additional weak diffraction peaks were observed which were
not present in the patterns of the as-received materials.  These peaks
                                    12

-------
Figure 7.  Monsanto char.  Probable pyrolyzed wood.   1000X
Figure 8.   Monsanto char.   Probable pyrolyzed wood.    3000X


-------
Figure 9.  Monsanto char.   Principally inorganic matter.      3000X
Figure 10.   Monsanto char.   Principally inorganic mater.     lO.OOOX
                               M

-------
Figure 11.   Monsanto char.   Principally carbonaceous material
                           1000X
                              15

-------
Figure 12.   Garrett char.   Inorganic matter.    1000X
Figure 13.   Garrett char.   Calcite crystal.     3000X
                         16

-------
Figure 14.  Sugar char.    100X
Figure 15.   Sugar char.    3000X
                L/

-------
 could not be identified as  any of the  iron-bearing  phases  listed  in  the
 ASTM index, and no further  attempts  at identification were made.

 Density and Surface Area

      Results of He density  and BET surface  area measurements  are  given
 in Table 5.  For purposes of comparison,  the  density of amorphous carbon
 commonly ranges from 1.8 to 2.1  g/cm ,  while  that of graphite is  about
 2.3 - 2.4 g/cm3.   The high  density of  the Monsanto  char may be attributed
 to its metal and mineral content.

 	TABLE 5.   DENSITY AND SURFACE  AREA OF  CHARS	

         Char	Density, g/cm3	Surface  area,
Monsanto
Garrett
Sugar
2.34
1.69
1.86
117 t 9
128 * 2
635 t 100
 Chemical Analysis

      In order to obtain representative samples for chemical analysis, the
 Monsanto and Garrett chars were coned and quartered following the ASTM
 Standard Method of Sampling Coke, Designation D-346-35.  The Monsanto
 char, which was high in moisture content (24.4 wt. %), was dried prior to
 analysis.  The Garrett char, which was low in moisture (2.87 Wt. X), was
 analyzed as sampled.

     Samples of the four PERC chars were obtained by splitting the chars
 as received with a riffle sampler.  The samples were then dried to
 constant weight and milled.  Some small flakes of metal that could not be
 effectively pulverized and homogenized in a mortar were dissolved and
 reconstituted into the sample.

     Results of the analyses for the Monsanto and Garrett chars are
 given in Table 6, and those for the PERC chars are in Table 7.  Elements in
 the tables are ordered according to the periodic table.  With the exception
 of H, C, N, and S, most of the elements were determined by quantitative
 spectrochemical analysis, supplemented in a few cases by neutron activa-
 tion analysis.  The sums of the elemental analyses of the Monsanto and
 Garrett chars were about 63.8 and 78.0 wt X, respectively, and those of
 the PERC chars were about 72.1, 74.7, 62.2, and 63.1 wt X.  The balances
 are assumed to be largely oxygen bound as metal oxides in the chars.
Other than the important differences in the amount of available carbon,
the most striking difference between the Monsanto char and the PERC chars
 is the fact that the Cl  content of the latter is an order of magnitude
higher.
                                    18

-------
     The only detectable metallic Impurity in the sugar charcoal was
0.002 wt. % Si.
                                 19

-------
       TABLE  6.   CHEMICAL ANALYSES OF MONSANTO AND GARRETT  CHARS
                           (percent by weight)

Element
H
Li
B
C
N
F
Na
Mg
Al
Si
P
S
Cl
K
Ca
Ti
V
Cr
Mn
Fe
Co
Ml
Cu
Zn
Br
Sr
Zr
Mo
Ag
Cd
Cs
Ba
Hf
Ta
W
Pb
Th
U
Monsanto Char
0.9
0.002
0.015
35.4
0.50
0.017
2.30
0.80
2.80
11.54
0.5
0.27
0.13
1.0
4.03
0.8
0.005
0.05
0.2
1.96
0.0015
0.01
0.1
0.2
0.002
0.05
0.02
0.003
0.004
0.002
0.0002
0.15
0.0007
0.0004
0.0008
0.07
0.0012
0.0020
Garrett Char
2.33
0.0015
0.004
60.02
0.30
0.005
0.31
0.32
1.39
6.04
0.2
0.04
0.016
0.6
1.58
0.6
0.005
0.015
0.15
0.77
0.001
0.02
0.01
0.03
0.0002
0.04
0.02
<0.001
0.0002
<0.001
0.0001
0.05
0.0002
<0.0001
0.0004
0.002
0.0002
0.0055
Ash
54.8
22.7
                                   20

-------
Elements below the limit of detection:

          <0.01      As, Ce, Pr,  Nd, Sm, Gd, Tb
          <0.003     Mb, Ru, Rh,  Pd, Sb, La, Dy, Ho, Tm, !r, Tl
          <0.002     Pt, Au
          <0.001     Sc, Ga, Ge,  Y, In, Eu, Er, Lu
          <0.0005    Hg
          <0.0003    Yb
          <0.0001    Be
                              21

-------
              TABLE 7.  CHEMICAL ANALYSES OF PERC CHARS
                          (percent by weight)

Element
H
Li
B
C
N
F
Na
Mg
Al
Si
P
S
Cl
K
Ca
Ti
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
Sr
Zr
Mo
Ag
Cd
Ba
Pb
\sh

#1221-5
0.52
0.005
0.003
41.8
0.69
0.065
2.46
0.53
3.24
6.2
0.05
0.1
2.3
2.0
6.84
0.4
0.002
0.3
0.2
3.3
0.004
0.04
0.2
0.2
0.05
0.03
0.2
0.0005
<0.001
0.1
0.25
46.3

#1222-6
0.81
0.0025
0.004
51.8
1.03
0.022
0.84
0.54
2.85
4.0
0.1
0.1
2.6
0.6
6.52
0.5
0.002
0.05
0.2
1.6
0.001
0.004
0.06
0.2
0.05
0.006
0.003
0.002
<0.001
0.1
0.15
34.2

#1223-1
0.32
0.0015
0.01
22.6
0.19
0.049
4.92
1.01
3.39
20.7
0.1
0.1
0.73
0.6
5.13
0.25
0.003
0.07
0.2
1.3
0.001
0.004
0.05
0.3
0.04
0.004
0.001
0.0003
<0.001
0.1
0.15
73.6

#1224-3
0.41
0.008
0.02
21.3
0.25
0.043
3.09
1.26
2.58
19.0
0.05
0.1
0.92
2.0
7.86
0.5
0.006
0.05
0.2
2.7
0.004
0.01
0.05
0.2
0.07
0.04
0.003
0.002
0.001
0.2
0.2
71.9
Elements below the limit of detection:
<0.1
<0.01
<0.005
<0.001
<0.0001
<0. 00002
Ta,
Ga,
Sb
Ge,
Be
Sn
W
Nb

In,





Bi


                                  22

-------
                              SECTION   V

                          EXPERIMENTAL METHOD

BACKGROUND

   The catalyzed or  uncatalyzed  gasification of various forms of carbon
with H20 or C02 is the  subject of a  very extensive  literature.  In
recent years major concerns  have been both to minimize the reactions, if
they are between impurities  in the helium coolant of high-temperature,
gas-cooled reactors  and core graphite,  or to maximize the reactions, if
they are used to gasify coals or chars  from various sources.  From all
this experience, including much  fundamental research, certain generaliza-
tions relating to the present work are  possible.4'5

   Under otherwise identical conditions, the rate of carbon gasification
depends both on the  gas composition  and on the nature of the carbon sur-
face.  When the overall rate of  gasification is determined by rates of
processes occurring  at  the carbon surface (which will be the case in this
work), rather than by rates  of transport of reactant gas to the surface,
the relative rates of gas-carbon reactions are roughly in the ratios
1 x 105 :3:1:3 x 10~3 for C-02,C-H20, C-C02, and C-H2.

   Rate expressions  for gasification by H20 and C02 are commonly written
in the forirr
                                   kiPrn
              Rate(C +  C02)  = _   2                 (6)
                                 1 *  «   P   * '
                                              3
                                      Pu
              Rate(C + H20)             _2                 (7)
                                1 + k? Pn  + k, Pu
                                     *  H2    3  H20

   These expressions can be derived by postulating reaction mechanisms
involving the formation and removal of surface complexes.  Their form
suggests inhibition of the rates of gasification by the products CO or
H2.  Retardation of both reactions by H2 is generally conceded, but
retardation by C02 is in some dispute.  Study of the above reactions
is complicated by the simultaneous occurrence of the water-gas shift
reaction (Reaction 3).  Rate data for the latter reaction have been
fitted to expressions of the form^»'
                                   23

-------
                       k  Prn  P
     Rate(CO + H20) =   ]       H2°         (forward reaction),    (8)
     Rate(C02 + H,,) - k, P

                                            (reverse reaction),    (9)

     Impurities can increase the rates of all these reactions.  A signifi-
 cant number of metallic elements or compounds (particularly those of the
 transition metals and the alkali metal carbonates) catalyze the gasifi-
 cation reactions.5'8  Also, carbon itself is reported to be a catalyst
 for  the water-gas shift reaction.  On the other hand, some of the reac-
 tions are subject to poisoning, e.g., by chlorine or boron.

   As noted, the purpose of the experimental part of this work is to
 apply what is known about catalysis of the gasification reactions to
 char produced by the pyrolysis of solid wastes, and to try to find a
 catalyst for the commercial application of the process.  The direct
 transfer of results with other sources of char is not immediately obvious,
 as the ash in the present chars contains both potential catalysts and
 potential poisons, and the form of the carbon may be different.  Because
 the  primary purpose of this research is to identify suitable gasification
 catalysts, we have not attempted to optimize methane production by the
 simultaneous addition of methanation catalysts.

 EQUIPMENT

   Some features of the experimental apparatus were modified during the
 course of work, and the following description refers to that used for
 most of the gasification runs.

   Runs were made with 1 gram samples mounted in the reactor tube
 sketched in Figure 16.  A photograph of the total system is shown in
 Figure 17.  The reactor tube was fabricated principally of 304 stainless
 steel and was installed in the system with Swagelok fittings.  The char
 bed was supported on a stainless steel screen, and gas flow was down
 through the reactor tube.  Heating was by a multi-zone furnace capable
 of temperatures to 800 C.  Inlet or reactive gases fed to the reactor
 tube were metered with calibrated rotameters.  Liquid water was pumped
with a calibrated Harvard Compact Infusion Pump through 304 stainless
 steel tubing and vaporized to steam on hot surfaces directly above the
 reactor tube.  Suitable valves prevented back-diffusion and condensation
of the steam in other parts of the system.  Water in off -gases from the
reactor was condensed out with an ice bath, and the flow rate of the
                                    24

-------
I" ODx 0.065" Wall
     Tube, SST-
       Char Bed
    SST Screen Sized
     for Material  Used-

    Removable  Portion-

J
-|

 t
Figure 16.   Schematic of gasification  reactor tube.
                       25

-------
Figure 17.   Char reactor equipment.

-------
remaining gases was measured with a following bubble flow meter.  Product
gases (except water) were periodically analyzed with a Hewlett-Packard
Model 5711A gas chromatograph.  Initially silica gel was used in the
chromatographic columns but Carbosieve B proved to be much more effective
in separating the components of the off-gases.

CALIBRATION OF THE CHROMATOGRAPH

   The response of the gas chromatograph was calibrated with pure H2, CO,
C02> or CH4 and a number of standardized mixed gases from among those
components.

   The quantification of chromatographic peaks is usually accomplished
by integrating the areas under the peaks.  When applied to the calibra-
tion of CO, C02, or CH/j, the result is a linear plot of peak area
versus concentration.  However, Ho is a special problem because the
thermal conductivities of H2 and the column carrier gas, helium, are
similar in magnitude.  In practice a "normal" positive peak appears for
Hp at low concentrations.  At some concentration, *• 15% in our equipment,
tne top of the peak flattens.  With further increasing H2 concentrations,
a double peak forms, the minimum between the peaks starts back down
toward the base line and eventually extends well into the negative
direction.  Initially we hoped to retain the convenience of a linear
plot of concentration versus peak area by defining the area as that
of the unfolded peak.  This procedure did not give a linear plot.
Although still other alternatives are available, we have adopted a non-
linear calibration plot of peak height versus concentration, where the
peak height is defined as the height of the unfolded peak. The problems
of measuring hydrogen by gas chromatography have been discussed by
Villalobos and Nuss.9

PROCEDURE

   The system was brought to temperature (650 C) in the presence of a
flowing inert gas, usually helium, before switching to the desired
reactant gas mixture.  Runs were made at an ambient pressure of about
0.8 atm.  Various sequences of gas flows and compositions were used, but
a typical run might last 4 hours and consist of the following parts,
each lasting one hour and in the indicated order (gas flows measured at 1
atm pressure and room temperature): (1) 10.8 ml/min of COo, (2) 10.8
ml/min of He + 0.0417 g/min of H20 (56.6 ml/min of "steam*1), (3) 0.0417
g/min of H20, and (4) 10.8 ml/min of CCL.  The purpose of the fourth
part with C02 in such a run was to monitor any decrease in the reactivity
of the chars over the indicated time interval.

   Information available from which to calculate gasification rates
includes the weights of char and water consumed as determined from
material balances, a comparison of tailpipe gas flows with and without
a carrier gas, and the rate of recovery of a gas concentration when
                                   27

-------
reactant gas composition is changes suddenly.

   Because the chars were not taken to exhaustion and each run was started
with a fresh batch, the average rates reported are probably higher than
would be observed for total gasification of the available carbon.
                                   28

-------
                              SECTION VI

                         EXPERIMENTAL RESULTS
CATALYST MATERIALS
   Decision as to which catalysts to test first was based on reported
results in the literature for similar systems with due regard for pro-
jected catalyst cost and ease of recovery.  Because it soon became
apparent that the alkali metal carbonates were indeed relatively more
effective than other catalysts tested, several attempts were made to
improve physical contact between the carbonates and char and hopefully
to improve thereby their performance further.  The modifications made
to the carbonate-char system are included in the following list of
catalysts studied:

   1)  KpCOo and CsoCOo (Some early work, not reported, was also done
       with Na2C03.)

   2)  VpOg (vanadium pentoxide)
       The reagent grade chemical was mixed dry with the char.

   3)  Fe (Metallic iron)
       The reduced, reagent grade chemical was mixed dry with the char.

   4)  CoMo04 (cobalt molybdate)
       This material was obtained as a commercial  catalyst, suspended
       on alumina, from the Houdry Process and Chemical Company.  Before
       use, the as-received pellets were ground and sieved to -35
       mesh.  The powdered catalyst was then mixed dry with the char.

   5)  NiMo04 (nickel molybdate)
       The source, received form, and preparation  of this commercial
       catalyst were the same as for the cobalt molybdate.

   6)  Fly ash
       This material had been collected at a commercial, coal-fired
       power plant and was mixed dry with the char.
   7)
       Equal  molar mixtures of Li2C03 and K2C03, which melt at about
       500 C, were used.   Therefore, at the experimental gasification
       temperature, the catalyst was molten and potentially provided
       better and more uniform contact with the char.  Catalyst was
       added  to the char in two different ways:


                                   29

-------
     a)  by slurring the mixture of carbonates in water, adding char to
         the slurry, and evaporating to dryness,
     b)  by first making a fused mixture of the carbonates, grinding
         and sieving the fused salts to -35 mesh, and adding dry to
         the char.

   8)    K2C03-DMSO
         Dimethyl sulfoxide (DMSO) has remarkable wetting and solubiliz-
         ing properties.  As an alternate method to enhance catalyst
         coverage of or contact with char, K2C03 was dissolved in an
         equal volume mixture of DMSO and water, the solution was added
         to a char sample, and the slurry was dried at 130 C for four
         hours (K2C03 does not have sufficient solubility in pure DMSO)

   For runs with all these materials, 0.1 g of catalyst was mixed with
0.9 g of char.  Runs with no catalyst and one run with DMSO alone also
used 0.9 g of char.

BLANK RUNS

   Two blank runs were made at 650 C with 10.8 ml/min of He (measured
at 1 atm and room temperature) and 0.0417 g/min of H20.  Thus the mol
fraction of He was 0.16 and that of steam was 0.84.  The first run was
made with an empty reactor and the second with 1 g of K2C03 in the
reactor.  The product gas concentrations quoted below refer to gas
chromatographic analysis of streams from which essentially all the H20
has been trapped out; i.e., they are percents of products referred to He.

EMPTY REACTOR

   The major product gas was H2-  At temperature just before addition
of water there was 0.2% H2 in the tailpipe He.  The amount increased
to 0.8% after 35 min of He-H20 flow, and then decreased to a steady
state 0.4%.  The C02 started out at 0.6% (presumably due to C02 impurity
in the He), increased to 0.7%.  There were traces of CO and CH4 (_<0.06%).

REACTOR PLUS K2C03

   Gas analyses were made for this run during furnace heat up and a  He
flow of 10.8 ml/min, during 100 min of He + H20 at temperature, and
during furnace cool down under He flow.  During furnace heat up and
before the addition of HpO there was a spike of C02 in which C02 started
from 0.6%, increased to j.8% and decreased to 0.7 + 0.1%. C02 remained
at the latter concentration during the remainder of the run.  The spike
is presumably due to C02 adsorbed on the carbonate.  The Ho concentration
increased to 0.5% at temperature before the addition of H2&, it had
increased to 1.4% and thereafter decreased slowly during the remainder of
the run, becoming undetectable shortly after the H20 and furnace were
turned off.  CO and CH4 were not detected.
                                   30

-------
CHAR GASIFICATION BY CARBON DIOXIDE

   Results for the gasification of chars by C02 are summarized in Table 8.
The results are arranged in the order of decreasing average percent con-
version of C02 to CO during the first hour of the runs.  The percent
conversion is defined as one-half the percent CO in the C02-  At equilib-
rium that quantity would be 21.2% under the conditions of the experiments.

   There is frequently, but not always, a relatively rapid decrease in
the percent conversion during the first 20 min of a run.  Also tabulated
are the percent conversions observed during the fourth hour for those
runs made according to the sequence outlined above as typical.  All
chars, including sugar char, showed a decrease in the conversion of
C02 to CO from the first hour to the fourth hour.

	TABLE 8.  GASIFICATION OF CHARS TO CO BY CO?	
Char
Catalyst
                                                       % Conversion to CO
1st hr
4th
Synthetic
Monsanto
Monsanto, acid leached
Garrett
Sugar
Monsanto, acid leached
Monsanto
Monsanto
Sugar
Monsanto
Monsanto
Monsanto, acid leached
PERC
PERC, acid leached
Monsanto
Monsanto
Monsanto
Garrett
Sugar
Sugar
Monsanto
Sugar
PERC
K2C03
pre-fused Li'2C03-K2C03
pre-fused Li'2C03-K2C03
K2C03
K2C03-DMSO
K2C03-
Li2C03-K2C03mixture
K2C03
pre-fused Li'2C03-K2C03
Fe
K2C03-DMSO
none
K2C03
none
V205
CoMoO
NiMoO
none
DMSO
fly ash
none
none
none
12.3
10.5
10.4
8.8
7.4
7.1
6.4
6.0
5.7
5.2
4.1
3.3
2.0
1.8
1.8
1.8
1.7
1.6
1.2
1.0
0.8
0.6
0.6
5.6
1.7
2.3
-
5.1
1.8
1.8
4.6
1.0
1.0
1.4
-
0.5
0.6
0.6
0.6
-
0.4
0.2
-
0.2


    The synthetic char referred to in Table 8 was a sample from a steam
pyrolysis experiment on a synthetic solid waste described later in this
Section.  Runs are also listed which were made with acid-leached Monsanto
                                    31

-------
 or PERC  chars  to  determine  the effect,  if any, of removing potentially
 catalytic  or poisoning  inorganic constituents of the chars.  The acid
 leaches  were made at  68 C with 60 or  120 ml of 1 N HN03 per gram of
 char followed  by  three  equal  volume water washes at the same temperature.

     The  following conclusions are drawn from the results in
 Table 8:
     The  inorganic constituents in the Monsanto, PERC, and Garrett chars
     have little or no catalytic activity for gasification by C02-

     Acid-leached  chars  in the absence of added catalyst have somewhat
     increased  reactivity with C02-

     The  alkali metal  carbonates are the best of the catalysts tested.
     With 10 wt %  catalyst,  rates of gasification by C02 are increased
     as much as 10 times.

     Procedures designed to  improve contact of alkali metal carbonates
     with char  do,  in  general, increase percent conversions.

     There  is a significant  and reproducible variation of results with
     chars  from different sources.  The reactivity of subject chars de-
     creased in the order Garrett > Monsanto > PERC.

 CHAR GASIFICATION BY  STEAM

     Results obtained  for the  rate of  gasification of chars by steam are
 given in Table 9.   The  data have been arranged in the order of decreas-
 ing gasification  rates  per  kg of char, and a column expressing the re-
 sults per  kg of available carbon has  been added.  Because the various
 measures of gasification given above  in Section V did not always agree,
 the tabulated  rates are probable good only to
    The PERC char proved somewhat difficult to work with in the steam
system.  Because the char contained large chunks of metal and other
matter, it was necessary to mill the material before all gasification
studies.  Furthermore, unless the pulverized char was agglomerated with
a small amount of water and then dried prior to steam gasification
runs, the flow system intermittently plugged.

    The major product of steam gasification was a mixture of H2 and C02
at a ratio of about 2:1.  With one exception, the amount of CO in the
product (measured on a dry basis) was < 5% and was usually < 2.5%.  The
exception was the run made with uncatalyzed, acid-leached Monsanto char,
in which CO was analyzed to be about 10% of the total (dry) gas.  Cfy
production was negligible in all cases.

    The small amount of CO produced is consistent with the observed
rates of steam gasification and the observed rates of steam gasification
and the assumption that, although the gases are not in equilibrium with
                                   32

-------
TABLE 9.  GASIFICATION OF CHARS BY STEAM

Liters of gas/mi n/
Char
Sugar
Sugar
Sugar
Synthetic
Monsanto
Monsanto
Garrett
Monsanto
Monsanto, acid leached
PERC
Monsanto
Monsanto
Monsanto, acid leached
Monsanto, acid leached
Monsanto
Monsanto
Monsanto
PERC
Monsanto
Sugar
Sugar
Sugar
Catalyst
K2C03
K2C03-DMSO
pre-fused Li'2C03-K2C03
K2C03
Li2C03-K9C03 mixture
pre-fused Li 2C03~K2C03
K?C03
pre-fused Li2C03-K?C03
K2C03
none
Fe
none
K2C03
Zeolite
K2C03-DMSO
CoMoO/i
none
V2°5
DMSO
none
fly ash
kg of
char
7.5
7.5
7.5
7.5
4.8
4.7
4.6
4.5
4.5
4.2
2.9
2.6
2.4
2.4
2.3
2.3
2.3
1.9
1.9
1.1
0.5
0.4
kg of
carbon
7.5
7.5
7.5
13.0
13.5
13.3
7.6
12.7
—
10.0
8.2
7.2
—
—
6.5
6.4
6.4
4.5
5.3
1.1
0.5
0.4
                    33

-------
 solid  carbon,  the water-gas shift reaction is at or near equilibrium.
 With the  steam flow rate used, the partial pressures of CO? and #2
 produced,  and  an extent of reaction corresponding to gasification rate of,
 say, 4.5  liters of  (dry) gas/kg of char/min, about 4% of the product gas
 would  be  CO  at equilibrium.  This might well be the maximum amount
 expected,  because as the product gases cooled, the amount of CO at
 equilibrium  with other products and steam would be expected to decrease
 with decreasing temperature until the process became rate limited.

    The relatively  rapid approach to equilibrium of the water-gas shift
 reaction  was also checked in the reverse direction by passing a 2:1
 mixture of \\%  and C02 over Monsanto char containing 10 wt % K2C03 at
 650 C.  The  predicted equilibrium constant was satisfied to with 10%.

    The following additional observations are made:

    The alkali metal carbonates are superior to any of the char
    catalysts  tested.

    Acid  leaching of chars appears to decrease slightly steam
    gasification rates.  This fact plus other comparisons of data
    in Table 9 support the conslusion that inorganic material in the
    chars, particularly in Monsanto char for which more data exist,
    is itself  acting as a catalyst for the reaction with steam.  For
    this  reason the increase of gasification rate obtained by adding
    10 wt  %  catalyst is smaller than for gasification with COg, and
    factors  of 2 to 3 are observed for Monsanto and PERC chars.

    Attempts to improve, contact of alkali metal carbonate catalysts
     with  chars and thereby enhance gasification rates by such tech-
    niques as  lowering the melting point of the catalyst below the
    gasification temperature were not particularly effective.

    When uncatalyzed the relative reactivities of the subject chars
    with steam again decrease in the order:  Garrett, Monsanto, and
    PERC.  When catalyzed, distinctions tend to disappear.

    The rates of steam gasification of catalyzed chars obtained here are
 similar to rates obtained elsewhere for other carbonaceous material
 under similar conditions; e.g., by Lewis, GiHi land,  and Hipkin on
 l<2C03-catalyzed wood charcoal,'0 and by Taylor and Bowman on uncatalyzed
 coal chars.  H

SCANNING ELECTRON MICROSCOPY OF PARTIALLY REACTED CHAR

    A series of partially reacted sugar chars was examined by scanning
electron microscopy.  The purpose was to document variations in the form
of the deposited catalyst and to correlate this, if possible, with any
evidence for a  corresponding variation in the nature of the surface
reaction.   Sugar char was used because the ash content of the other
                                   34

-------


Figure 18.   Reacted uncatalyzed sugar char.
                    100X
                    35

-------
chars made identification of surface features very difficult.

    The unreacted char (Figures 14 and 15) had large, smooth-edged
voids and cleavage-type fracture surfaces.  The unfractured surfaces were
smooth.  Figure 18 shows uncatalyzed sugar char that had been reacted
at  temperatures to 750 C in flowing C02-  Small, generally lenticular
voids are seen, and the cleavage markings on fractured surfaces are not
as  clearly delineated as in the unreacted sugar char.  The large void
edges are still generally smooth.  In none of the other samples were the
lenticular voids as apparent as in this case.  Figures 19 and 20 are
photomicrographs of sugar char that had been mixed with CS2C03 catalyst
prior to being reacted to temperatures up to 750 C in flowing C02-  The
catalyst was not uniformly distributed, and many char particles had no
catalyst on them.  The particles containing catalyst appeared to have
undergone a considerable amount of reaction adjacent to the catalyst.  A
few small lenticular voids were present, and cleavage marks were fairly
distinct.

    The remaining photomicrographs are of partially reacted sugar char
exposed to steam.  Figure 21 is a photomicrograph of a typical mode of
K2C03 deposition by recrystallization when the catalyst was dissolved in
water, the solution slurried with sugar char, and the slurry evaporated
to  dryness.  This particular sample had then been partially gasified with
a He-H20 gas mixture.  The catalyst was not uniformly distributed.  Some
particles had little, if any, catalyst, and the particles with catalyst
had a nonuniform catalyst distribution.  Both rough and smooth-edged
voids were present on the catalyst-containing particles.  The type of
edge, smooth or rough, did not correlate with the presentee or absence
of  catalyst.  The particles free of catalyst were generally smooth
edged.  No lenticular voids were found.  In addition to the deposition
of  the sort shown in Figure 21 the catalyst was also present in some areas
as  a rounded mass.

    Figure 22 is a photomicrograph of sugar char catalyzed with molten
Li'2C03-K2C03 and then reacted at 650 C with C02 and steam.  Catalyst was
present on all particles, but was not uniformly distributed.  Particles
low in catalyst had more rough-edged voids than those rich in catalyst.
No  lenticular voids were found.  The catalyst appears to have wet the
char, and no deposits of the sort shown in Figure 21 were found.

    Catalyst had been deposited on another sugar char sample by evaporat-
ing to dryness a slurry of char in a solution of K2C03 in a water-DMSO
mixture.  The char had then been partially gasified at 650 C with C02 and
H20.  Two types of catalyst particles were seen with the scanning
electron microscope.  Some were very small (^1 y); others were much
larger and appeared as columnar growths from the char surface and had
striations perpendicular to that surface (Figure 23).  The growths were
reminiscent of the catalyst particles shown in Figure 21.  In general,
the surface had a more roughened character than surfaces of the other
samples.   In addition to some small  lenticular voids (1-2 y) on the
                                    36

-------
Figure 19.  Reacted catalyzed sugar char.  100X
Figure 20.   Reacted catalyzed sugar char.   3000X
                      37

-------

    ,  •
•BBP^^" *



 Figure 21.   Sugar  char.   1^003  catalyst.  1000X
Figure 22.  Sugar char.
                                 catalyst.   3000X
                        ,,.

-------
surface, many other small narrower voids were visible.

    Although there are obvious differences in the character of the deposi-
ted catalyst in the various samples and some not so obvious divverences
in the char surfaces themselves, a correlation with char reactivity does
not seem possible at present.

LEACHING OF INORGANIC CONSTITUENTS FROM CHAR

    Reference was made above to the preparation of acid-leached Monsanto
and PERC chars by extraction with 1 N HN03 followed by water washes.  We
report here the percent of the initial content of several elements in
these chars which was so removed, as determined from analysis of com-
bined extraction solutions.  Leaching experiments were also made with
three equal volume water washes alone, also at 68 C.  Results are given
in Table 10.

    The cases in which greater than 100% of the initial elemental content
of the chars was recovered can probably be attributed to the difficulties
of sampling heterogeneous chars.

RECOVERY OF CARBONATE CATALYSTS FROM PARTIALLY GASIFIED CHARS

    Quarter-gram samples of partially gasified, catalyzed chars from
four runs were each leached with three 50-ml portions of distilled water
at 68 C as a preliminary test of the feasibility of recovering carbonate
catalyst from spent chars.  The type of char, catalyst used, and percent
catalyst recovered are tabulated in Table 11.  The amount of catalyst
recovered is based on analyses for lithium and potassium only.  The
potassium added as catalyst is 10 to 20 times the amount present in the
Monsanto char itself when the amount of catalyst added to 10 \nt% of the
char.

    Water is only partially effective for the recovery of catalyst.  It
is somewhat easier to remove catalyst with water from sugar char than from
Monsanto char.  The difference may perhaps be explained by the possibil-
ity of solid solution formation with oxides or other inorganic compounds
in the Monsanto char at gasification temperature.

STEAM PYROLYSIS OF A SIMULATED SOLID WASTE

    A single experiment was conducted to explore the response of a
solid waste mixture to pyrolysis in a steam atmosphere.  For this purpose
a cylindrical, electrically heated 1700 watt furnace was used to heat a
stainless steel reactor of about one liter capacity which contained 188 g
of a simulated solid waste.  Water was added to the reactor at the rate of
3.7 ml/min with a peristaltic pump.  Exit water and condensable products
were collected in a trap, and gaseous products were sampled and later
analyzed with a mass spectrometer.  The composition of the simulated
                                    39

-------
Figure 23.   Sugar char.   K2C03-DMSO catalyst.   3000X
                        40

-------
TABLE 10.  REMOVAL OF INORGANIC CONSTITUENTS FROM CHAR BY LEACHING





Percent of Initial Content Removed
PERC Char Monsanto Char
Water 1 N HNOq Water 1 N HNO^
Na
Mg
Al
Si
P
K
Ca
Fe
Cu
Zn
30.5
9.4
0.5
4.0
5.0
37.5
2.9
0.003
0.5
0.4
61.0
94.3
92.6
20.2
2500
62.5
87.7
75.8
175
175
4.4
6.2
0.3
2.2
0.5
15.0
3.7
0.02
0.5
0.2
26.1
93.8
71.4
13.0
90.0
40.0
111.7
51.0
45.0
125

            TABLE 11.  RECOVERY OF CATALYST FROM CHARS

Char
Sugar
Sugar
Monsanto
Monsanto
Catalyst
K2C03
Li2C03-K2C03
L?2C§3
Percent Recovered
33.2
36 '.8 K2C03
62.2 L12C03
25.2
26.1 K2COs
a r\ r\ i • /N/\
                                                        23
                                 41

-------
solid waste is given in Table 12.  Water feed was started at a reactor
temperature of 150 C, and the temperature was permitted to increase to
750 C over a period of 1 1/2 hours.

           TABLE 12.  COMPOSITION OF A SYNTHETIC SOLID WASTE

Material
Wt.,
grams
Material
Wt.,
grams
paper
food
packaging paper
lawn clippings
cloth
53
40
30
22
22
.5
.3
,2
.3
.3
wood
packaging plastics
rubber and leather
plastics
6.2
5.1
4.7
3.3
                           TOTAL:    188.1  g
    The composition of the pyrolysis gas as sampled at various times
during the run is given in Table 13.  At the lower temperatures, a small
amount of tar was trapped, but tar evolution ceased shortly after gas
sampling began.  At the end of the experiment, 11.3 g of char remained
which had the following composition on a dry basis (wt%):  C, 57.6;
H, 0.73; S, 0.47; N 0.13; ash, 38.6.

    It is concluded that solid wastes are amenable to steam reforming and
a synthesis gas composed of light hydrocarbons is thereby produced.  More
research is required to determine optimum temperature and yields.
                                   42

-------
                   TABLE 13.   ANALYSIS OF GAS EVOLVED FROM STEAM PYROLYZED  SYNTHETIC SOLID WASTE
CJ

Gas Composition, Vol%
Temp.
C
295
360
400
460
500
550
650
700
750
700
,
H2
4.2
21.0
54.6
57.2
57.6
55.9
43.9
61.5
54.8
37.6

CH4
11.5
21.5
13.3
8.5
4.3
2.6
2.3
1.9
2.3
1.0

C2
4.6
2.1
2.5
2.2
-
-
_
_
-
_

C3 C4
0.9 0.5
0.8 0.7
1.0 0.3
0.2
-
-
- -
_ -
-
-

CeHe
0.6
1.8
0.6
0.1
0.03
-
-
-
-
-

CO
21.1
16.9
4.4
5.6
14.9
14.6
13.2
12.9
18.8
6.0

C02
49.1
31.0
20.0
19.0
11.3
17.6
19.0
18.0
17.2
50.8

N2
0.7
0.87
0.52
0.93
0.28
0.87
0.86
0.45
1.0
1.5

02
0.19
0.25
0.15
0.17
0.07
0.25
0.26
0.13
0.24
0.42

H20
6.7
3.0
2.6
6.3
11.6
8.2
20.4
5.1
5.6
2.7
Time,
hr
0.20
0.53
0.72
0.87
1.02
1.12
1.27
1.33
1.47
1.97

-------
                             SECTION VII
                            SYSTEMS STUDY

    The objectives of the systems study are to determine the synthetic fuel
 production  potential of the solid wastes produced in this country and to
 isolate a system that would be suited to the production of hydrogen or
 methane! from  those solid wastes.  The system under investigation uses an
 external heat  source to supply the heat required by the endothermic
 gasification reactions.  For purposes of this study, the heat source is
 anticipated to be a power tower.  However, future developments may enable
 less  elaborate solar furnace designs to be used.  Also, the candidate solid
 wastes to be gasified are not limited to chars.

    Although the emphasis of the systems study has been on the production
 of methanol from solid wastes, the merits of producing hydrogen should
 not be underestimated.  The production of ammonia from methane represents
 between 3 and  4 % of our nation's natural gas demand.  The methane is
 reformed to produce hydrogen, and the hydrogen is used to produce ammonia
 via the Haber  process.  Thus, there presently exists a large demand for
 hydrogen, and  producing hydrogen from solid wastes would relieve some
 of the stress  on the nation's natural gas reserves.  Economic considera-
 tions will  ultimately determine whether solid wastes are used to produce
 hydrogen or methanol.

 SYNTHETIC FUEL PRODUCTION POTENTIAL OF SOLID WASTES

    At the  present time over 90% of our nation's energy demand is met
 with  fossil fuels.  During the next sixty-five years we will have ex-
 hausted our known natural gas and oil reserves, leaving only coal as a
 readily available fuel resource. 12  Primary energy sources, such as
 nuclear, thermonuclear, solar, and geothermal, are expected to make
 up the difference by the production of electricity and synthetic fuels
 Of course,  synthetic fuel production will have to depend upon some other
 source of raw  material.  But most sources of raw materials are also
 subject to  exhaustion.  Perhaps thermochemical hydrogen production
will solve  this problem by using water as a raw material, but this
 solution has not yet been proven practical.  Therefore, it seems wise to
 plan for the sustained production of synthetic fuels from our renewable
resources to avoid a future fuel crisis.

    Solid wastes constitute one of our major renewable resources.  Un-
fortunately, the energy content of solid wastes is small,13 so that they
can only be used to meet a few percent of our nation's energy demand. 14,15


                                    44

-------
However, solid wastes can also be used as a raw material for capturing
and storing the energy of an external heat source.  When used in this
manner solid wastes have the potential of meeting over half our entire
oil demand in the form of methanol.

    Over 50% (by weight) of the wastes discarded in this country consist
of hydrocarbons with a primary composition of CxHyOz.  Sulfur and nitrogen
are minor constituents of most solid wastes. 13,lo   under the proper
conditions solid wastes react with steam to yield hydrogen and carbon
dioxide:

             CxHyOz + (2x - z)H20 4 xC02 +  (y/2) + (2x - z)H2 .       (10)

    Hydrogen has received much attention during recent years as a poten-
tial universal fuel.I'-'9  Pyrolysis of solid wastes may provide an
ideal method for the production of hydrogen.  Alternatively, the effluent
gases may be used to produce methanol:

                        C02 + 3H2 •*• CH3OH + H20 .                      (11)

    Reaction 11 is discussed in Reference 20.  Intensive studies of
methanol as a gasoline additive are being conducted at several
institutions. 21-23

    Table 14 gives the average composition of municipal refuse, and this
composition can be used to calculate the hydrogen production potential
by using Equation 10.  For each kilogram of organic waste converted to
hydrogen by equation 10, 1.2 kg of water are required.  Thus the high
moisture content of organic matter is advantageous for the proposed
process.  Using the higher heating value of hydrogen, one ton of munici-
pal wastes can produce 7.17 x 10^ scf of hydrogen with a heating value
of 25 x 10^ Btu.  This hydrogen can be reacted with the carbon  dioxide
present in the gas stream to produce 2110 pounds of methanol with a
heating value of 20.6 x 10° Btu.  The 129 million tons of municipal
refuse produced in this country during 1971'4 could have been used to
generate 9.25 x 10^2 scf of hydrogen with a heating value of 3.3 x 10'5
Btu, roughly 15% of the nation's natural gas demand that year.   Alterna-
tively, methanol production from the refuse could have yielded 1.36 x 108
tons of methanol with a heating value of 2.67 x 1015 Btu, approximately
12% of our domestic crude oil production during 1971.  Tables 15 and 16
present the synthetic fuel production potential of all the solid wastes
generated in the U.S.A.  during 1971.  Clearly, solid wastes represent
a significant national resource.

    Equation 10 suggests that solid wastes are easily converted to
hydrogen and carbon dioxide.  Unfortunately, this is not the case.  When
pyrolyzed in an oxygen-free atmosphere, solid wastes decompose into a
variety of gases, tars, oils, liquors, and char.13  Theoretically, the
gaseous and liquid pyrolysis products can be cracked in the presence of
                                    45

-------
          TABLE 14.  AVERAGE ANALYSIS OF RAW MUNICIPAL REFUSE
                                     Ultimate per cent by weight
       Constituent	As received	Dry
Hydrogen
Carbon
Nitrogen
Oxygen
Sulfur
Ash
Btu per pound
Available Btu per ton
8.2
27.2
0.7
56.8
0.1
7.0
4827 f
9.65 x 106
6.0
47.6
1.2
32.9
0.3
12.0
8546 f
17.1 x 106

TABLE 15.  HYDROGEN PRODUCTION POTENTIAL OF WASTES PRODUCED IN THE USA
                                (1971)



Manure
Urban refuse
Logging and wood manu-
facturing residues
Agricultural crops and
food wastes
Industrial wastes
Municipal sewage solids
Miscellaneous
Totals
%U.S. natural gas demand
Wastes
generated
(106 tons)
200
129

55

390
44
12
50
880
(1971)
H2
(1015 Btu)
6.4
3.3

1.2

8.4
1.2
0.4
1.1
22.0
96
Readily
collectible
(106 tons)
26.0
71.0

5.0

22.6
5.2
1.5
5.0
136.3

H2
(1015 Btu)
0.83
1.80

0.11

0.49
0.14
0.05
0.11
3.53

                                   46

-------
 TABLE 16.  METHANOL PRODUCTION POTENTIAL OF WASTES PRODUCED IN THE USA
 	(1971)	
                             Wastes                 Readily
                           generated    Methanol  collectible  Methanol
                           (106 tons)   (1Q15 Btu) Q06 tons)  (1015 Btu)
Manure
Urban refuse
200
129
5.2'
2.7
26.0
71.0
0.68
1.48
Logging and wood manu-
 facturing residues             55         1.0         5.0         0.09
Agricultural crops and
 food wastes                   390         6.9        22.6         0.40
Industrial wastes               44         1.0         5.2         0.11
Municipal sewage solids         12         0.3         1.5         0.04
Miscellaneous                   50         0.9         5.0         0.09
Totals                         880        18.0       136.3         2.89
% U.S. oil demand                          58                      9.3
                                   47

-------
steam and catalysts at moderately high temperatures to yield hydrogen
and carbon dioxide.  Experimental verification of this cracking process
was recently given by J.L. Cox et a!.'6 at the University of Wyoming.
An experiment conducted at LASL to study steam pyrolysis of solid wastes
is described in Section VI of this report.

    As noted in Section III, the char that remains after pyrolysis of
the wastes can also be converted to hydrogen and carbon dioxide via
the Reactions 1-3.

    Reactions 1 and 2 are usually observed at temperatures above 800 C;
however, catalysts have been  used to significantly reduce this tempera-
ture.16,24  Equilibrium constants for the steam gasification reaction
(as calculated from thermodynamic data) were given above in Table 1.
Our progress in locating suitable catalysts for the practice of Reactions
1 and 2 is described elsewhere in this report.  Because approximately
20% of the hydrogen production potential of solid wastes is due to
Reactions 1-3, the identification of catalysts suited to the practice
of these reactions is an important task.

SYSTEMS FOR THE PRODUCTION OF SYNTHETIC FUEL FROM SOLID WASTES

    The following paragraphs discuss systems using steam as a reactant
for the production of synthetic fuels from solid wastes.  The conclusions
are equally applicable for carbon dioxide as a reactant.  Because the
gasification reactions are endothermic, a source of heat is required to
drive the reactions.  This heat can be supplied either by the sensible
heat of the steam or by the use of a heat exchanger to deliver heat
directly to the reaction zone.  The first method corresponds to adiabatic
conditions, and the second to isothermal conditions.  The following
paragraphs present an analyis of these two methods.

Adiabatic Reactor Pisign

    In the adiabatic case, steam is superheated by a solar furnace (or
some other heat source) and pumped through the chemical reactor.  Thus,
the steam serves as both a heat carrier and a reactant.  Characteristics
of solar furnaces place an upper bound on the steam outlet temperature
of about 700 C.  Higher temperatures are achievable, but cause severe
materials problems.  To provide the heat for Reaction 1 the steam will
drop in temperature by an amount AT.  To ensure that the char is gasi-
fied at the required rate the following relationship must be satisfied:

     Molar rate of steam flow
    	  x (% conversion at [700 - AT]) > 1.     (12)
     Molar rate of carbon flow

This assumes that sufficiently good catalysts are located so that  the
reaction proceeds to equilibrium.  Relation 12 puts an effective lower
                                   48

-------
bound on AT.  The magnitude of AT determines the amount of steam required
to provide the heat of reaction.

    A computer program was written to do heat and mass balances for com-
bined system.  A description and listing of this code are available on
request.  Representative results are given in Table 17.  The sto-
chiometric steam multiple  (SSM) is defined as

              CCM   ^ of  steam/kg of dry organic waste               ,  .
              SSM = 	^^   ,          (13)

                    1.2 kg steam/kg of dry organic waste

where the denominator represents the stoichiometric quantity of steam
required to gasify a kilogram of dry organic waste.  The initial steam
temperature and the amount of char residue remaining after pyrolysis
are seen to be critical parameters of the system.  If the char percentage
can be kept below 10%, or if steam should be available from the solar
collector at temperatures exceeding 700 C, the adiabatic system could
prove to be practical.  Certainly this is the desired conclusion, since
an adiabatic reactor would easily combine with present-day solar furnace
designs.

    Some mention should be made of the uncertainties present in the
calculations summarized in Table 17.  Only the average specific heat
of organic solid wastes was used to calculate the heat required for
pyrolysis.  Pyrolysis heats of reaction as a function of temperature
are not available, and neglect of these heats could change the values
of Tf is much greater than room temperature.

  TABLE 17.  MASS AND HEAT BALANCE CALCULATIONS FOR THE ADIABATIC SYSTEM

% Char
SSM

Ti
Temperature, C
Tr

Tf
    10                6             650              599             502
    10                3             700              597             429
    20               10             650              587             527
    20                6             700              597             501

SSM = stoichiometric steam multiple
 Ti = initial steam temperature
 Tc = temperature of gaseous products leaving char gasification zone of
      reactor
 T  = temperature of gaseous products leaving pyrolysis zone of reactor
                                    49

-------
 Isothermal  Reactor Design

     A significant fraction of systems study effort was devoted to the
 design of a  practical  chemical  reactor to be located at the focus of
 a tower top solar furnace.  This reactor design attempts to make use of the
 unique properties of a  solar furnace (which supplies all its heat in the
 form of radiant light energy) to simplify the intrinsic heat transfer
 problems and increase efficiency.  Although no tower top solar furnaces
 presently exist in this country, several design studies have been made.
 ERDA is presently planning for the construction of a 5 MWt^ test facil-
 ity and a 30 MWe production facility.  These two facilities are expected
 to be placed in operation during the next five years.  The White Sands
 Proving Ground in New Mexico has an experimental  solar furnace in opera-
 tion that could be employed for  design  studies until  the 5 MWth test
 facility becomes operational. Thus ERDA's emphasis on the  development
 of solar process heat technologies points to the timeliness of this
 research.

 Heliostat Economics
     Serious  design  studies  of a  tower  top  solar  furnace  have been made  by
 McDonnell  Douglas Astronautics Corporation and the University  of Houston. 25
 A review of  their economic  projections is  helpful in developing a better
 understanding  of the  constraints imposed upon a  chemical  reactor's  de-
 sign by the  solar furnace.  A heliostat consists of a  19 m2 back-silvered
 flat glass mirror supported by a welded steel frame and  a universal mount
 resting on a pedestal  of  reinforced concrete.  It is estimated that the
 heliostats comprising  the concentrator subsystem of the  solar  furnace will
 represent  85%  of the  total  energy collection system cost.  Therefore,
 it is essential  that  the  costs of the  heliostat  be minimized.  Since
 McDonnell  Douglas has  already worked on minimizing these costs, we  will use
 their heliostat  design  in this study.

     For a  hexagonal heliostat with a span  of 4.6 m and an  area of 19 m2,
 the  reflector  subassemby  cost is extimated to be $246.   This estimate
 includes handling,  assembly,  and installation costs.  The  total cost of
 the  heliostat  mounting, including guidance, tracking, and  control mech-
 anisms,  is projected to be  $345.  The  component  costs of this  subassem-
 bly  are:

             Electronics  and  sensors           $ 67
             Mounting and activators             104
             Pedestal                            95
             Handling, assembly  and installation 79
                                     TOTAL
    Other programmatics are expected to add 10% to these costs, bringing
the total cost to $650 per heliostat, or $34/m2.  More recent  studies
have suggested somewhat higher costs; however the figures given above  are
sufficiently accurate for our purpose.
                                    50

-------
    Because the size of the mirror determines the size of the solar
image at the focus of the solar furnace, and the size of this image
determines the size of the chemical reactor, it is of interest to under-
stand how the heliostat costs vary with mirror size.  We make the pessi-
mistic assumption that there would be no decrease in the cost of the
mounting subassembly for a smaller heliostat mirror.  With this assumption,
the cost per square meter for a heliostat with mirror span SH (in meters)
is given by
                       „                    18.16
                 cost/mz ($) = 1.1  12.95 +	               (14)
                                            $2/21.16
                                             H

    This formula illustrates the simple fact that if the heliostat mirror
had an area of 9.5 m2 rather than 19 m2, the cost per unit area would
Increase 60% due to the additional mountings required to achieve the
same area coverage (hence the same power output).  Thus it makes eco-
nomic sense to place as large a mirror as possible on each mounting.
Factors mitigating against the use of very large mirrors will be dis-
cussed in the following paragraphs.  Heliostat economics of scale are
illustrated in Table 18.

Image Size

    The chemical reactor should be located at the focus of the solar
furnace to facilitate transfer of heat to the reaction zone.   Thus the
chemical reactor functions as a heat exchanger in addition to its more
obvious role, and its size must be at least as large as the solar image
formed at the focus of the solar furnace.  The following considerations
show how this size limitation places further constraints on the design
of the chemical reactor.

                  TABLE 18.  HELIOSTAT ECONOMIES OF SCALE
             Heliostat mirror                 Heliostat cost,3
               Span, meters	$/m2	

                    1.0                             437
                    2.0                             120
                    3.0                              61
                    4.0                              41
                    5.0                              31
                    6.0                              26
                    7,0                              23
                    8.0                              21
             a
              Cost projections are pessimistic for spans less
              than 5 ITU optimistic for spans greater than 5 m.
                                     51

-------
    Relatively straightforward optical laws determine the diameter of
the sun's image DT to be
                                          2 ft
                           D! = (« + o.) 	+ SH  ,                 (15)

                                        cos 0

where a is the angle subtended by the solar disc, a is a measure of the
total optical errors associated with the beam reflection by the helio-
stat, k is the tower height, and 0 is the rim angle of the heliostat
field from the top of the tower.  This formula assumes the flat plate
heliostat mirror design studied by the University of Houston and the
McDonnell Douglas Corporation.  Smaller images are obtained using warped
mirror designs currently being developed by Martin Marietta Corporation
and the University of Georgia.  For these mirrors Djar SH-  For present
purposes the  values of a, a, and 0 are assumed to be 0.009 rad, 0.004
rad, and 63.4°, respectively.  These are the same values used by the
University of Houston and the McDonnell Douglas Corporation.  Substitut-
ing these values into Equation 15, we have

                          D! = 0.0581 k + SH     .                    (16)

    If a land area A£ is needed to collect the desired amount of solar
radiation, then the height of the tower required by the solar furnace is
given by
                           _, / A//TT   _  / A/
                          h^tan 0    '   3.54             '          (17)

Thus the height of the tower is determined by the power requirement
imposed on the solar furnace, and the only remaining "free" parameter
in Equation 15 is the span S^ of the heliostat's mirror.  But the eco-
nomics of scale discussed in the paragraphs above limit this span to 4 m
or larger; hence the image diameter is effectively determined as

                               0.0581 fc + 4   ,                       (18)

where h is given by Equation 17.

    Some reduction in the size of DI may be possible with more effort to
optimize the  design of the furnace.  The somewhat pessimistic value of
a chosen for  this analysis may be reduced as the heliostat guidance
system is improved.  The rim angle 0 cannot be made much larger because
of shading effects at the far edge of the mirror field, but some im-
provements may occur.  The analysis presented here is felt to represent
a realistic estimate of the solar image diameter as a function of the
tower height.

Reactor. Design

    One potential  design for an isothermal reactor uses a quartz "window"
                                     52

-------
in the chemical reactor to allow the light to pass directly into the
reaction zone where it is absorbed by the char and converted to thermal
energy.  Since char alproximates a perfect black body — it has been used
as a source of lampblack — it is an excellent absorber of solar energy.
A fluidized-bed chemical reactor is contemplated due to its excellent
heat transfer characteristics.  Two reactor geometries are described
here; more study will be required to determine the optimum design

    The more obvious design, given in Figure 24, uses a cylindrical reac-
tor surrounded by a cylindrical jacket with quartz windows at the bottom
of both cylinders.  The quartz windows are not intended to be a single
solid piece of quartz; rather they are expected to be many quartz plates
held together by a metal honeycomb structure.  The jacket window is air
tight under the modest pressures generated by the compressor.  The reactor
window however, is perforated to allow the gas reactant (H2<3 or CC^) to
fluidize the bed of char and solid wastes.  Focused sunlight passing through
both windows heats the fluidized bed and enables the pyrolysis and char
gasification reactions to proceed.

    The most significant drawback of this design is that the diameter DR of
the reactor is larger than DI and is given by
                                 DT
                         DR =     L     = 2.23 Dj   .                 (19)
                                 cos 0
The surface area SR is then given by
                         SR =   (DR/2)2 = 3.91 D2 .                    (20)
    Because the heat loss due to conduction and radiation is directly pro-
portional to the surface area, this design is potentially less efficient
than some others.  This design also requires more quartz than other de-
signs, and could be more expensive.  In spite of these disadvantages, the
design has many attractive features.  Conductive heat loss is minimized
by the gaseous layer separating the reactor from the jacket.  Radiative heat
loss is reduced by the absorptivity of quartz, steam, C02, CO, etc. in the
infrared.  The reactive bed is levitated above the quartz window, decreas-
ing the likelihood of chemical reactions between the quartz and the react-
ants that could cloud the quartz.  However, the reactivity of quartz with
H20, C02, CO, H2, Cfy, etc. at temperatures of  650 C needs further
investigation.

    A second design, intended to minimize the surface area of the reactor,
has also been studied.  This design is based on a hemispherical reactor
geometry, as shown in Figure 25.  Light enters the reactive bed through
the walls of the reactor as well as its bottom surface.  If we define
R! = Dj/2, then
                       r = RI sin 0
                       h = RI (1 - cos0)
                       y = ir - 0                                       (21)
                           2
                       H = Rj sin0 (2 - sin©)
                       r'= r + H sin "?


                                     53

-------
                      Steam
        C02 or H20 inlet
~l
      Heat exchanger-
en
      H20 C02  removal-
        V
                             IT
                               *
                         H2, CO and/or
                              CH
                 Storage--
                         Industrial or
                        commercial use
                                              Solid organic wastes
Fluidized
   bed
 reactor
                             •DR-
                             •DJ-
                                                                    - Refuse hopper

                                                                    •Seal

                                                                    Airlock
                 Seal
                                    \
                           Sunlight
                               »
                        [Solar  furnace
                                                                          Compressor
                                                                              Recirculates
                                                                               steam +
                                                                              pyrolysis gas
                   -Reactor wall

                   — Jacket wall
                   —Quartz window with
                       perforations
                   x— Quartz window
                                    Figure 24.   First reactor design.

-------
Figure 24.  Second reactor geometry
                 55

-------
 so that  the  total surface area of the reactor is given by

           SR = 2 R|(!  - cos e) + ir(r + r')/H2 + (r - r')2   .         (22)

 For  e  =  63.4° this becomes

                        SR = 11.06 R2 = 2.77 D2    ,                   (23)

 representing a 29% decrease in area over the previous case.  The volume
 of this  reactor is given by

             VR = ?irR3(l - cos e) + ^(r2 + rr' + r'2)
                  31              §
     This reactor design has most of the advantages attributed to the  pre-
 ceding design and has a smaller surface area.  However, it also has
 several  distinct disadvantages.  The construction of a hemispherical
 honeycomb of quartz windows does not appear to be particularly easy.
 The  reactive bed will now be in direct contact with quartz, causing
 abrasion on  the window  surface and clouding the quartz.  Finally, the
 reactor  has  to be full  of char at all times, which will be shown to
 be disadvantageous.

     Variations of these two designs are also possible.  The inner window
 of quartz could be replaced by a metal wall with a honeycomb radiant
 energy absorber attached to it (see Reference 26).  In this case the
 radiant  energy would be converted to thermal energy by the absorber and
 conducted through the metal wall.  This appears to be a less desirable
 method of heat transfer, but may be necessary if the quartz proves to
 be reactive  under working conditions.

     The  efficiency of this system will be determined primarily by the
 amount of heat lost from the reactor due to limitations in the reactor's
 design.   Conduction and radiation account for the major losses, with
 other  effects (such as  reflection from the quartz windows) contributing to
 a lesser extent.  Estimates of these effects are presented here to provide
 a better understanding of the system's strengths and limitations.

     To estimate conductive heat losses consider two quartz windows of
 thickness Wq separated by a layer of superheated steam at approximately
 atmospheric  pressure of thickness Ws (see Figure 26).  The inner surface
 of the second quartz window is at a temperature of 15 C.  Quartz has  a
 thermal  conductivity of 1.4 W/m-K and superheated steam a conductivity
 of   0.05 W/m-K at 300 C; therefore an estimate of the total conductive
 heat loss is given by

                        qc - 	635	  w/ 2    .             (24)
                              2Wq/1.4 + Ws/0.05    '

     For a reactor with WQ = 0.01 m and W  = 1.0 m, the conductive heat
 loss per square meter isM31.7 W/m2.  Heat loss through the sides and  top
of the reactor are made negligible by insulation.
                                    56

-------
1
Quartz V\l
Steam \
Quartz \
\
q
YS
Vq
Figure 26.  Schematic of window assembly.
                    57

-------
    If the quartz, steam, and other gases were transparent in the infrared,
the heat loss due to radiation would be given by

                       qr = Fo(T + 273)4 W/m2   ,                      (25)

where a is the Stefan-Boltzmann  constant and the angle factor F is
given by

                       F = ! (Z - / Z2 _ 4X2Y2   )   .                  (26)

Here X = D,/2W , Y = 2W /DR, and Z = 1 + (1  + X2)Y2.  Table 19 illus-
trates the effect of an increase in the reactor's temperature on radiant
heat loss.  Clearly there is a significant advantage in  practicing the
char gasification reaction at lower temperatures.  Since the quartz, steam
and other reactant gases are not transparent in the infrared and F   1,
the numbers in Table 19 significantly overestimate the actual heat loss
from the reactor.

     TABLE 19.  EFFECT OF TEMPERATURE ON RADIANT HEAT LOSS BY REACTOR


                   T.  C	Heat loss. W/m2
550
600
650
700
750
26.0 x 103
32.9 x 103
41.2 x 103
50.8 x 103
62.1 x 103
    A fraction of the incident light from the solar furnace will reflect
off the quartz window rather than pass through it.  Using Fresnel's equa-
tions, this loss is estimated to be 4% for light at normal incidence and
10% for light at an incidence angle of 63.4 , for an average loss of  6%.
Quartz also becomes less transparent to visible light at higher tempera-
tures, indicating that a non-negligible amount of the radiant energy
would be absorbed by the quartz rather than pass through it.

    Two solid waste gasification facilities are studied in the following
paragraphs:  one sized for a tower top solar furnace with a heliostat
field radius of 100 m and one with a radius of 200 m, covering land
areas of 3.14 x 104 m2 and 1.26 x lO^m2, respectively.  The first
yields approximately 7.5 MW^ and the second yields 30MWtn/  A 7.5
MW^n facility would be required to gasify the solid wastes of Los Alamos,
a community of some 17,000 residents.  The larger 30 MWth furnace has
been studied by scientists at the University of Houston.

    Using Equation 17, the tower height of the 7.5 MWth furnace is 50m,
                                    58

-------
and that of the 30 MWth furnace is 100 m.  The image diameter of the
first furnace is determined by Equation 16 to be 7.5 m, and that of the
second to be 10.4 m.  A chemical reactor sized to the 7.5 MWth facility
would reguire a quartz window surface area of 220 m2 and the second an area
of 423 m*.  Using these figures, heat loss from the two reactors is easily
calculated.  Comparing the results of Equation 24 with those of Equation
25, it is clear that conduction plays a negligible role in heat loss from
the system.  This fact illustrates one strength of the jacketed reactor
design.  Radiant heat loss from the smaller reactor is 8.84 x 106 W at 650 C
(assuming no absorption by the gases and quartz) and 16.9 x 10° W for the
larger reactor.  The smaller reactor loses 0.6 x 106 W due to reflection
off the quartz window, and the larger reflector loses 2.4 x 106 W by this
mechanism.  With these figures, the smaller system has a theoretical
efficiency of only 12% and the larger system has a theoretical efficiency
of 56%.  Fortunately, these figures are not realistic due to the large
absorption of infrared radiation by the quartz, steam, and the other
gases present in the reactor.  Thus the chemical reactor would be con-
siderably more efficient than indicated here.  A quantitative estimate
of these effects will be made later in this section.

    It is also of interest to determine if the kinetics of the carbon-steam
reaction are favorable at 650 C.  The reaction cpnsumes 0.038 kW-hr/mole
(32 kcal/mole) of heat at 650 C.  Assuming the 30 MW^u solar furnace
yields 108 MW-hr of usable 'heat during a 6-hour day (60% efficiency),
2.86 x 106 moles of C must be consumed each day, or 0.476 x 106 moles/hr.
Using a typical reaction rate of 0.09 mole/hr/gcat as determined by experi-
ments described above for synthetic char, the reactor must contain approx-
imately 5000 kg of catalyst and 45,000 kg of char to consume the solar
furnace's heat and produce .synthesis gas.  Using the Garrett char density
of 1690 kg/nr, the required amount of char would have a volume of 26.6 rip
and would then fill the cylindrical reactor to a depth of 0.063 m.  The
pressure exerted on the quartz window by the cayalyst and char in the
reactor would be 1210 nt/m2 (25.4 Ib/ft2).

    It should be emphasized that the preceding calculation is intended
only to indicate the feasibility of the design presently being considered.
As presented, the calculation must be qualified: 1) The amount of char
indicated is an overestimate.  Some of the furnace's heat would be used
to pyrolyze the solid wastes and provide char for the carbon dioxide -
carbon reaction;  2) the reaction rate used was obtained from a fixed-
bed reactor.  Data from a fluidized-bed reactor are required to make a
more realistic calculation.

    The preceding calculation also points out a major weakness of the
"spherical" reactor design pictured in Figure 25.  This reactor needs to
be full of char for optimum transfer of heat to the reactive bed.  How-
ever, a reactor of this design sized to the 30 MWth solar furnace would  have
a volume of over 900 m3, much larger than that required by the reaction
kinetics.  For this reason it seems clear that a design based on a
cylindrical reactor geometry is preferable.
                                      59

-------
    Further study of the double quartz window reactor design reveals certain
difficulties which could limit its practicality.  At high temperatures
(such as those encountered on the surface of the inner window) quartz
becomes less transparent and more reactive.  Without experimental data it
is difficult to predict the outcome of these effects on the reactor's
operation.  Nevertheless, it seemed wise to consider another design
patterned on a cavity absorber suggested by W. R. Powell.27  The following
paragraphs describe this new design in some detail.

    A brief discussion of the tubular absorber proposed by Powell will be
helpful in understanding the reactor described here.  As shown in Figure 27
focused sunlight enters the open face of a long quartz tube whose inner
surface is cbated with a reflective silver film.  The light reflects many
times down the length of the tube before it is eventually absorbed.

    Although the reflectivity p of silver is quite high (> 0.09), p11 is
small when the number of reflections n is large enough.  The tube loses
heat by conduction and infrared radiation; however much of this radiation
is absorbed by the quartz tube, so that little escapes out the open face.
The reader is referred to Reference 27 for a more detailed description
of Powell's invention.  The basic thought behind his idea is the use of a
reflective surface to achieve absorption of focused sunlight by a great
many reflections down the length of the tube.

    A chemical reactor having properties similar to the cavity-type
absorber may be described as follows.  Consider a group of nested annular
fluidized-bed reactors as pictured in Figure 28.  Each reactor is construct-
ed of a suitable metal with high reflectivity (polished stainless steel for
example).  Light entering through the open end of the reactor unit is
trapped in the cavities separating the annular reactors.  As in Powell's
design, the light reflects back and forth up the length of the annular
opening until it is eventually absorbed on the face of one of the reactor's
walls.  Thus the annular cavities separating the annular reactors play
the same role as the tubular cavity in Powell's design.  The reactor unit
is jacketed as before and has a single quartz window maintained at a rel
relatively low temperature to admit light and retain the reactant gases.

    Light enters the reactor with angles (measured with reference to the
reactor's cylindrical axis) varying between 0° and 63.4°.  If the angle 6
were too large, much of the light incident on the reactor would reflect
out without being converted to heat.  On the other hand, if 0 is very small
the annular reactors have negligible volume.  Clearly an optimum angle
exists which maximized the annular reactor's volume without sacrificing
incident sunlight.   A computer program written to determine the optimum
angle is described in Appendix A.  For a set of annular reactors of radii
R.J given by R,-  = 0.35 m + i(0.7 m), the optimum angle e is 1.26° and the
optimum height of the annular cone is 1.45 m.  The index i runs outward
from the center reactor.   If the total height of the reactor unit is 10 m,
then the ratio of each annular cavity's height to its width is ^ 15.
                                     60

-------
    COOLANT
   ENTRANCE
SILVER
 FILM
  I	
GLASS
 TUBE
        I
    COOLANT
    ENTRANCE
INSULATION
                       EXHAUST
                          INSULATION
                 Figure 27. Schematic of a tubular absorber,

-------
                                               H
C)
z
a:
P
o
UJ
tr

«
n
V
o
|
0
tr
_j
z
z



1




ro
n
V
i^







k k- R




eg
n
V












n
V







1 R, "











_
n
V
m.1
"1
R, '
                                          CVi
                                         V
 ro
«•
o
>-
r-
5
0
(T
<
_l
Z
<


*•
6
Z
(E
O
1-
UJ
(T


«•
n
V    V
Figure 28.  Schematic of a nested annular fluidized-bed reactor,
                                62

-------
Powell describes  this  ratio as  adequate  to  effectively trap radiation
incident  on  the cavity's  opening.

    The need to retain all  the  radiation incident  on the reactor's lower
surface severely  restricts  the  width  of  each  annular reactor and thus
limits the total  volume of  the  reactor unit.  The  size of the solar image
at the focus of a 30 MWth tower top solar furnace  was shown to be
23.2 m for the McDonnell  Douglas design. If  the spacing of the annular
reactors, R-j, is  as  given above, then 17 nested reactors are needed to
cover the image.   The  radius R, surface  area  A, and volume V of each of
the annular  reactors are  given  in  Table  20, together with sums for the
entire unit. The total volume  of the unit  is seen to be 300.2 m3, which
is more than adequate  to  contain the  26.6 m3  of char required to consume
the furnace's 30  MWth  of  heat,  taking a  catalyzed  gasification rate of
0.09 moles/hr/gcat and a  carbon/catalyst ratio of  10.

    If the reactor's walls  are  constructed  of 1/4" stainless steel, the
reactor would weigh  6,400 tons  and the char reactant would weigh 50 tons.
These numbers are based on  the  gasification rate obtained for the KoCOo
catalyzed C  + ^0 reaction  and  the McDonnell  Douglas solar furnace design.
Clearly the  reactor  is much too large for the amount of char required.
The size  of  the reactor is  determined by the  diameter of the solar image,
indicating the desirability of  using  focusing heliostats (such as those
being developed by Martin Marietta) to reduce the  image size.  Other
modifications to  present  solar  furnace designs may also be required to
optimize  the total system for synthetic  fuel  production.

    It is also of interest  to determine  the temperature drop across the wall
of each annular reactor.  The reactor unit  has a total area of 12,770 m2
(137,000  ft2).  If the wall is  constructed  of 1/4" stainless steel, a
temperature  drop  across the reactor wall of 0.46 C is required to trans-
mit 18 MW^h  in*0  tne reaction zone if it is assumed that the reactor
operates  at  60% efficiency.  Thus  the reactor's large surface area per-
mits the  transfer of large  quantities of energy with a negligible drop in
temperature.

    To limit radiation losses it is desirable to maintain the open end of
the reactor  unit  at  the lowest  feasible  operating  temperature.  This is prob-
ably best accomplished by introducing the "cold" reactant gas (steam, C02,
or some mixture thereof)  into the  bottom of the reactor.  Assuming the
reactor is used to gasify solid wastes (rather than just the char), the
pyrolysis reactions  could be practiced in the lower part of the reactor.
Elutriation  of the fine char particles could  provide a mechanism for their
movement to  the hotter portions of the reactor above the pyrolysis zone.

    It would be difficult,  if not  impossible, to fluidize a bed with a
height to width ratio  of  over 100.  Clearly the need is to incorporate sev-
eral fluidized beds, one  above  another,  in  each annular reactor.  If a
temperature  gradient exists along  the length  of the reactor, as was shown
to be desirable in the preceding paragraph, the use of stacked reactors
would provide a convenient  means for  practicing the various pyrolysis and
                                      63

-------
gasification reactions.
     It is to be emphasized that the reactor geometry presented in this
 section has not been fully optimized.  More data on the kinetics of char
 gasification are required to initiate a truly meaningful design effort.
 In addition, the intensity distribution of radiation over the solar
 image is needed to establish the proper location for each annular reactor.
 Determining the efficiency of this reactor design by estimating its
 radiant heat loss is an extremely difficult problem.  Powell's calcula-
 tions indicate that his tubular absorber can achieve efficiencies of 90%
 or better.  The purpose of this work was to establish the feasibility of
 such a reactor and lay the groundwork for a more detailed future study.

     It is probably desirable to isolate the chemical reactor from the
 environment by a glass (or quartz) window.  This window lowers thermal
 losses by conduction, convection, and by retaining a gas mixture of C02
 and  ^0 that is strongly absorptive in the infrared.  In addition, the
 glass window also absorbs strongly in the infrared.  These beneficial
 effects are partially counteracted by radiation losses due to reflection
 off  the two surfaces of the window, and absorption of radiation by the
 glass.  The result of these loss mechanisms is a decrease in the reactor's
 efficiency, which is traded off against the beneficial effects mentioned
 earlier.  The following paragraphs provide an estimate of the magnitude
 of these loss mechanisms which can be used to determine the desirability
 of a windowed reactor versus an open reactor.  Neglecting absorption, the
 transmittance of a single window xr i is given by Tr •] = (1 - p)/(l  + p),
 where   is the reflectance of a single glass sur ace!  For small to moderate
 angles of incidence e-j   FresneVs equations can be used to determine the
 reflectance p:         '
                           p =
                                     sin2(e2 - ei) + tan 2 (92 - el)
                                                     tan 2 (e2
                                                                      (27)
where ni/n2 = sin 02/sin e-j, n-j is the refractive index of air (n-|=l),
and n2 is the refractive index of glass (n2 = 1.526).  With a mirror
field with a rim angle of 63.4°. the average angle of incidence is 45°
for approximately equal mirror areas inside and outside the 45° cone.
The reflectance P for 45° incidence is found to be p = 0.054 and the
transmittance is then rrj-|  = 0.8975.  The transmittance due to absorp-
tion losses is given by ta= e~KL, where K is the extinction coefficient
for glass and L is the window thickness.  For a good quality clear glass
K = 0.04 cm~' and, assuming a window thickness of 5 mm, the transmittance
Ta becomes 0.972.  Light absorbed by the glass is turned into heat.  With
a radiation flux of 30 Mw over a window area of 423 m2, the heating
rate^of the window is 1.99 kW/m2.  The window radiates energy at a rate
           - T;
              surroundings
), where e-^0.94.  Using this expression, a
                                     64

-------
window temperature  of  105  C  is  required to radiate away the energy re-
ceived by absorption.

    The total transmittance  of  a single window is given by T = xr IT_.
For the conditions  assumed here, T = 0.873.  Thus the inclusion of the
glass window reduces the reactor's overall efficiency by some 12.7%.  To
be worthwhile,  the  window  must  reduce other thermal losses, and thereby
increase the reactor's overall  efficiency by at least this amount.

    Conduction  losses  are  minimized by surrounding the outer annular
reactor with insulation.   Assuming the temperature of the outer surface
to be 500 C, the  insulation  to  have a conductance of 0.02 Btu/hr-ft2.°F/ft
(glass wool) and  a  thickness of 0.7 m, the conduction loss is~ 1.6 kW
from the reactor.   This trivial amount indicates the importance of
minimizing radiation losses.

    Estimating  radiation losses from a geometry as complex as the
proposed design is  beyond  the scope of this report.  However, some in-
sight into the  absorption  properties of gaseous H20 and an H20.C02 mix-
ture can be gained  by  an analysis of a much simpler problem.  Consider two
flat, black-body  radiators maintained at temperatures T] and Tg, where
T2   T].  The two radiators  are separated by a region filled with gaseous
H20, or a mixture of H20 and C02.  If the gaseous H20 and C02 did not
absorb radiation, the  rate of heat transfer from radiator 2 to radiator 1
would be given  by

                               o(Tj - if)                           (28)

    Assuming radiative equilibrium exists, thereby neglecting conduction
and convection mechanisms, we may use the band approximation^ to estimate
the actual rate of  heat transfer.  The calculation is somewhat complex
and is presented  in Appendix B.  The results are impressive:  a one-
meter thickness of  gaseous J^O  at one atmosphere pressure absorbs 26%
of the radiation  flowing from radiator 2 to radiator 1 (To = 650 C,
Ti = 100 C).  For a mixture  of  gaseous ^0 and C02, over b9% of the radi-
ation is absorbed.  Clearly  the inclusion of these gases in the reactor's
design as a radiative  insulator could result in significantly improved
performance.

ECONOMICS OF THE  PRODUCTION  OF  SYNTHETIC FUELS FROM SOLID WASTES

    In view of the  relative  lack of information on the economics of solar
process heat, it  is difficult to provide a realistic estimate of the
proposed system.  Nevertheless, it is a worthwhile exercise because it
reveals what additional knowledge is required to provide such an estimate.
The following study presents the economics of a system scaled to the needs
of Los Alamos, a  community of some 17,000 residents.  Refuse collection
costs are based on  actual  experience.  If the residents conform to the
national  average  and discard 7  Ibs of refuse per day,14 50% of which is
organic,  the community must  dispose of 29.75 tons of organic solid wastes
                                    65

-------
per day.*  Presently this waste is disposed in a sanitary landfill.   We
now consider the possibility of using this waste to produce  hydrogen
for use as a fuel or chemical feedstock.

    The average yearly insolation a heliostat is expected to receive  in
New Mexico is 0.71 kWtn/m2.   Collector efficiencies are projected  to  be
                     hence
 approximately 78%;29 hence the power tower yields o.55 kW^h/ip2 of heliostat
 mirror  area.  Gasification of 30 tons of organic wastes is estimated to
 consume 1.8 x IQll j of heat over a 6-hour period and to require a fur-
 nace with 1.5 x 104 m2 of heliostats.  The furnace and associated facili-
 ties would cover about 8 acres of land, which is smaller than the present
 sanitary landfill.  Assuming that the heliostats cost $34.00/m2,30 the
 heliostat array is projected to cost $525 x 103.  If the tower would cost
 $345 x  103 and the furnace's total cost would be $870 x 103.

    Battelle32 estemates the cost of a pyrolysis unit and associated
 support facilities for a community this size to be $330 x 103.  Compressors
 to bring the gas mixture to 275 psia cost $400 x 103, and two shift
 reactors to remove the CO from the gas stream (if required) cost $600 x 103.
 Finally, C02 is removed by a hot potassium carbonate scrubber costing
 $1.4 x  106.  Some form of low pressure storage will be required to
 allow the chemical processing facility to operate 24 hours per day (rather
 than six);however, the cost of such a facility is not available.

    Los Alamos probably spends $56 x 103 on labor and $30 x 103 on equip-
 ment to collect its refuse per year.  It is standard procedure to have
 two men present at a hydrogen processing facility at all times, increasing
 the labor cost to $156 x 103.  Maintenance at 5% per year would cost $180 x
 103, and the compressor's electrical consumption is estimated to cost an
 addition $53 x 103.

    To  cover the project's capital costs, we assume the community would
 issue $3.6 x 106 of municipal bonds at 7% interest with an amortization
 period  of twenty years.  Principal plus interest payments on the bonds
 would cost $334 x 103.  Communities are presently33 generating their
 own electrical power from "municipalized" utilities, so the idea is not
 un—realistic.  With revenues of $6.00 per ton for solid waste collection
 paid by the residents, the community would break even if it sold the
 236 x 109 Btu of hydrogen produced yearly for $2.90 per million Btu.  These
 results  are summarized in Table 21.  If the solar furnace should cost three
 times the amount indicated in Table 21, hydrogen would have to be sold at
 $3.84 per million Btu to break even.
*A1though 7 Ibs of refuse per person per day may appear high when compared
with other studies, we are assuming that 3.5 Ibs of organic wastes are
available per day from the residents.  Because sewage sludge could be used
to supplement the refuse collected, this is a reasonable assumption.
                                    66

-------
              TABLE 20.  DIMENSIONS OF ANNULAR REACTORS

Index No. of
Annular Reactor
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

Radius, m
0.35
1.05
1.75
2.45
3.15
3.85
4.55
5.25
5.95
6.65
7.35
8.05
8.75
9.45
10.15
10.85
11.55
TOTAL
Area, m2
44.2
132.0
220.9
309.3
397.6
486.0
574.3
662.7
751.1
839.4
927.8
1016.2
1104.5
1192.9
1281.2
1371.0
1459.0
12,770
Volume, m3
1.04
3.13
5.21
7.30
9.38
11.47
13.55
15.64
17.72
19.81
21.89
23.97
26.06
28.14
30.23
32.3
34.4
300.2

TABLE 21.  ECONOMIC ANALYSIS OF A MUNICIPAL SYNTHETIC FUEL PLANT

Captial Costs ($106)
Heiiostats
Tower
Pyrolysis Reactor and Associated Plant
Compressors
Shift Reactors
Scrubber
Operating Costs ($106)
Principal and H interest
Labor
Equipment and Maintenance
Compressor (electricity)
Revenues ($106)
Refuse Co"l "lection
Hydrogen (269 x 109 Btu with 88% availability
at $2.90/106 Btu)
0.53
0.34
0.33
0.40
0.60
1.40
3.60
0.334
0.156
0.210
0.053
0.753
0.066
0.687
0.753
TOTAL
TOTAL
TOTAL
                                   67

-------
     To some extent the foregoing represents a "worst case analysis" due
to the dis-economies of scale present for such a small community.  Larger
communities could manufacture hydrogen less expensively because equip-
ment and operating costs would be proportionally less.  Nevertheless,
some pressing technological questions remain.  Power towers generate
steam at high pressures, but little is known about solid waste pyroly-
sis in a pressurized steam atmosphere.  Production of hydrogen under
pressure could greatly reduce compressor costs, but would increase the
cost of the pyrolysis reactor.  More accurate estimates of the heat
required for pyrolysis are needed, and the costs and recoverability of
catalysts for char gasification need to be investigated.  Some features
of present power tower designs will probably have to be altered to suit
the needs of chemical processing technology.  Although some pessimistic
assumptions were made during this study, solar process heat may prove
to be much more expensive than present projections.  Nevertheless, this
analysis suggests that the application of solar process heat to fuel
production may be a technology whose time has come.
                                   68

-------
                               SECTION  VII

                                REFERENCES


 1.  Huang, C. J., and Dalton, C., "Energy Recovery from Solid Wastes,"
     Summary Report, Volume 1, and Technical Report, Volume 2, NASA
     CR-2526, April 1975.

 2.  JANAG Thermochemical Tables. M. W. Chase, Project Director, Dow
     Chemical Company, Midland, MI, looseleaf and supplements.

 3.  Marvin, G., Inorganic Synthesis. Volume II. W. C. Fernelius, Editor,
     McGraw-Hill Book Company, Inc., New York (1946), page 74.

 4.  Walker, P. L., Fusinko, F., and Austin, L. G., "Gas Reactions of
     Carbon,"  Advances in Catalysis. Volume XI. D. D. Eley, P.  W. Selwood,
     and P. W. Weisz, Editors, Academic Press, New York (1959),  page 133.
 5.  Walker, P. L., Shelef, M., and Anderson, R. A., "Catalysis  of Carbon
     Gasification,"  Chemistry and Physics of Carbon, Volume 4  ,  P.  L.
     Walker, Editor, Marcel Dekker, New York (1969), page 287.
 6.  Graven, W. M. and Long, F. 0., "Kinetics and Mechanisms of the  Two
     Opposing Reactions of the Equilibrium CO + H20 = C02 + H2," J.  Amer.
     Chem. Soc. 76_, 2602, 6421 (1954).

 7.  Tingey, G. L., "Kinetics of the Water-Gas Equilibrium Reaction.  I.
     The Reaction of Carbon Dioxide with Hydrogen," J. Phys. Chem. 70..
     1406 (1966).

 8.  Kroger, C., "The Gasification of Carbon by Air, Carbon Dioxide  and
     Steam and the Effect  of Inorganic Catalysts,"  Z. Angew.  Chem. 52_,
     129 (1939).

 9.  Villalobos, R. and Nuss, G. R.,  "Measurement of Hydrogen in Process
     Streams by Gas Chromatography, " ISA Trans. 4_, 281 (1965).

10.  Lewis, W. K., Gilliland, E. R.,  and Hipkin, H., "Carbon-Steam Reac-
     tion at Low Temperatures,"  Ind. Eng. Chem. 45_, 1697 (1953).

11.  Taylor, R. W. and Bowman, D. W., "Rate of Reaction of Steam and
     Carbon Dioxide with Chars Produced from Subbituminous Coals,"
     Lawrence Livermore Laboratory Report UCRL-52002, January 1976.
                                     69

-------
 12.   Rubin,  E.  S.,  "Research  and  Development Needs for Enhancing U. S.
      Coal  Unti1ization,"   Proc. Ninth  Intersociety Energy Conversion Eng-
      ineering Conference,  San Francisco, Calif., Inst. Electrical and
      Electronic Engineers,  New York  (1974), page 997.

 13.   Saner,  W.  S.,  Ortuglio,  C.,  Walters, J. G., and Wolfson, D. E.,
      Conversion of  Municipal  Wastes  and Industrial Refuse into Useful
      Materials  by Pyrolysis," U.S.  Bureau of Mines Report of Investiga-
      tions 7428 (1970).

 14.   Anderson,  L. L.,  "Energy Potential from Organic Wastes:  A Review
      of the  Quantities  and  Sources," U. S. Bureau of Mines Information
      Circular 8549  (1972).

 15.   Maugh,  T.  H.,  "Fuel from Wastes:  A Minor Energy Source,:  Science
      178,  599 (1972).

 16.   Cox,  J.  L.,  Hoffman,  E.  J.,  Hoffman, R. W., Willson, W. G., Roberts,
      J.  A.,  and Stinson, D. L., "Gasification of Organic Waste," Amer.
      Chem. Soc.,  Div.  Fuel  Chem., Prepr. 18(1), 1 (1973).

 17.   DeBeni,  G.  and Marchetti, C., "Hydrogen, Key to the Energy Market,"
      Euro-Spectra 9_,_ 46  (1970).

 18.   Gregory, D.  P., "The  Hydrogen Economy," Sci. Amer. 228J1), 13 (1973).

 19.   Gregory, D.  P., Ng, D. Y. C., and Long, G. M., "The Hydrogen Econ-
      omy ," The  Electrochemistry of Cleaner Environments, J. 0'M. Bockris,
      Editor,  Plenum Press,  New York TT972), page 226.

 20.   Woodward,  H. F.,  "Methanol," Encyclopedia of Chemical Technology 13,
      379 (1967)

 21.   Reed, T. B.  and Lerner,  R. M.,  "Methanol:  A Versatile Fuel for
      Immediate  Use,"  Science 182, 1299 (1973).

 22.   Wigg, E. E., "Methanol as a  Gasoline Extender:  A Critique," Science
      186.  785 (1974).

 23.   Reed, T. B., Lerner, R.  M.,  Hinkley, E., and Fahey,  R.,  "Improved
     Performance of Internal Combustion Engines  Using 5 - 30%  Methanol
     in Gasoline,"  Proc. Ninth Intersociety Energy Conversion Engineering
     Conference, San Francisco, California  Inst.  Electrical  and Electronic
     Engineers, New York (1974),  page 952.

24.  Haynes, W.  P.,  Gasior, S. J., and Forney,  A.  J., "Catalysis of Coal
     Gasification at Elevated Pressure,"  Coal  Gasification,  L. G. Massey,
     Editor,  Adv. in Chem.  Series  131,  Amer.  Chem.  Soc.,  Washington,
     D. C.  (1974), page 179.
                                    70

-------
25.  Vant-Hull, L. L., "Solar Thermal Power Systems Based on Optical
     Transmission," Semiannual Progress Report, 15 June - 31 Dec.  1973,
     Office of Tech. Services Report PB-237005/4 (1974).

26.  Francia, G., "Pilot Plants of Solar Steam Generating Stations,"
     Solar Energy 12^, 51 (1968).

27.  Powell, W. R., "Absorber for Solar Power," Appl. Optics 13,  2430
     (1974).                                                 ~~

28.  Sparrow, E. M. and Cess, R. D., Radiation Heat Transfer. Brooks-
     Cole Publ. Co., Belmont, Calif. (1966).

29.  Blake, F. A., "Solar Augmentation of Hydroelectric Power Systems,"
     Energy Sources 1_, 361 (1973-74).

30.  Easton, C. R., Hallet, R. W., Gronich, S., and Gervais, R.  L.,
     "Evaluation of Central Solar Tower Power Plant,"  Proc. Ninth Inter-
     society Energy Conversion Conference, San Francisco, CallfTTlnst.
     Electrical and Electronic Engineers, New York (1974), page  271.

31.  Woodcock, G. R. and Gregory, D. L., "Economics Analyses of  Solar
     Energy Utilization," Proc. Nintji Intersociety Energy Conversion
     Conference,  San Francisco, Calif., Inst. Electrical and Electronic
     Engineers, New York (1974), page 306.

32.  Hammond, V. L., Mudge, L. K., Allen, C. H., and Schiefelbein, G. F.,
     "Energy from Solid Wastes by Pyrolysis-Incineration," Battelle
     Pacific Northwest Laboratories Report BNWL-SA-4471 (1972).

33.  Anon., "Is Municipal Power the Answer?," Business Week, Oan.  20,
     1975, page 50.
                                   71

-------
                            APPENDIX A

               OPTIMIZATION OF THE REACTOR'S VOLUME
     Light enters the cylindrical annular reactor with angles  varying
between 0 and 63.4  , as shown in Figure 29.  If the angle    is  small
enough, the light reflects off the wall of the reactor as shown  in
the figure and continues to progress deeper into the annular cavity
separating two of the reactors.  However, when e is too large,  the
light reflects out of the cavity rather than deeper into it.   The
volume of the reactor is also determined by e  .  A larger angle  e gives
a larger reactor volume.  The purpose of this appendix is to describe
an algorithm for maximizing the reactor's volume while retaining all
the radiation incident on it from the solar furnace.

     To solve this problem we examine the "worst case", as shown in
Figure 29.  Light strikes the "point" of the left annulus with  angle
e = 63.4°.  It is reflected with angle

                         *|l = 180° - 26 - <|>!j-   ,                 (29)

where <|>L = e = 63.4° is the angle of incidence on the left annulus for
the first reflection.  This beam of light is incident on the right
surface of the annular cone with angle    = 180° - 26 - $   .   it  reflects
off the right surface with angle

                         +* = 180° - 26 - **                      (30)
                                                I            R
and is incident on the left surface with angle 2 = 180° -  $•...  We  now
have an iterative scheme for determining the angles L, $, f-, R
as a function of the 2n - 1 and 2n reflection number.  Our  desire is
to keep either ^ or 4>L greater than 90°, which constrains  the number
of reflections n allowed.
                                    72

-------
Figure 29.  Light reflection between two mirror surfaces.

-------
              D
The distance X] is given by

                R     R sin (^ - 90°) sin (180°  . +L)    ,        (31)

               xl -            sin (^ -e)

and the height HR by


                           H-|  = x  cos e    ,                     (32)

with distance


                           AR? = x'i sin 9    .                     (33)

The distance xi is now seen to be


                       Xl = HR/sin (^ - 90°)    ,                (34)

and the values of xlj-, H-| , and  ARi are given by

                           T  sin (*? + ^ - 180°)
                     XL = xj	L,	
                      l      sin (180° - +R + e)
                                   L                              (35>
                                   !  cos e
                                   (,  sin
     If the total  height of the reactor is 10 m and R = 0.7 m,  we now
have an iterative  scheme for determining the volume of the reactor V
as a function of the number of reflections n and the angle of incidence
e.  Values given in the body of this report were obtained using a
computer program designed to optimize V for a given value ofe.
                                  74

-------
                            APPENDIX B

     RADIANT HEAT TRANSFER IN GASEOUS H20 AND C02 + H20 MIXTURES


     A description and derivation of the band model for the exponential
kernel solution to the radiative equilibrium problem is beyond the scope
of this report.  Such a derivation is given by Sparrow and Cess in their
book Radiation H&cut TuonA^e/i. on pp. 239-247.  28  Using their formulas,
gaseous water with the band structure
                          X                AX

                        1.38y            0.18y
                        1.37             0.3
                        2.7              0.29
                        6.3              2.0
                       20.               8.0

and a thickness of 1 m absorbs 26.6% of the infrared radiation leaving
a black body radiator at 650 C.  An equi-molar mixture of C02 and
gaseous water with a total pressure of 2 atm and a band structure

                         X                 AX

                        2.7y              0.6y
                        4.3               1.2
                       15.               10.9

with a thickness of 1 m absorbs 59.3% of the radiation leaving a black
body at 650 C.  These calculations indicate the effectiveness of C02
and gaseous H20 as a radiation insulator.
                                   75

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 1. REPORT NO.
   EPA-600/2-77-147
                                                            3. RECIPIENT'S ACCESSION NO.
 4. TITLE AND SUBTITLE

   SYNTHETIC FUEL PRODUCTION  FROM SOLID WASTES
             5. REPORT DATE
               September  1977(Issuinq Date)
                                                            6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
   Roy C.  Feber and Michael J. Antal
                                                            8. PERFORMING ORGANIZATION REPORT NO.
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
   Los Alamos Scientific Laboratory
   The University of California
   P.O.  Box 1663
   Los Alamos, New Mexico   87545
             10. PROGRAM ELEMENT NO.

              SOS #1 FY 76/Task  05
             11. CONTRACT/GRANT NO.

              EPA-IA6-D5-0646
 12. SPONSORING AGENCY NAME AND ADDRESS
                                                            13. TYPE OF REPORT AND PERIOD COVERED
   Municipal Environmental Protection Agency--Cin.,  OH
   Office of Research and Development
   U.S.  Environmental Research  Laboratory
   Cincinnati, Ohio  45268
              Final Report
             14. SPONSORING AGENCY CODE
              EPA/600/14
 15. SUPPLEMENTARY NOTES
   Project Officer:  Albert J.  Klee (513-684-7881)
 is.ABSTRACTj^g work described  in  this report has two objectives:   first, to evaluate
   potential  catalysts for the commercial  practice of the  gasification of chars produced
   by the pyrolysis of municipal  or industrial wastes; second,  to determine the potential
   for synthetic fuel production  from solid wastes produced  in  this country, and  to  ex-
   plore the  feasibility of providing the  heat required for  the gasification reactions  by
   coupling a chemical reactor to a solar  collector.

   To meet the first objective, a small  scale, fixed bed,  flow  through reactor was as-
   sembled, and a number of potential  catalysts were tested  on  chars from a number of
   sources.  The alkali metal  carbonates are superior to any of the catalysts for gasifi-
   cation with both steam and  carbon  dioxide at 650 C.  With these catalysts, rates  of
   gasification by steam are increased by  factors of two to  three, and rates of gasifica-
   tion  by carbon dioxide, by  factors  up to ten.  The rates  are comparable with those
   observed elsewhere with other  carbonaceous materials.

   To meet the second objective,  several possible schemes  for coupling a solar collector
   and a  gasification reactor  are suggested, and economic  analyses of the systems are
   attempted.   It is concluded  that a  feasible, economically attractive systems is
   possible.
 7.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b.IDENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/GlOUp
  Refuse
  Pyrolysis
  Catalysis
  Methane
  Hydrogen
  Carbon monoxide
 Pyrolysis  char
 Synthesis  gas
 Monsanto pyrolysis
  process
   7A
 8. DISTRIBUTION STATEMENT

  RELEASE TO PUBLIC
19. SECURITY CLASS {ThisReport)
 UNCLASSIFIED
21. NO. OF PAGES
   84
                                               20. SECURITY CLASS (Thispage)

                                                UNCLASSIFIED
                           22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE
                                             76
                                                     ll.S. GOVERNMENT PRINTING OFFICE:| 1977-757-056/6522 Region No. 5-11

-------