&EPA
             United States
             Environmental Protection
             Agency
             Municipal Environmental Research  EPA-600 2-80-1 21
             Laboratory          August 1980
             Cincinnati OH 45268
             Research and Development
Obtaining  Improved
Products from the
Organic Fraction  of
Municipal  Solid
Waste

-------
                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-600/2-80-121
                                    August 1980
   OBTAINING IMPROVED PRODUCTS FROM THE
 ORGANIC FRACTION OF MUNICIPAL SOLID WASTE
                    by

         N.L. Hecht, D. S. Duvall,
          A. A. Ghazee, B. L. Fox
  University of Dayton Research Institute
            Dayton, Ohio 45469
           Grant No. R-804421-01
             Project Officers

                Albert Klee
                    and
                Steve James
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 Environ-
mental 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 U.S. Environmental Protection Agency was created
because of increasing 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
testimonies to the deterioration of our natural environment.
The complexity of that environment and the interplay of its
components require a concentrated and integrated attack on
the problem.

     Research and development is that necessary first step in
problem solution; it involves defining the problem, measuring
its impact, and searching for solutions.  The Municipal Environ-
mental Research Laboratory develops new and improved technology
and systems to prevent, treat, and manage wastewater and solid
and hazardous waste pollutant discharges from municipal and
community sources, to preserve and treat 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 and provides a most vital commu-
nications link between the researcher and the user community.

     This report presents information resulting from a study of
chemical and thermal treatments to improve the quality of the
products derived from the organic fraction of municipal solid
waste.  Processes for obtaining powdered fuels are described in
depth.  It is hoped that the information provided in this
report will be of assistance to researchers and developers
concerned with resource recovery endeavors.
                                 Francis T. Mayo, Director
                                 Municipal Environmental
                                 Research Laboratory
                              111

-------
                           ABSTRACT

     Many  systems have been developed  for  steam or fuel produc-
tion from  Municipal  Solid Waste  (MSW), but the quality and
consistency of  such  fuels have not been completely satisfactory.
As conventional fuels become more difficult to obtain, the need
grows  for  processes  that can produce MSW-derived fuels of higher
quality and greater  consistency -

     One promising technology is for the production of powdered
fuels  from cellulose waste.  This project  has investigated
several processes for the conversion of the organic fraction of
MSW to a powder.  The study concentrated on two types of
processes:   (1)  conversion of MSW to a powdered icarbon char by
low-temperature pyrolysis, and  (2) embrittlement of cellulose
waste  by thermal-chemical treatment.   This report describes
the results of  these studies.

     The first  phase of the project was devoted to identifying
processes  that  offer a potential for enhanced product recovery,
an evaluation of chemical treatments to improve carbon recovery
from pyrolysis  processes, an  evaluation of laboratory processes
for the production of gaseous and liquid fuels, and a laboratory
investigation of embrittlement processes for cellulose wastes.
The second phase of  the program was concerned with further lab-
oratory studies of the embrittlement process, pilot studies of
the embrittlement process with shredded newsprint and refuse-
derived fuel, and an engineering and economic assessment for a
plant  to process powdered cellulose for use as a fuel.  A com-
prehensive description of Phase I was presented in an earlier
report entitled "Investigation of Advanced Thermal Chemical
Concepts for Obtaining Improved MSW-Derived Products" (EPA-600/
7-78-143).  This report summarizes the results of Phase I and
provides a comprehensive review of the work conducted in Phase
II of  the  program.

     This  report was submitted in fulfillment of Grant No.
R-804421-07 by  the University of Dayton Research Institute under
the sponsorship of the U.S. Environmental  Protection Agency.
This report covers the period September 1, 1976 to January 1,
1979,  and work  was completed as of January 1, 1979.
                               IV

-------
                           CONTENTS

Disclaimer	  ii
Foreword	 iii
Abstract	  iv
Figures	 vii
Tables 	  ix
Conversions	, . . , ,	  xi
Acknowledgment.,	 xii

     1.  Introduction	,	   1
     2.  Conclusions	   3
              Phase 1	   3
              Phase II	   4
     3.  Carbonization of Cellulose	   5
     4.  Powdered Cellulose	  10
              Introduction	  10
              Phase 1	  10
              Phase II	  11
              Pilot reactor studies	  21
              Analysis of reactor operations	  37
                   Reactor coating	  37
                   Teflon and silicone rubber parts	  40
                   Iron,  aluminum  and monel parts	  40
              Embrittlement mechanisms	  41
     5.  Engineering Analysis	  46
              Plant design	  46
                   Trommel - 12.7 cm  (5 in) openings	  46
                   Shredder	  52
                   Magnetic separator	  57
                   Trommel - 1.3 cm  (1/2 in) openings	  61
                   Air classification	  62
                   Drying	  66
                   Incineration and flue gas cleaning	  67
                   Reactor/heat exchanger/scrubber and
                     caustic tank	  71
                   Ball mill and rotary disc screen	  71
              Material and energy balance	  72
              Proposed plant layout	  81
                   Equipment specifications	  81
                   Floor plan	  81
                   Plant contingency  plans	  93
                   Health, safety, and environmental
                     protection considerations	  94
              Economic analysis	  96
                              v

-------
                     CONTENTS  (Concluded)

     5.  Engineering Analysis  (Continued)
                   Proposed future work	 97
References	101

Appendix A - Background for Economic Analysis	104
                               vx

-------
                            FIGURES

Number                                                     page

   1     Carbon char prepared by pyrolysis of MSW
         treated with NaaAlaOit .............................. 7
   2     Flow plan for carbon char production ............... 8

   3     Experimental arrangement used for embrittlement
         studies ............................................ 12

   4     Shredded newspaper treated with SOC12
         Cthiony 1 chloride ) ................................. 13

   5     Shredded newspaper treated with para-formaldehyde
         and SO 2 ............................................ 14

   6     Computer paper treated with chlorine ............... 15

   7     Shredded newspaper treated with formaldehyde
         in SOa ............................................. 16

   8     Shredded newspaper treated with hydrochloric acid.. 17

   9     Process flow diagram for pilot reactor ............. 23

  10     Pilot reactor ...................................... 24

  11     Effect of temperature  (shredded newsprint) ......... 31

  12     Effect of temperature  (RDF) ........................ 32

  13     Effect of treatment time  (shredded newsprint) ...... 33

  14     Effects of treatment time  (RDF) .................... 34

  15     Effect of HC1 concentration  (shredded newsprintl . . . 35

  16     Effect of moisture  Cshredded newsprint) ............ 36

  17     Effects of sample size and HC1 concentration  (RDF). 39

  18     Reduction of the degree of polymerization  of
         cellulose by hydrolysis ............................ 42
                              VI1

-------
                      FIGURES  (Concluded)

Number                                                     Page

  19     Possible embrittlement mechanism; hydrolysis and
         crosslinking of adjacent cellulose chains by
         oxygen bridges	45

  20     Process flow plan	47

  20a    Process flow plan  (metric)	48

  21     Cumulative distributions of raw waste	53

  22     Particle size distributions of raw refuse
         components	54

  23     Particle size distribution - shredded MSW*	55

  24     Cumulative distributions of shredded waste	59

  25     Particle size distributions of shredded waste	60

  26     Basic components of an incinerator	68

  26a    Basic components of an incinerator (metric)	69

  27     Material balance flow plan (metric	77

  28     Material balance flow plan (English
         equivalents)	78

  29     Energy balance flow plan  (metric units)	79

  30     Energy balance flow plan  (English
         equivalents	•	80

  31     Floor plan	84

  32     Office and maintenance areas	85

  33     Refuse receiving area and initial trommeling	86

  34     Magnetic recovery and shredding area	87

  35     Fuel recovery area	•	88
                              Vlll

-------
                            TABLES

Number                                                     Page

   1     Laboratory Results For HC1	19

   2     Laboratory Results for C12	19

   3     Laboratory Results for S02	20

   4     Qualitative and Semiqualitative ESCA Analysis of
         the Treated and Untreated Cellulose (Intensity
         in Terms of the Normalized Atomic Percent)*	22

   5     Chemical Analyses	22

   6     Ranges of Process Variables Evaluated	27

   7     Effect of Temperature on Shredded Newspaper*	28

   8     Effect of Temperature (RDF) *	28

   9     Effect of Treatment Time for Shredded Newsprint*...29

  10     Effect of Treatment Time for (RDF) *	29

  11     Effect of HC1 Concentration on Shredded Newsprint•-30

  12     Effects of Moisture on Shredded Newsprint*	30

  13     Effects of Sample Size (RDF)	38

  14     Comparison of Embrittlement and Viscosity of
         Cupriethylenediamine Solutions of Paper	43

  15     Comparison of Diameter to Capacity Relationship....50

  16     Distribution of Waste Components Minus 12.7 cm
         (5 in)*	56

  17     Typical Glass Particle Size Distribution*	58

  18     Size Distribution With a Williams Mill*	58
                               IX

-------
                      TABLES  (Concluded)



Number                                                     Page



  19     Distribution of Materials in the Ferrous

         Fraction*	62



  20     Cumulative Size Distribution of Waste Components

         Minus 1.3 cm (1/2 in)	,	63



  21     Waste Component Densities.	63



  22     p t  For Waste Components.. . ,	65
          s s


  23     Material Balance, Metric Tons	73



  24     Material Balance, English Equivalents	75



  25     Equipment List	82



  26     Estimated Capital Costs	98



  27     Estimated Annual Operating and Maintenance Costs...99



  28     Potential Revenue Sources	99



  A-l    Equipment and Installation Cost	105



  A-2    Plant Facilities and Auxiliary Equipment Cost	106



  A-3    Annual Manpower Costs	108



  A-4    Proposed Work Schedule	109
                               x

-------
           CONVERSIONS
Acres x 0.4 = hectares
Tons x 0.9 = metric tons
yd3 x 0.76 = m3
ft  x 0.3  = m
ft2 x 0.09 = m2
ft3 x 0.03 = m3
in  x 2.5  = cm
Ib  x 0.45 = kg
million BTU x 1.054 = mkj
gal x 3.8 = H
qt  x 0.95 = i
oz  x 28.0 = g
                XI

-------
                       ACKNOWLEDGMENT
    The authors wish to acknowledge the assistance of Albert
Klee and Steve James, U.S. Environmental Protection Agency
project officers.  In addition, the authors wish to acknowledge
the valuable assistance provided by the University of Dayton
students who worked on this project — Steve Hugenberg,
Ricardo King, Juan Torres, Colleen Kohn, Paul Murphy, and
Beverly Gantner.  The authors would also like to express
their appreciation to Patsy Collins, and Niki Maxwell for
typing the manuscript.
                              Xll

-------
                           SECTION 1

                         INTRODUCTION
     The purpose of this study was to investigate the potential
of known processes that could improve the quality of the fuels
or other products derived from the organic fraction in refuse.
Since cellulose products are the major constituents of the
organic fraction in refuse, it seemed likely that a number of
the processes employed in the pulp, paper, and textile indus-
tries would have considerable potential for refuse processing.

     To accomplish effectively the stated objective for this
study, a comprehensive review of processes for making refuse
a better fuel was performed.  Information was obtained from the
open literature and through personal contacts.  Possible pro-
cesses for improving the quality of products from the organic
fraction derived from municipal solid waste (MSW) were deter-
mined, and descriptions were developed for each process.

     This study concentrated on those processes designed to
produce a carbon char, a cellulose powder, and liquid and
gaseous fuels from the MSW.  The production of carbon char was
carried out by the use of chemical treatments that promote char
formation at lower pyrolysis temperatures.  For the production
of powdered cellulose, the chemical and thermal treatments that
cause cellulose embrittlement were of most interest.  For the
production of gaseous and liquid fuels, two processes were
evaluated — the Worcester Polytechnic hydrogenation-liquefac-
tion process (1) and the Wright-Malta gteam injection pyrolysis
process (2) .

     The program was conducted in two phases.  The first phase
was devoted to identifying processes that offer the potential
for enhanced product recovery, an evaluation of chemical treat-
ments to improve carbon recovery in pyrolysis processes, an
evaluation of the Worcester Polytechnic and Wright-Malta
processes, and a laboratory investigation of embrittlement
processes for cellulose wastes.  The second phase of the pro-
gram was concerned with further laboratory studies of the
embrittlement process, pilot studies of the embrittlement
process with shredded newsprint and refuse-derived fuel  (RDF),
and an engineering analysis and economic assessment for a plant
to process powdered cellulose for use as a fuel.  A comprehen-
sive description of the program conducted in Phase I was

-------
presented in a U.S. Environmental Protection Agency (EPA)
report entitled, "Investigation of Advanced Thermal-Chemical
Concepts for Obtaining Improved MSW-Derived Products." (3)
This current report summarizes the results of Phase I and
provides a comprehensive review of the work conducted in
Phase II.

-------
                           SECTION 2

                          CONCLUSIONS
PHASE I

     The technology of cellulose chemistry provides several
methods for processing the organic fraction of municipal solid
waste (MSW) to obtain improved refuse-derived products.  Lab-
oratory processes for obtaining a carbon char and a cellulose
powder have been identified.  In addition, laboratory processes
for obtaining liquid and gaseous fuels from solid wastes
described in the literature were reviewed.

     Several chemical treatments for promoting char formation
at low temperatures were demonstrated.  The patented process
developed by Harendza-Harinxma (4) using Na2Al2Oi» to promote
char production was evaluated for incorporation  into the
Black-Clawson (5) wet process.  Markets for the char product
were also investigated, but it was not possible to identify
firm markets for the char.  Lack of familiarity with the
product and the need for rigid product specifications limited
strong market acceptance.

     During the course of Phase I, two processes for the
production of liquid and gaseous fuels were also reviewed.
Both the Worcester Polytechnic Institute process and the
Wright-Malta Corporation process were selected because of their
unique potential for enhanced product recovery.  However, after
review, it was concluded that neither process was practical for
implementation on a large scale at present.

     A major accomplishment of Phase I was the identification
and laboratory verification of chemical treatments for cellu-
lose embrittlement.  As a result of preliminary laboratory
studies, the basic requirements were defined for producing
powdered material from the organic fraction of MSW.  Although
a number of the chemical treatments investigated were suitable
for embrittling cellulose waste materials, HCl was found to be
the most effective treatment agent.  However, more quantitative
measurements of the embrittlement process parameters were
needed.  The information obtained from these studies could
provide the basis for a detailed engineering and economic
analysis for a full-scale facility.

-------
PHASE II

     In Phase I several chemical treatments identified for
cellulose embrittlement were investigated.  Using a laboratory
reactor samples of shredded newsprint were exposed to different
concentrations of HC1, SO2, and Cl2 gases.  HC1 proved to be
the most favorable reagent for large-scale cellulose waste
processing.  To evaluate better the process variables for
cellulose embrittlement with HC1, a series of pilot reactor
studies was conducted with shredded newsprint and RDF.  From
these studies, it was found that acceptable operating results
were obtained at 149°C  (300°F), a 6-min reaction time, and
50% by volume concentration of HC1 in the reactive gas mixture.
It was further determined that the degree of embrittlement
could be improved by raising the treatment temperature to
177°C (350°F) and extending the reaction time to 8 to 10 min.
From the data obtained in these experiments, it can be conclu-
ded that the degree of embrittlement is maximized with high
treatment temperatures, long reaction times, minimum moisture
content, and moderate levels of HC1 concentration.

     A number of mechanisms have been considered to account for
the embrittlement observed in  the treated cellulose.  It is
believed that the embrittlement process is not a single mechan-
ism, but rather a complex combination of mechanisms.  Both a
reduction in the degree of polymerization and the formation of
cross-links between polymer chains are believed to be among the
dominant mechanisms for cellulose embrittlement in the presence
of a mineral acid.

     With data from the laboratory and pilot studies, a prelim-
inary design for a resource recovery plant to produce powdered
cellulose fuel (p-RDF) was developed.  The design plan was based
on the use of HCl gas for embrittlement.  A preliminary eco-
nomic analysis was also prepared.  The plant would process
907 metric tons  [mt]  (1000 tons) per day and process 399 mt
(440 tons) of p-RDF per day.   The capital cost of the facility
would be $30.3 million based on 1978 dollars.  From the data
developed, it was calculated that for the facility to break
even, the fuel would have to sell for about $0.064/kg
($0.029/lb) if the plant charged a tipping fee of $9.37/mt
($8.50/ton).

-------
                           SECTION 3

                  CARBONIZATION OF CELLULOSE
     Thermal decomposition of refuse in the absence or partial
absence of oxygen (pyrolysis) has been used for the production
of useful fuels.  Solid, liquid, or gaseous fuels can be ob-
tained by a variety of pyrolysis processes.  The quantity of
char, bitumen-like liquid, and gas varies and is a function of
the time-temperature sequence for each particular process.
Commercial, pilot, and laboratory processes that use a variety
of furnace designs have been developed to produce a wide range
of fuel products.

     The thermal processing employed can be designed to maximize
the end products desired from the cellulose wastes.  Higher
heating rates and higher temperatures produce large quantities
of gas and less char.  Conversely, lower heating rates and
lower temperature processes result in increased char production.

     The quantity of char, the composition of the gases and
liquids evolved, and the necessary reaction temperatures can be
significantly affected by the presence of chemical agents in
the cellulose materials.  A number of chemical treatments have
been identified that increase the quantity of char produced and
decrease the amount of combustible gases and tars formed.  Many
of these treatments were developed for flame-proofing cellulosic
materials.  The highly effective flame-proofing compositions are
all soluble in water, and most are salts of either strong acids
or bases.  In addition, many oxidizing agents attack cellulose
and have been found to be effective fire retardants.  Oxidized
cellulose pyrolyzes rapidly at lower temperatures, giving high
charcoal residues.

     Although a number of pyrolysis processes have been devel-
oped to produce carbon char, the literature search only produced
one process using chemical treatment to promote char formation.
In the patented process developed by A. Harendza-Harinxma  (4),
sodium aluminate  (Na2Al2Olt) is used to promote carbon formation
at temperatures below 349°C  (660°F).  In most of the con-
ventional pyrolysis processes reported in the literature,
pyrolysis is achieved with thermal treatments from 499°C to
899°C  (930°F to 1650°F).  In addition, the char yield is
reportedly increased for the sodium aluminate process  (4).

-------
     During  this  project,  three  char-promoting materials;
   Cl, CaCl2  and  Na2Al2Oit  were evaluated.  Thermogravimetric
analysis was  used to  study the pyrolysis of cellulose samples
treated with  each of  the chemicals  of  interest.  The results of
these preliminary studies  showed that  Na2Al201| and NHi»Cl are
promising char  promoting chemicals.  A sample of the carbon
char produced from the  pyrolysis of shredded newsprint treated
with a 2% solution of Na2Al2OIt is shown in Figure 1.

     In Dr.  Harendza-Harinxma's  process, municipal refuse and
sewage sludge are pyrolyzed to a char  and fuel gas.  In this
process, sewage sludge  is  used as the  solvent for a 2 to 4%
solution of  sodium aluminate.  This solution is mixed with
municipal refuse  (two parts sludge  to  one part refuse).  The
slurry mixture  is then  mechanically dewatered, resulting in a
thickened sludge  which  is  thermally dried to about 15% moisture
content.  The dried refuse-sludge mixture, impregnated with
sodium aluminate,  is  then  carbonized for about 1 hr at 249°C
to 299°C  (480°F to 570°F).

     A preliminary engineering study of a plant to produce car-
bon char from MSW and sewage sludge was conducted.  Based on the
information  accumulated, a design for  a low-temperature pyrol-
ysis facility was developed (see Figure 2).  The basic design
for this plant  is based on the Black-Clawson wet processing
system at Franklin, Ohio  (5).  The  process flow essentially
follows the Black-Clawson  process to the dewatering stage at
the hydrodenser.   After partial  dewatering at the hydrodenser,
the slurried  refuse would  be mixed  with sodium aluminate and
sewage sludge in  a flash mixer.   This  slurry mixture would then
be mechanically dewatered  in a cone press and dried in a rotary
carbonization kiln.

     A plant  processing 907 mt  (1000 ton) of refuse per day
could generate  up to  272 mt (300 ton)  of char per day or about
72,109 mt  (79,500 ton)  per year.  A preliminary economic anal-
ysis based on 1978 dollars was prepared for this proposed
pyrolysis plant.   The proposed plant would have a total capital
cost of $25,833,000.  Revenues for  the facility would be pro-
vided from several sources:  a) a tipping fee  ($8.50), b) sludge
disposal  ($5.00),  c)  sale  of aluminum  and iron  ($3.65), and
d) sale of carbon char.  The estimated total operating cost
would be $29.42/mt ($26.68/throughput  ton) of MSW with a
potential revenue of  $18.91/mt  ($17.15/throughput ton) plus the
revenue from  the  sale of carbon  char.   For the pyrolysis facil-
ity to break  even the carbon char would have to sell for
$0.035/kg ($0.016/lb) or $35.3/mt ($32/ton) in order to cover
the $9.53 deficit calculated between total operating cost and
identified potential  revenues.   Of  major concern is the market
potential for the char  product.   If large quantities are to be
produced, then  suitable long term markets must be available.

-------
Figure 1.  Carbon char prepared by pyrolysis of MSW treated with

-------

Water Return

Municipal
Solid Waste
1
1
Receiving floor
& Storage

^ _ Nonrerrous iv
I
	 •- Hvdr-mnl nnr- .. JuxiJs

i


Liquid Heavy fc Glass & Meta
Cyclone Fraction Recovery
!
1
_ . 1 Residue

Rejects
Thickening & R1 k rn -,„„_.,
Dewatering

ALTERNATE PROCESS

1 — 	 	 ^ Mixing Tank.K"

L
r—
1
1
1
Fue
Ga


Mechanical
Dewatering
1
^ Rotary
Dryer
1 1

*~ ^ _. Ccirfoonizalr ion
^
Furnace

[etal
~1
^ Magnetic
al Separator
i


— •» Glass Metal

U
RDF
	 [Sewage Sludge]
—I Sodium Aluminate

Figure 2.  Flow plan for carbon char production,

-------
     A number of potential markets were identified for the char
product from the pyrolysis process:  a) solid fuel, b) feed
stock for preparing gaseous or liquid fuels, and c) substitute
for the carbon now being used in carbon and graphite products
(activated carbon, charcoal, carbon fillers, carbon risers,
etc.).

     It appears that the char can be used as a solid fuel and as
feed stock for synthetic liquid and gaseous fuels at a market
value of about $0.022/kg ($0.01/lb).  With some additional
processing the char could also be utilized as an activated car-
bon,  a raw material for charcoal briquettes, a carbon riser in
iron and steel production and as fill in nonstructural rubber
products at a market value of $0.044 to $0.088/kg  ($0.02 to
$0.04/lb).  A major question regarding these latter applica-
tions is the availability of sufficient market demand for the
char product.  A plant processing 907 mt (1000 ton) of refuse
per day could more than saturate many of the potential market
opportunities identified (72,109 mt/year)  (79,500 ton/year).
Before a char-producing plant is built, therefore, it must
secure a certified market for its product from one of the
applications identified.

     Preliminary discussion with the companies considered
potential customers of the carbon char product revealed consi-
derable reservations concerning the use of this product.  These
reservations were based on a lack of familiarity with the car-
bon char product and concern about the use of an untried
material.  The inability to guarantee consistent product com-
position also presented considerable problems for these poten-
tial customers.  In addition, the inability to successfully
commercialize the char produced from conventional pyrolysis
processes developed to date  (.Baltimore, Occidental, etc.)
further moderates enthusiasm for the char product produced by
the low temperature pyrolysis process.

-------
                           SECTION 4

                      POWDERED CELLULOSE


INTRODUCTION

     The processes  for converting the organic fraction  (pre-
dominately cellulose) of MSW into a finely powdered material
offered a number of advantages for improving the quality and
marketability of the recovered products.  In a powdered form,
the MSW can be a more desirable fuel; it can be used in sus-
pension fired boilers; it can be slurried with oil for firing
in liquid fuel units; and it can be readily pelletized  (6).
In addition, the powdered MSW can be more readily used as a
filler material in  the fabrication of plastic and rubber
products.

     A number of thermal and chemical treatments have been
described in the literature for promoting the conversion of
the organic fraction in refuse to a fine powder (7) .  Since
the major constituent of the organic fraction is cellulose
(75%) , the treatments are based primarily on the technology of
cellulose processing., Most of these processes are chosen to
embrittle or degrade the cellulose by reducing the degree of
polymerization.  In addition to degradation, certain acid
treatments appear to promote crosslinking of adjacent molecules,
which makes the cellulose rigid and brittle.  Heating cellulose
in air up to temperatures of 204°C (400°F) also will result in
partial embrittlement.  In addition, cellulose treated with
formaldehyde also undergoes embrittlement.

PHASE I

     During Phase I a number of chemicals were screened for
their ability to embrittle paper and other cellulose wastes.
The chemicals studied are as follows:

                1.  Formalin
                2.  Formalin plus
                3.  Formalin plus
                4.  Formalin plus formic acid
                5.  Formalin plus acetic acid
                6.  Formalin plus SO2
                7.  Formalin plus HCl
                8.  Paraformaldehyde plus

                               10

-------
                9.   Methylan plus S02
               10.   Acetal plus SC>2
               11.   Hydrochloric acid
               12.   Chlorine
               13.   Thionyl chloride.

The results of these initial studies showed that hydrochloric
acid .(HC1) is an extremely good embrittlement agent.  Chlorine
gas (Cl) and thionyl chloride (SOCla) also were found to be
effective embrittlement agents.  Sulfur dioxide (SOa) proved
to be a moderate embrittlement agent.  The effectiveness of
S02 was improved by the inclusion of formaldehyde in the
reaction mixture.

     The experimental arrangements used for the chemical treat-
ment of the shredded samples is shown in Figure 3.  After chem-
ical treatment, the samples were pulverized and the powder
obtained was evaluated.  Examples of the powder obtained from
several of these studies are shown in Figures 4 through 8.

PHASE II

     A second series of experiments was conducted to define
better the embrittlement reaction of shredded newsprint.  Using
the laboratory reactor and the experimental arrangement shown
in Figure 3, 10 g (0.35 oz) samples of shredded newsprint were
exposed to different concentrations of HC1, SOa and Cl2 gases
for 10 min.  For the SOa treatment, the shredded newsprint was
first sprayed with formaldehyde before exposure to the sulfur
dioxide.  The reactor temperature was kept at 90°C  (194°F) by
the use of a heating lamp.  The quantity of reactant gases was
regulated by partially evacuating the reactor, then backfilling
it with the gas to atmospheric pressure.  In this fashion
various partial pressures of gases were evaluated and the mass
of the gases at these pressures was calculated using the ideal
gas equation.

     Unreacted gases were passed through a fritted glass dis-
persion tube into a scrubber unit consisting of a three-neck,
round bottom flask.  The effectiveness of the scrubber system
was determined by processing reactant gases through  an empty
reactor.  In this procedure the reactor with a volume of 4.2 Jl
(1.1 gal) was first evacuated and then backfilled with reactant
gases at atmospheric pressure and temperature.  The  total
number of moles in the reactor was calculated using  the ideal
gas equation.  The scrubber was filled with 1 S,  (1.06 qt) of
water to scrub HC1 gas, and SOz was recovered by reacting it
with 1  H  (1.06 qt)  of 0.8% potassium chlorate solution  in the
scrubber.  Recovery of chlorine gas required two scrubber units
in series.  The first unit was filled with a sodium sulfite
solution  (15% in excess of stoichiometric quantities).  The
second  scrubber contained  0.8% potassium chlorate solution to

                               11

-------

                                                           250 W
                                                         /Heat
                          .Gas  Inlet
                          Vacuum  Connection
                                                       3/ Reactor
                                                       Flask
                                                              - Ring Seal

                                                  ££*&<  ./xGlass Plate
                                                 Stopcock  Vent
Figure 3.  Experimental arrangement used  for embrittlement studies.

-------
Ul
          Figure  4.   Shredded newspaper treated with SOC19 (thionyl chloride).

-------
Figure 5.  Shredded newspaper treated with para-formaldehyde and S02•

-------
                                          r
                                t               *
                               .*
                              *     •     W
                                         m  *. m  m
Figure 6.  Computer paper treated with chlorine.
                        15

-------
a\
                                                           tff       V
                                                           •«-«-     v
             Figure  7.  Shredded newspaper treated with  formaldehyde in S09.

-------

Figure 8.  Shredded newspaper treated with hydrochloric acid.

-------
react with  the  SO2  formed in the first reactor.   The  following
stoichiometric  equations summarize the reactions  for  the
recovery  of C12 and S02.
Na2S03 + H20 + C12   	>-   Na2SO,»  + 2HC1
                                                  C12 recovery
     Na2S03  + 2HC1   	*-   2NaCl + H20
)
KC103  +  3S02  + 3H20 	>-   KC1  + 3H2SOU           S02 recovery

     The reaction between the gas and scrubber  solution was
carried  out by gently drawing the gaseous  content  of the
reactor  into  the scrubber for about one-half  hour.  In order
to maximize the recovery, the gas was dispersed as fine bubbles
in the scrubber solution by using a glass  frit  and stirring the
solution with a magnetic stirrer.  A 10 ml (0.34 oz) aliquot
sample was  taken after the completion of the  reaction and
titrated with O.lN NaOH to determine the strength  of the acid
formed as a result of the stated reactions.   Recovery of
95 to  98% of  the reactant gases  was achieved.

     Evaluation of the embrittlement process  by the various gas
treatments  was accomplished by placing an  8 cm  (3.1 in) length
of treated  paper and  a plastic breaker ball in  a plastic vial.
The vial was  placed in a Wig-L-Bug (Crescent  Dental Manufactur-
ing Company)  and the  time required to powder  the paper was
recorded.   Tables 1,  2,  and 3 summarize the results of this
second series of laboratory experiments.

     The data presented in Tables 1,  2, and 3 show that approx-
imately  2.9 g (0.1 oz)  (79 mmole)  HCl, 5.6 g  (0.2  oz)  (70 mmole)
C12 , 6.5 g  (0.23 oz)  (102 mmole)  S02  are consumed  in the treat-
ment of  10  g  (0.35 oz),  of shredded  newspaper for  embrittlement
in a batch  process.  The data from Tables  1 and 2  show that the
degree of embrittlement with HCl and C12 is not affected if the
initial  quantities of the reactant gases consumed  are in excess
of or  equal to the minimum required amount.   When  the quantity
of reactant gas consumed is less than this value,  poor embrit-
tlement  occurs.

     The SO2  consumed in the sulfur dioxide/formaldehyde
embrittlement (Table  3)  is almost twice the amount of HCl on
a weight basis.   In addition, the degree of embrittlement is
lower  as judged by the time required to powder  these samples.
The data also show that the degree of embrittlement can be
improved by the further addition of formaldehyde.   For example,
with an  SOa pressure  of 710 mm Hg (15.7 lb/in^) , doubling the
concentration of formaldehyde resulted in  a sample which was
powdered in one-third the time.   At the same  time,  the amount
of SO2 consumed per 10 g (0.35 oz)  of paper increased by 10 g
(.0.035 oz) .


                               18

-------
                    TABLE 1. LABORATORY RESULTS FOR HC1
  Partial
 pressure
  of HC1
 [mm Hg
 Initial mass
of HC1 in the
   reactor
   [g (02) ]
  Mass of HC1
 recovered in
 the scrubber
after reaction
   [g (oz)]
Amount of HC1
consumed per      Powder
10  g of paper     time
  [g (oz)]        (sec)
710 (13.7)

550 (10.6)

450 ( 8.7)

350 (6.8)

100 ( 1.9)
  5.9 (0.21)

  4.5 (0.16)

  3.7 (0.13)

  2.9 (0.10)

  0.8 (0.03)
  5.1 (0.18)

  3.8 (0.13)

  3.1 (0.11)

  2.3 (0.08)

  0.5 (0.02)
 0.8 (0.03)

 0.7 (0.02)

 0.6 (0.02)

 0.6 (0.02)

 0.3 (0.01)
 27

 32

 32

 32

210
                    TABLE 2. LABORATORY RESULTS  FOR

Partial
pressure
of Cla
[mm Hg
(lb/in2)J
710 (13.7)
550 (10.6)
450 ( 8.7)
350 ( 6.8)
100 ( 1.9)

Initial mass
of Cla in the
reactor
[g (oz)]
11.3 (0.4)
8.8 (0.31)
7.2 (0.25)
5.6 (0.2)
1.6 (0.06)
Mass of Cla.
recovered in
the scrubber
after reaction
[g (oz)]
8.2 (0.29)
5.4 (0.19)
4.2 (0.15)
4.0 (0.14)
1.0 (0.04)

Amount of Cl 2
consumed per
10 g of paper
[g (oz)]
3.1 (0.11)
3.4 (0.12)
3.0 (0.11)
1.6 (0.06)
0.6 (0.02)

Powder
time
(sec)
25
30
25
35
75
                                     19

-------
TABLE 3. LABORATORY RESULTS FOR S02

Partial
pressure
of S02
[mm Hg
to
o
710
710
550
450
100
(13.7)
(13.7)
(10.6)
( 8.7)
( 1.9)
Initial
wt
of SOa in
the reactor
[g (oz)]
10.3 (0
10.3 (0
8.0 (0
6.5 (0
1.5 (0
.36)
.36)
.28)
.23)
.05)
Initial mass
of formaldehyde
in the reactor
[g (oz)]
1.2
3.0
1.6
1.3
1.7
(0.04)
(0.11)
(0.06)
(0.05)
(0.06)
Weight
percent
formaldehyde
10
22
16
16
(53
.4
.5
.7
.7
.1)
Mass
of
S02 recovered
in the scrubber
[g (oz)]
8.7
7.7
6.4
5.2
1.2
(0
(0
(0
(0
(0
.31)
.27)
.23)
.18)
.04)
Amount of SOa
consumed per Powder
10 g of paper time
[g (oz)] (min)
1.6
2.6
1.6
1.3
0.3
(0.06)
(0.09)
(0.06)
(0.05)
(0.01)
3
1
2
2.6
(3.5)

-------
     In summary, these experiments showed that greater quanti-
ties of C12 were required to achieve equivalent degrees of
embrittlement to that achieved with HC1.  The S02 required for
the sulfur-dioxide formaldehyde embrittlement was almost twice
as much as that required with HC1.  In addition, the degree of
embrittlement by this process was not as complete as that
observed for the HCl and Cla treatments.  The time that was
required to powder the SOz treated samples was much greater
than that required for the HCl or Cl2 treated materials.  These
laboratory studies showed HCl to be  the most suitable reagent
compared to the C12 and SO2 treated samples.

     Treated samples were analyzed with Electron Spectroscopy
for Chemical Analysis (ESCA) and standard chemical analysis to
investigate the embrittlement reaction.  The ESCA results
(Table 4) show very little  (<1% atomic) chlorine is present
in untreated paper or in samples treated with HCl.  Similarly,
samples treated with SO2 and formaldehyde did not show any
significant increase in sulfur content.  However, significant
chlorine incorporation did result from C12 treatment.  These
results are in good qualitative agreement  with the chemical
analysis shown in Table 5.                          (

PILOT REACTOR STUDIES

     A third series of experiments studying the embrittlement
of shredded waste with HCl was also conducted in order to
evaluate process variables in a pilot size reactor.  Shredded
newsprint was used as the feed material for a major portion of
the testing period.  However, a number of embrittlement tests
were also conducted using refuse derived fuel (RDF) from the
Americology Plant in Milwaukee.  Operational problems with the
reactor system were analyzed through the study and reactor
performance was closely monitored.

     The reactor unit (shown in Figures 9 and 10), built by the
University of Dayton Research Institute (UDRI), was designed to
process 1.1 to 2.3 kg (2-1/2 to 5 Ib) of shredded waste  (news-
print or RDF) samples with a continuous flow of reactant gases.
The interior of the reactor was constructed from a 30 gal carbon
steel drum coated with a 20 mil thickness of Emralon* 314
(50% Teflon - 50% epoxy).  The reactor is supported in a Uni-
strut frame such that it could be loaded and unloaded by tilting
it about a horizontal axis between two vertical Unistrut chan-
nels.  The shredded waste was introduced at the top and the lid
was sealed with a specially designed silicone rubber gasket and
locking ring to prevent leakage.  Three 1800 watt drum heaters
were install-ed on the outside of the reactor for  heating.  A
stainless steel screen with 0.84 cm  (0.33 in) diameter holes
was installed at the bottom of the reactor  to ensure an  even
* Acheson Colloid Company, Michigan

                               21

-------
          TABLE 4. QUALITATIVE AND SEMIQUALITATIVE ESCA ANALYSIS OF THE
                   TREATED AND UNTREATED CELLULOSE (INTENSITY IN TERMS
                   OF THE NORMALIZED ATOMIC PERCENT)*
Element
Oxygen
Carbon
Chlorine
Sulfur
Level
Is
Is
2p
2p
Binding
Energy
(%)
285
533
197
165
Untreated
newspaper
(%)
24.83
74.68
0.49
—
Untreated
filter paper
(%)
53.70
45.66
0.64
—
Filter paper
treated
with HC1
(%)
50.07
49.50
0.42
—
Newspaper
treated
with Cl
(%) 2
9.19
63.74
27.08
—
Newspaper
treated with
SO and CH 0
2 (%) Z
13.50
85.81
0.32
0.37

to
to * These
analys
ses ignofe
atomic hydrogen
concentrations
since the ESCA
technique is

unable to analyze hydrogen.
                            TABLE 5. CHEMICAL ANALYSES


1.
2.
3.
4.
Sample
Newsprint treated with HC1
Newsprint treated with Cl
Newsprint treated with SO -CH 0
Untreated newsprint
% S
—
—
0.43
0.22
% Cl
1.05
5.49
—
0.09

-------
to
U)
                     SOLENOID VALVE
       N2 FLOW METER-
HCI  FLOW METER
             NEEDLE
             VALVE
        SOLENOID
         VALVE
                 POLY-
              ETHYLENE
                TUBING
              • HCI  CYLINDER
                                                   N,
                                          ••-PRESSURE REGULATOR
                                        TEFLON
                                      COATED DRUM
                                                         V BRAIDED WIRE, TEFLON
                                                         A
                                                LINED TUBING
                                  TEFLON TUBING
                                                              .SAFETY  RELIEF
                                                                 VALVE
                                                               SHREDDED
                                                               WASTE
                                                                DRUM HEATER
                                                                NaOH
                                                                SOLUTION
                                                THERMOMETER
                                                          SCRUBBER DRUM
                                                         FRITTED  GLASS
                                                           FUNNEL
                                                                                                  -AIR
                                                                                   RUBBER
                                                                                   HOSE
                                                                                   -, O J
                                                                 "^ \	
                                                                 I I  "o O ~O~ ~» O
                                                                   O 0 O o Ofl
                                                                  O  0 0 O  0
                                                                  oooo   <
                                                                                                   SAMPLE
                                                                                                    PORT
              Figure 9.   Process flow diagram  for pilot reactor.

-------
t-o
                                   Figure 10.   Pilot reactor

-------
distribution of the gas, and a Teflon relief valve was installed
on the lid to provide safety from excessive pressures.  All
temperature measurements were made with a digital thermometer.

     Hydrogen chloride was supplied from a 27.2 kg (60 Ib) HC1
cylinder, and Na was drawn from an existing N2 bank.   A 5 kW
heater was used to heat the N2.   Two glass rotameters
(0-35 ft3/min)  (0 99.1 £/min) were used to measure the rate of
the reactant gases flowing through the reactor.

     The scrubber unit consisted of a polyvinyl chloride-lined
208.2 H  (55 gal) drum containing 84 I (22.2 gal) of 0.15N
sodium hydroxide solution.  The unreacted gases were dispersed
through a 15.2 cm (6 in) diameter fritted glass funnel into the
NaOH solution.  Efficiency of the scrubber was tested by intro-
ducing a known quantity of HC1,  and titrating a 10 ml (0.34 oz)
sample from a 500 ml (16.9 oz) aliquot of the scrubber solution
with 0.IN HC1.  A series of tests showed that the titrimetric
data could account for approximately 95% of the quantity of
HC1 calculated from flow measurements.

     After the reactor was charged with 1.1 kg  (2-1/2 Ib) of
shredded waste, it was sealed and heated to the desired temper-
ate with the drum heaters.  Upon reaching the specified temper-
ature, a moderate flow of HC1 and N2 was begun.  The HC1 flow
was maintained for a short time to provide the desired quantity
of this reactant.  The reactor was then purged with Na for an
additional 0.5 hr.  Gas flow rates and times were recorded from
the control panel.  Unreacted HCl and the N2 passed out the top
of the reactor through a 1.9 cm (3/4 in) Teflon lined tube and
into the scrubber unit described above.  Two aliquots of the
scrubber solution were titrated with 0.IN HCl to account for
the amount of unreacted HCl.  The input to the reactor was
calculated from the flow rate of HCl at the measured pressure
and temperature.  Output of HCl from the reactor was calculated
from the change in normality of the NaOH solution.  The reten-
tion of HCl could be determined by the difference in these two
values.

     After chemical treatment, the shredded waste was ball
milled for 15 min.  The ball mill unit consisted of a 12.3 £
(13 qt) ceramic jar filled with 13.6 Kg  (30 Ib) of 1.3 cm
(1/2 in) x 1.3 cm (1/2 in) ceramic cylindrical grinding pellets.
The jar was rotated on a pair of 5.1 cm  (2 in) diameter hard
rubber rollers, one of which was chain driven by a 186.4 watt
(1/4 hp) electric motor.  Quartered samples of the powder
were screened on a RoTap shaker for 10 min.  The fractions
retained on the 354,  149,  and 74 y  (45   100   and  300 mesh)
screens were weighed and recorded.  The minus 354 y  (45 mesh)
fraction was selected as the criterion for comparing  the extent
of embrittlement which resulted from each of the treated
samples.

                               25

-------
     During  this  part of  the  study,  the  effects of several major
processing variables  on the embrittlement of  shredded waste were
studied.  The  roles of processing temperature, reaction time,
HC1 concentration, and moisture content  of  the sample were
studied using  the reactor system shown in Figure  9.  The ranges
evaluated for  each of the process variables are shown in Table
6.  Both shredded newsprint and RDF  from the  Americology Plant
were used for  this series of  tests.   Depending on the specific
circumstances,  each test  condition was evaluated  two to seven
times  for each of the process variables  studied.  Three runs
were made for  most of the test conditions.  Heating times of
about  15 min were required for the tests conducted at 149 °C
(300°F).  The  data recorded for each test series  (temperature
effects, reaction time, etc.)  are compiled  in Tables 7 through
12.  The results  from each test are  also plotted  in Figures 11
through 16.

     As shown  in  Tables 7 and 8,  and Figures  11 and 12, the
degree of embrittlement increases with increasing treatment
temperature; conversely,  the  amount  of HCl  consumed in the
reaction decreases with increasing temperature.   The decrease
in the quantity of HCl retained in the shredded waste with
increasing temperature may be due to physical desorption mech-
anisms which are  more active  with temperature (see Embrittlement
Mechanisms section).   Preliminary experiments also suggest that
the use of treatment  temperatures higher than those used would
result in even greater levels of embrittlement.   The treatment
temperature  selected  for  a full scale process should be a bal-
ance between the  level of embrittlement  required  and the energy
demands.  In addition, the reduced amount of  HCl  consumed at
higher reaction temperatures  might encourage  the  use of some-
what higher  temperatures  to achieve  greater embrittlement.

     As shown  in  Tables 9 and 10, and Figures 13  and 14, the
degree of embrittlement increases with increased  reaction time.
The longer exposure of the material  to heat and HCl treatment
also resulted  in  a greater amount of HCl consumed in the reac-
tion.  As with temperature, the reaction time used for a full
scale  process  would be a  balance between the  level of embrittle-
ment desired versus the energy demand and HCl requirements.

     The degree of embrittleraent and the consumption of HCl in
the reaction increased with increasing quantities of HCl in
the reactant gases (Table 11,  Figure 15).   The consumption of
HCl in the reaction followed  a uniform progression whereas
increasing embrittleraent  followed a  stepwise  progression.  The
maximum increase  in embrittlement was observed at the 40 volume
percent level.  Longer reaction times increased embrittlement
more effectively  than the higher HCl concentrations.

     The degree of embrittlement decreased  with increasing
initial moisture  levels in the shredded  waste samples studied.

                                26

-------
TABLE 6. RANGES OF PROCESS VARIABLES EVALUATED
     Variable
  Range
  Newsprint samples:
    Temperature, °C  (°F)


    HCl concentration, vol %



    Reaction time, min
  RDF samples:
    Temperature, °C  (°F)


    Moisture content, %



    Reaction time, min
97.8 (208)
122.8 (253)
148.9 (300)

15
25
48
75
1
2
3
4
5
6
154.4  (310)
176.7  (350)
203.9  (399)

10
15
25
40
1
2
3
6
8
                        27

-------
                                                                TABLE 7.  EFFECT OF TEMPERATURE ON SHREDDED NEWSPAPER*
Reactor conditions

Avg.
temp.
[°C (°F)]
97.8 (208)
122.8 (253)
150.6 (303)

Time to
temperature
(min)
8.4
10.0
15.4

Flow rate,
of HC1
[Vmin (cfm)]
17.6 (.62)
17.6 (.62)
17.3 (.62)

Flow rate
of N2
[fc/min (cfm)]
20.4 (.72)
20.4 (.72)
18.9 (.67)

Treatment
time
(min)
3
3
3

Concentration
of HC1
(vol %)
46.3
46.3
48.0
HC1
consumed
in process
q/g+
2.7
2.1
1.8
Product particle size distribution
(cumulative retained)

45 Mesh
(354 microns)
(wt %)
64.5
62.3
60.

100 Mesh
(149 microns)
(wt %)
70.8
69.1
67.6

200 Mesh
(74 microns)
(wt %)
86.9
86.6
86.3


Pan
(wt %)
100
100
100
to
00
                         * Sample size - 1.1-kg (2-1/2-lb) shredded newsprint.
                           The 303°F results are the average of 7 runs.
                           The 253°F results are the average of 3 runs.
                           The 208°F results are the average of 4 runs.
                         + g HCl/g sample
                                                                        TABLE 8. EFFECT OF TEMPERATURE  (RDF)*
Reactor conditions
Avg.
temp.
[°C (°F)]
154.4 (310)
176.7 (350)
203.9 (399)
Time to
temperature
(min)
11.1
12.1
18.9
Flow rate
of HC1
U/min (cfm)]
17.6
17.6
17.6
(.62)
(.62)
(.62)
Flow rate
of N;
[i/min (cfm)]
20.4 (.72)
20.4 (.72)
20.4 (.72)
Treatment
time
(min)
3
3
3
Concentration
of HC1
(vol %)
46.3
46.3
46.3
HC1
consumed
in process
9/9+
.029
.027
.024
Product particle size distribution
(cumulative retained)
45 Mesh
(354 microns)
(wt %)
46.3
45.8
41.3
100 Mesh
(149 microns)
(wt *)
55.9
56.6
51.1
200 Mesh
(74 microns)
(wt »)
75.7
79.0
71.9
Pan
(wt %)
100
100
100
                         * The 399°F results are the average of 3 runs.
                           The 310°F results are the average of 3 runs.
                           The RDF° sample weighed 1.1-kg  (2-1/2 Ib).

-------
                                                            TABLE 9.  EFFECT OF TREATMENT TIME FOR SHREDDED NEWSPRINT*
NJ
10
Reactor conditions
Treatment
tine
(min)
1
2
3
4
5
6
Flow rate
of HC1
Il/min (cfm)J
17.0 (.60)
17.6 (.62)
17.3 (.61)
17.3 (.61)
17.8 (.63)
17.3 (.61)
Flow rate
of N2
IH/min (cfm)]
20.4 (.72)
20.4 (.72)
18.9 (.67)
20.4 (.72)
20.4 (.72)
20.4 (.72)
* Sample size - 1.13-kg
The 1 minute results
The 3 minute results
All remaining results
+ g HCl/g sample



Avg.
temp.
[°C (°F)]
152.8 (307)
152.2 (306)
150.6 (303)
153.3 (308)
153.3 (308)
149.4 (301)
(2-1/2-lb) shredded
are the average of 2
are the average of 7
are the average of 3
TABLE 10
Concentration
of HC1
(vol %)
45.0
46.3
46.0
46.0
46.6
46.0
newsprint.
runs.
runs .
runs.
HC1
consumed
in process
g/g+
.007
.012
.018
.024
.028
.033

. EFFECT OF TREATMENT TIME FOR
Reactor conditions
Treatment
time
(min)
1.0
3.0
6.0
8.0
Flow rate
of HC1
[E/rain (cfm)J
17.6 (.62)
17.6 (.62)
17.6 (.62)
17.6 (.62)
Flow rate
Of NT
[e/min (cfm)J
20.4 (.72)
20.4 (.72)
20.4 (.72)
20.4 (.72)
Avg.
temp.
[°C (°F)]
151.7 (305)
154.4 (310)
158.9 (318)
160.0 (320)
Concentration
Of HC1
(vol %)
46.3
46.3
46.3
46.3
HC1
consumed
in process
9/9+
.011
.029
.037
.039
Product particle size distribution
(cumulative retained)
45 Mesh 100 Mesh 200 Mesh
(354 microns) (149 microns) £74 microns) Pan
(wt %) (wt *) (wt \) (wt %)
71.4 76.7 89.9 100
62.5 64.7 85.7 100
60.0 67.6 86.3 100
55.2 64.8 87.4 100
57.4 67.6 85.9 100
52.2 63.5 84.1 100

(RDF)*
Product particle size distribution
(cumulative retained)
45 Mesh 100 Mesh 200 Mesh
(354 microns) (149 microns) (74 microns) Pan
(wt %! (wt *) (wt %) (wt %)
COULD NOT BE BALL MILLED 100
46.3 55.9 75.7 100
35.6 44.6 65.6 100
35.2 44.8 64.7 100
                              * The 3-min results  are the average of 3 runs.
                                The RDF sample weighed 1.13-kg (2-1/2 Ib).

-------
                                                          TABLE 11. EFFECT OF HC1 CONCENTRATION ON SHREDDED NEWSPRINT
U)
o

Concentration
of HC1
(Vol %)
15
25
48
75
Reactor conditions
Avg. Flow rate Flow rate Treatment
temp. of HC1 of Na Time
I°C (°F)J IH/min (cfm)J [i/min {cfm)] (min)
149.0 (301) 8.5 (.30) 48.0 (1.7) 3
149.0 (301) 8.5 (.30) 25.5 ( .9) 3
150.5 (303) 17.3 (.61) 18.9 ( .67) 3
150.5 (303) 25.5 (.90) 8.5 ( .30) 3
Product particle size distribution
(cumulative retained)
HC1 con- 45 Mesh
sumed in (354 microns!
process [g/g]+ (wt %)
.012
.015
.018
.026
74.5
73.15
60.0
58.8
100 Mesh
(149 microns)
(wt %)
79.1
78.2
67.6
67.2
200 Mesh
(74 microns) Pan
(wt %) (wt %)
90 . 3 100
89.9 100
86.3 100
88.4 100
NOTE: Sample size - 1.13-kg (2-1/2-lb) shredded newsprint.
The 15% results are the average of 2 runs
The 25* results are the average of 2 runs
The 48% results are the average of 7 runs
The 75% results are the average of 3 runs

+ g HC1/ g sample

TABLE 12. EFFECTS OF MOISTURE ON SHREDDED

Moisture
added to
paper (%)
10
15
25
40
Reactor conditions

NEWSPRINT*




Product particle size distribution
(cumulative retained)
Flow rate Flow rate Avg. Concentration Treatment HC1 con-
of HC1 of N2 temp, of HC1 Time sumed in
U/min (cfm)] H/min (cfm)] [°C (°F)] (vol *) (min) process
18.1 (.64) 20.4 (.72) 154.4 (310) 47.0 3
16.6 (.62) 20.4 (.72) 151.1 (305) 46.3 3
17.6 (.62) 20.4 (.72) 156.1 (313) 46.3 3
17.6 (.62) 20.4 (.72) 157.2 (315) 46.3 3
0.025
0.028
0.047
0.056
45 Mesh
(354 microns)
[g/g]+ (wt %)
61.1
64.3
65.5
+
100 Mesh
(149 microns)
(wt %)
68.7
71.6
78.9
+
200 Mesh
(74 microns) Pan
(wt %) (wt %)
86.1 100
89.2 100
96.4 100
+ +
                   * Sample size - 1.1-kg  (2-1/2-lb) shredded newsprint.
                     The 10% moisture results are the average of 5 runs.
                     The 15% moisture results are the average of 3 runs.
                     The 25% moisture results are the average of 2 runs.
  Only 1 run was made at 40% moisture because embrittlement was
    inefficient, and ball milling of this sample was not possible.

+ Could not be ball milled.

-------
•P 50
to
CD
^
LTi
c:
CO
 CD
   40
 o>
 e
 CD
'E 30
 E
 CD
 CD
 1_
 en
 CD
O
37.8°C (100)          93.3°C(200)          I48.9°C(300)

                 Reactor Temperature (°F)

 Figure 11.  Effect of  temperature (shredded newsprint)

-------
          JC
           
           CD
           c 70
           CO
           OJ
           c
              60
00
to
           o>


           CD
50
           -540
           CD
           CD
           CD
           Q
                             148.9°C (300)  176.7°C (350)    204.4°C (400)
                                          Reactor Temperature (°F)
                                    Figure  12.   Effect of  Temperature  (RDF)

-------
          50
       CD

       §
       CO
       rtJ
          40
u>
tO
       CD


       I 30
       CD
       CD
       J_
       cn
       CD
                      Ficjure  13.
       3456.

    Treatment Time (Min)

Effect of treatment time (shredded  newsprint)

-------
oo
       .c
        CO
        CD
       UTN
       ITN

       370
        CD
          60
        £50
        CD
          40
        CD
        CD
        &_
        cn
        CD
                       0        2.0


                            Figure  14.
 3.0      4.0       5.0       6.0
Treatment Time (Mi n)
 Effects of treatment time (RDF)
7.0
8.0

-------
      t/j
      CD
     I
      CD
Cn
      O)

      E
      CD
      CD
      i_
      cn
      CD
         30 -
10
  20       10        "40

Concentration of HCL  in Feed Stream (Vol. %)
                    Figure 15.  Effect of  HCl concentration  (shredded newsprint)

-------
        50
     -C
      «/)
      CD
      c
      TO
      CD
         40
u>
en
      CD
      E
      CD
     -O
      E
     LU
      CD
      CD

      cn
      CD
     Q
         30
         10                   15                   20

       Moisture Added To Paper (Wt. %)

Figure  16.   Effect of moisture (shredded  newsprint)
25

-------
(Table 12, Figure 16).  Moreover, the retention of HCl is
increased, due probably to increased absorption on the moist
material.  The results from this series of tests demonstrates
the need to minimize the moisture content in the cellulose
charge.  Since shredded waste samples with 40% moisture could
not be ball milled, it may be necessary to include provisions
for drying the waste in a large scale operation.

     The results from the series of tests investigating the
effects of sample size showed that when sample size was in-
creased from 1.1 to 2.5 kg (2.5 to 5 Ib) for the same HCl con-
centration and reaction time, the degree of embrittlement
dropped significantly (see Table 13 and,Figure 17).  Consider-
able improvement in the degree of embrittlement was observed
when the reaction time was increased from 3 to 6 min.  When
both the reaction time and the HCl concentration were increased,
the degree of embrittlement was further increased.  These
results show that there is a definite relationship between the
total quantity of HCl required and the amount of sample to be
treated.  A minimum quantity of HCl per kilogram  (pound)  of
sample must be maintained for the desired degree of embrittle-
ment.  The appropriate ratio of HCl to waste will also have to
be maintained for all scale-up designs.

     In summary, acceptable operating results were obtained at
149°C  (300°F), 6-min reaction time, and 50% by volume concen-
tration of HCl in the reactive gas mixture.  From preliminary
tests conducted, it was concluded that the degree of embrittle-
ment could be improved by raising the treatment temperature to
between 177 and 204°C (350 and 400°F) and extending the reaction
time to 8 to 10 min.  At temperatures above these values,
processing problems may be encountered.  From the data obtained
in these series of experiments, it can be concluded that the
degree of embrittlement is maximized at the higher treatment
temperatures, longer reaction times, minimum moisture content,
and moderate levels of HCl concentration.

ANALYSIS OF REACTOR OPERATIONS

     The reactor system shown in Figures 9 and 10 was designed
as a pilot unit for embrittlement studies.  The materials used
for its construction were selected to withstand the anticipated
corrosive environment.  This system was in use for about 10
months and during this period no serious corrosion damage was
observed.  However, a few minor corrosion problems were encoun-
tered with several components in the system.  These problems
are discussed below.

Reactor Coating

     The Teflon-epoxy coating used for  the interior lining of
the reactor was recommended by the manufacturer for use to

                               37

-------
                                                                     TABLE 13.  EFFECTS OF SAMPLE SIZE (RDF)
to
00
Reactor conditions
Height of
sample
(kg tlb}]
1.1 (2.5)
2.3 (5.0)
2.3 (5.0)
2.3 15.0)
Treat.
tine
(min)
3.0
3.0
6.0
6.0
Flow rate
of HC1
[1/min (cfm)]
17.6 (.62)
17.6 (.62)
17.6 (.62)
26.3 (.93)
Flow rate
of N2
[ It /min (cfm)]
20.4 (.72)
20.4 (.72)
20.4 (.72)
10.2 (.36)
Concen-
tration of
HC1
(vol %)
46.3
46.3
46.3
72.1
Avg.
temp.
°C (°F)
154.4 (310)
157.8 (316)
153.9 (309)
150.0 (302)
Time
to temp.
(min)
11.1
10.2
11.3
12.5
HC1
consumed
in process
9/g+
.029
.01
.044
—
Product particle size distribution
(cumulative % retained)
45 Mesh
(354 microns)
(wt »)
46.3
83.5
55.4
32.1
100 Mesh
(140 microns)
(wt »)
55.9
88.3
62.9
42.1
200 Mesh
(74 microns)
(wt %)
75.7
93.4
73.7
63.1
Pan
(wt %)
100
100
100
100
                          +   g UCl/g sample

-------
         70
      c.
      ro
CO
vo
         60
          50
40
      CD
      OJ>
          30
      CD
      CD
      1_
      CD
      CD
      O
20
            Ukg(2.51b)
            3.0 minutes
            46 percent
2.3kg(5.0lb)
3.0 minutes
46 percent
2.3kg(5.0lb)
6.0 minutes
46 percent
2.3kg(5.0lb)
6.0 minutes
72 percent
                   Figure 17.  Effects of sample size  & HC1 concentration (RDF)

-------
260°C (500°F) in corrosive environments.  However, approximately
half the thickness of the coating was lost after 3 months of
operation leaving behind a thin black coating.  Although there
were some localized flaws in the coating, the reactor did not
show any significant corrosion damage.  The Teflon coating used
on the reactor lid lasted only 2 weeks in the HC1 environment
and elevated temperatures.  The lid was used for 7 additional
months after the coating was gone.  Although the bare metal lid
(carbon steel) was completely exposed to the operating environ-
ment, it showed no significant corrosion damage.

Teflon and Silicone Rubber Parts

     Teflon fittings exhibited excellent resistance to the HC1
environment and operating temperatures with no evidence of
corrosion damage.  However, a major problem with the Teflon
components was their low resistance to mechanical stress.  These
fittings had to be periodically replaced because of worn threads
which resulted in leaks in the system.

     Although all the sealing gaskets were made from a high
temperature silicone rubber, they had to be replaced about every
3 weeks.  Examination of these gaskets showed that the HCl
apparently reacted slowly with the rubber in the presence of
condensed moisture causing degradation of the rubber and sub-
sequent leaking.  Gaskets in the shielded area of the reactor
were replaced only once during the entire operating period.

Iron, Aluminum and  Monel Parts

     The cast iron fittings and other metal parts on the out-
side of the reactor showed only a moderate degree of uniform
corrosion.  Cast iron gradually rusts under normal atmospheric
conditions according to the following reaction:

          2Fe + 2H O + O  	** 2Fe(OH)

          2Fe(OH)  + H 0 + 1/2 0 	^2Fe(OH)

This reaction is pH dependent and the rate increases with
decreasing pH.  The condensed moisture on the reactor and
occasional contact of HCl results in an accelerated rate of
corrosion.  The stainless steel, gas-distribution screen at
the bottom of the reactor showed stress cracking due probably
to the interaction of the imposed stress and corrosion reaction.
This screen was replaced after 6 months of service.

     Considerable flaking was observed on the aluminum plate
which was bolted to the front of the stainless  steel cabinet
for instrument installation.  The coupling of two dissxmilar
metals, the occasional contact with HCl and the formation of
a moist dirt film could account for the flaking observed on

                                40

-------
the aluminum plate.  The monel fittings used in the system
showed good resistance to the HC1 environment and no significant
damage to these fittings was observed.

     Most of the typical corrosion problems were avoided in the
pilot unit by the proper selection of materials and operating
procedures.  Reactor operating temperature was maintained above
the dew point most of the time.  In addition, the reactor system
was purged with N2 before the system was shut down.  Using these
operating procedures, the carbon steel reactor provided satis-
factory resistance to corrosion.  Experience with this system
demonstrated that a large scale reactor could also be construct-
ed from carbon steel provided the operating temperature of the
reactor was maintained above the dew point and the system was
always flushed to remove residual HC1.  Monel could be used for
the pipes and fittings, and heating tape could be used to keep
the temperature of pipes and fittings above the dew point.  Tef-
lon gaskets are also recommended for use in a scaled-up system.

EMBRITTLEMENT MECHANISMS

     A number of mechanisms have been considered to account for
the embrittlement observed with the treated cellulose samples.
It is believed that the embrittlement process is not a single
mechanism but rather a complex combination of mechanisms.  Both
a reduction in the degree of polymerization  (DP) and the forma-
tion of cross-links between polymer chains are believed to be
among the dominant factors for cellulose embrittlement in the
presence of a mineral acid.

     The acid-catalyzed hydrolysis of cellulose is a well docu-
mented reaction, and the selection of HC1 was based upon this
fact.  Using standard procedures (TAPPI T-230), the viscosities
of cellulose solutions were measured, and it was confirmed that
a reduction of DP occurs for cellulose samples subjected to the
embrittlement process.  The reaction is shown conventionally in
Figure ISA and in a more cryptic but simpler format in Figure
18B.  It is believed that the water taking part in the acid
hydrolysis reaction is physically adsorbed on the cellulose.
Loss of this water of hydration can also contribute to cellulose
rigidity.

     In an effort to better understand the embrittlement process,
several samples of filter paper were exposed to different em-
brittlement treatment times and then dissolved in cupriethyl-
enediamine.  The viscosities of cupriethylenediamine and cupri-
ethylenediamine solutions with treated and untreated filter
paper were measured.  The results of these tests are compiled  in
Table 14.  The flow times of the solutions are the measure of
solution viscosity used for this analysis.   The cupriethylene-
diamine by itself had a flow time of 9.5 sec and in solution


                               41

-------
                A.
                                    CH2OH
                        H      OH
/°
^H
                                                     Ho
                                                                               CH2OH
it*
to
                 B.
 OH        OH
-1—0—[—®—L-®—p-®—
      OH        OH
        H20
       ~HCI
                                  OH
                                                                            HO-J—0  .  0—
                                                               ;Glycoside  Linkage
                   Figure  18.   Reduction  of the degree of  polymerization  of
                                cellulose  by hydrolysis.

-------
           TABLE 14. COMPARISON OF EMBRITTLEMENT AND
               VISCOSITY OF CUPRIETHYLENEDIAMINE
                      SOLUTIONS OF PAPER
                                            Flow Times (sec)
 Embrxttlement           Sample
                      Designation*
1 Untreated
2 LT
3 MT
4 IT
5 HT
270
22
17
17
12*
116
17
16
17
12+
                     Cupriethylene-
                     diamine reagent             9 . 5
  *
   LT-light   treatment, MT-moderate treatment, IT- intermediate
   treatment, HT-heavy treatment.

   S&S filter paper obtained from the Schleicher & Schuell Co.,
   New York, NY.

  'Incomplete solutions.  Approximately 10-40% solids by volume.
with untreated Whatman and S&S filter paper it had
average flow times of 270 sec and 115 sec respectively -

    A comparison of solution viscosity and degree of embrittle-
ment for various filter paper samples reveals that there is no
simple relationship between these properties.  Very mild treat-
ment with HCl results in a filter paper  (LT) which is only
slightly embrittled yet it yields a solution having a relative
viscosity of approximately 10% of that observed for solutions
of untreated paper.  More extensively treated and embrittled
filter papers on the other hand  (MT and  IT) failed to show
significant additional decrease in viscosity.  Highly embrittled
paper fails to completely dissolve suggesting a high DP, yet the
supernatant solutions in these experiments have viscosities
which approach that of the pure reagent.  Conversations with
scientists at the Forest Products Laboratory in Madison and the
Paper Institute in Appleton, Wisconsin,  had indicated that
overpulping (reduction in DP) should result in paper with
certain undesirable properties, but brittleness was not one of
them.  The results shown in Table 14 appear consistent with
these predictions.  Clearly, the precipitous drop in viscosity
for sample LT compared with untreated material indicates a
large decrease in DP.  Since sample LT is only poorly embrit-
tled,  and because the embrittlement changes markedly for the


                               43

-------
remainder of the samples with little change in viscosity being
observed, it is concluded that acid-catalyzed hydrolysis of
cellulose is not entirely responsible for embrittlement.

    The results of these limited studies were used for formu-
lating a working model for embrittlement.  However, the model
still requires experimental work for confirmation.  It is
believed that the lack of total solubility of the highly em-
brittled sample HT is particularly significant in formulating
a mechanism for embrittlement, and we interpret this result as
an indication of cross-linking between cellulose chains and/or
cellulose oligomers.  Assuming that this cross-linking occurs
via the formation of ether linkages as shown in Figure 19, the
following reactions in the cellulose can be predicted:

    1)  Rigidization with attendant ease of fracture;

    2)  Decrease in solubility as the molecular shape changes
        from a long extended form with readily accessible OH
        groups to a more spherical form with reduced surface
        area and inaccessible OH groups;

    3)  A rapid decrease in molecular weight or DP as the
        reactive glycoside linkages are cleaved, followed
        by a slowing of the decrease in DP as cross-linking
        takes place; and,

    4)  A decrease in the quantity of retained HCl as the
        process temperature is increased.  The ether cross-
        links are thought to be formed by the reaction of
        OH groups with chlorinated cellulose, and the sub-
        sequent loss of HCl.  Higher process temperatures
        should lead to more complete cross-linking and loss
        of HCl.

    Further elucidation of the mechanism for HCl-induced
embrittlement will require sophisticated instrumental analysis
An attractive but more pragmatic alternative would consist of
testing multifunctional cross-linking agents which are less
strongly acidic than HCl, but would be expected to induce
cross-linking under mild conditions.  Reagents such as
isocyanates, esters, silanes, etc., could be used for this
type of test program.
                               44

-------
   OH
   OH
   OH
       OH
        OH
        OH
OH
OH
OH
     OH
     OH
     OH
OH
OH
OH
Cl
OH
Figure 19.  Possible  embrittlement mechanism; hydrolysis  and
            crosslinking of adjacent cellulose chains  by
            oxygen  bridges.
                              45

-------
                            SECTION 5

                      ENGINEERING ANALYSIS


PLANT  DESIGN

     In a  third phase of this project  a  design was developed for
a plant to  process  907  mt (1000  tons) of refuse per day for the
production  of powdered  cellulose for  use as  a fuel.  The flow
plan finally selected is shown in Figure 20.  This plan is
based  on  the use  of commercially available and proven tech-
nology.   The basis  for  the selection  of each processing step
and  item  of equipment is described in the following sections.

Trommel - 12.7 cm (5 in)  Openings

     After the refuse is received and  the oversized bulky wastes
are  removed manually, the solid  waste would  be processed in a
trommel with 12.7 cm (5 in)  diameter  openings.  A trommel was
selected  as the first processing step to reduce the load to the
shredder, remove  a  majority of the glass and metal, reduce
shredder  wear,  and  facilitate metal and glass recovery.

     Trommels or rotary  screens are one  of the major types of
processing  equipment which can be utilized for recovery of the
valuable  components in  MSW.   Although the use of trommels in
the  mining  industry for ore recovery  dates back to the early
1800's, they have been  replaced  in recent years by vibrating
screens.  At the  present time, trommels are  used for sizing
sand and  gravel materials.   More recently, they have been
tried  on  a  limited  scale in some of the solid waste processing
facilities  (8,9,10).

     Trommels are  revolving screens consisting of perforated
cylindrical tubes mounted on drive units.  The trommel is
usually mounted on  an incline with the  feed  introduced at the
upper  end of the  tube.   The material  to be screened then
tumbles down the  tube.   The undersized  material passes through
the  openings in the cylinder wall,  while oversized material
exits  at  the lower  end  of the trommel.  The  material tends to
follow a helical  path through the length of  the tube.  The
operiings  in the trommel can be of a single size or there can
be two or three zones of different sized openings.  In addition,
compound trommels consisting of  two or  more  concentric screening
tubes  on the same axis  can  be used for  multifraction recovery.

                               46

-------
                   1000TPD
REFUSE  RECEIVING
AREA
                                               VERSIZE
                                                BULKY
                                                WASTE
                                      SHREDDER
                                      5" GRATE OPN
                                             SHREDDED
                                            , WASTE
                                         MAGNET
                                         2
                                             NONFERROUS
                                          AIR
                                       CLASSIFIER
1/2"

t
FERROUS
COMPACTO
X
1
V

"^lEAVIES

I



                                             LIGHTS
                                        CYCLONE
                                                   EXHAUST AIR
                               EXTRACTED MOISTURE
          Figure 20.   Process flow plan (English  equivalent)
                            47

-------
                                                    iVERSIZE
                                                     BULKY
                                                     WASTE
      REFUSE RECEIVING
   (-1.3 cm)
AIR
          lit

[
FERROUS
COMPACTOR
X1
1
N
HEAVIES

1



     LIGHTS
CYCLONE
                                                        EXHAUST AIR
                                    EXTRACTED MOISTURE
BALL
MILL

ROT
SCR
(0.3

MILLED
LIGHTS
ARY
EEN
cm)
FUEL J
"\
            Figure 20a.   Process flow plan  (metric).
                                  48

-------
Trommels with two or more zones of different sized openings or
compound trommels may be very effective in concentrating select-
ed components in municipal solid waste.  These units could also
provide a means for separating the combustible fraction, or a
glass-rich or metal-rich fraction.

     The design specification and operating parameters for trom-
mels are dictated by the physical characteristics of the mater-
ial to be screened, the required feed rate, and the separation
requirements.  The primary design parameters include trommel
length, diameter, slope, and aperture size and array to be
utilized.  The primary operating parameters are rotation veloc-
ity and feed rate.  The length of the trommel selected will
determine the retention of the material in the unit and affect
the screening efficiency.  The greater the trommel length, the
more complete will be the removal of undersized material; how-
ever, the majority of screening occurs in the first few meters
(feet) of the unit.  For screening raw MSW it must be remem-
bered that a considerable portion of the material will be in
plastic or paper sacks, and that in the first few feet of travel
the sacks will have to be broken open.  Thus, in the initial few
meters (feet) of processing only a fraction of the undersized
material will come into contact with the openings in the wall
of the trommel.  For most mineral ore dressing applications,
trommel length rarely exceeds 3 to 4.6 m (10 to 15 ft).  For
screening MSW, trommel length selection will be based upon the
nature of the material and the separation efficiency desired.
To date this selection has been based mostly on the trial and
error method.

     The diameter selected for a trommel unit will determine in
large part the capacity of the unit and the thickness of the
bed during processing.  For 'a fixed feed rate, the larger trom-
mel diameter will result in the thinner bed of refuse.  One
correlation developed for determining trommel diameter for given
capacities is described in Taggart (11) where D  (the diameter in
inches) is equal to 7.66 /C/Gs, where C is capacity in tons per
hour and Gs is specific gravity of the feed material.  For MSW,
specific gravity will range from 1 to 1.5.  It must be recog-
nized that this formula was developed for the screening of coal,
sand, and stone type materials and may not be appropriate for
MSW which is not as free flowing a feed material.  In addition,
an initial process required in the processing of MSW is the
breaking open of the plastic and paper sacks, the breaking
apart of agglomerates and the breaking of the large glass and
ceramics.  Trommel diameter and rotating speed must be care-
fully selected to allow the cascading and cataracting actions
to open the bagged and compacted material.

     Based on their experience as a manufacturer of screening
equipment, Triple/S Dynamics  (12) has developed a different


                               49

-------
approach for determining  the  diameter  to capacity relationship
for trommel units processing  MSW.  Their experience has shown
that the best  results  are obtained when the volume of MSW being
processed  is about  25% the available volume of the trommel.  To
calculate  the  desired  diameter,  Triple/S Dynamics assumes an
average velocity through  the  trommel of 4.6 m/min  (15 ft/min).
The resultant  formula  for determining  trommel diameter is then
D = 24 /C/2.12".  Using the formula presented in Taggart and the
approach and assumptions  developed by  Triple/S Dynamics, the
diameter to capacity relationships are compared in Table 15.

               TABLE 15.  COMPARISON OF DIAMETER
                    TO  CAPACITY RELATIONSHIP
         Diameter
          m  (in)
Capacity, [mt/hr (tons/hr)]
 Taggart         Triple/S
•
1.
1.
3.
91
22
83
05
( 36)
( 48)
( 72)
(120)
25
45
99
278
.4
.4
.8
.5
( 28)
( 50)
(110)
(307)
4.
8.
17.
48.
5
2
2
1
( 5)
( 9)
(19)
(53)

     In New  Orleans,  a  3  m (10  ft)  diameter  trommel installed
for processing MSW, has reported a  feed  rate capacity of over
91 mt/hr  (100 ton/hr)  (13) .   The data  from New Orleans show a
capacity  almost  twice that calculated  by the Triple/S approach,
and about one-third that  calculated by the formula in Taggart.

     The  slope of  the trommel employed will  affect the rate at
which the material travels through  the unit.  For a given feed
rate, the greater  the slope  the thinner  the  bed, thus providing
for higher efficiencies of screening.  Taggart states that for
punch-plate  trommels, a slope of 5° is recommended, and for
woven-wire units,  a slightly greater'slope is required.  In New
Orleans,  the slope selected  for the trommel  unit was 5°.
Trommel slopes of  3,  4, and  5°  were evaluated in a study for
the National Center for Resource Recovery (NCRR) and Tennessee
Valley Authority (TVA), and  little  difference was observed
(14,15).  At the University  of  California's  Solid Waste
Processing Laboratory,  Savage and Trezek (16,17) used a trommel
with an inclination of  15° to separate the fine fraction from
shredded  municipal solid  waste.  For a fixed feed rate, trommel
slope and rotation speed  have to be properly selected .to have
effective screening of  MSW materials.  Trommel rotation speed
will also affect both the capacity  and efficiency of the
trommel unit.  As  rotating speed increases,  trommel efficiency
will pass through  a maximum  and then decrease sharply  (11).
Maximum efficiency appears to correspond to  a rotating speed
                                50

-------
which causes the load to ride about two-thirds of the way to
the top of the screen.  The height to which the refuse will
travel along the side of the trommel before it tumbles down
(cataracting and cascading) is determined by the critical speed
for the particular unit.  The critical speed, defined by the
equation vc = 76.6// D , where D is the diameter (in ft) of
the trommel and vc is the velocity at which centrifugal force
on the waste in contact with the shell at the height of its
path equals the force on it due to gravity (18) .  At speeds
greater than v  the waste would be carried around the shell of
the rotating trommel.  Rotating speeds will normally range from
35 to 40% of critical velocity (11).  For a trommel with a
0.9 m (3 ft) diameter critical speed would be 44.22 rpm; for
a trommel with a 1.2 m (4 ft) diameter 38.3 rpm; and for a
1.8 m (6 ft) diameter trommel, 31.2 rpm.  In recent studies
with trommels processing municipal solid waste, rotational
velocities of from 9 to 30 rpm have been reported.

    In addition to length and diameter variations,  trommels can
also be designed with different configurations - cylindrical,
conical, or hexagonal.  Conical units have been used to a con-
siderable extent in gravel washing plants.  Hexagonal trommel
units have been used for screening of fine materials.  In a
study reported by Taggart, a hexagonal unit was compared with a
cylindrical unit and found to be more effective for the screen-
ing of pulp.  The higher efficiency of the hexagonal trommel is
attributed to the greater turnover of material as it flows
through the unit.

    The walls of the trommel can be constructed from wire mesh
or can be punched-metal screens.  Although punched-plate screens
have a longer life and can be prepared with a wide diversity
of aperture designs, they have a lower percentage of open area.
The opening in the woven-wire screen will either be square or
rectangular.   In the punched-plate screen the openings could
be round  (in a straight or staggered array), square, or rec-
tangular  (in a straight or staggered array), or oblong  (in a
straight, staggered, or diagonal array).  Taggart describes in
detail the correlation between aperture design and the max-
imum particle size transmitted  (11).

    Aperture size, spacing, and shape will be dictated by the
fraction identified  for separation and the feed rate required.
Material separation  efficiency is a statistical probability
function based on the number of collisions of undersized
material with the screen openings, and the capacity of the
bed for dry screening.  For coal,  .91 to  1.2 m  (3 to 4  ft) of
screen surface per ton  .9 mt per hr per in  (2.54 cm) of
aperture is used  (11).

    The screening efficiency of trommels  will vary from 70 to
90% depending on the interaction of many  of  the variables

                               51

-------
associated with the process.  Effective screening of MSW re-
quires a comprehensive understanding of the relationship between
the processed variables and municipal solid waste materials.

     The size distribution and shape characteristics of the
material to be treated will vary with the composition of the
waste.  The size range will extend from grains of sand and dirt
to the large bulky items of furniture and household appliances.
The morphology of these components can also vary greatly; how-
ever, the majority of the waste components will be either cylin-
drical, spherical, or platelike in shape.  A comprehensive
understanding of the size distribution, fluctuations, and the
range of shapes typical of the different components in the waste
stream is necessary for the proper design of a trommel unit.

     Several studies investigating the size distribution and
morphology of raw refuse are  reported in the literature.  The
most extensive appears to be  the Ph.D. thesis by J. Ruf  (19).
The cumulative distribution curves and the frequency distri-
bution curves  (Figures 21 and 22) for the particle size of com-
ponents in raw refuse developed by Ruf provides a fairly compre-
hensive evaluation of size distribution.  From the data in these
curves, which appear to be in good agreement with other pub-
lished data, it is possible to determine the average component
split of raw refuse which might be obtained for different
screening processes.  Assuming an optimistic 90% screening
efficiency the distribution which could be obtained for screen-
ing with a trommel having 12.7 cm  (5 in) diameter openings is
shown in Table 16.

Shredder

     The oversized raw refuse from the trommel would be proc-
essed in a coarse shredder with, grate openings spaced at
12.7 cm  (5 in) .  The coarse shredder was selected to provide
size reduction and homogenization of the oversized refuse from
the trommel.

     A wide range of equipment is available for reducing the
size of the received refuse to facilitate more efficient
separation and recovery processes.  The processes employed
include impacting, shearing,  tearing, grinding, milling,
shredding, pulverizing, and flailing.  Shredding is the most
commonly used term and is applied to the mechanical processes
which reduce the size and homogenizes the refuse.  The majority
of processes utilize either a grinding action or an impacting
action to reduce the refuse.  However, both tearing and
shredding are likely to occur in combination with the grinding
or impacting action.  Hammermills and ring grinders are the
most commonly used equipment  for dry size reduction of refuse.
                                52

-------
Ul
u>
                                      Nonferrous
                                         Textiles
                 25(10)
     2.5(1.0)
Particle Size Cm (In)
.25(0.1)
0.5 (.02)
                             Figure 21.  Cumulative distributions of raw waste.

-------
Ul
                             .^Xs*— Ferrous
                        5.1(2)    10.2(4)   15.2(6)   20.3(8)   25.4(10)  30.5(12)   35.6(14) 40.7(16)
                                      Particle  Size Cm (In)
                         Figure  22.   Particle  size distributions of raw  refuse
                                      components.

-------
Ul
Ul
              100
               80
               60

               40
               20
o
UJ
Q_
LU

|

O
o
o
0
8
6
                                                                     • ST.  LOUIS
                                                                     o WILMINGTON
                                                                     n MADISON
                                                                     • HOUSTON
                                                                     A VANCOUVER
                                                           I	I
                                                                    J	I
                                                             J	I
                          0.08         0.24  0.330.480.64  0.95 1.3   1.9 2.5   3.8 5.1
                                           LOG PARTICLE SIZE (CM)
                       Figure 23.  Particle  size  distribution  - shredded MSW*
                                                                     7.6

-------
               TABLE 16. DISTRIBUTION OF WASTE
               COMPONENTS MINUS 12.7 cm (5 in)*
           Components                Distribution (%}

           Cardboard                      10.8

           Paper        '                  45
           Plastic                        55.8

           Textile                        72
           Ferrous                        76.5

           Non-Ferrous                    86.4

           Wood                           76.5
           Food                           76.5

           Yard                           79.2

           Sand and Rock                  87.3

           Glass                          90
           Composite                      55.8


       * Estimated distribution recovered assuming 90%
         screening efficiency.

    Flail mills have also been used to a limited extent for
shredding refuse.  In the flail mill, two sets of articulated
flails on parallel shafts are rotated in opposite directions.
The flails tear the refuse as it passes between the rotors.
This equipment can process large quantities of refuse at very
low power levels.  The low power requirements are due to the
fact that the items difficult to shred  (auto tires, metal
appliances, etc.) are passed through the machine without being
significantly reduced in size.  In this kind of system, refuse
size is difficult to control; and, this hampers subsequent
processing steps.  Because of this inability to effectively
control particle size of the shredded refuse, the flail mill
(a low horsepower shredder) was not selected.

    A horizontal hammermill was selected for the proposed plant.
These mills must be capable of processing both the normal
municipal refuse (residential, commercial, institutional), as
well as the occasional bulky items such as household appliances
(washing machines, refrigerators, etc.), demolition waste
(2x4's, concrete rubble, etc.), automobile tires, water heaters,
tree limbs, furniture, etc.  These primary shredders should
reduce all refuse to less than 12.7 cm  (5 in) with 75% less
than 10.2 cm  (4 in).  The hammermills should also have the
ability to reject  (without being damaged) unshreddable items

                               56

-------
such as hardened steel shafts and gears.

    The particle size distribution of the shredded refuse is
quite variable and will be a function of the refuse composition,
the feed rate, the size of the exit grate in the hammermill and
the weight and speed of the rotating hammers.  Shredded munici-
pal refuse from five different sites was obtained by the
National Center for Resource Recovery (NCRR).  A plot of the
data reported is presented in Figure 23.  In the hammermill,
the metal will be mostly sliced and crumpled, the paper and
wood products shredded, and the glass and ceramics pulverized.
Because of its friable nature, nearly all the glass and ceramic
material will be reduced to particles smaller than 1.3 cm
(0.5 in) and the majority smaller than 0.5 cm (0.31 in).  A
compilation of the glass particles size distribution after
primary shredding is presented in Table 17.

    The particle size distribution data compiled by NCRR
(Figure 23) are based on only several hundred kilograms (pounds)
of samples collected during a one-day test period at each
facility.  Accordingly, more comprehensive data are required
about the particle size distribution of municipal refuse
shredded in a horizontal hammermill.  However, using the data
supplied by NCRR as a guide, it would appear that about 25 to
30% of the shredded refuse is minus 1.3 cm  (1/2 in) and that
the mean particle size is approximately 5.1 cm (2 in).  The
size distribution data supplied by Williams Patent Crusher and
Pulverizer Company for shredded municipal refuse through a
12.7 cm (5 in) cage bar opening is compiled in Table 18.  The
data provided by Williams appears to be in agreement with the
NCRR data.

    In his thesis (19), J. Ruf also investigated the size dis-
tribution of primary shredded waste components.  The cumulative
distribution curve and frequency distribution curve obtained
by Ruf for the range of components in primary shredded solid
waste is shown in Figures 24 and 25.  These data appear to be
in agreement with the other published data.

Magnetic Separator

    The raw refuse passing through the 12.7 cm (5 in) openings
in the trommel and the shredded refuse would then be processed
at a magnetic separator for ferrous recovery.  Two types of
magnetic separators are commonly used in these applications:
an overhead suspended belt magnet, and a drum magnet.  The
magnet manufacturers recommend that when either of the magnets
is used, it be suspended over the discharge of the conveyor
within 20 to 30 cm  (8  to 12 in) of the flowing refuse.  They
further recomment that the refuse conveyor  be traveling at
speeds up to 120 m/min  (400 ft/min) in order to achieve a
minimal bed depth 2.5  to 7.5 cm  (1-3 in).   A cleaner  ferrous

                               57

-------
          TABLE 17. TYPICAL GLASS PARTICLE
                 SIZE DISTRIBUTION*
        Opening size            Cumulative %
               (in)               passing
2.5
1.9
1.3
0.9
0.6
0.5
0.4
0.2
0.1
0.07
0.05
(1)
(3/4)
d/2)
(3/8)
d/4)
(3/16)
(0.14)
(0.06)
(0.04)
(0.03)
(0.02)
100.0
99.5
98.5
91.1
76.9
64.8
54.3
26.4
20.6
15.3
10.0

* NCRR private communication (1974).
          TABLE 18. SIZE DISTRIBUTION WITH
                  A WILLIAMS MILL*
        Opening size            Cumulative %
         cm    (in)               passing
15.2
12.7
10.2
7.6
5.1
3.8
2.5
1.9
1.3
0.9
0.6
(6)
(5)
(4)
(3)
(2)
(1-1/2)
(1)
(3/4)
(1/2)
(3/8)
(1/4)
90
80
75
65
52
44
32
27
20
15
13
   Private communications from Williams Patent
   Crusher and Pulverizer Company  (1974).
                          58

-------
                  100
vo
                           Ferrous

                           Miscelloneous
                            25.4(10)              2.54(1)
                                       Particle Size Cm (In)
0.25(0.1)
                       Figure 24.   Cumulative  distributions of shredded waste,

-------
a\
o
                   73.7
                 (187/2)
               c
               E
              o
               c:
               CD
               13
               o-
               o>
               CD
               on
                                              Cqrdboqrd x Ferrous
                               2.5(1)   5.1(2)   7.6(3)  10.2(4)  12.7(5)  15.2(6)   17.8(7)
                                               Particle Size Cm (In)
                          Figure  25.  Particle size  distributions of shredded waste,

-------
fraction has been reported for the self-cleaning belt magnet
unit compared with the drum magnet unit.  However, the belt
wear is quite high, resulting in excessive maintenance costs.
Drum magnets require little maintenance, except for the jamming
problems which can be caused by springs and other bulk metal
catching between the conveyor and drum.

     Although the amount of ferrous metal in the refuse is
variable, an average value of 7.5% has been established for the
ferrous fraction.  It is anticipated that about 80 to 85% of
the ferrous metal is removed in the initial magnetic separation
phase.  The major components of the ferrous fraction are the
tin-coated cans, the bimetal tin-coated cans, and the tin-free
bimetal cans.  These cans represent 60 to 90% of the ferrous
fraction.  Wire; bottle and jar caps; metal ends from paper
containers; metal from cookware, hardware, and appliances; and
ferrous metal furniture are the other types of ferrous scrap
found in refuse.

     The Bureau of Mines (20)  sampled gome 2.7 mt (6000 Ib)  of
municipal refuse and reported the distribution as shown in
Table 19 for the ferrous fraction.  In the Bureau of Mines
magnetic process, about 3.5% paper was found trapped with the
recovered metal.  This does not include the paper labels on the
cans nor the food wastes and other contaminants trapped in the
shredded cans.  Total contaminants in the recovered ferrous
fraction are likely to range from 5 to 15%.

     The ferrous fraction from the primary magnet is conveyed
to a cleaning magnet.  This final magnetic separation stage is
designed to provide a cleaner ferrous fraction for recovery.
The material is- conveyed to the drum and the nonferrous mater-
ials drop past the drum to an outfeed conveyor.  An aspiration
system can be incorporated with the drum magnet as an option
to pull off the light nonferrous fraction.  The ferrous frac-
tion picked up by the magnet is conveyed to a ferrous compactor
for densification (minimill).   Ferrous metal having a density
of 561 kg/m3  (35 lb/ft3) is compacted to about 1,202 kg/m3
(75 lb/ft3) for transport-to a ferrous processing plant.

Trommel - 1.3 cm (1/2 in) Openings

     The nonmagnetic material from the magnetic separator
processing the minus 12.7 cm (5 in) material from the trommel
will be processed in a trommel with 1.3 cm (1/2 in) diameter
openings.  The fine glass, ceramic, and stone particles tend
to get carried with the air-classified light fraction that is
to be used for fuel.  These particles, when in the fuel frac-
tion, can cause erosion problems in the pneumatic transport
system and add to the ash and slagging problems in the boiler.
It is, therefore, desirable to remove these fines from the
refuse.

                              61

-------
              TABLE 19. DISTRIBUTION OF MATERIALS
                   IN THE FERROUS FRACTION*
                                              Percent
          Tin-coated cans                       56

          Bimetal tin-coated cans               17

          Bimetal tin-free cans                 12

          Bottle and jar caps                    2

          Metal ends from paper containers       1-1/4

          Miscellaneous light metal
          Cpins, nails, wire, etc.)              7-3/4

          Miscellaneous heavy metal
          (appliances, furniture, etc.)          4


    * Ferrous fraction approximately 7.5% of raw refuse.

     A trommel 9.1 to 10.7 m  C30 to 35 ft) in length with
1.3 cm (1/2 in) diameter openings should facilitate the removal
of the fines in the unshredded refuse-  A vibrating flat screen
could also be used for this application; however, it is more
susceptible to jamming.

     An estimated compilation of the waste component size
distribution for the minus 1.3 cm  (.1/2 in) fraction that can
be obtained from the raw refuse is presented in Table^20.
Assuming 90% screening efficiency the distribution which
could be obtained for screening with a trommel having 1.3 cm
(1/2 in)  diameter openings is shown in Table 20.*

Air Classification

     The oversized refuse from the trommel, the nonferrous from
the shredder and from the cleaning magnet are to be conveyed
to the air classification system for recovery of most of the
organic materials.
     Air classification is the term used to describe a number of
processes which utilize gravity and air currents for material
separation.  The three main  factors that affect the separation
of particles in an air stream are particle size, shape, and
specific gravity.  Vertical, horizontal, and  inclined column
separator designs have been  developed which utilize these
parameters to separate the /different fractions in solid waste
(21) .  Component density is  the major factor  for component

   Tables 21 and 22 are based on data in Figures 24 and 25.
*  j.^,^	

                               62

-------
separation by gravitational processes.  A compilation of waste
component densities is presented in Table 21.

            TABLE 20. CUMULATIVE SIZE DISTRIBUTION
                      OF WASTE COMPONENTS MINUS
                      1.3 cm (1/2 in)
          Components
        Distribution for
         raw refuse (%)
Cardboard
Garden
Paper
Food
Plastic
Ferrous
Textile
Nonferrous
Wood
Glass
Sand and Rock
Miscellaneous
Composite
Negligible
57
Negligible
48
1
Negligible
1
4
6
1
79
—
6
(0)
(51)
(0)
(43)
(.9)
(0)
(.9)
(3.6)
(5)
(.9)
(71)
—


   NOTE:  Number in parentheses is the estimated screening
          distribution assumint 90% efficiency.
              TABLE 21.  WASTE COMPONENT DENSITIES
          Component
       Density
   [kg/m3 (lb/ft3)]
Paper
Plastic
Wood
Aluminum
Iron
Rubber and Leather
Food and Yard
Wastes
Other Nonferrous
Metal
705-1201.5
897-1602
304-897
2675
7866
993-2002

721

8811
(44-75)
(56-100)
(19-56)
(167)
(491)
(62-125)

(45)

(550)
          Glass
2307-2996   (144-187)
                               63

-------
      To suspend a particle in an air stream,  the  drag must equal
     weight:

                         Weight = Drag

                     V P_g = i PaCnv?So
                      s s    z  a D a s
where

      Vg = Volume of the material
      Ps = Density of the material

      Ss = Surface area of the material  exposed  to the
           air stream (lift area)
      Pa = Density of the air

      CD = Drag coefficient for the material
      va = Air stream velocity
       g = Gravitational force.

By  rearrangement:
                        2
                              P v
                               s s
                      2g   ~   CDSS

 If we assume  that most  shredded waste  is  "plate-like" then:

                        2
                     P v      p t
                      ad   	   S S
                     ~^
-------
                  TABLE 22. p t  FOR WASTE COMPONENTS
                           s s
                 p  Density
  Component     [g/cm3 (lb/in3)]
     ts

Average thickness
   [cm (in)]
  P t
   s s
[kg/m
Glass
Paper
Leather
Food and yard
was tes
Plastics
Wood
Aluminum
Iron
2
0
1

0

0
2
7
.5
.95
.4

.7
.12
.6
.7
.9
(0
(0
(0

(0
(0
(0
(0
(0
.0897)
.0345)
.0521)

.0260)
.0434)
.0231)
.0966)
.2841)
31.1
4.1
16.0

16.0
4.1
63.5
7.9
7.9
(0
(0
(0

(0
(0
(0
(0
(0
.125)
.016)-
.063)

.063)
.016)
.250)
.031)
.031)
7
0
2

1
0
4
2
6
.9
.4
.3

.1
.5
.1
.1
.2
(0
(0
(0

(0
(0
(0
(0
(0
.0112)
.0006)
.0033)

.0016)
.0007)
.0058)
.0030)
.0088)

for waste components are very limited or almost nonexistent.

     Both horizontal and vertical classifiers are available for
the dry separation of refuse.  In the vertical units the upward
flow of air in the enclosed column picks up and carries the
lower density materials as it contacts the downward falling
refuse.  Air stream velocity will dictate the separation ratio
(light versus heavies).  This separation is predominantly a
function of the density and surface area of the shredded refuse
particles.  To compensate for the day-to-day variations in
particle moisture content and morphology, a variable air veloc-
ity control is necessary.

     At the present time two types of vertical classification
units are on the market.  One of the systems incorporates a
zig-zag design in which the bulk shredded refuse is introduced
into a vertical air stream as it tumbles down the zig-zag
baffles.  Each zig-zag provides an additional point of decision
for the separation since each zig-zag is a point of turbulence
in the air stream.  The tumbling action at these points causes
the agglomerated materials to be broken up.  The second type
of unit is essentially a vertical column in which the shredded
refuse is injected midway rthto the column, which has an upward
air stream.  The column can be rectangular or cylindrical and
it may contain baffles that can be tilted.  These vertical
column units require more closely sized refuse than do the  zig-
zag units.  Manufacturer specifications require that the refuse
be minus 3.2 cm (1-1/2 in) in size.

     Several types of inclined and horizontal air classifiers
have also been developed.  In the horizontal unit the shredded
refuse is introduced  into a horizontal air stream and the refuse
particles, as a function of their weight and shape, will drop
                                65

-------
out of the air stream at different points along the flow path.
In most designs, bins are placed under the drop zones, and the
very light material is carried out of the classifier with the
air stream.  In the horizontal system, two to five separate
fractions can be recovered depending on the design.  The refuse
can be introduced into the classifier by dropping it past a
horizontal air stream or with a ballistic wheel which hurls the
refuse into the horizontal air stream.  Designs employing a
rotating inclined cylinder have also been developed to classify
municipal refuse.  In these systems, lifters within the cylinder
cause the refuse to cascade down the tube as an upward air flow
carries the lights out of the unit, while the heavy fraction is
discharged out of the bottom of the cylinder.

     Another type of horizontal unit is a vibro-aspirator based
on the modification of a gravity or air table.  In a vibro-
aspirator the shredded refuse is conveyed down a series of
stepped vibrating pans.  At each step, air passes through the
material carrying off the light particles.

     The light fraction from the air classifier is pneumatically
conveyed to a cyclone for deairing.  A cyclone is primarily a
settling chamber in which gravitational acceleration is replaced
by centrifugal acceleration.  (Cyclone design is based on the
vortex principle.)  The solid gas mixture enters tangentially
near the top and is forced down in an ever-decreasing spiral to
the solids outlet trap.  The solid material drops out of the gas
stream in increasing quantities as the apex of the cone of the
cyclone is approached.  The gas forms a vortex in the center and
travels upward into the air outlet.  The finer dust particles
are usually carried out with the gas ;and a dust filter is
usually required.  The light fraction would then be dried and
prepared for embrittlement.         -r-'
                                     j /
Drying

     The air classified light fraction is to be dried prior to
the embrittlement treatment.  The moisture content of municipal
refuse is quite variable.  Fluctuations between 15 and 50
weight percent of the refuse have been reported with the aver-
age being about 30.  The air classified light fraction has a
moisture content ranging from 20 to 25%.  This high moisture
content inhibits refuse separation processes and reduces BTU
content.  For the acid embrittlement process with HC1 it is
desirable to minimize the moisture content in the waste.
Furthermore, the large variation in moisture content from day
to day adversely affects the commercial marketability of the
organic fraction for use as a fuel.  It, therefore, would be
desirable to dry the refuse as part of the resource recovery
processing.
                               66

-------
     A number of different drying processes were investigated
to determine their applicability for use with refuse.  Freeze
drying, vacuum drying, chemical drying, hot gas drying, and
steam drying processes were considered.  Inquiries quickly
showed that both freeze drying and vacuum drying are prohibit-
ively expensive.  Several different chemical drying techniques
have been developed in the laboratory; however, they are not
commercially available for large facilities due to their high
cost.  Dessicants and chemicals forming a hydraulic reaction
with water have been used for moisture absorption and removal
in the laboratory; however, they are expensive and would be
difficult to separate from the refuse after they are employed.
It would appear that for the processes considered, hot gas
drying or steam heating are the most practical means for
removing the moisture from the refuse  (22).

     Both direct and indirect dryers can be used for moisture
removal from the waste.  Indirect heat dryers heat the waste
by radiation while direct heat dryers bring the waste into
direct contact with a drying medium.  A drying medium can be
hot air, steam, or hot exhaust gases from a combustion process.
For this proposed plan an indirect heated rotary dryer using the
hot flue gases coming from an incineration unit would be used.
In addition some of the hot flue gases would be diverted to a
heat exchanger to heat the HCl going to the reactor.

Incineration and Flue Gas Cleaning  (Electrostatic Precipitators)

     The heavy fraction from the air classifier, the minus
1.3 cm  (1/2 in) fraction from the trommel and the oversized
[+0.3 cm (+1/8 in)] from the rotary screen are to be inciner-
ated.  The hot flue gases from the incinerator will be used
as the heat source for the rotary dryers.

     Incineration is the general term applied to a variety of
combustion processes used for volume reduction of wastes.  The
basic components of an incinerator are shown schematically in
Figure 26.  Incinerators operate on a continuous and/or periodic
batch basis.  Continuous feed incinerators, e.g., the traveling-
grate, reciprocating-grate, ram-feed, and rotary-kiln are more
commonly used for incineration of municipal waste.  Instead of
quenching the flue gases as shown in Figure 26, the heat can be
recovered for steam generation.  A water wall system can be
used in place of refractories in the combustion chember, or a
boiler unit can be located after the combustion chamber for
heat recovery.

     In most of the systems described  in the literature, the
refuse is received in a storage pit and transferred to the in-
cinerator by an overheated crane and clamshell grab.  The crane
and grab can be operated at the crane  or from a remote station.

                                67

-------
                                 FURNACE FLUE GASES
Ol
00
*TT, _ FURNACE
/co 4.x COMBU£
(5.8mt) CHAM£
1
REFUSE
(.9mt)
>TION
JER
T
FURNACE
CHARGING
i
REFUSE
(.9mt)

REFUSE- 	 » UNLOADING
(.9ltlt) STORAGE

FLY ASH,1093°C)




EXCESS W/
(l.Smt
WATER ^ 	
RECYCLE
V
V
RESIDUE
(O.lSmt DRYJ

GAS
QUENCH
WET OR DRY
^TER
)
AI
FLUE GASES ^ POLLU
(315. 6°C) CONT
WET 0
PURGE TO SEWER OR
SETTLING TANK
WATER
TREATMENT
i
JASTE
JATER

RESIDUE
QUENCH
i
i
RE
(0.
WATER FLY ASH
fc O
I
FLY ASH
(8.2kg)
• WATFTC ST1
DISPI
i 1
SIDUE FLl
18mt DRY) (0.9
R
TION
ROL
R DRY
,

^CK
SRSAL
JE
1kg)
           NOTE:  Quantities- in parentheses are rough measures of flow  rates and temperatures,
                      Figure 26a.   Basic Components  of an Incinerator  - metric units.

-------
             AIR—
        (6.4 TONS)
   FURNACE
  COMBUSTION
   CHAMBER
                      REFUSE
                     (1 TON)
                                 FURNACE FLUE GASES
                                  (7.2 TONS,20 Ibm,
                                 FLY ASH,2000F)
                         FURNACE
                         CHARGING
en
vo
 REFUSE
(1  TON)
           REFUSE
          (1 TON)
    GAS
   QUENCH
                                                  WET OR DRY
                    EXCESS WATER
                       K2 TONS)
                  WATER,
                RECYCLE
FLUE GASES(600F^
   AIR
POLLUTION
 CONTROL
WET OR DRY
                                                       PURGE TO SEWER OR
                                                       SETTLING TANK
    WATER
 TREATMENT
WASTE
WATER
                                       RESIDUE
                                     (0.2 TONS DRY;
                                                                WATER FLY ASH
                                                                    T
     FLY ASH
    (18  Ibm)
                               RESIDUE
                               QUENCH
                                                                  • WATER
                               STACK
                             DISPERSAL
                                                     RESIDUE
                                                   [0.2 TONS DRY)
                                                           FLUE
                                                         (2 Ibm ASH)
           Figure 26.    Basic components of an incinerator.   (Note:   Quantities  in
                         parentheses are rough  measures of  flow rates  and temperatures.)
                         english  units

-------
The storage pit or bunker normally provides several days storage
capacity for the facility.  The grab deposits the refuse into
the furnace feed or charging chute, and then the refuse is nor-
mally gravity fed into 'the furnace.  The charging hoppers are
usually water cooled and are always kept filled to capacity in
order to prevent hot gases from entering the refuse storage
area.  Most incinerators are operated under negative pressure
to prevent the escape of hot gases and odors from the system.

     A variety of furnace designs have been developed for
burning refuse.  Most of the units are designed for mass
burning the raw refuse, and no special processing is required.
The\refuse is conveyed through the furnace by some type of
stoker system, e.g., traveling-grate, chain grate, reciprocating
stoker, stepped grate, shaking stoker, rocker grate, rotating
grate stoker  (rotating barrels) and rotary drum (rotary kiln).
The refuse is dried and burned on the stoker unit, with air
being introduced from both under the stoker and over the refuse.
In water wall units a minimum of excess air is required.  The
residue from combustion is normally carried by the stoker to a
water quencher.  The quench pit can serve as a water seal to
the furnace.

     The stoker unit is not only required to convey the refuse
through the furnace but also to agitate the refuse, permitting
more complete combustion.  Several furnace designs include
auxiliary burners to assist in combustion of the refuse.  This
may be required in cases where wet refuse is frequently
encountered.

     The residue from refuse incineration is usually a wet
complex mixture of metal, glass, slag, charred and unburned
organics and ash.  In some incinerator units the residue is
collected dry.  This residue  (.wet or dry) is usually disposed
of in a landfill.  In some units the residue is passed through
a rotary screen which separates the metal from the ash; the
metal is recovered, and the ash is used as a road bed fill.
The United States Bureau of Mines and the Raytheon Company
have developed a process  (23) which recoveres metal, and glass
from the incinerator residue.  However, this system has not
been commercially proven.

     The hot gases leaving the furnace chamber contain the
products of combustion, particulates, and other gases released
by thermal decomposition of the raw refuse.  These hot gases
can be passed through a boiler unit before being treated in
a gas cleaning unit.  Air pollution codes require the removal
of particulates from the gas before they can be released to
the atmosphere.  Although a number of different types of
particulate removal systems are availabe, electrostatic
precipitators are reported to be the most effective.  Reports
to date indicate that the precipitator units achieve 95 to 99%

                               70

-------
efficiency, and meet the required EPA performance levels.  The
fly ash collected is usually added to the incinerator residue
for disposal.

     In an electrostatic precipitator, solid particles in the
flue gas are electrostatically charged by a high-voltage dis-
charge and the gas is passed through a high voltage electro-
static field where the solid particles are attracted to a
negatively charged collection surface.  The collected particles
are either shaken or hosed down from the collecting surface
periodically.  Optimum operating temperatures for the precip-
itators are between 232 arid 288°C (450 and 550°F).   Since the
flue gases from the incinerator can average from 649 to 760°C
(1200 to 1400°F) it is necessary to cool the gases before gas
cleaning.  For the flow plan proposed the hot gases will be
cooled by passing them through a rotary dryer unit.

Reactor/Heat Exchanger/Scrubber and Caustic Tank

     The dried lights at an approximate temperature of 149°C
(300°F) will be transported a short distance to a rotary reactor
for embrittlement treatment with heated HCl.  The HCl gas will
be heated to 149°C (300°F) in the heat exchanger by a portion
of the hot incinerator flue gases.  To minimize corrosion
problems and to facilitate the embrittlement process, it is
desirable to maintain a processing temperature of 149°C  (300°F).
It is anticipated that the light fraction will be retained in
the reactor for about five minutes.  The treated refuse will
then be conveyed to a ball mill and the process gas recycled
for reuse.  It is estimated that no more than 2% of the HCl
would be retained with the treated waste.  For the proposed
embrittlement treatment a reactant gas mixture of 50% HCl/50%
air by volume is assumed.  However, to minimize the potential
for fire or explosion it may be necessary to consider the use
of an oxygen deficient gas (flue gas from the precipitator)
as the carrier medium for the HCl.  The recycled gas plus the
make-up gas would be heated in a heat exchanger using hot flue
gases from the incinerator.  A conventional heat exchanger of
the shell and tube type could be used for this application.
As backup, if the HCl cannot be recycled, a scrubber unit with
caustic solution is proposed.

Ball Mill and Rotary _Disc Screen

     The embrittled light fraction would be transported a
short distance to a ball mill for fine grinding and then to a
rotary disc screen to remove the oversized fraction.  Ball
mills are horizontal cylinders with a diameter about equal to
their length.  The mills are about half-filled with quartz
pebbles or iron balls.  The horizontal rotation of the mill
causes the balls and the charge to cascade from the side wall
back to the center of the mill.  The action of the balls

                               71

-------
rolling over each other cause the friable charge to be ground
into a powder.  A continuous feed process can be used with the
retention time controlled by the feed rate.  A rotary disc
screen is proposed for screening the discharge from the ball
mill because of its effectiveness in classifying shredded waste.
The oversized fraction would be conveyed to the incinerator for
combustion.  The powdered fraction would be conveyed to a
storage bin and then transported to the fuel customer.

MATERIAL AND ENERGY BALANCE

     Using the flow plan developed, a detailed material balance
and energy balance were developed.  The data developed were
based on a plant processing 907 mt  (1000 tons) of raw refuse
per day.  Also, it was assumed that a landfill was available
to serve as a backup to this facility.  The landfill would
provide disposal capability for waste deliveries in excess of
907 mt/day (1000 tons/day), for residue from the plant, and for
those times when the plant or part of the process was down for
repairs.  It should be recognized that the projected distribu-
tion for the refuse components at each stage of the flow is
based on that expected for an average refuse system.  The
fluctuations in refuse due to seasonal changes and the usual
daily variations result in moisture and compositional changes
that can alter the behavior of the waste products at each stage
of the resource recovery process.  The material balance pre-
sented in Table 23  (metric units) and Table 24  (English units)
and shown in Figure 27  (metric units) and Figure 28  (English
units) was based on data collected from University of Dayton
studies  (22), the Bureau of Mines pilot studies  (24), the
St. Louis Demonstration Plant  (25), NCRR reports (26), the
Ralph M. Parsons Company report  (27) , the thesis by J. Ruf  (19) ,
and from manufacturers of resource recovery equipment.  The
starting values used for the different refuse components were
taken from the data compiled in the thesis by J. Ruf.  The
values specified were used as a model for simulating the
resource recovery processes defined in the flow plan.  The
energy balance developed is presented in Figure 29  (metric
units) and Figure 30  (English units).

     As shown in Figures 29 and 30 the energy efficiency of the
proposed facility is 60.1%.  This calculation does not include
the energy that would be saved by recycling of ferrous metal.
A facility replacing virgin iron ore with 60.3 mt  (66.5 tons)
of ferrous scrap a day would save approximately 2033.6 mKj
(1,928 million BTU)  Inclusion of the energy conserved by
recycling the ferrous scrap would give the resource recovery
facility an energy efficiency of 82%.
                                72

-------
TABLE 23. MATERIAL BALANCE, METRIC TONS
Process
Trommel
(12.7-cm opn)
+12.7-cm
-12.7-cm
Shredder
Magnet1
Ferrous
Nonferrous
Magnet2
Ferrous
Nonferrous
Cleaning magnet
Ferrous

Nonferrous
Trommel
(1.3-cm opn)
+1.3-cm
-1.3-cm
Air classifier
Heavies
Lights
Dryer
Moisture
Solids
Reactor
Hcl pick-up
Ball mill
Paper and
paper
products

453.
299.
154.
299.
154.
1.
152.
299.
3.
295.
5.
0.

4.

152.
152.
—
452.
35.
417.
417.
59
358.
358.
366
366

5
3
2
3
2
8
4
3
6
7
4
9

5

4
4

6
0
6
6

6
6


Plastics

40
20
20
20
20
0
20
20
0
19
1
-

1

20
19
0
40
3
36
36
4
32
32
32
32

.9
.0
.9
.0
.9
.9
.0
.0
.9
.1
.8
-

.8

.0
.a
.2
.6
.9
.7
.7
.5
.2
.2
.6
.6
Textiles

13.6
4.5
. 9.1
4.5
9.0
0.9
8.1
4.5
.9
3.6
1.8
—

1 8

8.2
8.1
.1
13.5
2.2
11.3
11.3
1.8
9.5
9.5
10.7
10.7
wooa

54
13
40
13
40
-
40
14
-
14
-
-

-

41
39
1
52
25
27
27
3
23
23
25
25

.4
.6
.8
.6
.8
-
.8

-

-
-

-


.2
.8
.6
.4
.2
.2
.6
.6
.6
.2
.2
Food and
garden
wastes

163
45
118
45
117
3
114
45
1
43
5
1

3

114
62
51
109
22
87
87
45
41
41
42
42

.3
.4

.4
.9
.6
.4
.4
.8
.6
.4
.8

.6

.1
.5
.6
.7
.7

.1
.4
.7
.7
.6
.6
Glass

63.5
4.5
59.0
4.5
59.0
—
59.0
4.5
—
4.5
—
—

—

59
58.5
.5
63.1
62.6
.5
0.5
—
0.5
0.5
0.5
0.5
Sand
and
rock

40
4
36
4
36
-
36
4
-
4
-
-

-

36
11
24
16
11
5
5

•1
4
4
4

.8
.5
.3
.5
.3
-
.3
.5
-
.5
-
-

-

.3
.8
.5
.3
.3


.9
.1
.1
.1
.1
Ferrous
metal

68
18
50

.0
.0

18.0
49
41
9
18
15
2
56
56

-

9
8

11
9
2
2

1
1
1
1
.9


.1
.4
.7
.0
.0

-

.0
.9
.1
.5
.2
.3
.2
.5
.7
.7
.7
.7
Non-
ferrous
metal

9
1
7
1
7
2
5
1

1
2
1

0

5
5

7
6
1
1
1
1
1
1
1

.0
.8
.2
.8
.2
.0
.2
.8
.4
.4
.3
.4

.9

.4
.2
.2
.5

.5
.5
.5
.5
.5
.5
.5
Total

907
412
495
412
495
SO
445
411
23
388
72
60

12

445
366
79
767
178
589
589
115
473
473
484
484








.8

.8
.7
.1

.6




.4
.3
.1
.1
.7
.4
.4
.9
.9
Destination

Entering
To shredder
To magnet1
To magnet
Entering
To cleaning magnet
To 1/2" trommel
Entering
To cleaning magnet
To air classifier
Entering
To ferrous compactor
then ferrous bin
To air classifier

Entering
To air classifier
To incinerator
Entering
To incinerator
To dryer
Entering
Exhausted
To reactor
Entering
To ball mill
To rotary screen

-------
TABLE 23. MATERIAL BALANCE, METRIC TONS  (Concluded)
Paper and
paper
Process products
Rotary screen
40 . 3-cm
-0 . 3-cm
Incinerator
Moisture
Ash
Organics
Fly ash
Bottom ash
Combustible organics
and HC1 (2.2)
K Joules x 10
366
36.6
329.4
71.6
7.3
2.4
61.9
1.2
3.3

59.8
1075.9
Plastics
32.6
16.3
16.3
20.5
1.4
2.0
17.0
--
2.1

17.0
453.6
Textiles
10.7
2.17
8.53
4.4
0.9
.1
3.9
0.2
0.5

2.7
61.2
Wood
25.2
6.3
18.9
33.5
4.5
.9
28.1
0.9
1.3

26.8
501.0
Food and
garden
wastes
42.6
19.9
22.7
94.3
47.2
1.2
46
0.9
3

43.2
622.3
Glass
0.5
0.23
0.23
63.3
1.8
61.5
—
3.6
57.9

r 	
12.7
Sand
and
rock
4.1
2.3
1.8
38.1
0.9
37.2
—
2.1
34.5

~—
7.4
Ferrous
metal
1.7
1.9
0.8
10.4
0.2
10.2
—
0.5
9.8

--
17.9
Non-
ferrous
metal
1.5
1.9
.6
7.1
0.2
6.9
--
0.5
6.4

~~
11.6
Total
484.9
85.6
399.3
342.9
64.4
122.3
156.4
10.5
118.8


2763.6
Destination
Entering
To incinerator
To fuel storage
Entering









-------
TABLE 24. MATERIAL BALANCE, ENGLISH EQUIVALENTS
Process
Trommel
(5-in opn)
+5-in
-5-in
Shredder
Magnet1
Ferrous
Nonferrous
Magnet2
Ferrous
Nonferrous
Cleaning magnet
Ferrous

Nonferrous
Trommel
(1/2-in opn)
+l/2-in
-1/2-in
Air classifier
Heavies
Lights
Paper and
paper
products

500
330
170
330
170
2
168
330
4
326
6
1

5

168
168
—
499
38.6
460.4
Plastics

45
22
23
22
23
1
22
22
1
21
2
—

2

22
21.8
0.2
44.8
4.3
40.5
Textiles

15
5
10
5
10
1
9
5
1
4
2
2

—

9
8.9
0.1
14.9
2.4
12.5
Wood

60
15
45
15
45
...
45
15
—
15
—
—

—

45
43
2
58
28
30
Food and
garden
wastes

180
50
130
50
130
4
126
50
2
48
6
2

4

126
69
57
121
25
96
Glass

70
5
65
5
65
—
65
5
~
5
—
—

—

65
64.5
0.5
69.5
69.0
0.5
Sand
and
rock

45
5
40
5
40
—
40
5
—
5
—
—

—

40
13
27
18
12.5
5.5
Ferrous
metal

75
20
55
20
55
45
10
20
17
3
62
62

—

10
9.9
0.1
12.9
10.5
2.4
Non-
ferrous
metal

10
2
8
2
8
2
6
2
0.5
1.5
2.5
1.5

1

6
5.8
0.2
8.3
6.6
1.7
Total

1000
454
546
454
546
55
491
454
22.5
428.5
80.5
66.5

14

491
403.9
87.1
846.4
196.9
649.5
Destination Process notes

Entering (4,500 BTU/lb)
To shredder
To magnet1
To magnet^ •
Entering
To cleaning magnet
To 1/2" trommel
Entering
To cleaning magnet
To air classifier
Entering
To ferrous compactor
then to ferrous bin
To air classifier

Entering
To air classifier
To incinerator
Entering
To incinerator
To dryer

-------
TABLE 24. MATERIAL BALANCE, ENGLISH EQUIVALENTS (Concluded)
Process
Dryer
Moisture
Solids
Reactor
HC1 pick-up
Ball mill
Rotary screen
+1/8- in
-1/8-in
Incinerator


Moisture
Ash
Organics
Fly ash


Bottom ash

Combustibles , organics
& HC1 (2.2)
BTlI's x 106


Paper and
paper
products
460.4
65
395.4
395.4
403.5
403.5
403.5
40.3
363.2
78.9


8
2.6
68.3
1.3


3.6


66
1020


Plastics
40.5
5
35.5
35.5
36
36
36
18
18
22.5


1.5
2.3
18.7
—


2.3


18.7
430


Textiles
12.5
2
10.5
10.5
11.75
11.75
11.75
2.35
9.40
4.8


1
0.1
3.7
0.3


0.6


2.95
58


Hood
30
4
26
26
27.8
27.8
27.8
7.0
20.8
37.0


5
1
31.0
1.0


1.5


29.5
475


Food and
garden
wastes Glass
96
50
46
46
47
47
47
22
25
104


52
1.3
50.7
1.0


3.3


47.7
590


0.5
—
0.5
0.5
0.5
0.5
0.5
0.25
0.25
69.8


2
67.8
—
4


63.8


—
12


Sand
and
rock
5.5
1.0
4.5
4.5
4.5
4.5
4.5
2.5
2.0
42


1
41
—
3


38


—
7


Ferrous
metal
2.4
0.5
1.9
1.9
1.9
1.9
1.9
1
0.9
11.6


0.3
11.3
—
0.5


10.8


—
17


Non-
ferrous
metal
1.7
—
1.7
1.7
1.7
1.7
1.7
1
0.7
7.8


0.2
7.6
—
0.5


7.1


—
11


Total Destination
649 . 5 Entering
127.5 Exhausted
522 To reactor
522 Entering
534.6 To ball mill
534.6 To rotary screen
534.6 Entering
94.4 To incinerator
440.25 To fuel storage
378.4 Entering


71
135
172.4
11.6


131.0


164.8
2620


Process notes
Ash-6%; total moisture-
24%; moisture evapor-
ated- 19. 6%; heat for
drying- 367. 7xl06-BTU;

(HC1 gain 12.6)



(-6,516 x 106 BTU's)
Air required at 100%
excess=378.35 x 2 x 3=
2270 tons
(18.8%)
(35.6%)
(45.6%)
(3%) to electrostatic
precipitator (95%
trapped)
(34.6%) to landfill with
95% fly ash-141.95
(43.6%) total gas
2521 ton per day
(3460 BTU/%) about 60%
available for drying
at 1560 x 10 6 BTO

-------
N
F
13
FROM
CLEANING
MAGNET 'K
2059
AIR 85
FROM
ROTARY
SCREEN
10
FLY ASH FROM
ELECTROSTATIC
PRECIPITATOR
( RA
REF
V_!!
W ^
USE
2 I
SHREDDER
(12.7 cm
GRATE OPN)
412


MAGNET #2
412
ON
ERROUS



+12.7 cm


412
RECE1
ARE
907 n

VING
A
it/D*

TROMMEL
(12.7 cm OPN)
907
FERROUS -12.7 cm

389 LIGHTS
1
AIR
CLASSIFIER
768

INCINE
34
119
r
RATOR
3
589


"t
"V
23 v
N
'
+1.3
366
79
s
FLUE GAS
2283
GRATE
RESIDUE
(RESIDUE \
u. )

ACID
SCRUBBER
\
1
CAUSTIC
TANK
-*n
i
i
i
i 	


2055
I— ^-
228


495
s
MAGNET #1
495
3N
ERROUS
445
TROMMEL
(1.3 cm OPN)
445
-1.3

ROT
DRY
58



ER
9
473.5
REACTOR



HEAT
EXCHANGER



45
w



FER-
ROUS
50
(BULKY \
WASTES )
45 J

CLEA
MAG
7
3RROUS
NING
NET
3
60
COMPACTOR
60


NONFERROUS
13
TO AIR
CLASSIFIER
(FERROUS \
METAL I
60 J
115.5
(MOISTURE)
2055
s .
J
3.2
AIR
11.4
" HC1
v.^
_»' *>• -^ 2?B
i

HC1
SUPPLY



ELECTROSTATIC
PRECIPITATOR
2283

BALL
48

MILL
4.9
TO STACK
2273
10

ROTARY
SCREEN
484.9


85.6
TO
INCINERATOR

(POWDER \
FUEL ]
399.3 J
               * ALL PROCESS UNITS ARE METRIC TONS/DAY
Figure 27. Material balance flow plan  (metric  units)

-------
             RAW
             REFUSE
             1050TPD
                                                    BULKY
                                                    WASTES
                                                    50
                                                                NONFERROUS
                         FERROUS  -5"
                                                           66.5 CLASSIFIER
                                                      COMPACTOR
                 428.5 LIGHTS
            CLASSIFIER
  FROM
  CLEANING
  MAGNET (HEAVIES)
                                            (MOISTURE)
             INCINERA-
                                                      STATIC PRE-
FROM
ROTARY
SCREEN
(+1/8")
    11
                                                                TO RESIDUE
FLY ASH FROM
 ELECTROSTATIC
 PRECIPITATOR
                                 EXCHANGER
                                                        POWDER
                                                        FUEL
                                                        440.2
         CAUSTIC
         TANK
                                                            94.4
                                                            	1
                                                            TO INCINERATOR
                                *ALL PROCESS UNITS ARE TONS/DAY

         Figure  28.   Material balance flow plan
                        (English equivalents).
                                     78

-------
                              ENERGY BALANCE
(
**MSE
97.6
M&E
1.2

f ^\
RAH
REFUSE
V ^/

SHREDDER
(12.7 GRATE
OPN)

MAGNET »2

;
(FERROUS)

M&E
1.2
(LIGHTS)
M&E
40.1
FROM
CLEANING
MAGNET (H]
M&E l
130.8
AIR

FROM
ROTARY
SCREEN
FLY ASH FROM
ELECTROSTATIC
PRECIPITATOR
M&E
0.84
1
AIR
CLASSIFIER
NAVIES)
J89.2 , |
INCINERATOR
2567.6


^

ACID
SCRUBBER
I
CAUSTIC
TANK


(FLU
)
1
1
L

378.6

E GAS)
L*

M&E
0.84

RECEIVING
AREA
9303.8

TROMMEL
(12.7 OPN)

V '
MAGNET *1
7
V
M&E
M&E 0.95
8.1
(FERR-
OUS)




BULKY \
WASTES )
^ ^/

CLEANING
MAGNET


COMPACTOR
(NON-
- FERROUS)
TROMMEL
(1.3 OPN)


177°C
ROTARY

149°C
REACTOR
7434.4

HEAT
EXCHANGER


HC1
SUPPLY
M&E [
-Ti ^
(MOISTURE)
298
M&E
29 3
M&E
-29.3 L
J '
_ J
/
M&E \
0.53

(NONFERROUS)
TO
AIR
CLASSIFIER
M&E
12.9
FERROUS \
METAL J
103.6 /
^ ^x

ELEC
STJ
PRECIP

TIC
ITATOR

BALL MILL
7434.4

ROT
SCR
743


ARY
EEN
4.4

' POWDEP
FUEL
L 6334.6
^^ -^
TO STACK
TO RESIDUE
M&E
48.8
M&E
TO INCINERATOR
>M&E
3.84
                   * ALL PROCESS UNITS ARE MKJ/DAY




                   ** M&E - MECHANICAL AND ELECTRICAL ENERGY REQUIRED MKJ





                                  HEAT LOSSES

HSW
9303.8
2969.2
EMBRITTLE-
MENT
PROCESS

PROCESS
OUTPUT NET ENERGY
6334.6
MBTU
ENERGY FOR
PROCESS
5596.2
EFFICIENCY = 60.1%
                                     738.4
Figure 29.  Energy balance flow  plan  (metric units)
                                 79

-------
                                 ENERGY BALANCE
**M&E
92.5
M&E
1.1
M&E
38
FROM
CLEANING
MAGNET (HE3
^
AIR "4

FROM
ROTARY
SCREEN
FLY ASH FROM
ELECTROSTATIC
PRECIPITATOR
M&E
0.8
[ RAW
I REF
A
USE J

SHREI
(5" G
OPN)

DDER
RATE





RECE3
AREA
8820.

VING
4*

TROMMEL
(5"
(FERROUS)

MAGNET 2
i



M&E
1.1
V
1
OPN)

/
MAGNET 1
(LIGHTS)
1
AIR
CLASSIFIER
VIES)
032.6
INCI
TOR
2434
(
/ RESI


1
NERA
.2

DUE ]

ACID
SCRUBBER
*
l

CAUSTIC
TANK


(FLU
*-.
l
l

358.9

E GAS)


-»

M&E
0.8

i






M&E
"7.7





i
(FERR
OUS)




M
0


J BU
&_E ,
.9

I WAS

LKY 1
TES J

CLEANING
MAGNET
i

COMPACTOR
(NON
FERROUS)
TROMMEL
d/2"


350*
ROTj
DRY
733

OPN)

|C*
^RY
ER
0.7

300°F
REACTOR
7048.2

HEAT
EXCHANGER


HC1
SUPPLY
M&E
•-
3."
9


(MOISTURE)
282
.5
^M&E •
27.8
iM&E



^


/ 	





M&E
*o7so


(NONFERROUS)
TO
AIR
CLASSIFIER
M&E
12.2
/FERROUS\
METAL I
\9B.2 J


ELECTRO-
STATIC
PRECIPITA
TOR

BALL
7048

MILL
2

ROTARY
SCREEN
7048.2


f POWDER^
FUEL ]
x^eoos.sy
TO STACK
TO RESIDUE
M&E
46.3
"T.I
TO
INCINERATOR
1042.7
fE
8
                 * ALL PROCESS UNITS ARE MBTU/DAY
                ** M&E - MECHANICAL AND ELECTRICAL ENERGY REQUIRED MBTU
                                    HEAT LOSSES
                                     2814.9 MBTU
                                         PROCESS
                    MSW
                                                 NET ENERGY
                   8820.4 MBTU
                                                                  1.1%
                                      700.0 MBTU
Figure  30. Energy balance  flow  plan  (English equivalents).
                                     80

-------
PROPOSED PLANT LAYOUT

     Utilizing the flow plan developed, the material balance,
and available manufacturing data, a layout for a plant to pro-
duce powdered fuel from refuse was developed.

Equipment Specifications

     The equipment needed for each of the processing steps
specified in the flow plan are defined and listed in Table 26.
A potential supplier, model number,- power requirements for the
specific hourly capacity, and cost are listed.

Floor Plan

     Utilizing the data compiled in Table 26, a floor plan
arrangement was developed.  A schematic showing the proposed
plant layout is presented in Figure 31.  The arrangement shown
was selected through a trial and error process with the final
arrangement selected being the one most compatible with space
available and the special requirements needed for interfacing
the different items of equipment.  A detailed floor plan for the
office and maintenance area is shown in Figure 32.  Elevations
through different sections of the proposed plant are shown in
Figures 33 through 35.  The^receiving area, 45.7m x 24.4m
(150ft x 80ft), was determined by allowing both ample maneuver-
ing area for the front end loaders and a refuse storage area
large enough to accommodate one day of refuse.  The remaining
area of the plant was designed to accommodate the processing
equipment.  The processing building is to be constructed of
concrete and glass panels on an exposed steel structural system.
Concrete was chosen for its durability and chemical resisting
properties, glass panels to allow natural light into the plant,
while steel was used because of the long spans involved.

     The maintenance area provides space for repairing the
processing equipment, a locker room, restrooms, a lunch room,
and foreman offices.  The administrative area provides space for
payroll, billing, and supervisory personnel needed to allow the
plant to be self-sufficient.  Also included is a small quality
control lab, engineers office, and conference room.  The bag
houses and after-burner unit for odor control, the exhaust fans,
and the cyclone units will be located on the roof of the main
processing building.  A parking lot adjacent to the main build-
ing provides parking for plant and office employees.  The
buildings and grounds will be enclosed with a chain-link fence.
The security for the facility would be by television monitors
and security patrol.

     The proposed resource recovery plant is designed to
process 907 mt  (1000 ton) of municipal solid waste per day-
The plant would receive refuse 16 hr/day, 7 day/week.  The

                               81

-------
TABLE 25. EQUIPMENT LIST
Equipment identification*
1.
2.

3.

4.

5.

00 6.
to
7.

8.
9.
10.
11.
12.
13.



14.
15.
2-Trommels (12.7-cm) (5-in opn)
Coarse shredder (12.7-cm)
(5-in grate opn)
Magnetic separator #1

Magnetic separator #2

Cleaning magnet

Trommel (1.3-cm) (1/2-in opn)
2-Air classifiers with
cyclones
Metal compactor
2-Rotary dryers
2- Rotary reactors
2-Ball mill
2-Rotary disc screen
Incinerator (includes the grate
unit, wet scrubber, electro-
static precipitator, pit and
crane
Heat exchanger
Bank of HC1 tanks
Manufacturer
Triple/S Dynamics
Gruendler

Eriez Magnetic
Separation Division
Eriez Magnetic
Separation Division
Eri^z Magnetic
Separation Division
Triple/S Dynamics
Radar Systems, Inc.

Carborundum
C-E Raymond
C-E Raymond
Koppers Co . , Inc .
Radar System
Contractor Constructed



	
Vendor rented
Nominal
capacity
mt/hr
(ton/hr)
22
22

27

22

4

22
22

13
31
27
27
12
18



—
0
.7
.7

.2

.7

.5

.7
.7

.6
.7
.2
.2
.7
.1



-
.9
(25 ea)
(25)

(30)

(25)

( 5)

(25)
(25 ea)

(15)
(35)
(30)
(30)
(14)
(20)




( 1)
Model no.
6 x25
5xE

48 x48 SR-E

48 x48 SR-E

42 x42 SR-P

18 x6
T-79-184ADS

1000
150 x65 Style M
150 x65 Style M
8 x72
8 Dia.
	



___
	
1978
Equipment
Cost+
$ 53,
81,

23,

23,

17,

41,
201,

200,
350,
350,
65,
20,
5,000,



	
	
500 ea
800

525

525

823

658
500 ea

000
000
000
000
000
000





Total power
requirement
kw (HP)
18.6
447

5.2

5.2

4.5

18.6
105

373
67
67
112

600.3



	
	
(25 ea)
(600)

(7)

(7)

(6)

(25)
(141)

(500)
(90)
(90)
(150)

(805 ea)






-------
                                                       TABLE 25. EQUIPMENT LIST (Concluded)
00
OJ
Equipment identification*
16.
17..
18.
19.














Acid scrubber
Caustic tank
Powder storage bin
Conveyors
a. 2-Apron conveyors
b. 2- Apron conveyors
c. 2-Belt conveyors
d. 4-Belt conveyors
e. 3-Belt conveyors
f. 2-Belt conveyors
g. 2-Belt conveyors
h. 3-Belt conveyors
i. 1-Belt conveyor
j . 1-Belt conveyor
k. 2-Belt conveyors
1. 1-Belt conveyor
m. 1-Belt conveyor
n. 1-Belt conveyor
Manufacturer
Corning glass
Chemtronic System
Butler Manufacturing
Company
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Cindaco
Nominal
capacity
mt/hr
(ton/hr)
0.009
(.01)
Model no.
2
x8
1887-1 (500 gal) 	
453.5
40.8
40.8
40.8
40.8
40.8
40.8
36.3
40.8
18.1
18.1
18.1
18.1
4.5
13.6
(500)
(45)
(45)
(45)
(45)
(45)
(45)
(40)
(45)
(20)
(20)
(20)
(20)
( 5)
(15)
29
12
6
6
6
6
6
4
6
3
3
3
3
2
2
x64
x20
x36
x40
x20
x!5
Xl50
x20
x50
X252
x!30
x50
x!5
x!5
x50
1978
Equipment
Cost
$ 10
6
67
107
120
26
53
49
40
26
40
30
30
30
12
5
16
,000
,376
,500
,100
,500
,800
,600
,600
,150
,800
,200
,000
,000
,000
,000
,000
,600
Total power
requirement
kw (HP)
3.
3.
3.
15
30
11.
11.
11.
15
11.
22.
7.
7.
7.
3.
3.
3.
7
7
7


2
2
2

2
4
5
5
5
7
7
7
(5)
(5)
(5)
(20ea)
(40ea)
(15ea)
(15ea)
(15ea)
(20ea)
(15ea)
(30ea)
(10)
(10)
(lOea)
(5)
(5)
(5)
            This list also does not  include  the  following  items:   (a)  plant  collecting system;  (b)  waste water treatment system;
            (c) traveling bridgecrane;  (d) house air  supply  system;  (e)  truck  scales;  (f)  heating,  ventilating,  and air conditioning
            system;  (g) fire protection  system;  (h) electric power distribution system;  (i)  front-end loaders; (j)  remote control
            system; and (k) furniture,  fixtures,  and  laboratory equipment.

            Equipment cost does not  include  installation cost or  contractor  cost.

-------
00
                                                   I
I
                                                                                                          STORAGE BIN
                                                                                                         RESIDUE TRUCK
                                                          DRYER   Q   REACTOR
                                                                 !  I
                                                   Figure  31.  Floor plan,

-------
CO
U1
    3E1
                                                                                             MO.-S1
                                 Figure 32.   Office & Maintenance Areas

-------
00
              Figure 33. Refuse receiving area and initial trommelling.

-------
00
•vj
                   Figure  34.  Magnetic recovery and shredding area,

-------
00
00
                               Figure  35. Fuel  recovery area,

-------
processing facilities would operate 20 hr/day and the inciner-
ation facility would be operated on a 24 hr/day basis.  Assuming
an on-line availability of 85%, some 281,410 mt (310,250 ton)
can be processed annually.  It is assumed that the plant will be
located in close proximity to a sanitary landfill which will
receive the residue from the resource recovery plant and serve
as a backup to the plant during periods of shutdown or overage.

     Refuse trucks enter the plant and go first to a weigh
station and then to the refuse dumping area.  The weigh station
is used to control traffic into the plant, weigh the incoming
refuse and provide a printed record for both the driver and the
data processing system.  An automatic gate control system is
used to regulate the truck flow and prevent vehicles from enter-
ing the plant without weighing in.  An automatic/manual record-
ing system records the time, date, and gross weight of the truck
on the scale, along with the truck tare weight and identifica-
tion from the plastic data card provided by the driver.  Addi-
tional data such as billing code, truck route number, etc. can
also be recorded.  The information collected is retained for
billing and a receipt copy is given to the driver.  After the
cycle is completed, the driver proceeds to the designated tip-
ping stall while a new truck weighs in.  For trucks or private
vehicles without plastic data cards, a manual recording system
is used.  Trucks transporting the powdered fuel and ferrous
recovered from the refuse will exit through the weigh station
so that a record can also be maintained of production.

     The flow-through or dumping turnaround time will usually
vary from 5 to 10 min and will average about 8 min.  Turnaround
time extends from the time the truck enters the weigh station
to the time it leaves the refuse dumping area.  For the pro-
posed plant eight truck unloading stalls are proposed' providing
the facility with a capability to handle 60 trucks/hr.  With
this arrangement over 907 mt  (1000 ton) of refuse could be
handled during the 4 hr peak receiving period if necessary.

     The refuse trucks will dump the refuse onto a lower level
work floor, 3.7 m  (12 ft) below the truck unloading floor.
Front end loaders would be used to move the refuse to a storage
area or out to a conveyor for processing.

     A control tower, 4.6m x 9.8m  (15ft x 32ft), is located
above the work floor at the beginning of the process lines.  The
tower houses the electrical controls for all the processing
equipment and the traffic control system.  An operator stationed
in the tower will operate the processing equipment and manage
the traffic pattern of vehicles entering the truck receiving
area.  In addition, from this vantage point the tower operator
will provide visual inspection of refuse on the processing
                               89

-------
lines and coordinate  the  activities  of  the  front end loader
operators.

     Two processing lines are  proposed  for  the resource recovery
facility.  Each processing line would handle  22.7 mt/hr  (25 ton/
hr) .  Two front end loaders will  deposit  the  refuse onto two
apron conveyors.   These 3.7 m  (12 ft) wide  units would have a
belt speed of  1.5  to  4.6  m/min (5 to 15 ft/min).  The refuse
loaded onto  these  units would  be  conveyed onto two 1.8 m  (6 ft)
wide apron type conveyors having  a belt speed of 10.7 m/min
(35 ft/min).   The  assumed density of the  refuse feed onto the
conveyors is approximately 128.2  kg/m3  (8 lb/ft3).  The refuse
would be transported  up a 20°  incline to  the  first processing
step — trommeling.   Starting  at  the incline  all conveyors will
be enclosed with dust covers to minimize  plant housekeeping and
dust problems.  Refuse inspection would occur on the receiving
floor by the operators of the  front  end loader and on the first
stages of the  entrance conveyors. Waste  unsuitable for proces-
sing  (tires, tree  stumps, white goods,  etc.)  would be removed
at that time.

     Two trommels, 1.8m x 7.6m (6ft  x 25ft),  with 12.7 cm  (5 in)
circular openings  are the first processing  stage for the refuse.
The oversized  waste  (+12.7 cm)  (+5 in)  is conveyed to a hammer-
mill and the undersized  (-12.7 cm)  (-5  in)  to a magnet.

     A reversible  heavy duty hammermill with  a capacity of
27.7 mt/hr (30 ton/hr) will be used  to  reduce the oversized
fraction.  A heavy steel  hood  is  used to  confine heavy mater-
ials ballistically ejected by  the hammers.  A steel chain
curtain would  be used over the entrance of  the conveyor belt to
prevent damaging ejections.  The  grate  at the bottom of the
shredder would have 12.7  cm (5 in) openings to limit the size
of the exiting refuse.  The bulk  density  of the shredded refuse
is between 80.1 and 320.4 kg/m3  (5 and  20 lb/ft3) with an aver-
age of about 160.2 kg/m3  (10 lb/ft3).   The  shredded refuse is
discharged onto a  vibrating pan conveyor, which uniformly
feeds the refuse to a belt conveyor  going to  the magnetic
separator.

     Two drum  magnets, 1.2m dia x 2.1m  wide (4ft x 7ft) were
selected for ferrous  metal recovery.  The belts feeding the ref-
use are positioned about  25.4  cm  (10 in)  below the drum magnets,
and a belt speed of 33.5  to 47.2  m/min  (110 to 155 ft/min) is
required to maintain  a bed depth  of  about 3.8 cm  (1-1/2 in).
The recovered  ferrous is  conveyed to a  third  drum magnet which
serves as a cleaning  magnet to provide  a  cleaner ferrous frac-
tion for recovery.  An aspiration system  can  be incorporated
with the drum  magnet  to pull off  more of  the  light nonferrous
fraction.  The ferrous fraction with a  density of 560.7 kg/m
(35 lb/ft3) can be compacted to about 1201.5  kg/m3  (75 lb/ft )


                               90

-------
for transfer to a trailer or rail car for shipping to a ferrous
processor.  For ferrous compaction, a mini mill (a small verti-
cal hammer mill) has been selected.  The compacted ferrous will
be conveyed to a dispatch area for loading onto a trailer car
adjacent to the building.  The nonferrous fraction is conveyed
to an air classifier for further processing.

     The nonferrous material from the unshredded refuse
processing line (12.7 cm)   5 in) is conveyed to a trommel
with 1.3 cm (1/2 in) circular openings for processing to remove
the fine fraction.  The fine fraction (1.3 cm)  (+1/2 in) is
conveyed to the pit feeding the incinerator.  The 1.3 cm
(+1/2 in) nonferrous fraction from the trommel goes to a surge
bin feeding an air classifier.  The nonferrous shredded refuse
and the nonferrous fraction from the cleaning magnet are also
conveyed to a surge hopper for feeding to a second air
classifier.

     At the air classifier the light, predominantly organic
fraction is separated from the heavy fraction by an upward
air column.  Each classifier has a capacity to process
22.7 mt/hr (25 ton/hr).  The heavies dropping to the bottom
of the air classifier, are conveyed to the pit which feeds
the incinerator.  The light fraction is carried pneumatically
to a cyclone for deairing and transfer to the surge hopper
feeding a dryer.

     Two rotary dryers each with a design capacity of 31.7 mt/hr
(35 ton/hr) will be used to remove about 82% of the moisture
in the light fraction.  These dryers will be indirectly heated
to about 176.7°C  (350°F) by the combustion gases from the
incinerator.  From the rotary dryer the dried and heated
(148.9°C)(300°F) refuse will be dropped into the hopper to
the rotary reactor.  The evaporated moisture is vented from
the dryer through a filter to the atmosphere.  An alternative
option is to condense the heated water vapor with the HC1 and
air mixture used in the reactor.  There is considerable heat
in the water vapor and its utilization would raise the temper-
ature of the reactor gas and improve the energy efficiency of
the plant.  The condensed water can be treated and discharged
into the sewer system.

     Two rotary reactors with a capacity of 27.2 mt/hr
(30 ton/hr) will be used for embrittling the dried light
fraction of the refuse.  The units will operate at about
148.9°C  (300°F).  The heated and dried light fraction will
be tumbled through the reactor in a  flowing atmosphere of
air and gaseous HCl.  At the base of the rotary reactor,
the 50/50 gas mixture will be extracted from the top of the
cylinder and the treated refuse  from the bottom of the
cylinder.  The concentration level of HCl in the reactor will
be maintained as constant as possible.  During the reaction

                                91

-------
process the treated refuse adsorbs some of the HC1 gas.  Lab-
oratory data indicated that the cellulose waste adsorbs about
2.4% of its weight in HC1 during the process.  Therefore,
473.5 mt  (522 ton) of refuse can be expected to adsorb about
11.4 mt (12.6 ton) of HC1 gas.  However, by careful processing
it may be possible to reduce the amount of HC1 adsorbed by
the cellulose waste.  Preliminary study indicates that the HCl
adsorption can be limited to about 1% of the weight of the
cellulose waste.  The reactant gases will be pulled through
a filter and moisture trap then mixed with make-up gases and
reheated for reuse.  The embrittled light fraction will be
conveyed to a ball mill for further processing.  A caustic
scrubber unit would be available to treat the reactant gases
if they cannot be recycled.

     Two ball mills with 27.2 mt/hr  (30 ton/hr) capacity will
be used to reduce the embrittled light fraction to a powder.
As the material is reduced in size, it flows toward the exit
end of the mill.  Under the mill is a rotary disc screen to
remove the oversized plastics and other organics which are not
pulverized.  The rotary disc screens will also have the capacity
to process 27 mt/hr  (30 ton/hr).  The oversized material is
conveyed to the incinerator pit and the powder is transported
to an air lock feeder for pneumatic conveyance to a storage bin.
About 399 mt  (440 ton) of powdered fuel will be produced per day.
A cyclone separator is mounted on the storage bin to deair the
powder.  From the storage bin the powdered fuel will be trans-
ported to the fuel customer.  Trucks will be loaded from the
base of the storage bin.

     The heavy fraction from the air classifier, the fines from
the 1.3 cm  (1/2 in) trommel and the oversized from the rotary
disc screen will be conveyed to the incinerator pit.  About
342.7 mt  (378 ton) of refuse will be transported to the
incinerator pit for combustion.  The incinerator pit, 7.6m x
22.9 x 5.6m (_25ft x 75ft x 25ft 1 will have sufficient capacity
to store one day's load, 435.4 mt  (480 ton).  An overhead crane
will transport the refuse from the pit to the incinerator feed
hopper.  The incinerator will be a conventional moving grate
unit operated 24 hr/day with a capacity of 18.1 mt/hr  (20-ton/
hr).  During the 20 hr that the processing lines are in oper-
ation, the incinerator will process about 15.6 mt/hr
17.2 ton/hr).  During the 4 hr that the process lines are
down the incinerators will process about 7.8 mt/hr  (8.6 ton/hr).
The combustion gases from the incinerator  will be piped to the
two rotary dryers and from the dryers to the electrostatic •
precipitators.  During the 4 hr down period the combustion 'gases
will be piped directly to the precipitators.  A wet scrubber
unit is incorporated for quenching the exhaust gases prior to
transporting them to the electrostatic precipitator when they
are not used in the dryer.  Because the gas temperature cannot


                              -92

-------
exceed 315.6°C (600°F) at the precipitator unit, water from the
scrubber unit and from the bottom ash quench will be treated and
disposed of into the main sewer line.  The bottom ash from the
incinerator and fly ash from the precipitator will be conveyed
to a truck for delivery to the landfill.

     In addition to the processing equipment, considerable
auxiliary equipment and support facilities are required for the
proposed plant.  A waste water treatment system is required to
treat the drainage water from washing down the tipping floor
and process floor and the quench water from the incinerator.
This system would remove the scum and settleable solids prior
to pumping the water into a main sewer line for further treat-
ment at a municipal facility.  An overhead traveling bridge
crane is required for maintenance on the shredder and other
processing equipment.  An electrical substation is also needed
to meet the electrical requirements of the proposed plant.
The substation would be located adjacent to the processing
plant.

     A dust control system is another major support facility
required by the plant.  Air from the dust collection hoods over
the selected processing equipment, such as the shredder and
other high dust areas like the tipping floor as well as the air
from the cyclones have to be treated before it can be released
to the atmosphere.  Two bag houses and two afterburner units
will be required to remove the particulate matter and odors
from the plant air before it can be released.  These units will
be located on the roof of the processing plant.  Fire protection
and explosion protection units for the processing plant will
also be required.  These items will be discussed in greater
detail in the section on environmental, health, and safety
requirements.

     Other items of auxiliary equipment also required for the
proposed facility include a house air supply, a TV monitoring
system, a fuel supply for rolling stock and a heating, ventila-
ting, and air conditioning system.

Plant Contingency Plans

     As stated, the proposed processing plant is designed to
process 907 mt (1000 ton) of refuse a day.  However, it is
anticipated that the plant will not be able to operate at full
capacity through the year.  An 85% on-line availability is
assumed for the proposed facility.  When the plant is down
completely, the raw refuse would be sent directly to a nearby
landfill.  In addition it is recognized that the quantity of
refuse generated is subject to considerable fluctuations due
to seasonal variations as well as daily variations.  Daily
fluctuations can be managed by maintaining limited storage of
the refuse to smooth out peaks and valleys in the delivery

                               93

-------
 cycle.   Limited quantities of overage due to seasonal changes
 can be  managed by burning increased loads in the incinerator
 unit.   One wall of the incinerator pit fronts on the tipping
 floor to permit the front end loaders to push part of the refuse
 into the pit.  Large amounts of overage due to seasonal highs
 would have to be taken to the landfill.

     In the proposed facility the processing lines would be
;closed  4 hr/day for maintenance functions.  An aggressive
•program of daily preventive maintenance should minimize
 unscheduled down time.  However, it is recognized that some
 unanticipated equipment breakdowns will occur.  Two processing
 lines receive the refuse and separate it into two fractions:
 +12.7 cm C+5 in) and -12.7 cm 1-5 inl.  Each fraction is then
 processed on a separate line.  If any of the equipment in
 either  line breaks down, it is possible to continue the
 processing of the other fraction.  The refuse from the proc-
 essing  line that is out of order would be passed directly
 to the  incinerator pit.  This would be accomplished by bring-
 ing in  a portable conveyor unit and attaching it to the
 opposite end of either the 12.7 cm (+5 in) or the 12.7 cm
 (-5 in) processing line at the trommels.  The conveyor would
 carry the refuse back to the tipping floor for a front end
 loader  to move it to the incinerator pit.

 Health, Safety, and Environmental Protection Considerations

     This proposed resource plant will conform to all OSHA and
 EPA regulations.  All pit areas will be properly railed and all
 catwalks and ladders will be enclosed.  A fire control water
 sprinkler system is proposed for all of the buildings.  In
 addition, halogenated hydrocarbon explosion suppression systems
 are proposed for the primary shredders.  A similar halogenated
 hydrocarbon system is recommended for the control tower and
 motor control centers.  Carbon dioxide fire suppression units
 are recommended for the refuse conveyor lines, in the trommels,
 in the  air classification system, in the ball mills and
 screening units, cyclones, dust collection units, and in the
 storage bins.  This system should provide total plant fire and
 explosion protection.

     Recovery of powdered fuel and ferrous metal results in a
 reduction of solid waste for disposal and the associated
 environmental problems.  However, processing, use and disposal
 of the  products from this proposed facility could result in
 other types of environmental pollution problems.  Waste to
 energy  conversion facilities and the products generated can
 contribute to land, water, .and air pollution.  The potential
 pollution problem and associated control technologies for an
 embrittlement processing facilities have been considered and
 the areas of major concern are discussed in the following
 paragraphs.
                                94

-------
     Dust and gas emissions from incineration are two of the
potential pollution problems which have been identified for the
proposed process.  Dust emission is a problem that plagues most
phases of solid waste processing.  Shredding, air classification
and screening are major sources of dust emission.  Bacteria and
virus emissions are closely associated with dust since both
generally reside on the surface of the dust particles.  Effect-
ive dust control for the processing requirements of the proposed
facility does not appear to be a problem.  Care mustbe taken to
enclose all processing equipment and conveyors.  In addition,
effective exhaust systems must be utilized at the processing
equipment and for the plant in general.  Fabric filtration units
(baghouses) can be readily employed for cleaning the exhaust
gases, since the gas is at ambient temperature and the dust is
not very abrasive.  Pilot studies demonstrated that 99.9% par-
ticulate removal efficiencies can be obtained with the use of
fabric filters.

     Odors from the putrefaction of the food wastes can be
picked up by the fluidizing air.  Afterburner combustion units
are proposed for the exit end of the dust collection system for
controlling odors in the exhaust air.  Noise control systems
do not appear to be required for any of the processing areas,
although ear protection is recommended for workers in the
shredding area.  If the noise from the hammermills does prove
to be excessive after start-up, soundproofing paneling can be
installed.  Hammermill manufacturers do not feel that sound-
proofing is necessary as part of the installation.

     The other cause of air pollution is the exhaust gases from
incineration of the heavy and fine fractions which provide the
heat source for embrittlement.  The uncontrolled emissions from
this process should be similar to those reported for most incin-
erator operations:  particulates, HC1, S02, NOX, CO, and un-
burned hydrocarbons.  CO, NOX, SO2, and unbruned hydro-
carbons do not appear to be a problem nor are any special
controls anticipated for these pollutants.  HC1 removal can be
achieved with a scrubber unit; however, appropriate materials
selection in the design are necessary to inhibit acid corrosion.
Particulate control would probably require an electrostatic
precipitator.  In addition, since up to 30% of the particulates
could be less than 1 ym, the use of a fabric filter system may
also be required.  Disposal of the fly ash collected is also
of concern due to the presence of trace elements  (Cl, Pb, Cu,
Zn, etc.) found in the fly ash.

     The proposed facility for producing a powdered fuel relies
primarily on dry processing.  The only water required is for
quenching the incinerator residue and washing down processing
equipment.  This water picks up a considerable amount of con-
taminants, organics, trace metals, acids, etc., and will require
preliminary treatment before it can be sent to the waste water

                                95

-------
facility-

     The residue from the proposed facility will come from the
incineration process and is relatively inert.  The captured fly
ash and quenched bottom ash will contain a small amount of un-
burned organics  (2.5%) and some trace elements  (Be, Hg, Cl, Pb,
etc.) requiring proper sanitary landfill procedures for its
disposal.

     The powdered fuel obtained by the embrittlement process
can cause environmental problems in its use.  Combustion of this
powdered fuel will produce air pollutants  (particulates, trace
metals, and gaseous emissions) similar to those described for
the incineration of the heavies and fine fraction.  Of partic-
ular concern are the fine particulates, trace metals and acid
vapors.  Similar precautions  to those described for incinera-
tion will have to be taken to control these pollutants.

     A problem with the powdered fuel which should also be
addressed, although it is not usually covered under environment-
al concerns, is the explosive potential of the powdered fuel.
Fine powders can readily detonate and great care must be exer-
cised in handling and storage of the powdered fuel.  It has been
reported that several explosions were experienced by A.D. Little
and CEA in their work with powdered fuels.  It is estimated that
the explosive power of the powdered cellulose was equivalent to
half that attributed to powdered grain  (28).  The standard
safety practices established  for grain should be followed in
the handling of powdered fuels.

     Another approach to handling the powdered fuel would be to
suspend the treated light fraction in a light fuel oil.  Grind-
ing and storage of the treated organics in a light fuel oil will
inhibit explosions and could  facilitate introduction of the
powder into the combustion system of a furnace or boiler  (29) .

ECONOMIC ANALYSIS

     An economic analysis was developed for a plant designed to
convert 907 mt/day  (1000 ton/day) of municipal refuse into a
powdered fuel.  The plant is  based on the flow plan developed
in Figure 20.  The plant is to be staffed to receive refuse
16 hr/day for 365 days a year.  The refuse will be processed
for 20 hr/day and the remaining 4 hr would be used for equip-
ment maintenance and house cleaning chores.  It has been assumed
that a landfill is within 5 miles to serve as a backup to the
facility.  On an annual basis the facility will process
281,180 mt (310,000 ton)  (1000 x 365 x 0.85) assuming ah on-line
availability of 85%.  The analysis that was used followed the
format published in EPA document SW-157-6  (30).  The data pro-
vided are based on 1978 dollars for a site in the East North
Central United States.  Where data were not available, the

                               96

-------
guidelines provided in SW-157.6 were used to estimate line item
costs.  A summary of the capital costs, annual operating and
maintenance costs and required revenues is discussed in the
following paragraphs.  The basis for the data used and the
detailed calculations are presented in Appendix A.

     A summary of the estimated capital costs for the proposed
facility is presented in Table 26.  The capital eost is esti-
mated to be $30,319,000.  The total interest for revenue bonds
on a loan calculated at 8% is $30,585,807 over 20 years.  This
results in a total cost of $60,904,807 or $3,045,240 annually.
The estimated capital cost per ton is $10.84/mt ($9.83/ton).

     A summary of the estimated annual operating and maintenance
(O&M) costs are presented in Table 27.  The total annual O&M
cost is estimated  to be,$7,980,000.  This results in an O&M
cost per ton of $28.31/mt ($25.51/ton).

     The estimated total cost per ton which is the sum of the
O&M and the capital cost is $38.96/mt ($35.34/ton).  To cover
these costs, revenues are anticipated from a tipping fee for
waste disposal and the sale of recovered metals and the powdered
fuel.  A summary of the potential revenue sources required is
presented in Table 29.  As shown in Table 28, the powdered fuel
has to have a market value of $63.46/mt ($57.56/ton) or
$0.064/kg ($0.029/lb) in order for the process to break even.
With a heating value of about 16,281 kg (7000 BTUs/lb) the
powdered fuel would cost $3.89/million KJ ($4.10/million BTU).

Proposed Future Work

     Laboratory studies at UDRI have demonstrated that a number
of treatments can be used for the embrittlement of cellulose
wastes to produce a powdered fuel  (P-RDF).  Although the process
variables have been determined and an engineering design and
economic analysis have been established for the production of
a P-RDF, there is a need for further research studies.  A num-
ber of studies to better characterize the P-RDF as a fuel or
fuel component are needed.  In addition, studies to extend the
utilization of embrittlement in resource recovery processes are
also needed.

     A compilation of the tasks which should be initiated would
include:

      (a)   Combustion analysis of P-RDF fuel.
      (b)   Oil/P-RDF slurry development.
      (c)   Embrittlement studies on unprocessed raw refuse.
      (d)   Embrittlement studies on wood product and crop wastes.
      (e)   Investigation of partial embrittlement techniques
           for preparing RDF flakes.


                               97

-------
               TABLE 26. ESTIMATED CAPITAL COSTS
     Item
  Amount
Land
Site Preparation
Construction
Equipment and Installation
Auxiliary Equipment and Facilities
          Subtotal
Engineering and Design

          New Subtotal
Contingencies
Financing and Legal
Start-up and Working Capital
          Total Initial Capital Investment

Estimated useful life of facility  (years)
Total interest to be paid
Total capital cost
Average annual capital cost
Annual throughput  (tons)

Capital cost per ton
$   967,000
    339,000
  4,856,000
  9,272,000
  6,179,000

$21,613,000

  1,297,000

$22,910,000

  2,291,000
    505,000
  4,613,000

$30,319,000


20 years
$30,585,807
$60,904,807
$ 3,045,240
281,180 mt
 (310,000 ton)
$10.84/rat
 ($9.83/ton)
                               98

-------
             TABLE 27. ESTIMATED ANNUAL OPERATING
                     AND MAINTENANCE COSTS
        Item
          Amount
Salaries and Benefits
Fuel
Electricity
Water and Sewer
Maintenance
Replacement Equipment
   (straight line depreciation)
Residue Removal
Materials and Supplies
Taxes
Insurance
General and Administrative Costs
          Total Annual O&M Costs

          Operating and Maintenance Cost/ton
       $1,815,000
           70,000
        1,000,000
           45,000
          932,000

        2,422,000
          314,000
          904,000
          190,000
          126,000
           90,000

       $7,908,000


       $28.13/mt
        ($25.51/ton)
              TABLE 28. POTENTIAL REVENUE SOURCES
       Source
       Dollars/ton
   refuse processed
Tipping Fee
Ferrous Metal
Powdered Fuel
  $0.064/kg  ($0.029/lb)
          Total Required Revenue
$ 9.37/mt (  8.50/ton)
  1.65/mt (  1.50/ton)
 27.93/mt (25.34/ton)


$38.96/mt (35.34/ton)
                              99

-------
     (f)    Reevaluation of the economics for the embrittlement
           processes.

     Although P-RDF has been characterized, performance in burn
tests has not been well established.  Therefore, combustion
studies of the powdered fuel prepared from the embrittlement of
RDF are proposed for future study in a test unit with a burner
for pulverized coal.  Another area of major interest is the
suspension of the P-RDF in an oil.  The handling and shipment
of a powdered fuel to its point of use presents some problems—
a major one being safety hazards arising from the potential
explosion characteristics of a finely powdered material.  A
possible solution to this problem is storing, transporting, and
burning the powdered fuel in an oil slurry.  The techniques for
powder suspension can be initially based on the work done in
developing mixtures of powdered coal in oil.  Once an effective
P-RDF/oil slurry is developed, mixtures of 10, 25, and 50
percent powder/oil slurries can be prepared for combustion
tests.  It is expected that an air atomizing burner unit would
be used for the proposed combustion tests.
                                100

-------
                         REFERENCES

1.  Kaufman, J. A. and A. H. Weiss.  Solid Waste Conversion:
    Cellulose Liquefaction.  EPA-670/2-75-031, U.S. Environ-
    mental Protection Agency, Cincinnati, Ohio.

2.  Hooverman, R. H.  Rotary Kiln Gasification of Solid
    Wastes and Sewage Sludge:  Experimental Work.  U.S.
    Environmental Protection Agency, Cincinnati, Ohio.

3.  Hecht, N. L., D. S. Duvall, and B.  L. Fox.  Investigation
    of Advanced Thermal-Chemical Concepts for Obtaining
    Improved MSW-Derived Products.  EPA-600/7-78-143, U.S.
    Environmental Protection Agency, Cincinnati, Ohio,
    August 1978.

4.  Harendza, Harinxma, A.  Method of Carbonizing a Substance
    Comprising Cellulose.  U.S. Pa,tent 3,961,025, 1976.

5.  Wittman, T. J.,  et al.  A Technical,  Environmental and
    Economic Evaluation of the Wet Processing System for
    the Recovery and Disposal of Municipal Solid Waste.
    U.S. Environmental Protection Agency, Washington, D.C.
    1975.  217 pp.

6.  Hasselriis, F.  Private communication.  Combustion
    Equipment Associates.  July 1979.

7.  Brennemon, R. S. and J. J. Clancy.   Process of Treating
    Organic Wastes and Products Thereof.   U.S. Patent
    3,961,913.  June 8, 1976.'

8.  Patrick, P. K.  Waste Volume Reduction by Pulverization,
    Crushing and Shearing.  The Institute of Public Cleansing,
    69th Annual Conference, Blackpool.  June 5-9, 1967.

9.  Warren, J. L.  The Use of a Rotating Screen as a Means
    of Grading Crude Refuse for Pulverization and Compression.
    Resource Recovery and Conservation.   (3):97-111, 1978.
                            101

-------
10.  Drobny, N. L., H. E. Hull, and R. F. Testin.  Recovery
     and Utilization of Municipal Solid Waste.  U.S. Environ-
     mental Protection Agency, Solid Waste Management Office.
     1971.

11.  Taggart, A. F.  Handbook of Mineral Dressing Ores and
     Industrial Minerals.  J. Wiley & Sons, New York.
     pp. 7-27 to 7-34 and 5-03 to 5-10, 1956.

12.  Hill, R. M.  Personal communication.  Triple/S Dynamics,
     Dallas, Texas.  July 1979.

13.  Bernheisel, J. F., P. M. Bagalman, and W. S. Parker.
     Trommel Processing of Municipal Solid Waste Prior to
     Shredding.  National Center for Resource Recovery.- Inc.,
     presented at Mineral Waste Utilization Symposium, Chicago,
     Illinois.  May 2, 1978.

14.  Woodruff, K. L.  Preprocessing of Municipal Solid Waste
     for Resource Recovery with a Trommel.  Transactions of
     SME.  September 1976.

15.  Woodruff, K. L. and E. P- Bales.  Preprocessing of Munici-
     pal Solid Waste for Resource Recovery with a Trommel—
     Update.  Proceedings of the 1978 National Waste Processing
     Conference, Chicago, Illinois.  May 1978.

16.  Savage, G. M., L. F. Diaz, and G. J. Trezek.  RDF:  Quality
     Must Precede Quantity.  Waste Age.  April 1978.

17.  Savage, G. M. and G. J. Trezek.  Screening Shredded
     Municipal Solid Waste.  Compost Science.  January/
     February 1976.

18.  Perry, R. H. and C. H. Chilton.  Chemical Engineers
     Handbook.  McGraw Hill Book Company, Fifth Edition.
     pp. 8-25 to 8-28, 1973.

19.  Ruf, J.  Particle Size Spectrum and Compressibility of
     Raw and Shredded Municipal Solid Waste.  Ph.D. Thesis.
     The University of Florida.  1974.

20.  Kenahan, C. B., et al.  Composition and Characteristics
     of Municipal Incinerator Residues.  U.S. Department of
     the Interior, Bureau of Mines, Washington, D.C.  RI 7204.
     December 1968.

21.  Sweeny, P- J.  An Investigation of the Effects of Density,
     Size, and Shape Upon! the Air Classification of Municipal
     Type Solid Waste.  Ph.D. Thesis.  University of Dayton.
     April 1977.


                               102

-------
22.  Hecht, N. L., et al.  Design for a Resource Recovery
     Plant for AMAX Inc.  University of Dayton.  1974.

23.  Sullivan, P. M., M. H. Stanczyk, and M. J. Spendlove.
     Resource Recovery from Raw Urban Refuse.  Bureau of
     Mines, Washington, D.C.  RI 7760.  1973.

24.  Anonymous.  Edmonston (Md.) Solid Waste Recycling Project.
     Bureau of Mines, Fact Sheet, Bureau of Mines, U.S.
     Department of the Interior.

25.  Gorman, P. G., et al.  St. Louis Demonstration Project
     Final Report:  Power Plant Equipment, Facilities, and
     Environmental Evaluations.  EPA Contract No. 68-02-1871.
     MRI Project No. 4033-L, Environmental Systems Section,
     Midwest Research Institute, 425 Volker Blvd., Kansas
     City, Missouri 64110.  1977.

26.  Anonymous.  National Center for Resource Recovery,
     Washington, D.C.  Volume 1 - Materials Recovery System
     (Engineering Feasibility Study), December 1972.
     Volume 2 - New Orleans Resource Recovery Facility.  1977.

27.  Wilson, E. Milton.  Engineering and Economic Analysis of
     Waste to Energy Systems.  Prepared for Industrial Environ-
     mental Research Laboratory, Office of Research and Develop-
     ment,  U.S. Environmental Protection Agency, Cincinnati,
     Ohio 45268, Ralph M. Parsons Co., Pasadena, California
     91124.  Contract No. 68-02-2101, Parsons File No. 5495-1.
     June 1977.

28.  Rigo, H. G.  Private communication.  Systems Technology
     Corporation, Xenia, Ohio.  December 1977.

29.  Sussman, D.B.  Resource Recovery Plant Implementation:
     Guides for Municipal Officials.  EPA Publication SW-157.6.
                               103

-------
                           APPENDIX A

                BACKGROUND FOR ECONOMIC ANALYSIS"*"


CAPITAL  COSTS

 1.  Land  requirements

     a)  Required acreage for plant;
         30  acres @  $12,500/acre = $375,000

     b)  Required acreage for landfill;
         process  residue: 142 TPD x 365 = 51,830  TPY
         raw refuse:  (1000 x 365)  - [(1000 x 365).85] =
                      54,750 TPY

         Total  waste  for landfill 106,580 TPY.  Assuming 1000
lbs/yd3, gives  213,160  yd3/yr.   Allowing 25% for  fill gives
266,450  yd3/yr.   This converts to 7,194,150  ft3/yr.  Assuming
two tiers  at 20 ft/tier for the landfill results  in a require-
ment of  179,854 ft2/yr  for 20 years or a total  of 3,597,080  ft2.
In addition  allowing  for a 10% buffer  zone increases the
requirement  to  3,956,788 ft2 which converts  to  91 acres.  At an
estimated  cost  of $6,500/acre the cost for landfill would be
$592,755.  Total  land costs are estimated to be $967,755.

 2.  Site  preparation

     Allowing 35% of  land costs for site preparation gives
.35 x $966,500  =  $338,275.

 3.  Construction costs

     Average construction costs for. the North East Central
states was given  at $42/ft2 by a local A&E firm in the field.

     The proposed plant would occupy a building complex of
114,900 ft2; therefore,  $42 x 114,900  = $4,825,800.  In addition
about $30,000 is  required for the  scale  house.  Total construc-
tion costs would  be $4,855,800.


+A11 calculations  are given in  engineering units  (see conversion
 chart,  page A-10).

                              104

-------
 4.  Equipment and installation

     Based on manufacturer's data a complete list of the
required equipment for the proposed plant was developed.  This
list itemizes all equipment costs and installation costs (see
Table A-l).  Total cost would be $9,272,000.

          TABLE A-l. EQUIPMENT AND INSTALLATION COST


1.
2.
3.
4.
5.
6.
7-
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Equipment
Description
2-Trommel (5")
Coarse shredder
(5" grate OPN)
Magnetic separator
#1
Magnetic separator
#2
Cleaning magnet
Trommel (1/2" OPN)
2-Air classifiers
with cyclones
Metal compactor
2 -Rotary dryers
2 -Rotary reactors
Ball mill
2-Rotary disc
screens
Incinerator system*
Heat exchanger
Acid scrubber
Caustic tank
Powder storage bin
Pneumatic powder
conveying system
2 7 -Conveyors
TOTAL
Equipment
cost
$107,000
81,800
23,525
23,525
17,823
41,658
403,000
100,000
700,000
700,000
130,000
40,000
—
20,000
10,000
6,376
67,500
26,000
Installation
cost
$ 28,900
38,400
11,475
11,475
8,705
11,248
166,800
51,500
350,000
350,000
65,000
20,000
—
10,000
5,000
3,624
45,350
8,000
Total
cost
$ 135,900
120,200
35,000
35,000
26,528
52,906
569,800
151,500
1,050,000
1,050,000
195,000
60,000
5,000,000
30,000
15,000
10,000
112,850
34,000
588,400
$9,272,084
 electrostatic precipitator, pit and crane,
                              105

-------
 5.  Auxiliary equipment and facilities

     A similar compilation of all required auxiliary equipment
and facilities has also been compiled  (see Table A-2).  Total
cost is $6,179,000.

   TABLE A-2. PLANT FACILITIES AND AUXILIARY EQUIPMENT COST

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Equipment description
2 Scales
2 Fork trucks
1 Pickup truck
3 Frontend loaders
Explosion protection system
Miscellaneous conveyors
2 Bag house systems
Electrical substation and
motor controls
HCL system
Small tools
T.V. security system
Power sweeper
Office furniture
Bathroom fixtures and
accessories
Sump pumps
Lab equipment
Fire protection
Heating, ventilation and
air conditioning system
Wastewater treatment system
Overhead crane
Total cost
$ 52,000
24,000
4,000
128,500
88,500
500,000
1,120,000
1,191,000
74,000
62,000
47,000
24,000
17,000
4,000
2,000
10,000
206,000
125,000
1,000,000
1,500,000
          TOTAL                           $6,179,000
                               106

-------
 6.  Engineering and design

     It is assumed that the engineering and design costs are
based on the average engineering and design assignment rather
than a first of a kind facility-  A conventional 6% fee was
used for A&E services.  .06 x $21,613,000 = $1,296,780.

 7.  Contingencies

     A value of 10% has been allowed for contingencies.
$22,910,000 x .1 = $2,291,000.  (The new subtotal for items 1-7
= $25,201,000.)

 8.  Financing and legal fees

     Referencing SW-157.6 a value of 2% of the capital costs
should be used for the cost of a bond counsel, legal fees,
financial management consultants, etc. to cover financing the
loan for the facility-  Therefore, $25,201,000 x .02 = $504,020.

 9.  Start-up and working capital

     Seven months payroll  (4 mo start-up and 3 mo working capi-
tal) has been selected for this item.  (Table 28 in the text
shows the annual O&M costs.)  Therefore, $7,908,000 x 7/12 =
$4,613,000.  Total initial capital investment rounded to the
nearest $1,000 is $30,319,000.

10.  Interest to be paid

     For a 20 yr loan the monthly payment  (per $1,000 loan) at
8% interest is reported in the tables to be $8.37.  Per yr this
would be $100.44 and for 20 yrs would be $2008.80.  For a loan
of $30,319,000, the total payment would be:  $2008.80 x 30319 =
60,904,807 - 30,319,000 = $30,585,807.

11.  Total capital cost - $60,904,807-

12.  Average annual capital cost - $60,904,807^20 = $3,045,240.

13.  Capital cost per ton - $3,045,240^310,000 = $9.83.

OPERATING AND MAINTENANCE COSTS

14.  Salaries and benefits

     A complete list of all personnel required to operate the
proposed plant has been developed.  Job descriptions for all
positions have also been prepared.  This information has been
used to develop yearly manpower costs.  A  summary of yearly
manpower costs is presented in Table A-3.
                              107

-------
Position
Weighmaster
Front-end loader operator
Control tower' operator
Process operator
Assistant process operator
Incinerator operator
Assistant incinerator operator
Maintenance man
Maintenance helper
Utility man I
Utility man II
Receiving area attendant
Janitor ;
Crane operatbr
Foreman
Plant manager
Assistant plant maria~ger"
Billing clerk .
v".
Accounting clerk
Secretary
Quality control technician
TOTAL ;; :•;
Annual cost for
each position
$16,117.23
18,488.30
18,488.30
18,488.30
16,592.12
18,488.30
16,592.12
21,245.88
18,849.13
18,849.13
15,722.91
13,069.16
14,997.65
18,488.30
24,522.02
32,955.40
29,341.30
11,577.31
13., 277. 66
11,577.31
14,978.02-
Number
required
3
8
8
8
12
4
8
12
12
4
3
1
1
4
4
1
1
1
1
1
1
98
Total
cost
$ 48,351.69
147,906.40
147,906.40
147,906.40
199,105.44
73,953.20
132,736.96
254,950.56
226,189.56
75,396.52
47,168.73
13,069.16
14,997.65
73,953.20
98,088.08
32,955.40
29,341.30
11,577.31
13,277.66
11,577.31
14,978.02
$1,815,386.50

     Yearly manpower costs  -  $1,815,386.80.
                  JC.                    '-
     The wage rate-used  in  this  analysis' is based  on  the pro-
jected rates in;effeet during 1978  in  the northeast central
United States.  A proposed  work  schedule is summarized in Table
A-4.  In order to; maintain"24 hr staffing of the foreman, incin-
erator operator-: and utility man  positions, a 4-man crew could
be scheduled to work 6 days,  be  off 2  and then  rotate shifts.
This scheduling procedure necessitates a planned overtime of 50
                               108

-------
                                                     TABLE A-4. PROPOSED WORK SCHEDULE
O
10
Day
Shift
Weigh master
Front end loader operator
Control tower operator
Process operator
Assistant process operator
Assistant incinerator operator
Maintenance wan
Maintenance helper
Utility man I
Utility man II
Receiving area attendant
Janitor
Crane operator
Foreman
Plant manager
Assistant plant manager
Billing clerk
Accounting clerk
Secretary
Quality control technician
Monday
1st
1
2
2
2
3
2
3
3
1
1


1
1
1
1
1
1
2nd
1
2
2
2
3
2
3
3
1









3rd

2
2
2
3
2
3
3


1







Tuesday
1st
1
2
2
2
3
2
3
3
1
1


1
1
1
1
1
1
2nd
1
2
2
2
3
2
3
3
1









3rd

2
2
2
3
2
3
3


1







Wednesday
1st
1
2
2
2
3
2
3
3
1
1


1
1
1
1
1
1
2nd
1
2
2
2
3
2
3
3
1









3rd

2
2
2
3
2
3
3


1







Thursday
1st
1
2
2
2
3
2
3
3
1
1

1
1
1
1
1
1
1
2nd
1
2
2
2
3
2
3
3
1









3rd

2
2
2
3
2
3
3


1







Friday
1st
1
2
2
2
3
2
3
3
1
1


1
1
1
1
1
1
2nd
1
2
2
2
3
2
3
3
1









3rd

2
2
2
3
2
3
3


1







Saturday
1st
1
2
2
2
3
2
3
3
1









2nd
1
2
2
2
3
2
3
3
1









3rd

2
2
2
3
2
3
3










Sunday
1st
1
2
2
2
3
2
3
3
1









2nd
1
2
2
2
3
2
3
3
1









3rd

2
2
2
3
2
3
3










                                    Total  30  23  22   30  23  22   30  23  22   30  23  22   30  23  22   23  23  22   23  23   22

-------
shifts a year per crew.  The men on each of these crews will
have to work an average of 12.5 overtime shifts.

     The wage rate for each position is multiplied by 2080 hr
in order to obtain the annual cost for that position.  This
rate includes vacation and sick leave pay, company-paid medical
and dental insurance, disability income and long-term disability
insurance, a company-paid retirement plan, and normal state and
federal payroll taxes.  A shift differential of $.12 for the
second shift and $.18 for third shift has also been included and
the average value for the differential is added to the wage
schedule.

15.  Fuel

     Estimated fuel costs for the 2 front-end loaders are
$5.20/hr x 16 hr x 2 units = $160/day x 365 days = $58,400 or
rounded off = $60,000.  Estimated cost for other plant vehicles
is 300 gal/wk @ $.60/gal = $180/wk x 52 a $10,000.  Total cost;
$70,000.

16.  Electricity

     Total estimated power requirements for the plant is 700 x
10  BTU/day.  This translated to 700 x 106/3 x 1/3412.2 = 68,382
KWh.  Using an electric rate of $.04/KWh gives a cost of 68,382
x 365 x .04 = $998,378.40.

17.  Water and sewer

     Based on a flow requirement of 175 gal/min the combined
sewer and water rate was calculated to be $44,496/yr.

18.  Maintenance

     Maintenance material and outside service costs are esti-
mated at 5% of the total installed cost of all equipment.  In
addition, 1% of the construction cost for all buildings is esti-
mated for additional building maintenance costs.  Additional
maintenance costs can also be expected for the shredder and
rolling stock.

     extra shredder maintenance cost                   $ 96,496
     rolling stock maintenance cost                      14,483
     equipment $13,561,000 x .05                        678,050
     buildings $4,855,800 x  .01                          48,558
                                                       $837,587
                               110

-------
19.  Replacement equipment

     A straight line depreciation formula was used for replace-
ment of equipment:

     Buildings 20 yr = $4,855,800 ^20               $  242,790
     Rolling stock - 3 yr = $156,500 T 3                 52,167
     Shredder - 10 yr = $120,200 * 10                    12,020
     Scale - 10 yr = $52,000 T 10                         5,200
     Electrical equipment - 10 yr = $1,191,000 * 10     119,100
     All other equipment - 7 yr = $12,088,100 * 7     1,726,871

                                                     $2,158,148

20.  Residue removal costs

     142 TPD
       x 365

      51,830 z 52,000 TPY

     Land disposal fee @ $4.50/ton       $234,000
                    Transportation         8 0,0 0 0

                                         $314,000

21.  Taxes

     Estimated taxes @ 0.75% of the capital cost of the facility
are $22,999,000 x .0071 = $172,492.

22.  Insurance

     Estimated insurance @ 0.05% of the capital cost of the
facility is $22,999,000 x .005 = $114,995.

23.  Materials and supplies

     General supplies estimated @ $2000/mo             $ 24,000
     Anhydrous HC1 quoted at $225/ton delivered
       is $225/ton x 12.6 tons/day x 310 days ~         880,000
                                                       $904,000

24.  Annual general and administrative costs

     Postage and office supplies                        $ 1,500
     Janitorial services                                  2,000
     Telephone                                            3,500
     Monthly accounting                                   4,500
     Year end audit                                       3,800
     Laboratory supplies                                  3,000
     Landscaping                                          3,500
                              111

-------
24.  Annual  general  and administrative  costs  (Continued)

     Travel                                                7/500
     Public  relations                                      7,200
     Employee  services                                     7*200
     Subscriptions and  misc.                               7,200
     Security  services                                    38,000

                                                        $89,000

25.  O&M Recount                                      $1,815,000
                                                          70,000
                                                      1,000,000
                                                          45,000
                                                        932,000
     Per ton cost                                    2,422,000
                                                        314,000
                           51                            904,000
                                                        126,000
                                                        190,000
                                                          90,000

                                                      $7,908,000

REVENUES

26.  Tipping fee

     Waste disposal  costs  are  subject to  considerable variation.
Using a  range  from $4.50 to $12.50  per  ton  (landfill  vs.  incin-
eration)  an  average  value  of $8.50  was  selected.

27.  Ferrous metal sale

     66.5 tons are recovered/day with, an  average market value
of $22.50/ton.  This gives a revenue of .0665 x $22.5 = $1.50.

28.  Powdered  fuel sale

     Sale of the  fuel has  to cover  the  remaining per  ton  pro-
cessing  costs  of  the facility.

     Total processing cost - $35.34/ton
     Revenues  - $8.50 + 1.50 =  $10.00

     Required  cost for powdered fuel is $35.34 - $10.00 =
$25.34/ton.  The  fuel recovered is  44.02% of the raw  refuse
delivered; therefore, $25.34 *  .4402 =  $57.56/ton.  The cost/lb
is $57.56 T  2000  = $0.029/lb.   The  heat content of the powder
was measured at 7000 BTU/lb which gives a cost per 10^ BTU of
$0.029/7000  =  $4.10.
                              112

-------
                                   TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
1. REPORT NO.
    EPA-600/2-80-I21
                             2.
                                                           3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
      Obtaining Improved Products  From The Organic
      Fraction Of Municipal  Solid  Waste
              5. REPORT DATE
                 August 1980  (Issuing Date)
               6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

    N.  L.  Hecht, D. S. Duvall ,  A.A.  Ghazee, B.L. Fox
              8. PERFORMING ORGANIZATION REPORT NO.
3. PERFORMING ORGANIZATION NAME AND ADDRESS
     University of Dayton Research  Institute
     300  College Park Ave.
     Dayton,  Ohio  45469
               10. PROGRAM ELEMENT NO.

                 C73D1C
               11. CONTRACT/GRANT NO.

                 R-804421-01
12. SPONSORING AGENCY NAME AND ADDRESS
     Municipal  Environmental Research  Laboratory- Cin., OH
     Office of Research and Development
     U.S.  Environmental Protection  Agency
     Cincinnati, Ohio   45268
               13. TYPE OF REPORT AND PERIOD COVERED
                 Final
               14. SPONSORING AGENCY CODE


                 EPA/600/14
15. SUPPLEMENTARY NOTES
     Project Officer  : Stephen  C.  James  (513) 684-7881
16. ABSTRACT
 This project has  investigated several processes  for the conversion of the organic
 fraction of municipal  solid waste (MSW) to a  powder.  The study concentrated on  two
 types of processes:  (1)  conversion of MSW to  a powdered carbon char by low-temperature
 pyrolysis, and  (2)  embrittlement of cellulose waste by thermal-chemical treatment.
 This report describes  the results of these studies.
 The first phase of  this  project was devoted to identifying processes that offer a
 potential for enhanced product recovery, an evaluation of chemical treatments  to
 improve carbon  recovery  from pyrolysis processes,  an evaluation of laboratory
 processes for the production of gaseous and liquid fuels, and a laboratory
 investigation of  embrittlement processes for  cellulose wastes. The second phase of the
 program'was concerned  with further laboratory studies of the embrittlement  process,
 pilot studies of  the embrittlement process with  shredded newsprint and refuse  derived
 fuel, and an engineering and economic assessment for a plant to process powdered
 cellulose for use as fuel. A comprehensive description of Phasel was presented in an
 earlier report  entitled "Investigation of Advanced Thermal Concepts for Obtaining
 Improved MSW-Derived Products" (EPA-600/7-78-143). This report summarizes the
 results of Phase  I  and provides a comprehensive  review of the work conducted  in
 Phase II of the program.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
  b.lDENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/Group
     Liquefaction
     Municipal Solid  Waste
     Pyrolysis
     Refuse
   Carbon Char
   Cellulose Embrittlement
   Fuel
   Hydrogenation
   Resource Recovery
   Solid Waste  Management
                 13B
18. DISTRIBUTION STATEMENT

     Release to  Public
  19. SECURITY CLASS (This Report)

   Unclassified	
               21. NO. OF PAGES
                  125
  20. SECURITY CLASS (Thispage)
   Unclassified
                             22. PRICE
EPA Form 2220-1 (Rev. 4-77)
113
it U.S. GOVERNMENT PRINTING OFFICE: 1980--657-165/0023

-------