United States
Environmental Protection
Agency
Water Engineering
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S2-86/034 June 1986
Project  Summary
STORS:  The  Sludge-to-Oil
Reactor  System

P. M. Molton, A. G. Fassbender, and M. D. Brown
  Direct, continuous thermochemical
liquefaction of primary, undigested mu-
nicipal sewage sludge was carried out
to produce a heavy oil and char product
suitable for use as a boiler fuel. The
work was carried out in a prototype
sludge-to-oil reactor system (STORS)
capable of processing sludge with 20%
solids at a rate of 30 L/hr. Up to 73% of
the energy  content of the feedstock
was recovered as combustible products
(oil and char). The oil product had a
heating value of 80% to 90% that of
diesel fuel. These products are capable
of supplying the energy requirements
for dewatering and liquefaction, so that
a wastewater treatment plant based on
the liquefaction concept can, in princi-
ple, be  energy-self-sufficient.
  A standard BOD determination indi-
cated that the  wastewater from the
process was biodegradable. The only
other significant byproduct was a gas
composed of more than 95% carbon
dioxide. The process is therefore rela-
tively nonpolluting.
  STORS was operated for more than
100 hr without any sign of corrosion or
char buildup on the inside walls. Com-
plete feedstock conversion was
achieved at 300°C, with a nominal 1.5-hr
residence time  or 5.8 kg/hr of solids
throughput. An economic assessment
prepared for a conceptual commercial
liquefaction reactor indicated a sludge
disposal cost of $43/dry tonne for a city
of 1 million. This figure is highly com-
petitive with incineration costs. Pro-
jected capital construction costs for a
STORS  unit were $6.1 million, which is
much lower than that for an incinera-
tion plant. The process therefore has
considerable promise for many poten-
tial sites in the United States.
  This Project Summary was devel-
oped by EPA's Water Engineering Re-
search Laboratory, Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).

Introduction
  A current need exists for implementa-
tion of alternative disposal technologies
for sewage sludge. Alternative tech-
nologies need to be energy-efficient,
and, if possible, some product of value
should be recovered from the sludge.
Direct thermochemical liquefaction can
meet these requirements.
  Investigations over many years have
shown that organic material can be con-
verted to bitumen, heavy oil, and distil-
late oils with combustion heats  ap-
proaching those of petroleum products.
An organic solvent liquefaction process
investigated on behalf of the U.S. Envi-
ronmental Protection Agency (EPA) also
proved capable of producing  liquid
fuels, but it was abandoned because of
poor economics and its need for a dry
feedstock.
  The work reported here describes re-
sults  from continuous  liquefaction ex-
periments  using primary sewage
sludge (20% solids) as feedstock. The
sludge was  converted to fuel oil and
char products with up to 73% of the heat
energy of the dry feed organic material,
or it was converted to oil with 80% of the
heat value of diesel fuel. Separation of
products from the wastewater is gener-
ally spontaneous; this fact coupled with
the use of a wet feedstock makes the
energy balance of the direct liquefaction

-------
process particularly favorable when
compared with incineration, for exam-
ple.
  This project, which was based on ear-
lier autoclave (batch  process) work to
make a synthetic asphalt, was designed
to determine whether a continuous
process would  work, whether  useful
products could  be obtained, and
whether the overall process would  be
environmentally acceptable. Answers
to these questions are all positive, but
further research is required before the
technology can be efficiently commer-
cialized.

The Prototype  Sludge-to-Oil
Continuous Reactor System
  Previous work on sewage sludge liq-
uefaction was performed in 1-gal auto-
claves. For this project, we made use of
a prototype continuous unit designated
as STORS  (sludge-to-oil-reactor-
system), which was capable of process-
ing up to 30 L/hr of 20% solids sludge.
The  reactor is a 15-cm O.D., vertical,
stirred tube with sludge injection from
the bottom by means of a metering  in-
jection pump. The unit  is electrically
heated with a 21.4-kW ceramic heater.
Products were water-cooled before col-
lection.
  Seraped-surface  heat exchange was
selected after consideration  of  immer-
sion  heaters and  steam  injection. The
scrapers were driven by a hydraulic
motor inside the  reactor. A complex
gas-handling system was used to trans-
fer products between collection vessels
and  to separate  gaseous and  liquid
products.
  Reactor  instrumentation consisted of
19 thermocouples, 3 pressure gauges,
and 4 pressure transducers. The pres-
sure  transducers were connected to the
intermediate line,  reactor headspace,
injector water feed line,  and storage
tank. The pressure gauges were located
on the storage tank, injector feed line,
and a connecting  line from the reactor
headspace and intermediate line. Data
were collected and stored in a Hewlett-
Packard* data acquisition system with
real-time video display output. This  in-
formation  was used in conjunction with
the gauges to operate and control the
reactor system.
  On leaving the reactor vessel, the hot
pressurized  reaction  products  passed
•Mention of trade names or commercial products
 does not constitute endorsement or recommenda-
 tion for use.
into one of two pressure let-down ves-
sels after being cooled.  The gas-
handling system  was used to effect
product transfer to a collection drum
and to control gases evolved from the
reaction and pass them to a  gas collec-
tion vessel.

Experimental Procedure

  STORS operated with a 20%-solids
sludge feedstock and 5% anhydrous
sodium carbonate.  Primary settled
sewage sludge from  the City of Rich-
land, Washington, was mixed with 1%
polymer and centrifuged to produce the
18% to 20% solids material.
  Bulk products obtained from continu-
ous reactor experiments were collected
in 55-gal drums, which were thoroughly
stirred. A 200-mL sample of the homo-
geneous product was taken for analysis.
After weighing, the mixed product con-
taining water, oil, and  char was ex-
tracted by dichloromethane. The oil and
char suspension/solution in dichloro-
methane was separated from the water
phase, which was evaporated, and the
residue  was weighed, ashed, and
weighed again to obtain total solids, or-
ganics, and ash. The char was separated
from the oil solution by mild  centrifuga-
tion. The oil solution was poured off and
evaporated. The char was air-dried and
then dried in a 110°C oven  overnight.
Both oil and char were weighed to ob-
tain yields.
  Results corresponded with the defini-
tions of char and oil—materials that are
insoluble and  soluble in  dichloro-
methane, respectively.  In some mea-
surements involving oil,  a sample was
physically skimmed off the  surface of
the settled product, and water was de-
canted to leave the char.  This  char
product contained considerable
amounts of entrained oil and is there-
fore different from the char obtained by
extraction. Since a commercial applica-
tion of this technology would use skim-
ming and decantation rather than sol-
vent extraction for product separation,
the comparison of results between
skimmed and extracted products  is of
some value.
  The City of Richland primary sewage
sludge used  for the liquefaction experi-
ments was analyzed for ash, water, fat,
and  heat content, and  its  elemental
composition was  determined. The oil
products obtained from each experi-
ment were analyzed for:
  Water content.
  Elemental  analysis (C, H, N, 0),
  X-ray fluorescence analysis  (S, P,
  metals),
  Heat content (ASTM-D3286-77),
  Ash content,
  Thermogravimetry  (in air and nitro-
  gen),
  Differential  scanning  calorimetry (in
  air and helium), and
  Viscosity, flash point, pour point.
Char yield, ash content, and elemental
analysis were also determined. Waste-
water analysis was  performed on  se-
lected  samples from the liquefaction
runs and included dichloromethane ex-
traction and  determinations of BOD,
COD, suspended solids, evaporation
residue, and pH.
  A mid-run gas sample was taken for
each  run and analyzed by gas chro-
matography to complete the product
analysis scheme.

Results and Discussion
  The first four experiments in the con-
tinuous reactor demonstrated  the es-
sentially complete conversion of sludge
to oil, char, water-soluble organics, and
gas at temperatures as low as 275°C
with a 1.5-hr residence time and with up
to 73% energy yields for char and oil
combined. The spontaneous separation
of oil and char from the wastewater that
was observed during autoclave experi-
ments did not occur consistently. Sub-
sequent runs were therefore performed
with an  increased nitrogen overpres-
sure to determine whether this condi-
tion would improve separation  by sup-
pressing local  boiling. This  result
occurred spontaneously within 24 hr in
most cases, though some oil was still
entrained with the char. The wastewater
was a transparent light brown. The mix-
ture of sludge, water, gas, and product
in the reactor at temperatures above
275°C behaved like a liquid of very low
viscosity. A calculation based  on gas
production during one run indicated
that the dissolved and undissolved gas
concentration in the reactor was about
1.3 times the reactor volume at standard
temperature and pressure.
  No corrosion was observed inside the
reactor, even on surfaces that were not
scraped. Also, no buildup  of char or
coke occurred on any heated  surface
during the approximately 100 hr of reac-
tor operation. We conclude that corro-
sion and coking should not present a
problem in commercial practice,  pro-
vided that reactor materials are care-
fully selected.
  Severe erosion of teflon seals  oc-
curred on the pressure relief valves,  |

-------
however. This erosion was caused by
high-velocity mineral particles that
were entrapped in the outflowing gas.
  Weight and energy yields of oils and
chars extracted from the  liquefaction
products from STORS are listed in
Table 1.
  Theoretical oil and char mass yield
from sewage sludge is 60% to 70% of
the starting material weight because of
water and carbon dioxide  losses from
the feed during liquefaction. Actual
yields varied between 18% and 70%
based on organic material.  At the same
time, energy yields varied between 22%
and 79%,  excluding  one result above
100%.
  Eight continuous runs were per-
formed during  this project, including
five full-length runs and three in which
conditions were changed during the
run. Below 250°C, no noticeable sludge
reaction occurred. A preliminary run at
250°C yielded no oil, though the solids
had been  digested in the  alkaline
medium.  Above 265°C, a rapid en-
dothermic reaction set in, temporarily
exceeding the capacity of  the heaters.
At 275°C, where all but two of the runs
were  performed, feed conversion to
products was total. In the  last 3 runs,
steady-state conditions probably were
not achieved because of mixing inside
the reactor. Product yields  varied little,
whether the reaction was run at a low or
high temperature, or for 1.5 or 3 hr. This
result indicates that the conversion of
sludge to oil occurred in less than the
shortest residence time (i.e., in less than
1.5hr).
  Differences in composition between
the oil and char products were readily
apparent. Carbon, hydrogen, and nitro-
                             gen contents were all increased in the
                             oil relative to the feed, and oxygen was
                             significantly reduced (by a factor of ap-
                             proximately 4). Except for nitrogen, all
                             of the other elements were present in
                             the char in amounts similar to the val-
                             ues in the feed.
                               Because liquefaction is an alkaline di-
                             gestion, it has the potential for exten-
                             sive degradation of toxic organic mate-
                             rials that may be  present in the feed.
                             Preliminary laboratory  work has
                             demonstrated this  degradation for chlo-
                             roform,  lindane, and a 2,4,5-T analog
                             (2,4,5-trichlorophenylpropionic acid).
                             Chromium was concentrated  about
                             twofold in the char, zinc was twofold to
                             fourfold, and lead  was threefold to six-
                             fold.
                               The overall  average heating value of
                             the oil products was 8460 cal/g, with a
                             range of 7678 to 8970 cal/g on  an as-
                             recovered basis. These values compare
                             with n-octane (11,410 cal/g) and are be-
                             tween 67.3% and 78.6% of the n-octane
                             value as recovered. On a  dry, ash-free
                             basis, for example, the heating value
                             translates to 86.8% that of n-octane for
                             run 208 oil. Water content  of the oils
                             varied widely  between less than 1% to
                             more than 12%, and ash content ranged
                             from 2.1% to  above  16%. These varia-
                             tions represent  differences  in product
                             homogeneity  rather than changes re-
                             sulting from reaction parameters.
                               Distillation characteristics  of the
                             product were determined by differential
                             scanning calorimetry (DSC) and ther-
                             mogravimetric analysis (TG). The TG re-
                             sults were far different from fractional
                             distillation. For example, a typical oil left
                             only a  10.6% residue after  heating to
                             530°C in helium, and only a  14.2%
Table 1.
Product Yields and Production Conditions
                                                        % Yield'
Run
No.
208
305
306
313
418
502-1
502-2
502-3A
507-4
507-5
507-6A
509-6
509- 7 A
509- 7 B
Conditions
°C
305
275
305
275
275
275
275
275
275
275
275
275
275
275
min
90
180
180
90
90
105
114
200
165
152
261
252
57
60
psi
1700
1700
1750
1750
2000
2150
2150
2150
1650
1250
2150
2050
2050
2060
%Feed
Solids
20.7
23.6
18.8
19.8
20.2
20.2
20.2
20.6
20.7
20.7
20.7
19.4
19.4
19.4
Weight
Char
23.65
11.29
15.08
12.01
24.14
5.68
4.18
5.91
4.65
28.97
16.22
4.81
—
3.70
Oil
10.23
20.18
24.90
16.91
17.62
9.83
15.71
8.91
17.35
19.63
36.30
14.01
16.90
7.10
Energy
Char
17.8
4.5
8.1
11.5
20.9
3.5
2.0
3.0
3.1
22.8
17.3
2.6
0.
3.0
Oil
26.7
53.4
64.8
44.2
47.7
28.9
45.7
27.9
46.7
56.2
100.
33.8
45.2
19.0
'Yields were calculated from feed solids content on a dry-weight basis.
residue in air. This result can  be con-
trasted with the excessive foaming and
conversion of the sample to a non-
volatile plastic material  that  was ob-
served when fractional vacuum distilla-
tion was attempted. Apparently, the
bulk oil had undergone some changes
during storage outside for 6 months.
The sample given  as an  example was
162 days old at the time of analysis. Vis-
cosities of extracted oils were all above
10,000 centipoise (cp), whereas viscosi-
ties determined on skimmed oils were
below 1000 cp.
  These viscosity data indicate that the
oils may be too unstable for use if they
are stored for longer than 2  months
(60 days), unless the increase in viscos-
ity can be retarded by use of an antioxi-
dant. Use in  a sludge treatment plant
would be expected to be within the 60-
day period, however. An  alternative
would be to perform a fractional distilla-
tion immediately  after production if
foaming problems can be overcome.
  Pour-point determinations were car-
ried out on extracted oils, but none be-
came pourable. Softening began at 107°
to 111°C and continued to 125°C. At
130°C, the oils became thicker. On con-
tinued heating to 155°C, the thickening
continued, and on cooling to room tem-
perature, the  originally soft products
became hard, though still plastic.  In
contrast, pour-point values for oils
skimmed physically from the surface of
the aqueous layer were all about
~2.5°C, indicating a considerable differ-
ence  in properties  between extracted
and skimmed oil products. Corrected
flash points of extracted oils according
to ASTM D-97 were between 168° and
182°C, in the range expected for  a vis-
cous oil.
  Char was produced during these ex-
periments in yields of 14% to 36%. Char
heating values ranged from 1100 to
3100 cal/g, and ash contents  ranged
from 43% to 80%. This variation is  prob-
ably due to incomplete extraction of the
oil from the char. A small-scale combus-
tion test with a char sample showed that
it will burn  and can thus  be used as a
source of fuel for the liquefaction reac-
tor.
  Gas production accounted for 13% of
carbon input, principally  as  carbon
dioxide. This figure did not vary enough
to warrant a  separate calculation  of
mass for each sample. Instead, a fixed
value of 13%  was used  in  calculating
balances.
  The major  component  of the by-
product gas was carbon dioxide, ex-

-------
eluding the nitrogen used for pressure
regulation.  In  practice, the gas pro-
duced in liquefaction would be used for
pressure control. Some hydrogen was
also generated in the early stages of a
run, but in quantities insufficient to be
significant.
  A series of tests were performed on
selected process wastewaters to deter-
mine total dissolved solids, BOD, COD,
and treatability (aerobic and anaerobic).
The pH of the crude  wastewaters
ranged  from 8.1 to 8.4, rising to 8.6 to
8.7 after  dichloromethane extraction.
Solvent-soluble material was  1.7  to
1.9g/L,  suspended solids (char and in-
soluble ash) were 0.2 g/L, and total dis-
solved material  was 67 to 88 g/L.
Biodegradability of organic material in
wastewater was determined as the ratio
         of BOD:COD expressed as a percent-
         age, and it ranged 68% to 74% for mate-
         rial with COD values between 40 and
         62 g/L.
           The result of a preliminary experi-
         ment to determine anaerobic digestibil-
         ity of the wastewater was negative, but
         no problem should  occur in recycling
         the wastewater from a liquefaction unit
         with the primary sludge input into an
         aerobic treatment plant, as the addi-
         tional loading would be minor.

         Liquefaction Process
         Preliminary Economics
           A preliminary economic evaluation of
         the liquefaction process was developed
         for small towns  with  a population  of
         10,000, medium cities of 100,000, and
         large metropolitan areas of 1  million.
                                   The assumed baseline was sludge pro-
                                   duction value of 0.08 kg/capita per day
                                   (dry basis) consisting of 35% ash and
                                   65% organics. The wet sludge, when
                                   mixed with  5%  sodium carbonate,
                                   would have a solids content  of 25%, a
                                   density of 1.1, and a volume of 3.82,
                                   38.2, and 382 m3 for the small, medium,
                                   and large populations, respectively. The
                                   feed input to each size of STOPS would
                                   therefore be 76.2% water, 12.4% organ-
                                   ics, and 11.4% inorganics. Because of
                                   the differences in  size, the basic design
                                   for each STORS would be quite differ-
                                   ent. Cost estimates are compared for
                                   the three  facilities sizes in Table 2.
                                   Table 3 presents a thermal analysis of
                                   the overall  process, and Table 4  esti-
                                   mates the amount and value of excess
                                   oil for each facility. These figures were
Table 2.   Cost Estimate Comparison for Three Facility Sizes
                                    Town of 10,000
                                    City of 100,000
                                                 Metropolis of 1 million
           Item
 Percent of
Capital Cost
 Estimated
   Cost
 Percent of
Capital Cost
 Estimated
   Cost
 Percent of
Capital Cost
Estimated
   Cost
Capital cost element for
  STOPS:
    Engineering Design
    Purchased Materials
    Construction Labor
    Construction Manage-
      ment
    Start-Up Materials
    Indirect Capital Costs
    Capital Cost

Operating cost element for
  STORS:


    Debt Service at 8%,
      20 years
    Maintanance Labor and
      Supplies
    Operator Labor
    Fuel and Soda Ash
Miscellaneous Supplies
    Indirect Maintenance
    Indirect Operating Cost
    Annual Operating Cost
    Annual Operating
      Revenue
    Net Annual Cost
    Sludge Disposal Cost $/T
      (dry)

Incremental annual treatment
  plant costs:

    Operating and mainte-
      nance
    Amortized capital
      Total incremental cost

    Incremental sludge dis-
      posal cost, $/T(dry)
15
68
11
4
1.02
1.46
100
Percent of
Operating Cost
$100,000
$470,000
$ 75,000
$ 25,000
$ 7,000
$ 10,000
$687,000
Estimated
Annual Cost
   39.9

   14.2
   22.8
    8.5
   14.2
    0.1
    0.2
   100
$ 69,972

$ 25,000
$ 40,000
$ 15,000
$ 25,000
$   200
$   300
$175,472

$     0
$175,472

$   601
               $ 56,000
               $ 37,000
               $ 93,000
                    318
13
66
16
3
0.64
1.48
100
Percent of
Operating Cost
$ 200,000
$1,031,500
$ 250,000
$ 40,000
$ 10,000
$ 23,000
$1,554,500
Estimated
Annual Cost
   41

    6
   21
   26
    6
    0
    0
   100
$  158,329

$  25,000
$  80,000
$  100,000
$  25,000
$     600
$    1,000
$  389,929

$  55,672
$  334,257

$     114
                               $ 130,000
                               $  85,000
                               $ 215,000
8
69
15
2
3
3
100
Percent of
Operating Cost
$ 500,000
$4,466,500
$1,000,000
$ 150,000
$ 200,000
$ 170,000
$6,116,500
Estimated
Annual Cost
     35

      5
     13
     40
      5
      0
      0
    100
$ 660,664

$ 100,000
$ 250,000
$ 750,000
$ 100,000
$    3,600
$    2,800
$1,867,064

$ 556,719
$1,310,345

$      45
                                               $ 600,000
                                               $ 400,000
                                               $1.000,000

-------
 Table 3.    Thermal Analysis of the Sludge-to-Oil Process for Three Facility Sizes
 Energy in oil

 Energy of
  excess oil
                          Heating Values for
                                      Heating Values for
8590
6.27E+08

6.27E+08
8590
6.27E+09

3.51E+09
                                                          Heating Values for
Operating
Element
Energy input
of sludge
Heat used in
process
Energy in
char
Town
calories/g
3300
—

7770
of 10,000*
kcal/year
9.64E+08
3.54E+08

7.76E + 07
City of 100,000*
calories/g kcal/year
3300 9.64E+09
3.5E+09

1770 7.76E + 08
Metropolis of 1
calories/g
3300
—

1770
million
kcal/year
964E+10
3.54+10

7.76E+09
                                                                      8590
6.27E+10

3.51E+ 10
 "Excess oil is 83, 416, and 4,160 tonnes/year for the small, medium, and large facilities, respectively. Value of the excess oil is $4/million Btu, or
 $11,135, $55,672, and $556,719 for the small, medium, and large facilities, respectively.
 Table 4.   Estimate Amount and Value of Excess Oil for Three Facility Sizes

     Item           Town of 10,000       City of 100,000      Metropolis of 1 million
Amount of ex-
  cess oil, ton-
  nes/year

Value of excess
  oil @ $4/mil-
  lion Btu
  83
$11,135
     416
    $56,672
       4,160
     $556,719
obtained by assuming  a  char and oil
yield of 15% and 25% of the weight of
the dry feed sludge, with heating values
of 1,770 and 8,590 cal/g, respectively.
The energy required by the process was
obtained by assuming a 150°C heat
input and an overall furnace and heat-
ing system  efficiency of 65%. Also, it
was assumed that all of the char would
be burned before the oil, since surplus
oil has a value of $4/million Btu. Note
that the energy recovered from this
process is  more than three times the
theoretical  amount  needed to heat the
sludge a net of 150°C. This fact provides
considerable  flexibility in furnace
design.
  This summary cost estimate  repre-
sents our best judgment based on our
results. The final result is an ultimate
sludge disposal process  costing $43/dry
tonne for a large city, which is equal to
$11/tonne of 25% solids  sludge. This
cost is almost competitive with  ocean
dumping and is less expensive than in-
cineration. Smaller plants  are less eco-
nomically feasible.

Recommendations
  The project described in this report
demonstrated the feasibility of the con-
tinuous liquefaction process. The early
                 optimism about this process was thus
                 fully justified. We continue to feel that
                 liquefaction offers a cost-effective
                 alternative to incineration—one that
                 would be environmentally acceptable in
                 practice. However, the behavior of the
                 sludge  at elevated  temperature and
                 pressure was unexpected. Based on our
                 experience with the prototype, further
                 work should be performed with a modi-
                 fied reactor to refine the parameters for
                 a pilot plant.
                   To ensure optimum product recovery
                 and quality, further work should also be
                 done on product separation based  on
                 the pressure effect. Solvent extraction is
                 unacceptable in a commercial process
                 for sludge conversion  to fuel. Sponta-
                 neous separation of oil and char was
                 observed in  some experiments and  not
                 in others, but a detailed examination of
                 separation was outside the scope of this
                 project.
                   Liquefaction has shown the potential
                 for concentrating  metals in the inor-
                 ganic fraction of the char and also  for
                 destroying organic materials through
                 alkaline digestion. Further study should
                 be conducted on the potential of the  liq-
                 uefaction process as a novel technique
                 for concentrating heavy metals and ren-
                 dering organic wastes  harmless. Accu-
       rate data on mineral balances and de-
       struction rates should also be obtained.
         An additional research project should
       be performed using a smaller-diameter
       reactor so that extended  runs  can  be
       performed. With  the  use of  sludge
       "spiked" with various  metals and or-
       ganic nonradioactive tracers, the work
       outlined here on process parameter re-
       fining, separation,  and detoxification
       potential can be performed in  a cost-
       effective manner.

       Conclusions
         The results  of experimental studies
       on primary sewage sludge liquefaction
       in a prototype continuous reactor have
       amply confirmed the feasibility of the
       process on a continuous scale  for the
       production of fuel oil. The major objec-
       tive of the  project has  therefore been
       met. New information has also been ob-
       tained on the behavior  of aqueous pri-
       mary sewage sludge slurries at elevated
       temperature and pressure. As in previ-
       ous batch studies, sludge liquefaction
       occurred rapidly above 265°C. Complete
       conversion  to  products (oil, char, and
       gas) occurred within 1.5 hr at 300°C, the
       shortest residence time  achievable with
       the reactor in its present configuration.
         After more than 100 hr of operation,
       there was minimal char buildup, corro-
       sion, and plugging. Some erosion  of
       teflon valve seals did occur, presumably
       because of high-velocity sand and ash
       particles, but this was solved by includ-
       ing a steel wool particle trap in the line.
       Corrosion, plugging, and  erosion are
       not likely to be obstacles in further de-
       velopment of the liquefaction process.
        Product mass and energy yields were
       high, with typical yields of up to 36% for

-------
oil and 18% for char, and combined en-
ergy yields of 73% of the feedstock heat-
ing value. Spontaneous product separa-
tion from wastewater did not occur in all
experiments. Later experiments per-
formed with higher pressures of inert
gas (nitrogen) did achieve spontaneous
separation of a higher-quality  oil
product from wastewater. Further work
needs to be done on product separation
to confirm this result.
  BOD determinations were performed
using an  unadapted microbial seed
from  a  local sewage treatment plant.
The results suggest that the wastewater
from the process would be highly
biodegradable with aerobic digestion.
Byproduct gas was also a minor envi-
ronmental pollution concern.  Metal
concentrations were as much as 27-fold
lower in the product oil than in the start-
ing sludge, but they were more concen-
trated in the char. Any toxic organic ma-
terials produced in this process are
likely to be concentrated in the oil and
would be destroyed during combustion.
Disposal of ash from char combustion
remains a potential problem.
  The preliminary economic compari-
son  of the liquefaction process with
conventional incineration indicates that
liquefaction  is economically more  at-
tractive, both on an initial capital invest-
ment and  on operating  cost basis. For
example, the sludge disposal  cost is
$43/dry tonne for a city of 1 million, and
the capital  equipment cost is only
$6.1 million.
  The full report was submitted in fulfill-
ment of Cooperative Agreement No.  CR
810690-01-0 by Battelle-Northwest
under the sponsorship of the U.S. Envi-
ronmental Protection Agency.
                                                                          GOVERNMB
-------
      P. M. Molton, A. G. Fassbender, and M. D. Brown are with Battelle-Northwest.
        Richland, WA 99352.
      Howard Wall is the EPA Project Officer (see below).
      The complete report, entitled "STOPS: The Sludge-to-QH Reactor System," (Order
        No. PB 86-175 684/AS; Cost: $16.95, subject to change) will be available only
        from:
              National Technical Information Service
              5285 Port Royal Road
              Springfield. VA 22161
              Telephone: 703-487-4650
      The EPA Project Officer can be contacted at:
              Water Engineering Research Laboratory
              U.S. Environmental Protection Agency
              Cincinnati. OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S2-86/034
                  0000329   PS
                  U S  ENVIR  PROTECTION  AGENCY
                   CHICAGO

-------