EPA-660/2-74-090
DECEMBER 1974
Environmental Protection Technology Series
Conversion of Cattle Feedlot Wastes
to Ammonia Synthesis Gas
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Corvallis, Oregon 97330
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped into
five series. These five broad categories were established to
facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in
related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY STUDIES series. This series describes research
performed to develop and demonstrate instrumentation, equipment
and methodology to repair or prevent environmental 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 report has been reviewed by the Office of Research and
Development, EPA, and approved for publication. Approval does
not signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement or
recommendation for use.
-------
EPA-660/2-74-090
December 1974
CONVERSION OF CATTLE FEEDLOT WASTES TO
AMMONIA SYNTHESIS GAS
By
James E. Halligan
Karl L. Herzog
Harry W. Parker
Robert M. Sweazy
Department of Chemical Engineering
and
Water Resources Center
Texas Tech University
Lubbock, Texas 79409
Grant No. R801065
Program Element 1BB039
ROAP 21-BES Task 031
Project Officer
R. Douglas Kreis
Robert S. Kerr Environmental Research Laboratory
National Environmental Research Center
P. 0. Box 1198
Ada, Oklahoma 74820
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CORVALLIS, OREGON 97330
For saie by the Superintendent of Documents, U.S Government Printing
Washington. D.C. 20402 - Stock No. 5501-00992
-------
ABSTRACT
A study was undertaken to determine the potential of a process to con-
vert cattle feedlot manure into anhydrous ammonia. Due to the fact
that ammonia is currently produced on a large scale using natural gas
and air, only the processing associated with a reactor system to con-
vert the manure into a suitable synthesis gas was considered in this
study. The synthesis gas can be further processed to anhydrous am-
monia using existing technology.
During this phase of the project, a 4.1-cm diameter fluidized-bed
reactor system was designed, constructed, and operated. Significant
yields of synthesis gas were obtained when this system was fed a mix-
ture of manure, air, steam, and a very small stream of carbon dioxide.
This synthesis gas was deemed to be compatible with existing ammonia
plant technology and suitable for subsequent conversion to anhydrous
ammonia. The upper limit of the production rate of this gas was not
determined in this study; however, the ammonia equivalent of manure
was shown to be in excess of one-half metric ton of ammonia per metric
ton of dry, ash-free animal waste. Assuming that on an as-is basis
the manure contained 15% moisture and 25% ash, this would translate to
approximately 317 kg of ammonia per metric ton of manure received.
This report was submitted by Texas Tech University under the partial
sponsorship of the Environmental Protection Agency, Program Element
1BB039 in fulfillment, of Grant No. R-801065. Work was completed as of
December 31, 1973.
ii
-------
CONTENTS
Page
Abstract ii
List of Figures iv
List of Tables v
Acknowledgements vi
Sections
I Conclusions 1
II Recommendations 3
III Introduction 4
IV Experimental Equipment, Procedures, and Materials 11
V Discussion of Results 20
VI References 43
m
-------
FIGURES
No. Page
1 Partial oxidation of manure-flow diagram 12
2 Reactor details 13
3 Top plate of reactor, bottom view 15
4 Ultimate hydrogen yield from manure 31
5 Hydrogen yields from manure 33
6 Projected ultimate hydrogen yield from manure 36
7 Carbon distribution between gas and char 39
IV
-------
TABLES
No. Page
1 Summary of Reported Batch Pyrolyses of Manure 8
2 Characterizations of Feedlot Cattle Manures 19
3 Summary of the Partial Oxidations of Manure 22
4 Synthesis Gas Ratios of the Product Gases 37
5 Energy Balances 41
-------
ACKNOWLEDGEMENTS
The financial and moral support provided by the following people and
organizations is gratefully acknowledged:
Mr. K. Burt Watson President, Pioneer Natural Gas Company
Mr. Douglas Kreis Project Officer, U.S. Environmental
Protection Agency
Mr. Thurman Whitis Senior Vice President, Pioneer Natural
Gas Company
Mr. Charles E. Ball Executive Vice President, Texas Cattle
Feeders Association
Mr. Glenn D. Bickel Agricultural Consultant, Southwestern
Public Service Company
-------
SECTION I
CONCLUSIONS
1. For a 4.1-cm in diameter fluidized-bed reactor system similar to
that used in this study, the partial oxidation of cattle feedlot manure,
if followed by conventional desulfurization and reforming steps, can
result in a synthesis gas suitable for the production of anhydrous am-
monia.
2. An upper limit for the synthesis gas production rate was not ob-
served over the temperature range investigated in this study; however,
the ammonia equivalent of the reactor gases was shown to be in excess
of one-half metric ton of ammonia per metric ton of dry, ash-free animal
waste. Assuming that on an as-is basis a manure contained 15% moisture
and 25% ash, this would translate to approximately 317 kg of ammonia per
metric ton of manure received.
3. Tentative energy balances indicate that the partial oxidation step
of a synthesis gas from manure process would require the addition of
little or no chemical and Tatent energy. However, sensible energy con-
servation would be required to make the step adiabatic unless the energy
in the char were to be recovered.
-------
4. At a mean reactor temperature of 800° C, approximately 18% of
the carbon in the manure feed exited the reactor as char. This char
was 54% by weight ash and had a heating value of 2290 cal/gr. On a
dry mass basis, approximately 22% of the manure fed to the reactor
exited in the char stream. Due to the large steam demands associated
with an ammonia plant, this material can probably be used for boiler
fuel. However, disposal of the ash will still be required. If no
other suitable use is developed, this material could be used for
landfill.
-------
SECTION II
RECOMMENDATIONS
1. A larger scale reactor system should be built in order to establish
a design basis for a metric tons-per-day scale plant.
2. The temperature within the reactor should be increased in an effort
to determine the limiting synthesis gas yield.
3. The new reactor system should be constructed with pressure-holding
fittings which would allow investigations at modest pressures.
4. Electrostatic precipitators should be added to the reactor system
to allow greater closure of material and energy balances.
5. The future fluidized-bed system should have provision for char re-
moval by some means other than entrainment so that superficial velocities
within the reactor can be lowered.
6. The development of technology compatible with or downstream of a
manure to anhydrous ammonia should be carefully monitored. Programs are
underway to remove sulfur from power gas at elevated temperatures and to
utilize the ash from cattle feedlot manure. Careful monitoring will in-
sure that developments within this project are consistent with objectives
within those programs.
-------
SECTION III
INTRODUCTION
Approximately 25 million head of feedlot cattle are marketed annually
to supply the beef demands of the United States. The most negative
aspect of feedlot operations is the animal waste* particularly the
manure which can harm the environment through the watershed and through
2
the atmosphere. When cattle are fattened by grazing, the large area
of land involved dilutes the effects of pollution, but this is not so
at the feedlot. Land spreading, still the most used method of manure
management, cannot be economically attractive when large quantities
of manure must be trucked any distance to achieve dilution. Great
piles of manure beside the feedlot result.
The trend toward large feedlots is evidenced by the High Plains of
Texas where in the Hereford-Dimmitt area the estimated capacity of 30
lots is in excess of 800,000 head. Such large numbers of cattle
require that many thousands of acres be in grain crops. In turn, these
crops require such fertilizers as anhydrous ammonia. Therefore, large
ammonia plants are not uncommon in the vicinity of extensive, fed-cattle
operations.
Most of the ammonia.produced in the United States is based on reforming
natural gas with steam to produce the hydrogen-nitrogen mixture required
for ammonia synthesis. Because the present critical shortage of
-------
natural gas may develop into a disastrous shortage, many ammonia pro-
ducers are making provisions to replace natural gas with other hydro-
carbons for various fuel requirements. Obviously, the energy situa-
tion would not preclude the complete elimination of natural gas if
cheaper sources of synthesis gas could be made available. During the
course of this project, we have been seriously contacted by more than
six firms relative to the commercialization of this project.
Halligan and Sweazy have compared previously proposed conversions of
o
solid waste to methane and to oil with their own concept of ammonia
synthesis gas from cattle manure. Their estimates show that each of
these three conversions would have a product value of about $9,000 per
day given 1900 metric tons of manure, the estimated daily production
of 600,000 cattle. However, due to its relative simplicity and its
obvious consistency with the economy of cattle-producing areas, the
conversion to ammonia synthesis gas appeared to be the most attractive
for future development.
The thermochemical calculations made by Halligan and Sweazy were for
the conversion of manure to a gas containing H~ and CO together in a
3:1 molar ratio with Np. This gas mixture would be suitable as feed
to the shift converters preceding an ammonia synthesis loop. The
nitrogen required in the synthesis was to be provided by adding air to
a manure pyrolysis reaction. Although the addition of air would result
in energy release through partial oxidation ot.the manure, the overall
step to synthesis gas was shown to be endothermic. The continuous
reaction scheme; and the fast-heating environment operative within a
fluidized bed led to the selection of this technology for application
to the partial oxidation of manure.
As envisioned by Halligan and Sweazy, ammonia synthesis gas, a relative-
ly pure mixture of hydrogen and nitrogen, would be produced, at least in
-------
part, by the partial oxidation of the animal waste with air. Partial
oxidation is intermediate to pyrolysis and combustion, the two extremes
of thermochemical processing. Pyrolysis is simply the thermal decom-
position of an organic material in the absence of oxygen. When less
than stoichiometric quantities of oxygen are added to a pyrolysis re-
action, a partial oxidation, or partial combustion, of a portion of
the pyrolysis products occurs.
The results of any reported thermochemical processing of manure are
subject to many variables and must be interpreted with care. Although
the dry, ash-free portion of cattle manure is composed largely of cel-
lulose, hemicellulose, lignin, and ligno-protein complexes, the rela-
tive amounts of these constituents depend on the feed ration. There-
fore, the temperature range in which exothermal decomposition begins,
as well as the yields and compositions of the gases, liquids, and chars
resulting from pyrolysis, can differ substantially from one of these
12
natural polymers to the other. The reactions involved in the pyroly-
sis and combustion of cellulose alone have been found to be influenced
by the temperature and the period of heating, by the supply of oxygen,
steam, and other reacting or inert gases, and by the presence of in-
13
organic impurities. Despite the complexity of the situation, trends
can be gleaned from the literature. Albeit with different goals and
intentions several groups of investigators have conducted batch pyroly-
ses of manure.
14
White and Taiganides performed laboratory-scale batch pyrolyses of
beef, dairy, swine, and poultry manures in an effort to establish the
heating value of the gaseous products. Samples were heated at con-
trolled rates to 800° C at which temperature the retort was maintained
until gas evolution ceased. The gas included H2> CO, C02, combustibles
(methane and ethane), and illuminants (olefins and aromatic hydro-
carbons). The gas from beef manure contained an average of 1206 kg cal/kg
-------
of combustion energy per kg of dry, volatile solids charged. No
detailed analyses were performed on the manure samples or on the resid-
ual chars and liquids.
In an effort to maximize yields of liquid organic compounds, Garner
and Smith conducted a series of batch pyrolyses of steer manure.
Using thermogravimetric analysis and differential thermal analysis to
identify a temperature range for conducting the pyrolyses and employ-
ing Evolutionary Operation to guide the experiments, these investigators
determined that low pressures and maximum temperature of 400° to 500° C
were optimum for the production of liquid products.
Due to the estimated high costs for separating the complex liquid pyroly-
sate, which included alcohols, aldehydes, ketones, acids, amines, and
phenols, into its components, the authors ruled out manure pyrolysis for
the value of the liquids. Estimated costs for predrying the manure,
which was assumed to be 80% moisture, were also prohibitive. It should
be interjected that Halligan and Sweazy had previously discounted a
solid waste-to-liquid conversion which would require few separation
Q
steps, the cellulose-to-oil scheme proposed by Appell and others.
Garner and Smith concluded their studies with a simulated, continuous
pyrolysis at 500° C and atmospheric pressure. A purge stream of helium
was passed through the retort to rapidly remove the primary products
of pyrolysis and minimize secondary reactions. The results of this
experiment have been summarized in Table 1.
15
At the Bureau of Mines, Schlesinger and others pyrolyzed feedlot
manures under batch conditions with a terminal temperature of 900° C.
These investigators were interested in converting agricultural wastes
to gas, liquid, and low volatile solids to facilitate incineration.
The authors reported that the volatile material was reduced from 66.5%
in the manure charged to 2.7% in the char. A summary of this pyrolysis
has also been included in Table 1.
7
-------
Table 1. SUMMARY OF REPORTED BATCH PYROLYSES OF MANURE
Investigator
Garner and Schlesinger and
Smith11 others15
Temperature
Analysis of dry manure
(weight percent)
Carbon
Hydrogen
Oxygen
... *. W3v-.,
Sulfur
Ash
Yields (weight percent)
dry, ash-free feed)
Gas
Ash-free char
Liquid organics
Aqueous phase
Gas composition
(volume percent)
H2
CO
co2
CH4
C2+
500° C
-
-
.
_
9.2
23.7
29.2
23.7
18.4
11.0
20.4
49.6
16.4
2.7
900° C
42.7
5.5
31.2
2.4
0.3
17.8
48.6
23.6
7.3
20.1
27.5
18.0
24.5
22.7
7.3
8
-------
Garner and Smith minimized the production of gas by purging the retort
with helium, while in the Bureau of Mines pyrolysis there was no purge
with inert gas; therefore, high temperature can be inferred as important
in maximizing gas and minimizing organic liquids and char. The impor-
tance of high temperatures and heating rates for attaining high yields
of gas from cellulosic materials has been stressed elsewhere. *
As gas yields will undoubtedly be of importance in establishing the
feasibility of synthesis gas from manure, continuous reaction schemes
that employ fast-heating environments appear to be the most promising
for future study.
18
Moving beds have been used in the gasification of coal. The principal
advantages of moving beds are maximum heat economy and, for coal, a
high conversion of the solid to gas. At Texas Tech University the
mechanical feasibility of a moving-bed, solid-waste retort has been
19 20
established with manure as the feed. ' This unique retort operates
in a two-step cycle involving partial oxidation and pyrolysis. The
manure undergoes drying, heating, pyrolysis, partial oxidation, and,
finally, cooling as it moves down the reactor. In the pyrolysis zone,
the manure is heated slowly at temperatures no greater than 427° to
538° C resulting in relatively high yields of liquid organics (13.2 kg/
45.4 kg dry, ash-free manure fed). A moving bed appears, then, to
create less than the best conditions for high gas production from a
highly volatile material such as manure.
Rapid heating is one of several unique features of the fluidized bed
that has prompted the application of this technology to the pyrolysis
and incineration of solid wastes. ' ' 2> Additional features
include: efficient gas-solid contact and extended particle surface
which result in high specific gasification rates; rapid solids back-
mixing which creates a thermal flywheel and which would facilitate the
-------
processing of materials with varying moisture content and composition;
and simple construction involving no internal parts. These aspects
make the fluidized bed a potentially attractive means for effecting
the partial oxidation of manure. Historically, the first commercial
application of the fluidized bed was the Winkler process for producing
24
raw synthetic gas from brown coal via partial oxidation.
on OC O~7
Bailie and other investigators ' have used bench-scale fluidized
beds, such as the one employed in this work, to combust manure. In the
38.1-cm diameter, fluidized bed used in the Bailie study, the bed mate-
rial was sand fluidized with a preheated gas mixture of nitrogen and
carbon dioxide containing traces of oxygen. At a feed rate of 11 kg/hr
of wet manure (7.8% moisture) and at a bed temperature of 760° C, the
gas generation was 280 £/0.454 kg of dry manure fed. The pyrolysis
gas had the following composition: 31% H2, 38% CO, 20.6% C02> 7.7% CH-,
1.9% C2H4, and 0.3% C2H6. The mass of gas evolved represented 73% of
the mass of the dry, ash-free manure fed to the bed.
10
-------
SECTION IV
EXPERIMENTAL EQUIPMENT, PROCEDURES, ^ND MATERIALS
This section includes a description of the laboratory-scale facility
for the gasification of manure, recounts operating and analytical pro-
cedures, and characterizes the manures fed in the partial oxidation
studies.
EQUIPMENT
The flow diagram of the bench-scale apparatus is shown in Figure 1.
Central was the fluidized bed reactor which was constructed from 3.8 cm,
Schedule 40, Incoloy 800 pipe to permit operation at temperatures up
to 87° C. The reactor was partially enclosed by a 1000-watt tubular
Hoskins Type FH305 Electric Furnace to supply heat for endothermic
reactions and to compensate for heat losses.
Reactor details are given by the cross-sectioned view of Figure 2.
The fluidizing airstream mixture entered the conical base section of
the reactor through a 1.7-cm outside diameter (OD) Incoloy tube. The
reactor contained two 0.32-cm diameter, swaged magnesia, Type K thermo-
couples. The sheath of the lower thermocouple came up through the
gas inlet and coincided with the center line of the 4.06-cm inside
diameter (ID) reactor. The Chrome!-Alumel junction of this thermocouple
was 13.46 cm above the reactor base, while the upper thermocouple junction
was situated 3.05 cm below the entrance to a 1.27-cm OD Incoloy dip tube which
11
-------
INS
C02
Feed
Hopper
Char
Collector
Reactor - •
Furnace
Furnace
Controller
I
1
- -J
team
era tor
i _
I
i
— >
i
L KV
1
1
1
J
1
> 1
1
1
1
L. _
Water-cooled
Condenser
Condensate Collection
Flasks
Temperature
Recorder
Rotameter
Figure 1
Partial oxidation of manure - flow diagram
-------
10.2 cm
10.7 cm
3.1 cm
13.0 cm
9.7 cm
3.8 cm
Top Plate
(0.25 cm thick)
Flange
(0.95 cm thick)
Dip Tube
(1.3 cm OD
Incoloy Tubing)
Thermocouples
(0.32 cm OD)
Reactor details
Reactor
(4.1 cm Sch. 40
Incoloy 800 Pipe)
Entrance Section
Gas Inlet Tube
(1.3 cm OD
Incoloy Tubing)
Approximately 1/2
Scale
-------
provided an exit for the product gases and for the reacted solids by
entrainment. Each thermocouple was connected to four points of a
twelve-point Honeywell temperature recorder with a range 0° to 1200° C.
The upper thermocouple served as the sensor for a Thermoelectric Model
32422 Indicating Thermocouple Controller which regulated the power to
the furnace.
The reactor contained no inlet gas distributor because its initial
design and operation called for bottom feeding in which manure was
transported by the fluidizing gas up into the bed. However, this mode
of feeding was abandoned due to frequent plugging of the transport line.
In the top feeding arrangement shown in Figure 1, manure was allowed to
fall by gravity from a star-type solids feeder, driven by a variable-
speed Bodine gear motor, directly into the top of the reactor. A small,
metered stream of carbon dioxide was passed through the manure feed
line to prevent condensibles, such as steam, from back-flowing in this
line and creating plugging problems. A manually-operated ram was in-
stalled so that the vertical section of the feed line could be cleaned
of plugs and deposits during a run. Figure 3 shows the positioning of
the manure inlet, the dip tube, and the upper thermocouple with respect
to the reactor center line.
Other equipment items included a steam generator which was simply a
copper box in an electric furnace. A portion of the air feed was passed
through the steam generator to provide smooth vaporization. The total
air rate, as well as the CQ~ rate to the manure feed line, was indicated
and maintained at the preset level by a Brooks-Sho-Rate Rotameter equip-
ped with an integral flow controller.
Downstream from the reactor was a char collector which employed velocity
reduction to separate the reacted solids from the gas stream. A
Magnehelic gauge with range 0 to 12.7 cm of water gave the differential
pressure between the gas inlet to the reactor and the char collector.
14
-------
Flange
Bolt S.
Hole V—v
(4 each) ()
Manure
Inlet \,
o
Reactor
(4.1 cm ID)
-i-.. o
- Dip
Tube
Openine
Upper
Thermocouple
O
9.7 cm
9.7 cm
,
Scale: Approximately Full
Scale
Figure 3
Top plate of reactor, bottom view
15
-------
Two air-cooled flasks followed the char collector to receive condensing
steam and organic vapors. A water-cooled condenser dried the gas to its
saturation level before it exited through a Precision Wet Test Meter
to vent.
OPERATIONAL PROCEDURE
A partial oxidation run was initiated by bringing the empty reactor up
to temperature with air and steam alone flowing. A small back pressure
was temporarily imposed on the system to insure the absence of leaks.
Manure feeding was begun, and the system was allowed to reach a steady
state as determined by the leveling of the reactor temperatures. This
leveling period normally required about 30 minutes. The char and con-
densate collectors were then emptied and a material balance period was
begun. During the course of the next one to two hours, exit gas rates
were determined and recorded. Gas samples were taken with 50-ml syringes
at a point just upstream from the wet test meter. Although the manure
feeder was roughly calibrated for the milled particles fed, the manure
rate was determined using the run time and the difference between the
initial and the final weights of the feed hopper. Likewise, the total
i
condensate rate was determined by the increase in the weights of the
condensate flasks over the material balance period. The char collector
was dumped and thoroughly brushed to obtain the entire solids production
of a period.
A Carle Model 8000 Basic Gas Chromatograph, equipped with a Carle Number
6547 Oxidation Gas Analysis Package, was used to routinely analyze gas
samples. With helium carrier gas flowing at 40 ml/min and with a column
temperature of 38° C, this thermal conductivity chromatograph separated
02, N2, C02, CH4, C2Hg, C2H4, and CO in less than 10 minutes. Mixtures
of these gases were made and peak height versus volume percent cali-
bration curves were prepared for each specie. The chromatograms were
given by a Sargent Model SR strip chart recorder.
16
-------
The water content of a gas sample was estimated by assuming that the
partial pressure of the water was equal to its vapor pressure at the
ambient temperature. The percentage of H~ in a gas sample was esti-
mated by the difference between the total volume and the volume of gas
unaccounted for by this analysis. This differencing technique was
tested by using the Carle chromatograph but with nitrogen as the car-
rier gas to determine the hydrogen concentration of some duplicate
samples taken in a preliminary run. Using the estimated water vapor
content, from 97% to 98% of the gas could be accounted for by the two
analyses, indicating that, when the actual percentage of hydrogen is
20, the percentage determined by difference would be from 22 to 23.
This accuracy was deemed to be sufficient for the purposes of this
study.
In two runs, a Unico Precision Gas Detector fitted with Kitagawa
Detector Tubes, Catalog Number 120a, was used to determine the con-
centration of H2S in the gas exiting to vent.
The chars and the manures fed were characterized as to moisture con-
tent, ultimate analysis, and heating value. The moisture content of
a solid sample was determined by the weight loss upon drying overnight
at 102°-103° C. A Hewlett-Packard Carbon-Hydrogen-Nitrogen Analyzer
was employed to obtain an elemental analysis exclusive of oxygen. The
28
reader is referred to the C-H-N Manual for the details of this tech-
nique. The ash content of a sample was taken as the material remain-
ing upon ignition in a muffle furnace at 750° C. A Parr Peroxide Bomb
Calorimeter was used to determine the thermal values of manure and
char samples. The use of this instrument followed the operating ,in-
?Q
structions given by Parr. As recommended in the instruction manual,
a hydration correction of 0.111° C per gram of sample charged to the
bomb was assumed in calculating all heats of combustion. A Parr
Turbidimeter was used to obtain the total sulfur content of manure
samples.
17
-------
The chemical oxygen demand and the total, volatile, and fixed residues
of a condensate were obtained by the techniques described in Standard
Methods.30
FEED MATERIALS
The cattle manures used as feeds in the partial oxidation studies were
obtained from the nearby Texas Tech experimental feedlots. Two batches,
hereafter designated as Manure "A" and Manure "B", were collected.
Each batch was milled so as to completely pass a No. 40 USS Sieve.
The -40 +60 portion was retained as feed material. Samples in this
particle-size range were taken for the solids characterizations described
in the preceding section. Table 2 presents the results of these tests.
Also presented are the ranges of values for four pen and lot manure
samples obtained from commercial feedlots in the Hereford-Dimmitt area.
Except for having lower moisture contents, the experimental manures
appear to be representative of High Plains cattle feedlot manures.
18
-------
Table 2. CHARACTERIZATIONS OF FEEDLOT CATTLE MANURES
Experimental feedlot
manures
Moisture content
(weight percent)
Ultimate analysis u
(weight percent)
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Heating valueh r
(kg cal/kg)D'c
aAs received basis
Dry basis
"A"
9.6
42.6
5.5
2.8
-
24.9
3528
"B"
7.6
39.7
4.8
2.9
0.5
26.6
3256
Commercial feedlot
manures
15.3 -
35.1 -
5.3 -
2.5 -
0.4 -
23.5 -
3194
36.7
39.6
5.9
3.1
0.6
29.2
3739
19
-------
SECTION V
DISCUSSION OF RESULTS
Bench-scale, fluidized-bed, partial oxidations of feedlot cattle manures
were conducted to investigate the feasibility of transforming this
solid waste to ammonia synthesis gas. Bottom feeding of manure to the
reactor was initially selected in an effort to obtain full use of the
unique advantages of the fluidized bed. Bottom-fed manure particles
would be rapidly heated by a hot, back-mixed bed and would be forced
to have some residence time in the bed before being entrained from the
reactor. Although numerous attempts were made, this mode of feeding
was never fully established at the laboratory scale. Something as
simple as a few hairs bridging across the manure transport line was
enough to cause plugging of this line and terminate operation. The
first 32 runs in this series were mostly total failures with perhaps
3 or 4 very partial successes.
Little information was gained from the bottom-fed runs because mechani-
cal operability was the primary concern. Although top feeding present-
ed other problems, it was a reliable means of admitting manure to the
heated environment of the reactor, and the six top-fed runs discussed
in this chapter offer considerable insight into ammonia synthesis gas
from manure. First, discussion of the overall aspects of the partial
oxidations will be given.
20
-------
PARTIAL OXIDATION RUNS
The operating conditions and some important results of the top-fed
partial oxidations are presented in Table 3. As indicated, a signifi-
cant feature of operation was the sometimes large temperature gradi-
ents within the reactor. Although the upper reactor temperature was
constant during a run because the upper reactor thermocouple was the
sensor for the furnace controller and variations in the lower reactor
temperature were small to moderate (± 15° C), the absolute instantaneous
difference between the two temperatures ranged up to 99° C. Such
temperature gradients are not to be expected in a fluidized bed; the
large inventory of solids in violent agitation that characterizes the
fluidized bed creates a thermal flywheel resulting in isothermal oper-
91
ation. Except during Run 5 and during a part of Run 2, the upper
section of the reactor was at a lower temperature. A "cold" upper
section would suggest that the top-fed manure was creating a heat load
which could not be absorbed by th€ available bed. The greatest tem-
perature differences occurred during Run 1 in which the manure rate of
0.5 kg/hr was more than twice as high as the feed rate in any other
run.
Although the C02, which was passed through the manure feed line, did
aid in transporting char from the bed, this stream was not considered
to contribute to fluidization. The fluidization velocity was to
be the sum of the steam and the product gas superficial velocities.
(The product gas is the exit gas free of the added C02 and water vapor;
the pyrolysis gas is the product gas free of N2 and 02.) The fluidiza-
tion velocity was set at from 0.29 m/sec to 0.40 m/sec in the 6 success-
ful top-fed runs.
An earlier attempt to fluidize with a velocity of 0.11 m/sec was un-
successful as no char was entrained from the reactor until the latter
21
-------
Table 3. SUMMARY OF THE PARTIAL OXIDATIONS OF MANURE
ro
PO
Manure fed
Reactor tempera-
tures (°C)
Upper
Lower
Mean
Superficial ve-
locities
(m/sec)
Steam a
Product gas
Total
Material balance:
Inputs (kg/hr)
Wet manure
Air
Steam
co2
Total
Run 1
"B"
686.1
770 (±15)
728
0.17
0.23
0.40
0.403
0.095
0.150
0.040
0.748
Run 2
"A"
696.1
690 (±6)
693
0.34
0.05
0.39
0.159
0.039
0.330
0.020
0.548
Run 3 Run 4
"A" "A"
715
727.2(±2.7)
721
0.24
0,07
0.31
0.164
0.039
0.225
0.020
0.448
730
782.8
757
0.25
0.12
0.37
0.212
0.039
0.225
0.020
0.496
Run 5 Run 6
"A" "A"
775
757.1(±2.2)
766
0.25
0.09
0.34
0.147
0.039
0.225
0.020
0.431
777.8
813.9(±7.2)
797
0.21
0.08
0.28
0.107
0.039
0.180
0.020
0.346
-------
IV)
CO
Table 3 (continued). SUMMARY OF THE PARTIAL OXIDATIONS OF MANURE
Manure fed Run 1
"B"
Material balance:
Outputs (kg/hr)
Total gas 0.314
Wet char 0.101
Condensates 0.231
Bed accumula- 0.045
tion
Total 0.691
Percent recovery 92.4
of mass
Run 2 Run 3 Run 4 Run 5 Run 6
"A" "A" "A" "A" "A"
0.088 0.113 0.168 0.121 0.119
0.070 0.076 0.105 0.085 0.058
0.363 0.226 0.201 0.221 0.084
0.021 0.026 0.019 . . 0.012
0.542 0.441 0.493 0.427 0.273
98.8 98.3 99.2 >99.2 80.0
Excludes added C02
Run 1
Run 2 Run 3 Run 4 Run 5 Run 6
Average product
gas yield9
(I at 0° C
and 1 atm. kg
dry, ash-free
manure fed)
807
603
721
862
975
1229
-------
Table 3 (continued). SUMMARY OF THE PARTIAL OXIDATIONS OF MANURE
ro
Average product gas
analysis9'0
(volume percent)
hL-by difference
CO
co2
CH4
C2H4
N2
°2
Char analysis
(weight percent)
Moisture
Carbon
Hydrogen
Nitrogen
Ash
Run 1
25.6
16.4
18.3
7.3
2.7
28.4
1.4
• *
. .
* •
Run 2
26.2
12.6
13.1
6.2
2.7
38.3
0.9
43.2
40.4
1.4
2.5
36.7
Run 3
20.0
14.3
20.1
7.8
3.3
34.0
0.4
62.9
41.2
1.0
2.5
39.5
Run 4
18.1
15.4
21.4
7.9
4.1
30.7
2.5
55.0
36.4
0.3
2.4
Run 5
26.7
17.3
14.7
6.4
3.5
31.0
0.3
• *
• t
* •
• 4
Run 6
24.2
11.5
18.7
6.8
3.7
34.5
0.6
62.7
32.0
1.6
2.0
53.9
Dry char heating
(Btu/lb) . . 5535 5502 . . . . 4132
-------
Table 3 (continued). SUMMARY OF THE PARTIAL OXIDATIONS OF MANURE
ro
in
Run 1
Dry bed analysis
(weight per-
cent)
Carbon . .
Hydrogen . .
Nitrogen . ,
Ash . .
Total condensate
characteristics
(mg/1)
Chemical oxygen . .
demand
Total solids . .
Volatile TS . .
Fixed TS . .
Suspended - . .
solids
Fixed SS . .
Run 2
• »
• •
* *
4 *
40925
12868
12433
570
450
120
Run 3
20.8
0.7
0.7
87.5
35800
1 5830
14915
2010
1955
55
Run 4 Run 5 Run 6
10.5
^0
0.2
98.7
47000
30930
29150
11595
11180
415
'Excludes added C02.
Average of 3-4 gas samples except Run 4 where only one gas sample was taken.
^
'Dry basis, except for percent moisture which is on a wet basis.
-------
part of the run when raw, unpyrolyzed manure was carried over to the
char collector. An examination of the reactor showed that a fixed, non-
flowing bed had built up to the dip-tube level.
The higher velocities which were found to be required for successful
operation were not entirely consistent with fluidization theory. A
24
correlation, given by Kunii and Levenspiel and due to Wen and Yu, was
used to estimate the minimum fluidization velocity. For 40 - 60 mesh
3
char particles of density 256.8 kg/m this correlation predicts a minimum
fluidization velocity of .0024 m/sec if the fluidizing gas is steam at
704.4° C. The char density was estimated by combining the experimental
observation that a particle retained only about 20% of its mass in being
transformed from wet manure to dry char with a reported literature value
3 2
of 642.0 kg/m for compacted manure containing 10% moisture- With the
same set of conditions a graphical correlation, again given by Kunii and
24
Levenspiel but due to Pinchbeck and Popper, predicts that char entrainment
from the 4.1-cm ID reactor would begin at a velocity of 0.18 m/sec. The
higher velocities that were set in the successful runs led to increased
char entrainment, but entrainment was necessary as it provided the only
•means of egress for the reacted solids. These velocities were about 150
#
times greater than the estimated minimum velocity of 0.0024 m/sec for char
particles. However, examinations of the bed materials did show the bed
particles to be agglomerates of from five to ten char-sized particles.
The experimental fluidization velocities were at least ten times the mini-
mum for these agglomerates.
There was no evidence that ash fusion was the cause of these easily
crushed agglomerates. The beds often contained 0.64- to 1.3-cm agglom-
erates which were ashen in appearance. Some, but not all, of these
larger agglomerates are believed to have arisen from the infrequent use
of the ram to clear the manure feed line of plugs. The formation of these
unfluidized particles may have contributed to the plugging problems en-
countered in the bottom-fed runs.
26
-------
The steam flow in the top-fed runs was used more as a means of setting
the fluidization velocity than as a reactant. The use of steam arose
from the belief that the problems with bottom feeding were due, in
part, to irregularities in the manure feeding. The star feeder in-
he^ently dumped discrete batches of manure into the transport line lead-
ing to the base of the reactor. When, as in the bottom-fed runs, the gas
evolved from manure constituted the bulk of the gas within the reactor,
the bed could collapse between "slugs" of highly reactive manure. The
high reactivity of the manure was evidenced by the observation that
surges of vapors to the condensate collection flasks were in phase with
the batches of manure dumped from the feeder. The high reactivity of
manure was further evidenced by the 90% ash content of one bottom-fed bed.
Therefore, to dampen the effects of intermittent feeding and stabilize
the gas velocity within the reactor a steady flow of steam was added to
the airflow. The steam was generated under the system back-pressure of
no more than 25.4 cm H20 and carried little superheat. It should be
emphasized that it is not anticipated that such a large amount of steam
would be required at a larger scale. Its use was principally due to the
scale of the reactor system.
The effects of steam addition to a partial oxidation reaction must be
considered. Experiments with municipal solid waste (MSW) have shown that
the steam-carbon reaction, the steam-hydrocarbon reactions, and the
water-gas shift reaction all take place, and appropriate shifts in the
product distribution have been observed. The endothermic steam-carbon
and steam-hydrocarbon reactions were used to explain increased water
consumption as the pyrolysis temperature was -increased from 482.2° to
982.2° C. However, the consumption of water was low and relatively
constant at about 5 kg FLO/lOO kg MSW at temperatures below 815.6° C.
The water consumption did increase drastically to about 13.6 kg/45.5 kg
between 815.6° and 982.2° C . Studies made with coal char have shown
that the rate of the steam-carbon reaction is quite low at temperatures
32
below 926.7° C. The use of steam is not, then, thought to have led
-------
to increased production of ultimate hydrogen (hU + CO) in the lower
temperature runs of this study.
As shown in Table 3, the total output of gas condensates and char was
not sufficient to account for the total input of mass. Accumulation
of solid material in the bed accounted for the remainder. The runs
were, then, not strictly steady-state runs, but, if a bed remains inert,
such accumulations will not lead to transients in the various product
rates and compositions. The beds of Runs 3 and 4 were found to be largely
inert ash. These bed analyses each sum to greater than 100% because a
portion of the total carton given in the C-H-N analysis remains as in-
oq
organic carbonate carbon in the residue of the ash determination. A
bed analysis is probably not the true inrun bed composition because the
reactor had to be cooled before bed samples could be taken. Devolatilization
of the bed material could have occurred during the initial states of this
cooling period. However, as previously alluded to, the high inertness of,
or lack of, the beds was evidenced by imperceptible gas generation during
periods of no manure feeding. Despite accumulation in the beds, the bed
material collected from the one- to two-hour runs weighed no more than
22 to 47 grams. The expanded beds during a run were then quite dilute.
Dilute beds would be consistent with large temperature gradients within
the reactor and with fluidization velocities which were theoretically high.
The solids characterizations showed the chars to be considerably less inert
than the beds. The chars had heating values of 7167 to 7611 kg cal/kg of
total carbon in the char. These heats of combustion compare with that of
graphite carbon at 7833 kg cal/kg. As the carbon exiting with the char
ranged from 16.8% to 26.2% of the carbon in the manure fed, substantial
portions of the incoming combustion energy were unutilized in the reactor.
There was some concern, due to the high reactivity of the manure and the
dilute beds obtained, that the top-fed manure particles were being en-
trained from the reactor by the gas evolved from the manure itself before
being completely pyrolyzed. However, the yields of ash-free char were,
28
-------
in fact, lower than, those of prolonged batch pyrolyses of manure that
are reported in the literature (See Table 1). Apparently, the reactor
with its large surface-to-volume ratio was being supplied with enough
heat from the furnace coils to compensate for the lack of bed material.
For unheated reactors much greater amounts of hot, back-mixed bed mate-
rial will undoubtedly be required to obtain the fast heating rates
simulated in these experiments.
As shown in Table 3, high strength condensates were produced by the pro-
cess. It should be noted, however, that excessive amounts of steam
were employed in order to stabilize the fluidization in the reactor, an
unnecessary operation for larger reactors, thus accounting for the re-
latively large volumes produced. Nonetheless, significant volumes of
high strength condensates will be produced and must be handled effective-
ly.
From Figure 7 it is seen that at 001.67° C, the anticipated Phase II
reaction temperature, about 70% of the product carbon is in the gas and
solid phases, leaving 30% in the liquid or condensate phase. If it is
assumed that the manure as received contains 40% carbon on a dry weight
basis and 20% moisture, that all the carbon containing condensates are
3
alkanes weighing 960 kg/m , that all the water in the manure exits the
reactor, and that 0.1 kg steam/kg manure will be required, then approxi-
3
mately 41.2 kg or .041 m of condensates will be produced per 100 kg of
manure reacted. Referring to the previously mentioned Hereford area with
3
a capacity for feeding 600,000 cattle/day, about 953.8 m of condensate
will be produced daily.
Because of the exceedingly high organic strength of the condensate, con-
ventional treatment of this emulsified waste would prove inadequate.
Therefore, stage condensation will be employed in Phase II to collect the
condensible by-products in relatively pure states. These will then be
recycled or marketed as heating fuel. The water, approximately 696.44 m ,
29
-------
can be utilized for steam production, treated, if necessary, and used
for irrigation purposes, or permanently impounded and allowed to evap-
orate.
AMMONIA SYNTHESIS GAS FROM MANURE
The stated goal of this study was to determine the feasibility of pro-
ducing a gas suitable for the synthesis of ammonia from cattle manure.
The yields and compositions of the product gases were then of primary
importance in this work.
As expected, the gases evolved from manure were composed largely of FL,
CO, C02> and CH^. The unsaturated hydrocarbon ethylene (C2H4) was
the only other gas of pyrolysis present in sufficient concentration to
be detected in the routine gas analysis. The percentage of methane in
the Np- and O^-free pyrolysis gas was essentially constant as it ranged
only from 9.3 to 11.9. The ethylene concentration in the pyrolysis gas
followed no trend during the course of the runs but ranged from 3.8 to
6.1 volume percent. The gas from the fluidized-bed pyrolysis of manure
at 760° C by Bailie25 contained 7.7% CH4, 1.9% C2H4, and 0.3% C2Hg; how-
ever, the low temperature, batch pyrolysis of Garner and Smith^ contained
more ethane ^Hg) than ethylene. In Runs 5 and 6 the routine gas analy-
sis was supplemented with a semi-quantitative, hydrocarbon gas analysis
given by a chromatograph equipped with a hydrogen flame ionization de-
tector. All samples were shown to have ethane and propylene in amounts
less than 10% of the ethylene.
In Figure 4 are presented the yields of ultimate hydrogen obtained direct-
ly from the partial oxidations of manure. The ultimate hydrogen yield
was defined as the sum of the pound moles of H? and CO produced per pound
of dry, ash-free manure fed to the reactor and is plotted versus the mean
reactor temperature. Carbon monoxide was considered to be hydrogen in that
it directly results from the partial oxidations because, in a conventional
30
-------
2.5
CM
O
0)
s-
13
C
fO
O)
O)
t/)
re
£
CD
O
Ol
o
-a
'al
OJ
a;
4->
(O
E
2.0
1.5
1.0
.5
0.0
250
200
150
100
OJ
S-
Z3
C
ra
a)
01
to
to
c:
O
50.0
en
r\
-o
i—
0)
c:
01
cr
LU
QJ
-!->
E
•^
-*->
0.0
675
700
725
750
800
825
Mean Reactor Temperature ° C
Figure 4
Ultimate hydrogen yield from manure
(before reforming)
31
-------
ammonia plant, CO is shifted to H~ via the water gas reaction:
CO + H20 £ H2 + C02 (1)
The reactor temperature, since it fixes the heating rate to particles
of a given size, was expected to be a significant variable. The mean
reactor temperature, which is the arithmetic average of the upper and
the lower reactor temperatures, was found to be the best means of cor-
relating the various yields of all runs. Although this means of data
correlation led to considerable scatter, higher temperatures are shown
to result in greater yields of ultimate hydrogen. On a per kg of dry,
ash-free manure basis, the ultimate hydrogen yield increased from 19.1
to 35.9 kg over the 90° C range investigated. Again on a per kg basis,
the ammonia equivalent of the ultimate hydrogen yielded is from 108.2
to 203.2 kg.
Of course, a variable in addition to the reactor temperature which could
have influenced the yields of any combustible gas was the amount of air
added to the partial oxidation reaction. The relative air rate in these
runs did range from 0.27 to 0.53 kg air/kg dry, ash-free manure fed.
However, the highest relative air rate was in Run 6 in which the highest
yields of all combustible gases were realized.
As shown in Figure 5, the yields of the hydrocarbons, methane and ethylene,
were well correlated by the mean reactor temperature. The yield of each
of these hydrocarbons more than doubled as the mean reactor temperature
was raised from 693.3° to 795.6° C.
The results of these bench-scale partial oxidations can be put on basis
more meaningful for commercial process development when the basic techno-
logy of ammonia plants is considered. A synthesis gas that is introduced
as make-up into an ammonia synthesis loop must be relatively free of
inerts to avoid product loss to purge. All hydrocarbons are inerts
32
-------
.4
CM
O
OJ
s.
..u
a>
in
to
T3
cn
to
O>
r
Methane
-------
in ammonia production. A process log sheet provided by a local ammonia
34
producer shows that the make-up synthesis gas contains less than 0.5%
CH. and no other hydrocarbons. The high hydrocarbon contents of the
experimental product gases would make these gases unacceptable to a
conventional ammonia plant.
Excess nitrogen in a synthesis gas is undesirable because reacted
nitrogen remains in the synthesis loop unless purged. Ammonia synthesis
gas normally contains no less than the stoichiometric ratio of 3 moles
H^/mole N2. The H2 + CO to N2 ratios of the experimental gases were
unacceptably low. These synthesis gas ratios were all less than 1.5:1.
A logical means of increasing the synthesis gas ratio, while, at the same
time, purifying the gas from hydrocarbons, would be to thermally reform,
with steam, the gas of partial oxidation. The steam reforming of methane
and ethylene would produce ultimate hydrogen through the reactions:
PLJ _i_ 14 n j~ rr\ -t ^u f9a^
Ln. T H0U -«- LU T orlo \<-a.)
and CH + 2H0 + 2CO + 4H (2b)
With each mole of methane and ethylene yielding four and six moles,
respectively, of ultimate hydrogen via steam reforming, the synthesis
gas ratio could be increased significantly by the application of these
reactions.
35
The asoects of steam-hydrocarbon reforming have been discussed elsewhere. '
3fi 37 38
* ' The aspect that would appear to most limit the usefulness of
reforming the gas from manure is the low sulfur requirement. The total
sulfur in the gas to a reformer must be less than 0.5 parts-per-million
(ppm) to avoid frequent regeneration of the nickel catalyst. Spot deter-
minations of the sulfur content of the exit gases of Runs 5 and 6 were
34
-------
made using Kitagawa H2S Detector Tubes. The H2S concentration in the
added C02-free product gas ranged from 232,000 to 294,000 ppm.
Although it was recognized that a desulfurization step must, in practice,
precede a catalytic reformer, reactions (2a) and (2b) were used to con-
ceptually convert the experimental yields of hydrocarbons to H2 and CO.
The sum of the ultimate hydrogen resulting from the reforming reactions
and that yielded directly from manure constitutes the projected ultimate
hydrogen from manure. These projected ultimate H2 yields are plotted
versus the mean, bench-scale, reactor temperature in Figure 6. The mean
temperature is shown to be an excellent means of correlating these pro-
jected H2 yields. On a per 909 kg of dry, ash-free manure basis, the
projected ultimate hydrogen yield increased continuously from 39.1 kg at
a mean temperature of 693.3° C. The upper yield is equivalent to 486 kg
of ammonia.
Despite the fact that the char analyses were consistent with reported
batch pyrolyses of manure, there was some concern, due to the rather ill-
defined state of fluidization, that these projected yields would not
be representative of those from future fluidized beds. However, as shown
in Figure 6, the projected yield calculated from the results of the
pilot-scale, fluidized-bed pyrolysis of manure by Bailie ~ is very con-
sistent with the bench-scale data. This only indicates that this study
was consistent with other small-scale experiments. Scale up of these data
to larger-scale systems will require additional experimentation.
Synthesis gas ratios of the experimental product gases would be in-
creased by factors greater than two if these gases were to be reformed.
Table 4 gives the synthesis gas ratios of the gases as they exited the
bench-scale reactor and the projected synthesis gas ratios if the-gases
were to be passed through a catalytic reformer to convert the methane and
ethylene to ultimate hydrogen. It should be emphasized that this ratio
may be easily manipulated by changing reactor conditions.
35
-------
5.0
4.0
3.0
OJ
o
O)
s_
«o
01
tu
03
«S
TJ
CT
U>
O)
'o
^ 2.0
CU
ro
o
+J
CU
en
•a
'cu
-M
cu
ra
200 -q
This study
Pilot-scale
fluidized bed
25
100
0
cr
LaJ
CO
tO
.f—
+J
5
"O
O)
a>
o
G.
675 700 725 750 775 800
Mean Reactor Temperature, ° C
Figure 6
Projected ultimate hydrogen yield from manure
(after reforming)
825
36
-------
Table 4. SYNTHESIS GAS RATIOS OF THE PRODUCT GASES
Run Number
Ultimate H:N 123456
Before reforming 1.5 1.0 1.0 1.1 1.4 1.1
After reforming - based on 3.1 2.1 2.5 2.9 2.9 2.5
gas analysis
37
-------
The projections given in Table 4 do show that, if reformed, the experi-
mental product gases would have, in general, synthesis gas ratios con-
sistent with conventional ammonia technology. Of course, the ratios
could have been increased or decreased simply by adjusting the amount of
air added to the pyrolysis reactions.
The partial oxidations which have been conducted show no limit to the
amount of ultimate hydrogen that can be projected from manure. Greater
insight into the situation of gas production can be had from Figure 7
which gives the carbon distribution between the gaseous and solid pro-
ducts of partial oxidation as a function of the mean reactor temperature.
Increased temperature definitely shifts the carbon distribution toward
the gas. However, at no temperature is more than 67% of the carbon that
was fed in with the manure accounted for by the gas and the char analyses.
It was assumed that the remaining carbon left the system in the form of
oils and tars. Since these tended to coat the transfer lines, it was
almost impossible to accurately measure the amounts generated.
The amount of carbon appearing in the gas phase increased by over 30 per-
centage points as the mean reactor temperature was raised from 693.3° to
795° C, and the amount of carbon remaining in the char decreased by less
than 10 percentage points. These facts, taken together, suggest that
increased rates of secondary reactions involving the higher hydrocarbons
that normally condensed in the recovery train were the primary sources of
increased gas production at the higher temperatures. The yields of gas
from secondary reactions could be increased by raising the temperature or
by lengthening the gas residence time in the reactor. The reverse of
the latter was done by Garner and Smith who sought to maximize the pro-
duction of liquids by quickly removing the primary products of pyrolysis
from the heated environment.
38
-------
-o
oj
u_
o
-Q
ta
o
OJ
o
i_
a>
a.
CO
•a
o
O.
o
-Q
S-
«c
CJ
5.5
5.0
4.0
3.0
2.0
1.0 _
0
675
700 725 750 775
Mean Reactor Temperature, ° C
Figure 7
Carbon distribution between gas and char
800
39
-------
Since chars and condensates constituted considerable portions of the pro-
ducts of the partial oxidations, and since the gases of partial oxida-
tion contained hydrocarbons, it is to be expected that the energy require-
ments for these conversions would be less than that for the complete
conversion of the reactive portion of the manure to ammonia synthesis gas.
The energy requirements were estimated for each of the four runs in which
the characteristics of the chars were known. Tentative energy balances
for these runs are presented in Table 5. Each balance assumes that the
steam fed to the reactor was saturated at atmospheric pressure, and, for
the purposes of estimating the latent and sensible energies, that the
condensible fraction of the vapors exiting the reactor had the properties
of steam at the upper reactor temperature. For the condensible vapors,
the gross heat of combustion in the condensed state which is listed as
the chemical energy was assumed to be 7628 kg cal/kg of carbon. This
latter assumption has given closure to a previous energy balance in-
volving a pyrolysis of municipal waste. The carbon content of the
condensible vapors was determined by difference.
The reported chemical energy of the manure and of the char is the measur-
ed gross heat of combustion for that particular solid. Since this quanti-
ty was not measured, the char of Run 4 was assumed to have a gross heat
of combustion of 7222 kg cal/kg of total carbon in that char.
The specific heat of each char was assumed to be 0.3 kg cal/kg0 C.
Finally, the gross heats of combustion of the product gases, which are
the chemical energies of those gases, were calculated from the gas analy-
ses and tabulated combustion data.
Because the estimated contributions of the vapors are large, the energy
balances can be considered as only approximately, but they do give indica-
tions. Although the net energy requirements of the partial oxidations are
relatively larger, the combined chemical and latent energy effects, which
40
-------
Table 5. ENERGY BALANCES
(kg cal/kg, ash-free manure fed)
2
Run Number
3 4
6
Inputs:
Manure
Chemical energy
Sensible energy
Latent energy
Steam
Chemical energy
Sensible energy
Latent energy
4697
0
0
0
229
1644
4697
0
0
0
152
1094
4697
0
0
0
117
841
4697
0
0
0
186
1337
Outputs:
Product gas
Chemical energy
Sensible energy
Latent energy
Char
Chemical energy
Sensible energy
Latent energy
Condensable vapors
(estimated)
Chemical energy
Sensible energy
Latent energy
Combined chemical and
latent energy effect
Sensible energy effect
Net energy requirement
1312
191
0
1131
74
2277
1197
1965
+ 342
1653
237
0
777
53
2099
839
1337
i- 73
2059
279
0
858
69
1562
627
981
- 78
2830
431
0
684
67
1372
623
912
- 236
+1232
1574
+ 976
1049
+ 858
780
+ 934
698
41
-------
exclude the sensible energy losses from the reactor, are small. Even the
most endothermic requirement of 342.2 kg cal/kg of dry, ash-free manure
fed to the reactor is less than 8% of the chemical energy incoming with
the manure. Considerable portions of the incoming chemical energy exits
unutilized with the chars, and the recovery of this energy could make the
net energy requirement of the partial oxidation reaction small.
42
-------
SECTION VI
REFERENCES
1. Uvacek, E., "economics of Feedlots and Financing," The Feedlot,
Dyer, I. A. and O'Mary, C. C. (ed.), Lea and Febiger, Philadelphia,
pp 11-27 (1972).
2. Hart, S. A., "Manure Management," The Feedlot, Dyer, I. A. and
O'Mary, C. C. (ed.), Lea and Febiger, Philadelphia, pp 162-175
(1972).
3. Whetstone, G. A., H. W. Parker, and D. M. Wells, "Study of Current
and Proposed Practices in Animal Waste Management," Environmental
Protection Agency Contract, Number 68-01-9785, Water Resources Center,
Texas Tech University, Lubbock, Tex., pp 7-17 (1973).
4. Tnomas, Sam A., G. D. Bickel, and R. Flood, "Fed Cattle Industry,
1973 Report," Agricultural Development Department, Southwestern
Public Service Company, Amarillo, Tex. (March 1, 1973).
5. Mark, H. F., J. J. McKetta, and D. F. Othmer (ed), "Ammonia," Kirk
and Othmer Encyclopedia of Chemical Technology, 2nd Ed., John Wiley
and Sons, New York, p 274 (1963).
6. Sloan, C. R. and A. A. McHone, "The Effect of the Energy Crisis on
Ammonia Producers," Ammonia Plant Safety (and related facilities),
15, Prepared by the editors of Chem. Engr. P_r_oc[., American Institute
of Chemical Engineers, New York, pp 91-95 (1973).
7. Halligan, J. E. and R. M. Sweazy, "Thermochemical Evaluation of
Aninial Waste Conversion Processes," Paper presented at the 72nd
National American Institute of Chemical Engineers Meeting, St. Louis
(May 21-24, 1972).
S. Feldmann, H. F.. ''Pipeline Gas from Solid Wastes," Chemical Engineering
Applications in Solid Waste Treatment, Weismantel, G. E. (ed.),
American Institute of Chemical Engineers Symposium Series Number 122,
68. AIChE, New York, pp 125-131 (1972).
43
-------
9. Appall, H. R., Y. C. Fu, S. Freidman, P. M. Yavorsky, and I. Wender,
"Converting Organic Wastes to Oil: A Replenishable Energy Resource,"
Bureau of Mines Report of Investigations Number 7560, Pittsburgh
Energy Research Center, Pittsburgh (1971).
10. Grub, W., R. Albin, D. M. Wells, and R. Z. Wheaton, "Animal Waste
Management," Paper presented to Cornell University Conference on
Animal Waste Management, Ithaca, N. Y. (1969).
11. Garner, W. and I. C. Smith, "The Disposal of Cattle Feedlot Wastes
by Pyrolysis," Environmental Protection Agency Contract,
Number 14-12-850, Midwest Research Institute, Kansas City, Mo.
(1973).
12. Nikitin, N. I., Chemistry of Cellulose and Wood, (trans. J. Schmorak),
Israel Program for Scientific Translations, Jerusalem, pp 585-594
(1966).
13. Safizadeh, F., "Pyrolysis and Combustion of Cellulosic Materials,"
Advances in Carbohydrate Chemistry, 23, Wolfrom, M. L. and Tipson,
R. S. (ed), Academic Press, New York, pp 419-474 (1968).
14. White, R. and E. P. Taiganides, "Pyrolysis of Livestock Wastes,"
Livestock Waste Management and Pollution Abatement, American Society
of Agricultural Engineers, St. Joseph, Mo. (1971).
15. Schlesinger, M. D., W. S. Sanner, and D. E. Wolfson, "Energy from
the Pyrolysis of Agricultural Wastes," Bureau of Mines, Pittsburgh
Energy Research Center, Pittsburgh (1972).
16. Bailie, R. C. and M. Ishida, "Gasification of Solid Wastes in Fluidized
Beds," Chemical Engineering Applications in Solid Waste Treatment,
Weismantel, G. E. (ed.), American Institute of Chemical Engineers
Symposium Series Number 122, 68, AIChE, New York, pp 73-80 (1972).
17. Burton, R. S., Ill and R. C. Bailie, "Municipal Solid Waste Pyrolysis,"
West Virginia University, Morgantwon, W. Va. (1973).
18. Mark, H. F., J. J. McKetta, and D. F. Othmer, (ed.), "Manufactured
Gas," Kirk and Othmer Encyclopedia of Chemical Technology, 2nd Ed.,
10, John Wiley & Sons, New York, pp 359-372 (1963).
19. Massie, J. R., Jr., "Continuous Refuse Retort—A Feasibility Investi-
gation," Unpublished M. S. Thesis, Texas Tech University, Lubbock,
Tex. (1972),
20. Massie, 0. R., Jr. and H. W. Parker, "Continuous Refuse Retort—A
Feasibility Study," Paper presented at 74th National American
Institute of Chemical Engineers Meeting, New Orleans (March 12-15,
1973).
44
-------
21. Bailie, R. C. and R. S. Burton, III, "Fluidized Bed Reactors in
Solid Waste Treatment," Chemical Engineering Applications in Solid
Haste Treatment. Weismantel, G. E. (ed.), American Institute of
Chemical Engineers Symposium Series Number 122, 68, AIChE, New York,
pp 140-150 (1972). ~~
22. Copeland, G. G. , "Industrial Waste Disposal by Fluidized Bed Oxida-
tion," Chemical Engineering Applications in Solid Waste Treatment.
Weismantel, G. E. (ed.), American Institute of Chemical Engineers
Symposium Series Number 122, 68, AIChE, New York, pp 63-72 (1972).
23. Ruisard, B. E. and J. E. Hanway, Jr., "Role of Fluidized Beds in
Pollution Abatement—Present and Future," Paper presented at 75th
National American Institute of Chemical Engineers Meeting, Detroit
(June 3-6, 1973).
24. Junii, D. and 0. Levenspiel, Fluidization Engineering, John Wiley
and Sons, New York, pp 16-22, 73, 77-78 (1969).
25. Bailie, R. C., Personal communication to H. W. Parker, Department
of Chemical Engineering, Texas Tech University, Lubbock, Tex.
(May 22, 1973).
26. Davis, E. G. , I. L. Feld, and J. H. Brown, "Combustion Disposal of
Manure Wastes and Utilization of the Residue," Bureau of Mines
Solid Waste Research Program, Metallurgy Research Laboratory,
Tuscaloosa, Ala., Technical Progress Report 26 (1972).
27. Ishida, M. and T. Shirai, "Fluidized Incineration of Chicken
Droppings," Kagaku Kogaku, 32 (5), Abstracted in Chem. Abstr. , 70 :
14238r
28. Hewlett-Packard, Operating and Service Manual for Model 185 C-H-N
Analyzer (1965).
29. Parr Instrument Company, Peroxide Bomb Calorimetry, Manual 122
(1951).
30. American Public Health Association, Standard Methods for the
Examination of Water and Wastewater, 12th Ed., New York,
pp 510-514, 422-426 (1965).
31. McFarland, J. M., C. R. Glassey; P. H. McGauhey, D. L. Brink,
S. A. Klein, and C. G. Golucke, "Comprehensive Studies of Solid
Waste Management," Environmental Protection Agency 2RO-1-EC-00260-05,
Sanitary Engineering Research Laboratory, College of Engineering
and School of Public Health, University of California, Berkeley.
Final Report (1972).
45
-------
32. Feldkirchner, H. W. and J. Huebler, "Reaction of Coal with Steam-
Hydrogen Mixtures at High Temperatures and Pressures," I & EC
Process Design and Development, 4, Number 2, pp 134-142 (1965).
33. Wilson, D., "Mathematical Determination of Total Oxygen in Solid
Wastes," Physical, Chemical, and Microbiological Methods of Solid
Waste Testing, Bender, D. G., J. L. Peterson, and H. Stierli, (ed.)
Environmental Protection Agency, National Environmental Research
Center, Cincinnati (1973).
34. Green, R., Personal Communication to James E. Halligan, Department
of Chemical Engineering, Texas Tech University, Lubbock, Tex. (1973).
35. Tuttle, H. A., "Preparation of Crude Synthesis Gas from Hydrocarbons,"
Chem. Engr. Prog., 48, Number 6, pp 273-275 (1952).
36. Kitzen, M. R. and J. Tielrooy, "What's New in Steam Methane Reformers,"
Petroleum Refiner, 40, Number 4, pp 169-174 (1961).
37. Thomas, C. L., Catalytic Processes and Proven Catalyst, Academic
Press, New York, pp 106-107 (1970).
38. Kataloco Corporation, Bulletin on Kataloco 22-6 Catalysts, Chicago
(1973).
46
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. RE = OFIT NO.
EPA-660/2-74-ODO
2.
4. TITLE AND SUBTITLE
Conversion of Cattle Feedlot Wastes to Ammonia
Synthesis Gas
7. AUTHORiS)
James E. Halligan, Karl L.
and Robert M. Sweazy
Herzog, Harry W. Parker
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Texas Tech University
Lubbock, Texas 79409
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
P. 0. Box 1198, Ada, Oklahoma 74820
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
December 1974
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
>
10. PROGRAM ELEMENT NO.
1BB039 (21 AYV-13)
11. CONTRACT/GRANT NO.
R-801065
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A study was undertaken to determine the potential of a process to convert cattle
feedlot manure to anhydrous ammonia. Due to the fact that ammonia is currently pro-
duced on a large scale using natural gas and air, only the processing associated with
a reactor system to convert the manure into a suitable synthesis gas was considered
in this study. The synthesis gas can be further processed to anhydrous ammonia using
existing technology.
17.
a. DESCRIPTORS
*Cattle; Farm wastes; Gases
13. DISTRIBUTION STATEMENT
Release unlimited
KEY WORDS AND DOCUMENT ANALYSIS
b.lDENTIFI
*Fluidi
Anhydrc
Synthes
19. SECURI
20. SECURI
ERS/OPEN ENDED TERMS c. COSATI Held/Group
.zed bed reactor; 02 01
us ammonia;
is gas; *Feedlots
TY CLASS (This Report) 21. NO. OF PAGES
55
TY CLASS (This page) 22. PRICE
EPA Form 2220-1 (9-73)
J. S. GOVERNMENT PANTING OFFICE: 1975—697-604/75 REGION id
------- |