FY 2005 FoodPAC Final Report

Project Title: Combustion of Poultry Fat for Plant Heat and Steam

Principal Investigator:		Thomas T. Adams, Ph.D., P.E.

Engineering Outreach Coordinator

Faculty of Engineering

Driftmier Engineering Center

The University of Georgia

Athens, Georgia 30606-4435

706-542-0793

  HYPERLINK "mailto:tadams@engr.uga.edu"  tadams@engr.uga.edu 

Primary Industry Partner: 	Ron Welk, Plant Manager

					Cagle's Inc. 

					14075 Highway 116 

					Pine Mountain Valley, GA 31823

					Phone: 706-628-4251

					Fax: 706-628-4403

Research Team:

Daniel Geller, M.S. UGA Faculty of Engineering   HYPERLINK
"mailto:dgeller@engr.uga.edu"  dgeller@engr.uga.edu 

John Goodrum, Ph.D. UGA BAE   HYPERLINK "mailto:jgoodrum@engr.uga.edu" 
jgoodrum@engr.uga.edu 

Bryan Graffagnini UGA Engineering Outreach   HYPERLINK
"mailto:bryan@engr.uga.edu"  bryan@engr.uga.edu 

Other Industry Advisors:

Andy O’Neal, Pilgrim’s Pride Continuous Improvement   HYPERLINK
"mailto:AOneal@pilgrimspride.com"  AOneal@pilgrimspride.com 

Fred Phillips, Director of Research, Clean Burn   HYPERLINK
"mailto:fredcb@cleanburn.com"  fredcb@cleanburn.com 

EXECUTIVE SUMMARY

Presently, most chicken fat from poultry processing plants is sold to
rendering facilities at very low prices. This method provides an easy
way of handling disposal of the fat, but neglects the fact that animal
fat can actually be a valuable byproduct of the food processing
industry. The goal of this study was to examine the possibility of
creating an in-house application for this by-product of the Georgia
poultry processing industry while simultaneously reducing the need to
purchase high-cost, outside energy by providing an alternative fuel
produced on-site. The predicted result of this study was a reduction in
overall costs associated with poultry production and reduced dependence
on external energy sources.

The use of animal fat as an industrial boiler fuel has already been
extensively documented by University of Georgia Engineering Outreach
(Adams, et al., 2002). Rendered animal fats have been shown to be clean,
efficient and effective fuels for such applications. This study sought
to scale down this application and use poultry fat as a fuel for smaller
scale boilers to provide heat or hot water within the facilities that
process poultry and produce the fat itself. By eliminating the need for
transportation and third party processing of the fat, a cost effective
fuel was developed. Clean Burn, a leading manufacturer of recycled oil
burning boilers, offered to donate a boiler for the study. These boilers
are traditionally fueled with petroleum based waste oil such as crank
case oil. This study showed that animal fats provided an effective fuel
for the Clean Burn boiler. Emissions were also generally reduced when
using biologically derived oils as compared to the emissions experienced
when using oil types normally burned in Clean Burn furnaces and boilers.

Cagle's, Inc. Pine Mountain Valley, GA functioned as an industrial
partner in this study. The particular operation we observed produces an
estimated 22,000 lbs. of fat per day for which they desire a value-added
application. Team members studied the recovery and use of their fat as a
fuel for a Clean Burn boiler. In turn, Cagle’s provided fat containing
byproduct for our study. The Athens, GA Pilgrim’s Pride poultry
processing facility also provided byproduct which contained significant
amounts of fat and offered advice, expertise and guidance during the
study.

Multiple processes for extracting useable fat fuel from different
by-products of poultry processing were examined. These included
extracting low fat sources such as offal as well as fat rich materials
such as leaf fat and saddle fat. Aerobic and anaerobic fermentations
were studied for efficiency of extraction along with traditional thermal
rendering techniques. It was determined that traditional thermal
rendering processes provided the most simple, cost effective and
efficient method of fat extraction from these sources.

A process was developed which Cagle's, Pilgrim’s Pride and similar
poultry processing facilities can implement on-site to utilize their fat
byproduct as a fuel. Fat was extracted and combusted in a small (350,000
btu/hr) industrial boiler. Twenty-five individual runs were executed
using a variety of fuels during the study. Critical parameters such as
fuel consumption and efficiency, emissions and performance were tracked
during each run. Financial feasibility of the process was determined to
be dependent on both the current price of heating fuel and the value of
the byproduct fat in other markets. Due to increasingly higher prices of
petroleum based heating oils such as No. 2 Diesel and Kerosene, the
economics of this process are even more viable at the present time than
at the start of this project. Ultimately, results of this study suggest
technical feasibility of such systems in the Georgia poultry processing
industry. Additionally, the economic benefits of these systems will only
continue to grow as the price of petroleum based energy sources
continues to rise.

PROBLEM DEFINITION AND PROJECT GOALS

The intent of this study was to work with the Georgia poultry processing
industry to develop a method of on-site utilization of chicken fat as an
alternative heating fuel. Waste poultry fat is a plentiful commodity in
Georgia. Although the fat has financial value, disposal is generally
accomplished by sale to rendering facilities at undervalued prices. As a
result of poultry processing, more than 44.6 million gallons of chicken
fat, become available in Georgia each year. In its rendered, food grade
state this material can be valued at up to $0.18/lb. ($1.33/gal).
However, much of this fat ends up in waste streams and is sold to
rendering facilities at $.03/lb ($.22/gal). The value of this poultry
fat, based on the current market price for chicken fat sold into the
food market, is $59.3 million. The equivalent in heating value to 44.6
million gallons of chicken fat is 39.9 million gallons of #2 diesel
fuel. Assuming the current price of #2 diesel fuel is approximately
$2.60/gallon this results in $103.8 million in offset fuel costs.
Therefore, utilizing this chicken fat as a fuel could represent a price
differential of up to $44.5 million for the Georgia poultry industry.
Recent increases in petroleum prices have started a rise in fuel costs
which shows little sign of disappearing; making this study even more
relevant. Since the completion of another recent University of Georgia
study (Adams, et al., 2002) which explored the use of animal fats as
fuel for industrial boilers, several industries in Georgia have started
purchasing rendered chicken fat as a less expensive replacement for #2
diesel fuel and natural gas. 

It is logical that on-site rendering would reduce the cost of poultry
fat based fuels for producers leading to additional fuel cost savings
and the creation of a more efficient and environmentally friendly
rendering process for poultry processing by-products. Other domestic
meat processing facilities such as those in the beef industry already
render onsite as do many non-US poultry processing facilities. Savings
in transportation and overhead costs would both benefit the bottom line
and reduce environmental impacts associated with transit of this
material. Additionally, the proximity of the onsite facility would allow
rapid rendering after collection of fatty materials from the waste
stream of poultry production. This would lead to a higher quality fat
for fuel use which would have reduced levels of free fatty acids and
other decomposition products. Resulting fuels would have higher energy
content, storability and oxidative stability.

Due to its relevance in utilization and value enhancement of a byproduct
in an important Georgia food industry, this study directly addresses the
Process and Product Improvement – Value Added Byproduct priority of
FoodPAC as outlined in the FY2005 call for proposals. Not only does this
study result in the production of a value added product in a Georgia
food industry, it reduces the cost of this product by including means
for its production at the same site it will be utilized.

The overall goal of this project was to develop a method by which a
poultry processing facility could extract fat from their waste stream on
site and utilize it in existing oil burning furnaces and/or boilers.
Additional analysis of the product and its properties were also
performed. The specific objectives of this study were:

1. Develop an effective method of fat recovery that poultry processing
plant may accomplish from by-products on site: fuel preparation for
combustion including pre-heating and filtering.

2. Develop efficient in-plant delivery of the fuel into the on-site
boiler including pumping and selection of injection nozzles.

3. Examine the fuel characteristics of the prepared fat.

4. Demonstration of prepared poultry fat biofuel combustion in a Clean
Burn boiler.

5. Determine legal limits on use of in-plant fat as alternative heating
fuel. 

6. Deliver plan for onsite utilization of poultry fat as a heating fuel
at Cagle's, Inc. (Pine Mountain Valley, GA).

7. Publish results of this study and procedures for use of this biomass
as an alternative fuel.

PROJECT SUMMARY AND FINDINGS

This project utilized UGA’s extensive experience in the Biofuels field
and facilities already existing in the UGA Biofuels Laboratory.
Analytical and emissions detection equipment and expertise available at
UGA were also utilized in this project. Donation of a 350 BTU/hr (Clean
Burn Model #CB 350 CTB) waste oil burning boiler from Clean Burn (Leola,
PA) was invaluable in conducting this study. Previous studies by the UGA
Biofuels Team on large scale use of poultry fat in an industrial boiler
provided a solid background for this project. Objectives (1-7) presented
in the previous section were addressed as follows:

Objective 1: Fat extraction by several methods was examined during this
project. Anaerobic and aerobic fermentations as well as direct heat
rendering and autoclave methods were investigated. Additionally, several
waste products were examined for fat content and viability for fuel
applications. Offal (backdoor waste streams), leaf fat (fat from the
upper part of the bird) and saddle fat (fat associated with the hind
halves of the bird) were the three main products examined. 

First, the feasibility of on-site fat extraction from Cagle’s offal
stream was examined. Successful separation of fuel quality fat from this
stream would be a valuable upgrade to this by-product. Initial analysis
of this material was achieved by a thermal rendering process. 150g of
offal was heated in a beaker to 212oF for one hour after which time fat
melted and rose to the top of the beaker. Water in the system was
vaporized and vented. The fat was poured off and filtered through a 400
mesh filter. Final material weights were calculated. It was determined
that this material was composed of approximately 10.4% fat, 47.9% water
and 41.7% solid residues.

Two fermenting extraction methods were also examined; aerobic
fermentation using microorganisms inherent in the offal material and
anaerobic fermentation using silage cultures. Aerobic fermentation was
achieved by incubating offal at 82 oF for ten days without any
inoculation. Anaerobic fermentation was achieved by mixing 5-6% wheat
silage, 1-2% glucose and 90-93% offal in ~100g quantities. Two
treatments were blended for 30 seconds in a food processor and two
treatments were left unblended. These treatments were also incubated for
ten days at 82 oF. At the end of the fermentation period, all treatments
were autoclaved to kill any harmful microorganisms. The material was
then chilled to 3 oC at which point fat solidified and was easily
removed from the top of the digestion by scraping the congealed fat with
a spatula. Fat yields averaged 5.8% for aerobic fermentations and 10.0 %
for anaerobic fermentations. Blending had no significant effect on fat
yields. As total fat in the offal used for the fermentations was taken
from the top of the storage container it was determined that the
material contained 21% fat, on average fermentation yields were able to
recover 28% (aerobic) and 48% (anaerobic) of total fat in each
treatment.

Fermentation proved to be a low energy input method for recovering fat
from poultry plant offal. However, high volumes, long residence times
and low yields prevent this method from being economically viable.
Aerobic fermentation also has the disadvantage of producing a
contaminated byproduct. Visual examination suggested that most of the
separation had occurred after two days of incubation. Assuming two days
residence time and 5% fat recovery (48% recovery of material containing
10.4% fat), anaerobic fermentation would require ~400 gallons of storage
for every 1 gallon of fat collected. This fact and the challenges
involved in recovering the fat from the digestion limit the feasibility
of this process on-site in poultry processing plants. 

Thermal processing of offal requires less total capacity than
fermentation; however the low concentration of fat in offal would
require the heating of massive amounts of material. Approximately 3.4
gallons of water would have to be removed for every 1 gallon of fat
recovered. At 8092 btu/gallon of water vaporized, this would require
27,512 btu per gallon of fat recovered. Each gallon of fat provides
approximately 124,780 btu resulting in net energy recovery of
approximately 78%. 

Leaf fat and saddle fat samples were obtained in 50lb quantities from
Cagle’s (leaf) and Pilgrim’s Pride (saddle). These were pulled on
the line at each facility for the expressed purpose of providing a high
fat containing by-product for this study. Leaf fat is generally kept on
the final product but it often falls off during processing ultimately
ending up in offal. Saddle fat is also treated in a similar manner as it
is desired that it remains attached to the end product. However, it too
often ends up in the waste stream. This material is not 100% fat, but is
much higher in fat content than offal. Thermal rendering process
analysis showed leaf fat as delivered to be 75.25% fat, 8.74% water, and
14.58% other solids. As delivered, saddle fat was 43.43% fat, 29.53%
water, and 27.05% solids. Only thermal rendering was examined for these
highly concentrated fat source byproducts. 25-50lb. batches of these
materials were heated to 212 oF in large steam kettles. Once product
temperatures began to elevate above the boiling point of water, product
was filtered through paper shortening filters resulting in the
separation of liquefied fat from solids. This is the only method that
provided large enough quantities of fuel grade fat for testing in our
industrial boiler. Additionally the energy yield on this process was
much higher than that of offal processing as dewatering of leaf fat has
nearly a 99% energy yield and that of saddle fat is about 96%. 

After the apparent success of thermal rendering, the group attempted to
render fat using an autoclave. 1 pound samples of Cagle’s Leaf Fat
were autoclaved for one hour intervals at 230oF/15psi. Fat content of
solids were examined by visual inspection at the end of each heating
cycle. At the end of four one hour cycles, fat was still visibly
entrained in the original solid material. Pressure from the autoclave
clearly kept the fat solid under autoclave conditions. While this method
was clean and simple, its lack of efficiency and failure to completely
extract fat from the solid material makes it an unlikely candidate for
fat recovery on-site in poultry processing facilities.

Ultimately, it was determined that non-pressurized, thermal extraction
of concentrated fat containing waste material such as the leaf fat and
saddle fat examined in this study was determined to be the most likely
candidate for fuel fat production at a poultry processing plant.
Fermentation methods, while technically feasible, were deemed
economically unviable due to large treatment volumes and extended
residence times. Autoclave methods were also deemed unfeasible both
technically and economically as the high pressure used in these systems
resulted in incomplete extraction even with extended treatment times.

Fat Recovery points were determined from examining the process stream of
both Cagle’s and Pilgrim’s Pride facilities. Optimum recovery points
were determined to be pre-waste stream. Fat that reached the waste
stream required much more energy, residence time and storage volume to
extract due to the large amounts of water introduced during processing.
Recommendations to both facilities involved in this study suggest an
analysis of all locations that introduce fat into waste streams and
implementation of fat recovery processes at these points. In many cases
recovery will be as simple as a line employee routing the fat into a
collection vessel instead of dropping the material into waste. It was
suggested that during processing up to 60% of saddle fat is lost to
offal. Material that would otherwise end up in offal could be recovered
by this method during the processing of hind quarters at the Pilgrim’s
Pride facility quite easily. Other mechanical methods such as screens to
catch large fat particles and skimmers to collect floating fat could
also decrease water content in processed by-product dramatically.
Ultimately, any process that can inexpensively prevent introduction of
water into concentrated fat containing materials and prevent these
materials from entering offal will facilitate the efficient collection
of this potential fuel.

 

Objective 2: Pump and nozzle selection was accomplished through the
advisement of Clean Burn technicians and product developers. In house
BTU analysis determined energy value per volume of biofuel was
determined to be within 11% of that of petroleum diesel and other liquid
fuels used in these boilers suggesting similar volumes of fuel would be
required for optimal operation of the boiler when running on biofuel.
Two Clean Burn stock 9-5 injectors were used in the study and were
traded out periodically for cleaning. A continuous 2.5 gallon per hour
pump was included with the CTB350 boiler and was used for all fuels. An
optional oil regulator was left inline on the system as it allowed us to
make minor adjustments to oil flow when necessary. 

Objective 3: Fuel properties were measured in three laboratories:

University of Georgia Biofuel Testing Laboratory: Energy content,
specific gravity and viscosity.

PSC Analytical Services Laboratory (Hatfield, PA): Ultimate Elemental
Analysis (C,H,O,N,S – ASTM D5291 and ASTM D4239)

Eurofins Scientific Laboratory (Des Moines, IA): Triglyceride Profile,
Moisture (AOCS Ca 2b-38), Insolubles (AOCS Ca 3a-46), Unsaponifiables
(AOCS Ca 6A-40) and Free Fatty Acids (AOCS Ca 5a-40).

The results of these analyses are summarized in Table 1. Triglyceride
profiles are shown in Table 2.

Table 1. Fuel Properties of Tested Fuels (percent composition unless
otherwise noted)

Fuel	Stored Fat	80%Fat:20%D2	Saddle Fat	Leaf Fat	D2	UMO

Ash	0.1	0.193	0.12	0.076	0.01	0.807

Carbon	79.2	77.5	76.77	70.7	86.2	81.1

Hydrogen	12.5	12.4	12.40	12.2	12.8	13.5

Nitrogen	0.14	0.02	0.20	0.024	0.04	0.26

Oxygen	7.96	9.83	10.43	10.7	0.841	0.95

Sulfur	0.1	0.01	0.01	0.01	0.0488	0.372

MIU	1.22	14.83	1.38	1.19	n/a	n/a

Moisture	0.1	1.27	1.09	0.36	n/a	n/a

Insoluble	0.44	0.05	0.16	0.1	n/a	n/a

Unsaponifiable	0.68	13.51	0.13	0.73	n/a	n/a

FFA	9.4	10.5	0.45	0.4	n/a	n/a

Viscosity (cP)	15.60	9.24	15.90	14.40	1.42	18.00

Specific Gravity (g/mL)	0.887	0.88	0.88	0.885	0.852	0.87

Energy Content (BTU/lb)	17047	17547	16488	17062	19144	19155



Table 2. Triglyceride Profiles of Tested Fats (percent composition)

Fuel	Stored Fat	Saddle 1	Saddle 2	Saddle Top	Leaf

C14:0	0.57	0.5	0.51	0.5	0.64

C14:1	0.22	0.22	0.24	0.25	0.25

C16:0	23.5	23.63	24.31	23.9	25.48

C16:1	8.33	8.91	9.28	9.52	8.78

C18:0	5.37	5.14	4.97	4.61	5.52

C18:1	41.71	44.68	44	45.12	42.5

C18:2	16.86	14.37	14.22	14.17	13.8

C18:3	1.01	0.81	0.82	0.79	0.77

C18:4	0.18	0.14	0.15	0.15	0.14

C20:1	0.54	0.56	0.54	0.57	0.65

C20:2	0.18	0.15	0.14	0.14	0.13

C20:3	0.18	0.12	0.13	0.12	0.1

C20:4	0.41	0.15	0.15	0.14	0.12

Unknown	0.74	0.62	0.54	0.63	0.61



Interestingly, fat that was stored for over one year had similar energy
content to the leaf fat provided by and freshly extracted in UGA
laboratories. Saddle fat obtained from Pilgrim’s pride had slightly
less energy than the other two examined fat sources; it also contained
significantly more water as delivered, which may have had a diluting
effect on the energy content of this material. Stored fat clearly had
much higher free fatty acid content which is attributable to oxidative
effects associated with long term storage. However, this increase in FFA
did not have a large impact on energy values as the products of
oxidation have similar energy content to the native triglycerides found
in these oils. The 80% poultry fat/20% diesel mixture also had high free
fatty acids and slightly higher energy content than pure fats as diesel
fuel contains more energy than fat. Carbon and hydrogen levels were
consistent throughout all fuels, but fats had much more oxygen which
generally enhances combustion and reduces emissions. Petroleum based
fuels (UMO, D2) had much more sulfur which is the source of sulfur oxide
emissions.

The consistency seen in fatty acid profiles among the three different
fats studied here is significant. Poultry from different parts of the
bird, differently handled birds and variable storage conditions all
generally had similar fatty acid composition. This is important to note
as it suggests variability of fuels will not be dependent on the source
of the fat, but on extraction methods and handling procedures.
Additionally, saddle fat samples from two different extraction batches
(1 and 2) had consistent fatty acid profiles. Saddle fat separated into
two distinct layers after about 24 hours. A liquid top layer and a
semi-solid bottom layer. The top layer was sampled separately and
analyzed using the same procedures as the entire fat sample. It was
found that there was no significant difference in the composition and
fuel properties of this liquid top layer and the rest of the fat sample
suggesting similar performance characteristics can be expected from
separate layers of stored fats.

American Proteins rendered fat was stored 1.5 years at room temperature
for a completed storage study before the start of testing. Conveniently,
this provided a high free fatty acid product which represented stored
fuels for the purpose of this study. 

Objective 4 – Extracted poultry fat samples were combusted in the
Clean Burn CB 350 CTB boiler. The fueling system, hydronic setup and
combustion setup are illustrated in Figure 1.

 

Figure 1. Bloc diagram of CB 350CTB Coil Tube Boiler, fueling system and
hydronic system. 

Stack emissions were measured using an ENERAC 3000E. The team recorded
both average and instantaneous measurements of flue gas concentrations
for oxygen, carbon monoxide, carbon dioxide, combustible gases, excess
air, nitric oxide, nitrogen dioxide, NOx (NO + NO2), and sulfur dioxide.
The analyzer software program enabled the recording of emissions data
directly to a spreadsheet file on the hard drive of a laptop computer.
Data was recorded during steady state operations for each fuel tested,
at both maximum and part loads and at each FGR damper setting. The
ENERAC 3000 portable emissions analyzer is a self-contained, extractive
flue gas monitoring system utilizing electrochemical sensors with an
internal sample pump designed for 600-900 cc/minute. A separate vacuum
pump extracted flue gas from a breaching port and discharged it to the
ENERAC. Teflon tubing interconnected a filter probe in the breaching
through two moisture condensers to the vacuum pump and then to the
analyzer. The ENERAC sensors use an electronically controlled circuit to
minimize zero drift and reject cross interference from other compounds,
in compliance with EPA Conditional Test Methods (CTM) –022, -030 and
–034. Performance specifications of the CTM-022 method are equivalent
to US EPA Method 7E requirements. Accuracy of the sensors is +/-2%, and
they are capable of operating at 1.5 orders of magnitude of gas
concentrations. Airflow was low in the system so precise measurements
were made using a Dwyer micrometer and a pitot tube at locations shown
in Figure 1.

Boiler efficiency, temperature differential, and boiler parameters were
also recorded every 10 minutes during testing. Oil flow volumes and air
intake were kept consistent throughout the study to assure similar
boiler conditions, and therefore performance measurements, were
comparable between fuels. Water flow rates were measured using an
in-line flow meter produced by Universal Flow Meters specifically for
this application. 100% diesel fuel runs were executed for 1 hour before
each test fuel was examined to provide baseline measurements. Test
results including emissions and performance data are summarized in Table
3.

Table 3. Emissions and performance of various fuels in CB350 CTB
boiler. *Average refers to the average of the three poultry fat samples.
Numbers in parentheses indicate number of runs per fuel type.

Fuel	100 PRF(4)	Leaf Fat(1)	Saddle Fat(4)	Average*	80/20(2)	UMO(3)
D2(11)

Stack Temp (oF)	352	362	364	360	356	367	360

Oxygen (%)	4.00	4.03	5.17	4.67	3.98	3.35	4.30

Carbon Monoxide (lb/min)	0.0059	0.0062	0.0065	0.0063	0.0078	8.5029
0.0049

Carbon Dioxide (%)	14.98	16.56	13.94	0.00	15.00	15.56	13.83

Combustibles (lb/min)	2.47E-07	1.03E-06	1.41E-06	1.02E-06	1.31E-06
2.42E-06	1.02E-06

Excess Air (%)	22.12	22.78	31.16	27.38	21.81	17.89	24.13

Nitric Oxide (lb/min)	0.00073	0.00037	0.00043	0.00051	0.00036	0.00099
0.00030

Nitrogen Dioxide (lb/min)	0.00022	0.00019	0.00022	0.00021	0.00023
0.00006	0.00017

Oxides of Nitrogen (lb/min)	0.00095	0.00056	0.00065	0.00072	0.00059
0.00105	0.00047

Sulfur Dioxide (lb/min)	0.00007	0.00000	0.00006	0.00005	0.00008	0.00053
0.00006

Consumption (gal/hr)	2.43	2.38	2.51	2.46	2.42	2.37	2.22

Efficiency (%)	78.19	72.67	79.87	78.32	76.79	81.77	79.18

∆T (oF)	24.69	24.17	25.77	25.11	24.92	28.52	24.81



In all cases, poultry fat had reduced emissions as compared to used
crankcase oil (UMO), the fuel the CB 350CTB was designed to burn.
However, when poultry fat emissions were compared to diesel fuel
emissions, diesel had slightly lower emissions although the difference
was not significant. The 80% poultry fat/20% diesel fuel mixture tended
to have emissions between that of the two component fuels suggesting an
averaging effect of properties associated with these two fuels. These
results were in contrast to those observed in a previous University of
Georgia study on larger, 100,000 lb/hr boiler. In this study poultry fat
had reduced emissions as compared to petroleum diesel fuel resulting in
a cleaner burn.

∆T, which is the difference in temperature between incoming water and
outgoing water, was comparable between poultry fat (25.11oF) and diesel
fuel (24.81oF). The higher BTU value of used crankcase oil resulted in a
significantly higher ∆T of 28.52oF. Efficiency was also observed to be
the highest when the boiler was fueled on UMO, its intended fuel. Diesel
fuel and poultry fat had almost identical efficiencies of 79% and 78%
respectively.

A brief study on the effects of regulation of oil and air flow on
emissions and efficiency was conducted with little significant impact on
boiler performance. Changes to aspiration pressure, oil pressure and air
intake led to unpredictable changes in emissions but resulted in no net
change in performance or efficiency. As the focus of this study was on
performance and it was determined that emissions on this scale would not
be regulated, this part of the study was terminated before adequate data
was collected to predict the effects of such regulation on emissions.

Objective 5 - Legal limits of in-plant use of poultry fat as a boiler
fuel were explored by Bob Synk an engineering consultant contracted by
the University of Georgia to aid in this study. Research revealed no
legal conflicts in using this fuel in on-site boilers and heaters in the
State of Georgia. The scale of such applications in poultry processing
environments is small enough such that regulations over emissions and
other environmental impacts would not extend to these systems.

Objective 6 - A plan for onsite utilization was developed and is being
written for use by both Cagle’s Poultry and Pilgrim’s Pride Poultry.
Implementation of this technology is being carefully considered by
Pilgrim’s Pride’s Continuous Improvement Division. Cagle’s is also
ready to implement this system when the plan is delivered. If fuel costs
continue to be elevated, it will be economically advantageous for other
poultry processors throughout the state to implement similar systems.

Objective 7 - The above plan along with this report will be published on
the University of Georgia Faculty of Engineering, Engineering Outreach
website (  HYPERLINK
"http://www.engr.uga.edu/service/outreach/index.html" 
http://www.engr.uga.edu/service/outreach/index.html ). Findings from
this study have already been shared directly by Engineering Outreach
employees with interested parties in the state of Georgia. Engineering
Outreach will continue to share the findings of this study and transfer
the technology studied here within the state of Georgia via publications
and presentations.

CONCLUSIONS AND EXPECTED IMPACT ON THE FOOD INDUSTRY

This project was successful in that it appears the collection and
extraction of concentrated-fat containing materials from poultry
processing lines is a feasible method of providing an alternative to
petroleum based #2 fuel oil for industrial boilers. The methods
established here require the collection of these materials inline which
can be accomplished manually or with mechanical defatters. In other
situations, this material can simply be redirected from waste streams.
Regardless of the collection method this material must then be thermally
processed and filtered to separate fat from water and solids. This is a
straightforward process requiring the introduction of steam heated
retort or kettle systems. As poultry processing generally requires steam
heat, this should be readily available for introduction into the new
processing environment. The only capital investment needed in these
applications is the purchase of heating vessels used to dewater and melt
the poultry fat and a filter system to remove solids after water is
removed.

On-site fat extraction from offal and other waste streams proved to be
resource intensive as compared to simple extraction from higher-grade
fat-containing products. The low yield of fat from offal (approximately
10%) necessitates high residence volumes and high energy inputs for
extraction. Dewatering of offal accounts for most of the energy needed
as the product examined was about 48% water. Ultimately, the processing
of offal was beyond the scope of this study. Processing fat after it has
entered the wastewater stream would require the implementation of large
volume, large footprint rendering equipment more suited to an actual
rendering facility than on-site in a poultry processing facility.
Initial capital investment and required real estate would make it a
long-term return project, whereas simple on-site extraction of fat-rich
materials is quite feasible.

Even with the limitation of using high-fat containing material captured
before waste streams in poultry fat a significant financial impact can
be realized. If only 20% of the 44.6 million gallons of poultry fat
created in the state is recovered this could displace 8.9 million
gallons of diesel fuel used in poultry processing. At a possible income
sacrifice of $1.33/gal, its use will displace $2.60/gal in fuel costs
for a net savings of $1.27/gal or $11.3 million for the state poultry
industry. The capital investment for such a recovery system is
relatively small as only steam jacketed kettles or retorts are needed
with a simple filtration system. The required energy input of the system
is only about 20% of that recovered. Taking this energy expense into
account, the energy return is still near $9.0 million a year after
initial capital investment is recovered. This is a conservative estimate
as the price for petroleum has reached $70+ per barrel driving the price
of diesel over $3.00/gal in some parts of the state. Additionally, the
$.18/lb ($1.33/gal) price for poultry fat assumes the fat itself would
have been kept in the box and sold for human consumption or other
high-value use. Likely, much of the fat (up to 60%) would end up in
offal for a price near $.03/lb ($.22/gal). If this system were to be
implemented state-wide, the actual economic benefit to the state poultry
industry could ultimately exceed the $11.3 million predicted above.

The economic benefit of this type of system will vary between facilities
and will be dependent upon the processing taking place at any given
time. Optimum efficiency of this operation will be realized when
concentrated fat sources can be easily diverted from waste streams and
collected for processing to fuel. As fuel prices continue to increase
and the threat of possible shortages emerges the use of any and all
domestic sources of fuel becomes increasingly important. Along with
providing reduced fueling costs, this project increases domestic energy
security by displacing foreign petroleum with domestic poultry fat.
Additionally, since this material is actually produced in the state of
GA, it takes advantage of Georgia’s own energy resources. Finally,
since this material is used in-house it eliminates transportation costs
and conserves the fuel necessary to transport petroleum products to
poultry processing facilities where the fuels are needed.

COST SUMMARY

Industry Capital In-Kind Contributions

	Source/Description			Amount		

Cagle's Inc. - Poultry Fat			$     500		

Cagle's Inc. - Labor				$  1,000

Pilgrim’s Pride – Labor			$  1,000

Pilgrim’s Pride Poultry Fat			$     500

Clean Burn –Accessories			$  1,000

Clean Burn – Laboratory Equipment	$  1,000		

Clean Burn - 350,000btu/hr boiler		$11,000

Free Heat Services, Inc. – Labor		$  1,000	

TOTAL					$17,000

FoodPac Funding Distribution

University of Georgia – Engineering Outreach   100%

Cagle’s, Inc. provided initial offal and poultry fat samples and
initial consultation and other personnel services. Pilgrim’s Pride
later assumed guidance and supply roles providing the rest of the
poultry fat samples and invaluable technical assistance. Clean Burn was
also essential to this study providing the major equipment needed for
this study a selection of necessary accessories and laboratory equipment
specialized for measuring exhaust emissions from their boiler. In
addition, Free Heat Services, Inc. of Henagar, AL, a partner of Clean
Burn, provided extensive on-site technical support in the operation and
assembly of the Clean Burn boiler.

As the University of Georgia’s Engineering Outreach was the only
participating project organization, 100% of the funding went to this
unit to pay personnel costs, supply costs, overhead and outsourced
chemical analyses fees.

TECHNOLOGY TRANSFER ACTIVITIES

Results and technology developed in this study were presented at the
Georgia Environmental Partnership Spring (GEP) 2005 meetings. These take
place twice annually in seven locations across the state of Georgia and
allow University of Georgia Engineering Outreach to share their progress
and developing technologies with interested parties across the state.
The Georgia Environmental Partnership includes P2AD, The University of
Georgia, and Georgia Tech. The locations and dates of their
presentations were as follows:

Adams, T.T., 2005. RENEWABLE ENERGY IN GEORGIA. Georgia Environmental
Partnership. Regional Environmental Network Meeting. Macon, GA, May 3.

Adams, T.T., 2005. RENEWABLE ENERGY IN GEORGIA. Georgia Environmental
Partnership. Regional Environmental Network Meeting. Tifton, GA May 4.

Adams, T.T., 2005. RENEWABLE ENERGY IN GEORGIA. Georgia Environmental
Partnership. Regional Environmental Network Meeting. Savannah, GA, May
10.

Adams, T.T., 2005. RENEWABLE ENERGY IN GEORGIA. Georgia Environmental
Partnership. Regional Environmental Network Meeting. Augusta, GA, May
11.

Adams, T.T., 2005. RENEWABLE ENERGY IN GEORGIA. Georgia Environmental
Partnership. Regional Environmental Network Meeting. Gainesville, GA,
May 12.

Adams, T.T., 2005. RENEWABLE ENERGY IN GEORGIA. Georgia Environmental
Partnership. Regional Environmental Network Meeting. Newnan, GA, May 17.

Adams, T.T., 2005. RENEWABLE ENERGY IN GEORGIA. Georgia Environmental
Partnership. Regional Environmental Network Meeting. Dalton, GA, May 19.

Additional information on the project and the results and conclusions
will be disseminated in the fall GEP meetings beginning on November 2nd,
2005 by Daniel P. Geller. Additionally, scholarly journal articles are
being assembled to deliver the funded research to the scientific
community. Finally, this report and the final report to Pilgrim’s
Pride and Cagle’s will be posted on the University of Georgia’s
Engineering Outreach website:   HYPERLINK
"http://www.engr.uga.edu/service/outreach" 
http://www.engr.uga.edu/service/outreach .

REFERENCES

Adams, T., J. Walsh, M. Brown, J. Goodrum, J. Sellers, K. Das, 2002. A
Demonstration of Fat and Grease as an Industrial Boiler Fuel. University
of Georgia Outreach Report.
http://www.engr.uga.edu/service/outreach/Articles.htm#BioProducts.

University of Georgia Engineers work with Engineers from Clean Burn and
Underwriter’s Laboratories to optimize an instrumented Clean Burn used
oil burning boiler to efficiently burn freshly rendered poultry fat.

T1= Inlet H2O Temperature

„@

„Ð^„@

1

2

Y

Z

[

m

n

o

p

q

€

‡

ˆ

Š

‹

¿

À

Ó

ë

í

o

p

q

Š

‹

摧๚᠀í

î

ï

Ä

ñ

h

h

h¹

Ö欀핤

Ö欀㉤

hG

r	ð

r	ð

r	ð

r	ð

r	ð

r	ð

h…

I

:O Temperature

Temp Port

Enerac Port

Flow Meter

T2

T1

UO

CF

D2

Fuel

Air Line

Air Pump

Fuel Line

Fuel Pump

H2O Inlet

H2O Exhaust

Circ pump

Gas Exhaust

CTB350

