Anaerobic Respirometry Studies of Fats, Oils, and Grease

Bryan Lipp and Christopher Schmit, P.E., Ph.D

Executive Summary

Fats, oils, and grease (FOG) from grease traps and interceptors is an abundant waste and

is increasingly considered as a candidate for co-digestion to increase biogas production. The City

of Sioux Falls, SD Water Reclamation Facility (SFWRF) initiated this project to investigate the

addition of FOG to their anaerobic digesters. Specifically, this study aimed to perform a

literature review to determine best practices, characterize FOG, and determine three different

parameters of the FOG and test digester sludge. A recent literature review on the co-digestion of

FOG with municipal primary and secondary sludges determined that the addition of FOGs to

municipal anaerobic digesters can result in a substantial increase in methane production (Long et

al., 2012).

This study utilized a respirometer operated in anaerobic mode to determine the three

FOG and digester sludge parameters. The sludge culture used was collected from the anaerobic

primary digesters at the SFWRF. The three tests performed were: biochemical methane potential

(BMP), anaerobic toxicity assay (ATA), and specific methane production (SMP). The tests were

performed at mesophilic temperatures (35 °C) to match the conditions at the SFWRF.

During this research project, the volume of methane that could be produced from adding

FOGs to the SFWRF anaerobic digesters was found, along with the inhibitory concentration of

FOG and the SMP of the test sludge before and after a 30 day daily addition of FOG to the test

anaerobic digester sludge. A FOG sample was also collected from a grease interceptor in Sioux

Falls and characterized in terms of pH, total solids, volatile solids, and COD.

2

The FOG portion of the grease interceptor waste sample collected is a very concentrated

waste, with the potential to create large amounts of methane in an anaerobic digester. The waste

had a low pH, high percentage of volatile solids, and an extremely high concentration of COD.

When FOG was added to the anaerobic digester sludge in a batch test, the reactor produced

approximately 760 mL methane per 1.0 mL FOG added, 860 mL methane per gram of VS added,

or 335 mL methane per gram of COD added. The activity of the biomass did not significantly

change over a period of 30 days in which FOG was added at a rate of 0.35 kg VS/m3·d. Finally,

an inhibitory concentration of FOG was reached at a concentration of 4.52 g COD/L or 1.77 kg

VS/m3.

The results of this research indicate that adding FOGs to the SFWRF anaerobic digesters

will most likely increase methane production significantly. While it is possible to add too much

and inhibit the anaerobic bacteria, this should not be a problem if the digester and digester feed

are closely monitored.

Introduction

FOG waste is a byproduct from the preparation of certain foods by restaurants, schools,

hospitals, and other food service facilities. FOG can be defined as the top floatable layer of

wastewater rich in lipids that is generated through food processing and cooking (Long et al.,

2012). FOG includes material that is composed of animal, vegetable, and mineral origin (US

EPA, 2004a).

In most municipalities, discharging FOG directly into the sanitary sewer system is

considered illegal because of problems associated with pipe blockages (Long et al., 2012). If

FOG enters the collection system, solid deposits can form on pipe walls, in manholes, and on

3

other sewer components through a physical or chemical reaction with calcium (He et al., 2011;

U.S. EPA, 2004a). These deposits reduce the cross sectional area of the pipes, therefore reducing

the flow capacity. This can ultimately lead to pipe blockages and sanitary sewer overflows

(SSOs), incidents that occur tens of thousands of times per year in the United States, exposing

untreated wastewater to the public (US EPA, 2004b). The U.S. EPA has attributed 47% of

pipeline blockages and 23% of SSOs to FOG accumulation (U.S. EPA, 2004b). Not only are

FOG deposits in sewers a public health risk because of potential SSOs, they are a financial

burden to municipalities. Future costs associated with clearing sewer blockages in the United

States are expected to exceed $25 billion (Ducoste, 2013).

To remove FOG before entering the collection system, grease abatement devices (GADs)

such as grease traps and grease interceptors are installed in the sewer lines and at food

preparation facilities. These grease abatement devices use gravity to separate FOG from the rest

of the wastewater. Grease traps are typically installed directly below the sink inside of a facility,

and average 50 gallons in volume (Long et al., 2012). Grease interceptors are much larger than

grease traps, averaging between 1000 – 2000 gallons in volume. Grease interceptors are typically

installed outside of the facility below ground (Long et al., 2012).

GADs are placed in a manner that allows wastewater to flow through from one side to the

other, with enough detention time to allow solid particles to settle to the bottom and FOG to float

to the top. The three layers that form inside a GAD are food particles and other solids on the

bottom, wastewater in the middle, and FOG at the top (Wang et al., 2013). Collectively, the

contents of GADs are referred to as grease interceptor waste (GIW) or grease trap waste (GTW).

While the terms FOG and GTW have been used interchangeably, for the purposes of this

4

research GTW will refer to the entire contents of the GAD while FOG will refer to only the top

floatable layer inside of a GAD.

In a study completed by the National Renewable Energy Lab, the amount of FOG

generated per capita was determined by examining the FOG recovered from grease traps and

interceptors, as well as at local WWTPs in 30 United States metropolitan areas. The study

concluded that the average person produces 16 pounds of FOG per year, and this estimate may

be applied to the entire United States (Wiltsee, 1998). Assuming the specific gravity of FOG is

0.92, this equates to approximately 1.8 gallons per person per year. Because this estimate

includes FOG from GTW and FOG that has been mixed with wastewater that arrives at local

WWTPs, this is likely an overestimation of FOG that can be recovered. Once FOG is allowed to

mix with wastewater, it can be difficult to separate (Long et al., 2012). Another study of the

Raleigh, NC metropolitan area concluded that the average person in that area produces 18.7

gallons of GTW per year (Austic, 2010). This estimate includes the entire contents of the GAD,

of which an estimated 5% is FOG, equating to 0.94 gallons of FOG per person per year (Austic

2010). Averaging the results of these two studies yields 1.37 gallons per person per year.

Studies have shown that co-digestion of FOG in municipal anaerobic digesters can

increase biogas production from 32-82% (Bailey, 2007; Cockrell, 2008; Muller et al., 2010),

with smaller laboratory studies showing nearly 200% increase in digester gas production

(Kabouris et al., 2009a,b). The excess methane produced from the co-digestion of FOGs can help

offset the external energy use required by wastewater treatment plants. Along with the benefit of

increased biogas production, however, some detrimental effects have been observed during

studies such as inhibition of methanogenic bacteria, digester foaming, and pipe and pump

blockages, among others (Long et al., 2012).

5

Martin-Gonzalez et al. determined that FOG waste from WWTF increased the methane

formation potential when co-digested with the organic fraction of municipal solid wastes (2010).

Under mesophilic conditions and a feed ratio of 1:7 (FOG to organic fraction of municipal solid

waste), methane production increased from 0.38 L/g VSfeed to 0.55 L/g VSfeed in a 5-liter lab-

scale reactor (Martin-Gonzalez et al., 2010). No inhibition due to long chain fatty acid (LCFA)

accumulation was observed (Martin-Gonzalez et al., 2010).

With the abundance of FOG presumed to be present in the City of Sioux Falls for use at

the SFWRF, the first goal of this study was to obtain a typical sample and characterize it in terms

of total and volatile solids, pH, and chemical oxygen demand (COD). Characterizing the FOG in

this manner would help the City determine how much FOG to add and how much methane could

be expected. The FOG sample collected for this task would also be used for testing in the

methods listed subsequently.

The first test performed that would utilize the anaerobic respirometer was be the BMP

test. This test yields the volume of methane produced from a certain volume of FOG. Next, the

SMP test was run to determine the activity of the anaerobic biomass (anaerobic digester) sample

being used in the study. The results are given in terms of COD equivalent of methane produced

per gram of biomass in the reactor per day. The next test run was the ATA. This test determines

the concentration of FOG which becomes inhibitory to the methanogenic bacteria. These results

will be helpful to the personal at the SFWRF because it will give a maximum concentration of

FOG that can be fed to the anaerobic digesters. Finally, the last test run involved the daily

addition of FOG to the anaerobic digester samples for 30 days. The SMP and rate of methane

produced from FOG can then be compared before and after the 30 day test to determine if the

methanogenic bacteria were able to acclimate to the FOG waste.

6

Materials and Methods

This study was conducted at the Water and Environmental Engineering Research Center

(WEERC) on the campus of South Dakota State University (SDSU) in Brookings, SD, beginning

during the summer of 2013. The bench-scale test utilized an RSA PF-8000 respirometer, shown

in Figure 1, operated in anaerobic mode with a heated water bath and reactor vessels filled with

SFWRF primary digester sludge as the test culture. Each reactor vessel also contained a 5 cm

magnetic stir bar, both nutrient and mineral bases, and sodium bicarbonate pH buffer solution.

The respirometer control module kept a record of gas production in each vessel via tubing from

each vessel to the control module. Since methane was the product of interest in this study, carbon

dioxide and moisture scrubbers were placed in the tubing line between each reactor vessel and

the control module. Moisture scrubbers were necessary to protect the control module from

corrosion. Test substrates were added to bottles through rubber septum caps by syringes

equipped with 20 gage needles.

Figure 1. Respirometer Setup

7

A grease interceptor waste (GIW) sample was collected from the grease interceptor of a

restaurant in the city of Sioux Falls, SD. The GIW sample, shown in Figure 2, formed into three

layers after collection: a top layer of FOG, middle layer of wastewater, and bottom layer of food

particles and other settleable material. In order to ensure the material was able to be moved

through the 20 gage needles and maintain consistent injections into the test vessels, only the top

layer of GIW (the FOG portion) was tested in this research. The FOG sample was characterized

in terms of total solids (TS), volatile solids (VS), pH, and chemical oxygen demand (COD).

Three tests were performed using the respirometer: biochemical methane potential

(BMP), anaerobic toxicity assay (ATA), and specific methane production (SMP). To perform the

BMP test, a precisely measured volume of FOG was injected into stable reactor vessels and

allowed to be utilized by the anaerobic culture. The methane produced from the FOG was then

Figure 2. GIW sample taken from a restaurant grease interceptor in the city of Sioux Falls, SD.

Bottom layer: settleable

particles

Middle layer: wastewater

Top layer: FOG

8

compared with the methane produced from a like amount of COD in the form of acetic acid,

which acted as a control. This test not only yielded the volume of methane that could be

produced, but also gave an estimate of the percent removal of COD that can be expected.

The SMP test was used to compare the activity of the existing biomass with the activity

of biomass that had been fed FOG daily for a period of 30 days. The test was completed by

adding a high concentration of readily biodegradable substrate (acetic acid) to stable reactor

vessels and measuring the methane production rate. The rate of methane production is the SMP,

and is expressed in terms of g CODCH4 /g VSS·d. The SMP results can also be used to determine

the percent of acetoclastic methanogens by weight in the biomass sample tested. The last test

performed in this study was the ATA. This test was completed by adding increasing

concentrations of FOG to the reactor vessels until a noticeable inhibition of methane production

was observed. This concentration is then the level at which the FOG becomes inhibitory to the

test biomass.

To ensure each sample reactor was stable, a measured amount of a readily biodegradable

substrate was added to each vessel, which was acetic acid in this case. Stable reactors should

produce between 350 – 400 mL methane per gram of COD added. Any reactor that yielded a

methane volume not within this range or not within 5% of this range was discarded. The reactors

within the acceptable range could then move on to the specific respirometer tests.

Results and Discussion

FOG Characterization

The FOG characterization results are shown below in Table 1. The FOG TS and VS were

determined to be 97.1% and 97.0% of wet weight, respectively, yielding a VS/TS ratio of 99.9%.

9

Table 1. FOG Characterization Results

Parameter Result

pH (entire GIW sample, pH units) 4.32

Total Solids (% of wet weight) 97.1

Volatile Solids (% of wet weight) 97.0

VS/TS (%) 99.9

COD (mg/L) 2,260,000

The COD of the FOG was 2,260,000 mg/l and the pH of the GIW sample as a whole was 4.32.

This is a very strong waste in terms of volatile solids and COD, has the potential to create a large

amount of methane when anaerobically digested. Based on methanogenic bacteria yielding 350 –

400 mL methane per gram of COD destroyed, this waste has the potential to create up to 900

liters of methane during anaerobic digestion per liter of FOG.

These results compare well to previous studies. Long et al. summarized different FOG

characterization studies and found that all of the GTW samples studied had a pH range of 3.90 –

6.20 (2012). The total solids, volatile solids, and COD results were very different, but this may

be explained by them testing the entire contents of the grease trap or interceptor, while this study

only tested the top layer of FOG. The percentage of solids that were volatile (VS/TS) ranged

from 90% - 100% (Long et al., 2012).

Biochemical Methane Potential

A typical BMP test of the FOG sample is shown in Figure 3. For this test, 0.25 mL FOG

was injected. This test resulted in approximately 760 mL methane per 1.0 mL FOG added, 860

mL methane per gram of VS added, or 335 mL methane per gram of COD added. During this

10

Figure 3. BMP of FOG sample. These curves are the average of the methane

production curves of all six vessels, and are plotted on top of each other to

show the average difference between the methane production of the control

substrate (acetic acid) and the test substrate (FOG).

test, six reactor vessels were used for test substrates and two were used as blanks. The six test

vessels were first injected with acetic acid to ensure stability, followed by FOG. All six test

vessels were found to be stable, and all six were then injected with the FOG sample. Figure 3

shows the results of this test in the form of methane production curves for the acetic acid control

and FOG sample. The two curves are an average of all six vessels and are plotted on top of each

other to show the difference in methane production between the control substrate (acetic acid)

and the test substrate (FOG). The FOG BMP curve in Figure 3 shows that not all of the FOG in

the form of COD was converted to methane. This is illustrated by the FOG BMP curve not

reaching the level of the acetic acid control curve. Figure 3 also shows that the FOG BMP

methane production curve is not as steep initially as the acetic acid control methane production

curve, meaning that methane was produced at a lower rate for FOG than it was for acetic acid. It

0

50

100

150

200

250

300

350

400

0 10 20 30 40 50 60 70 80

Me

than

e P

rod

uce

d (

mL/

g C

OD

ad

de

d)

Time (Hours)

Acetic Acid Control

FOG BMP

11

is also observed that while initially the methane production rate is high during the FOG BMP,

the methane production rate begins to slow at a time of 30 hours after the injection.

Specific Methane Production

Figure 4 shows the initial and final SMP of the test sludge. The initial SMP was

determined from injecting six reactor vessels with acetic acid before the start of the 30 day daily

addition of FOG, averaging the results, and is represented by the blue data series. From the

beginning of the test, the SMP sharply increases to over 0.025 g CODCH4 / g VS·d before

decreasing to below 0.010 g CODCH4 / g VS·d. It is not believed that this initial spike or the

initial spike in any of the three data series is the actual maximum SMP because the spikes don’t

fit the trend of the rest of the data. The initial SMP then slowly increases and reaches the

maximum of approximately 0.018 g CODCH4 / g VS·d at a time of 21 hours. From the maximum

SMP of approximately 0.018 g CODCH4 / g VS·d, the percentage of acetoclastic methanogens by

weight in the biomass is found by dividing by the maximum specific growth rate for acetic acid

conversion to methane, which is 6 g COD/g VSS·d (Young and Cowan 2004). This calculation

results in 0.30% acetoclastic methanogens by weight in the biomass. Because the initial amount

of acetoclastic methanogens present in the biomass is known, it can be compared to the SMP at

the end of the 30 day daily FOG injection to determine any change in the population of

acetoclastic methanogens.

During this 30 day period, 0.20 mL FOG was injected every 24 hours for 30 days, which

represents 0.18 g VS and a loading rate of 0.35 kg VS/m3·d. In the 30 day period, the methane

production in three of the reactor vessels significantly dropped, which may have been caused by

an over-loading of FOG. The other three reactor vessels finished the 30 day period without

significant methane production drops, and were again injected with acetic acid for SMP

12

Figure 4. SMP of the sludge sample before and after the 30 day daily FOG addition.

determination. One of the reactors, reactor #1 did not produce an acceptable amount of methane

and was not included in the results. The other two reactors, #3 and #6, were within the acceptable

range and are therefore included in Figure 4. It can be observed that the results of the SMP in

reactors #3 and #6 were significantly different. Because of this, their data series were not

averaged as they were in the initial SMP, and are shown separately. The SMP of reactor #3

reaches its maximum point of approximately 0.012 g CODCH4 / g VS·d at a time of

approximately 22 hours. The SMP curve of this reactor is more drawn out and lower than the

other two SMP curves, meaning it took the methanogenic bacteria longer to consume all of the

acetic acid. The percentage of acetoclastic methanogens by weight in the biomass was calculated

to be 0.20%, lower than the initial 0.30% acetoclastic methanogens by weight.

0.000

0.005

0.010

0.015

0.020

0.025

0 10 20 30 40 50 60 70 80 90

SMP

(g

CO

DC

H4/g

VS/

d)

Time (Hours)

Initial, Average

Final, Bottle #3

Final, Bottle #6

13

The SMP of reactor #6 reaches its maximum point of approximately 0.023 g CODCH4 / g

VS·d at a time of approximately 13 hours. The SMP curve of this reactor is higher than the other

two SMP curves, meaning it took the methanogenic bacteria less time to consume all of the

acetic acid. The percentage of acetoclastic methanogens by weight in the biomass was calculated

to be 0.38%, which is higher than the initial 0.30% acetoclastic methanogens by weight.

The results of the specific methane production tests are somewhat inconclusive. Of the

six reactor vessels at the beginning of the 30 day daily FOG injection, only two could be used to

determine the final SMP. These two vessels had significantly different results, with the

maximum SMP of reactor #6 more than double than that of reactor #3, and the SMP of reactor

#6 was higher than the initial SMP while the SMP for reactor #3 was lower. This may indicate

that the FOG loading rate during the 30 day daily FOG addition was too high.

Anaerobic Toxicity Assay

The results of the ATA are shown in Figure 5. For this test, five increasing amounts of

FOG were injected into different reactor vessels. The amounts of FOG injected and their

corresponding equivalents and loading rates are listed in Table 2. It can be observed that the

smallest amount of FOG added resulted in the most methane produced per gram of COD added,

indicated by the blue data series. As the concentrations of FOG increased, the volume of

methane produced per gram of COD added continued to decrease. It can also be noted that the

final volume of methane produced per gram of COD added for the 1.13 g COD/L concentration

data series is nearly identical to the volume produced in the BMP test, as shown previously in

Figure 3. When the FOG concentration is increased to 4.52 g COD/L, the volume of methane

produced per gram of COD added decreases from 330 mL to 250 mL, a 24% reduction, which

14

Figure 5. ATA conducted with five different FOG concentrations.

Table 2. FOG Loading for ATA

mL FOG g COD/L g VS kg VS/m³

0.25 1.13 0.22 0.44

1.00 4.52 0.88 1.77

2.50 11.3 2.21 4.42

5.00 22.6 4.42 8.84

10.0 45.2 8.84 17.7

indicates inhibition of the methanogenic bacteria is occurring. When the concentration of

FOG is increased to 11.3 g COD/L, the volume of methane produced per gram of COD added

decreases from 330 mL to 30 mL, which is an 82% reduction.

The next reactor vessel, which had a FOG concentration of 22.6 g COD/L, produced little

methane due to the large amount of FOG added. At the end of the test, this reactor vessel had a

ball of foam floating at the top of the test sludge that was about the size of a quarter. The final

0

50

100

150

200

250

300

350

0 50 100 150 200 250 300

Me

than

e P

rod

uce

d (

mL/

g C

OD

ad

de

d)

Time (Hours)

1.13 g COD/L

4.52 g COD/L

11.3 g COD/L

22.6 g COD/L

45.2 g COD/L

15

reactor vessel, which had FOG concentration of 45.2 g COD/L, experienced significant foaming,

and nearly stopped producing methane after a thick layer of foam had collected on the surface.

The foaming was so significant that it filled the entire headspace of the reactor, and had the

consistency of a light paste.

The data clearly indicates that the lowest FOG concentration of 1.13 g COD/L (0.44 kg

VS/m3) produced the most amount of methane per gram of COD added. When the concentration

of FOG reached 4.52 g COD/L (1.77 kg VS/m3) , the methane produced per gram of COD added

decreased by 24%. This is important because it gives a concentration of FOG that can be used

effectively, and concentrations of FOG that should be avoided in the SFWRF anaerobic

digesters.

Wang et al. determined that the optimal feed rate for co-digestion of FOG and municipal

sludges to be 20% FOG, 80% thickened waste activated sludge (TWAS), which resulted in a VS

loading of 2.16 kg VS/ m³ (2013). They also found that inhibition occurred at a feed of 40%

FOG, 60% TWAS, or 3.54 kg VS/ m³ (Wang et al., 2013).

Conclusions

The goal of this research was to use anaerobic respirometry to determine the volume of

methane that could be produced from adding FOGs to the SFWRF anaerobic digesters. This was

found, along with the inhibitory concentration of FOG and the SMP of the test sludge before and

after a 30 day daily addition of FOG to the test anaerobic digester sludge. A FOG sample was

also collected from a grease interceptor in Sioux Falls and characterized in terms of pH, total

solids, volatile solids, and COD. Analysis of the data leads to the following conclusions:

16

1. The FOG portion of the grease interceptor waste sample collected is a very concentrated

waste, with the potential to create large amounts of methane in an anaerobic digester. The

waste had a low pH, high percentage of volatile solids, and an extremely high

concentration of COD.

2. When FOG was added to the anaerobic digester sludge in a batch test, the reactor

produced approximately 760 mL methane per 1.0 mL FOG added, 860 mL methane per

gram of VS added, or 335 mL methane per gram of COD added.

3. The activity of the biomass did not significantly change over a period of 30 days in which

FOG was added at a rate of 0.35 kg VS/m3·d.

4. An inhibitory concentration of FOG was reached at a concentration of 4.52 g COD/L or

1.77 kg VS/m3.

From the testing completed in this research project, I do believe that a full-scale FOG co-

digestion implementation could proceed. Foaming was only observed at very high concentrations

of FOG that could be easily avoided, and the potential for capturing more methane for heating or

electricity generation is very high. To ensure that inhibitory concentrations of FOG are not

reached, I would begin co-digesting at a FOG feed rate of 0.22 kg VS/m3·d. This rate was chosen

after close examination of Figures 3 and 5. As the figures show (red data series in Figure 3, blue

data series in Figure 5), the 0.25 mL of FOG that was injected in each case was nearly all utilized

by the sludge culture by approximately t = 50 hours. This can be seen by the curves leveling off

at these times. Therefore, an amount of FOG that would be utilized in a period of 24 hours would

be approximately half of the 0.25 mL of FOG that was used in these two instances, assuming that

17

the sludge culture would use the FOG waste at the same rate. Half of 0.25 mL FOG added to

each reactor vessel is equivalent to 0.22 kg VS/m3·d.

One of the challenges of implementing this plan would be monitoring the amount of FOG

entering the digesters. While it would be easy if it was only FOG being added, there would most

likely be large amounts of water from grease abatement devices as well. It may be difficult to

keep a homogenous mixture and know the concentration of volatile solids and/or COD because

these concentrations would change each time more waste is delivered to the holding tank.

Acknowledgements

I would like to thank the City of Sioux Falls for sponsoring this interesting research. It

was a good learning experience and yielded useful results. I would also like to thank Dr.

Christopher Schmit for his guidance during this project, Beverly Klein of the WEERC lab, and

Graduate Assistant Shahnaj Akter.

(Bailey, 2007; Cockrell, 2008; John C Kabouris et al., 2009a; John C. Kabouris et al.,

2009b; Martín-González, Colturato, Font, & Vicent, 2010; Muller et al., 2010; Young,

2004)

WORKS CITED

Austic, G. (2010). Feasibility study: Evaluating the profitability of a trap effluent dewatering

facility in the Raleigh area. For ECO Collections through the Biofuels Center of North

Carolina, 1-21.

Bailey, R. S. (2007). Anaerobic digestion of restaurant grease wastewater to improve methane

gas production and electrical power generation potential. Proceedings of the Water

Environment Federation, 2007(11), 6793-6805.

Cockrell, P. (2008). Greasing Digester-Gas Production. Water environment & technology, 20(1),

70-73.

Ducoste, J. (2013). Factors that influence the Formation of FOG Deposits in Sewer Collection

Systems. Paper presented at the EPA Research Forum, Arlington, VA, USA.

18

EPA, US. (2004a). Local Limits Development Guidance. (EPA 833-R-04-002A).

EPA, US. (2004b). Report to Congress: Impacts and Control of CSOs and SSOs. (EPA 833-R-

04-001).

He, X., Iasmin, M., Dean, L. O., Lappi, S. E., Ducoste, J. J., & de los Reyes III, F. L. (2011).

Evidence for fat, oil, and grease (FOG) deposit formation mechanisms in sewer lines.

Environmental science & technology, 45(10), 4385-4391.

Kabouris, J. C., Tezel, U., Pavlostathis, S. G., Engelmann, M., Dulaney, J., Gillette, R. A., &

Todd, A. C. (2009a). Methane recovery from the anaerobic codigestion of municipal

sludge and FOG. Bioresource technology, 100(15), 3701-3705.

Kabouris, J. C., Tezel, U., Pavlostathis, S. G., Engelmann, M., Dulaney, J. A., Todd, A. C., &

Gillette, R. A. (2009b). Mesophilic and Thermophilic Anaerobic Digestion of Municipal

Sludge and Fat, Oil, and Grease. Water Environment Research (10614303), 81(5), 476-

485. doi: 10.2175/106143008X357192

Long, J. H., Aziz, T. N., Reyes III, F. L., & Ducoste, J. J. (2012). Anaerobic co-digestion of fat,

oil, and grease (FOG): A review of gas production and process limitations. Process

Safety and Environmental Protection, 90(3), 231-245.

Martín-González, L., Colturato, L., Font, X., & Vicent, T. (2010). Anaerobic co-digestion of the

organic fraction of municipal solid waste with FOG waste from a sewage treatment plant:

Recovering a wasted methane potential and enhancing the biogas yield. Waste

Management, 30(10), 1854-1859.

Muller, C., Lam, P., Lin, E., Chapman, T., Devin-Clark, D., Belknap-Williamson, J., & Krugel,

S. (2010). Co-digestion at Annacis Island WWTP: Metro Vancouvers Path to Renewable

Energy and Greenhouse Gas Emissions Reductions. Proceedings of the Water

Environment Federation, 2010(14), 2706-2722.

Wang, L., Aziz, T. N., & de los Reyes III, F. L. (2013). Determining the Limits of Anaerobic

Co-digestion of Thickened Waste Activated Sludge with Grease Interceptor Waste.

Water Research.

Wiltsee, G. (1998). Waste grease resource in 30 US metropolitan areas. Paper presented at the

The Proceedings of Bioenergy 98 Conference, Wisconsin.

Young, J. C., and Cowan, R.M. (2004). Respirometry for Environmental Science and

Engineering. Springdale, AR: SJ Enterprises.