Accelerat ing t he world's research.

Biogas prediction and design of a

food waste to energy system for the
urban environment
Nathan Curry

Renewable Energy

Cite this paper Downloaded from Academia.edu 

Get the citation in MLA, APA, or Chicago styles

Related papers Download a PDF Pack of t he best relat ed papers 

Modeling and Design of a Food Wast e t o Energy Syst em for an Urban Building
Nat han Curry

A review of simple t o scient ific models for anaerobic digest ion

Georgios Florides

Biogas from Kit chen Food Wast e: Est imat ion, Applicat ions and Economic analysis
Ajmal Saeed, Neeladri Sekhar Kachari, Jit ender Maharshi, Sachin Kamboj
 Renewable Energy xxx (2011) 1e10

Contents lists available at SciVerse ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Biogas prediction and design of a food waste to energy system for the urban
environment
Nathan Curry*, Pragasen Pillay
P.D. Ziogas Power Electronics Laboratory, Department of Electrical and Computer Engineering, Concordia University, 1455 de Maisonneuve Blvd. W., Montreal,
Quebec H3G 1M8, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Anaerobic digestion applied to the organic waste produced in urban environments could provide
Received 18 July 2011 a critical solution to growing garbage problems while simultaneously reducing external energy
Accepted 23 October 2011 requirements. As landfills across Canada and the rest of the world are filled to their limits, a carbon-
Available online xxx
neutral process which can locally generate electricity and heat while providing up to 50% volatile
solid reduction is something to be seriously considered. This paper investigates the feasibility of urban
Keywords:
anaerobic digestion, presents four techniques for biogas estimation e ultimate analysis, yield from
Anaerobic digestion
molecular formula analysis, a novel computer simulation technique using Anaerobic Digestion Model #1
Urban environment
Waste to energy
(ADM1), and a literature review of experimentally determined biogas yields. In addition, a case study for
Food waste small-scale anaerobic digestion system design is presented for an urban building.
Renewable energy Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction
1.1. Population growth and increased urbanization
Urban waste generation and disposal will continue to be a major
global issue as the world’s population grows past the 7 billion mark
and more people move to urban areas. As of 2008, the number of
people living in cities surpassed those living in rural areas and it has
been estimated that by 2050, 6 billion people will be living in cities
compared with 3.5 billion now [1].
Global population has more than doubled since 1960 and by
2050 it will have more than tripled to almost 9 billion [2]. Ninety-
nine percent of projected population growth and 50% of the urban
growth is predicted to occur in the developing world e currently
considered to be Africa, Asia, Latin America, and the Caribbean - at
a pace of almost 50% per year while North America and European
population growth will only increase by 2% and total urbanization
will increase by 10% [2,3].
1.2. The problem with landfills
The waste generated by this increasing urbanization of humans
and industries will have to be sorted and processed. The most
common waste management solution is landfilling. The main
problem with this is that landfills around the world are running
out of space. The last landfill in the greater New York City area
closed in 2001 and now 12,500 tonnes of garbage are transported
daily outside the state by truck, barge, and train [4]. London
currently sends 20 million tonnes of annual waste to 18 different
landfills that are running out of space [5]. Montreal, Canada, has
a contract to send 1.3 million tonnes of waste each year to
a landfill located 40 km away with permits that extend only to
2012 [6]. In 2006, nearly a million tonnes of the waste generated
in Toronto, Canada, was trucked to landfills across the U.S. border
into Michigan. As a solution, the City of Toronto has purchased
a landfill site that is over 200 km away from the downtown area
that opened January, 2011 [7]. Mexico City produces 12,500 tonnes
of trash per day and sends it to a sprawling, polluted landfill that is
running out of space [8]. As of 2011, two thirds of China’s cities are
overrun with garbage and millions of tonnes of waste are sent to
non-sanitary landfills with one quarter of cities having no place to
dispose of trash [9].
There is no simple solution to this waste problem. This is
a pressing global issue that requires design and consumption
paradigm shifts as well as new waste management solutions.
1.3. Food waste available for anaerobic digestion

A study of global food waste published in 2011 by the Food and

* Corresponding author. Tel.: þ1 514 848 2424x7089, þ 514 802 9099 (Cell). Agriculture Organization of the UN found that roughly one third of
E-mail addresses: n_curry@encs.concordia.ca, nathanacurry@gmail.com (N. Curry). all food produced for human consumption each year goes to waste

0960-1481/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.renene.2011.10.019

Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
2 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10
totaling 1.3 billion tonnes [10]. This waste is distributed fairly
evenly between developing and industrialized nations with 40% of
the food waste in the developing nations occurring in the
production and processing phases of consumption while in the
industrialized nations, 40% occurs at the retail and consumer levels
of consumption [10].
Applying anaerobic digestion (AD) to this amount of food waste
has the potential to generate 367 m3 of biogas per dry tonne at
about 65% methane [11] with an energy content of 6.25 kWh/m3 of
biogas [12] yielding 894 TWh annually. This represents almost 5% of
the total global electrical energy utilization of 20,181 TWh in 2008
[13]. In addition, where anaerobic digestion technology is applied,
food waste would not be sent to landfills reducing transportation
costs and greenhouse gas emissions.
In 2009, the United States generated 243 million tonnes of
municipal solid waste (MSW) comprised mainly of food scraps,
yard waste, plastic packaging, furniture, tires, appliances, paper,
and cardboard. The discarded MSW came from two main sources:
Residential (55e65%) and Commercial/Institutional (35e45%) with
construction and hazardous wastes not considered in the grouping.
Nearly half of this waste was recycled or reclaimed with 132 million
tonnes (54%) going to landfill [14]. Organic materials represented
the largest component of MSW with 34.2 million tonnes of food
waste and yard waste disposed of each year. Waste breakdown is
shown in Fig. 1.
Fig. 1. MSW composition 2009 before recyclying, USA EPA.
In Canada, the total amount of waste sent to landfill in 2008 was
26 million tonnes [15]. It has been estimated that almost 50% of
food produced for human consumption in Canada is discarded
totaling 6 million tonnes annually and representing 23% of total
waste [16].
Using the figures from the US and Canada and the biogas and
energy content discussed at the beginning of this section (2.3 MWh
per dry tonne of food waste), the annual energy available from food
waste is in Table 1.
Table 1
Energy available from AD of OFMSW in US and Canada.
Food waste (t) Biogas (m3) Energy content (MWh)
US 34,200,000 3,765,420,000 23,533,875
Canada 6,000,000 660,600,000 4,128,750
This paper is organized as follows: Section 2 has an overview
of the anaerobic digestion process and introduces food waste as
Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
a viable substrate and presents different biogas estimation
techniques, Section 3 presents an overview of the design
process for a small-scale food waste to energy system, Section 4
is a case study of the design of a small-scale food waste to
energy system for an urban building, and Section 5 presents the
conclusions.
2. Anaerobic digestion in the urban environment
2.1. Anaerobic digestion overview
The first anaerobic digestion plant built to convert waste to
energy occurred in 1859 at a leper colony in Bombay, India -
although references to the process that produces a fuel gas from the
putrification of organic materials was observed and harnessed in
much earlier centuries. Since the turn of the 20th century, it has
become a widely used process and can be found everywhere from
small farms in Africa, villages in India, integrated networks of farms
in Germany and Denmark, and MW farm installations in the US.
In recent years, anaerobic digestion technology has seen rapid
growth. Biogas plants around the world have experienced a 20 to
30 percent increase each year with the most experienced and well-
developed markets being in Germany, Denmark, and Austria [17].
As of 2007, Germany has 3700 biogas plants in operation [17],
Denmark has 20 centralized plants and 35 farm scale plants in
operation [19], and Austria has 323 plants with an electrical
capacity of 81 MW [18].
In other parts of the world, AD technology also flourishes as
a waste-to-energy solution but on a smaller scale and in a decen-
tralized manner. In 2007, China had an estimated 18 million biogas
digesters and in India there are currently over 5 million small-scale
biogas plants in operation. The volumes of the digesters range from
2 m3 to 20 m3 and they are usually fed household and agricultural
wastes. The biogas is predominantly used for personal cooking and
lighting purposes [17].
Despite all of these advances, anaerobic digestion has yet to
make a large migration to the urban environment. Historically,
landfilling has been a cheaper option for urban development but as
landfills continue to run out of space and move farther from the
cities, the cost of landfilling will continue to rise, both monetarily
and environmentally. If the organic waste produced in the cities was
source-separated from the recyclable materials and digested on-site
in small-scale anaerobic reactors, it could provide a critical solution
to growing garbage problems while simultaneously reducing
external energy requirements and greenhouse gas emissions.
2.2. The anaerobic digestion process
Anaerobic digestion is a process in which microorganisms
breakdown biodegradable material in the absence of oxygen. The
process is widely used to treat wastewater sludges and industrial
and farm wastes because it provides volume and mass reduction of
the input material. Anaerobic digestion is considered a renewable
energy source because the methane-rich biogas produced is suit-
able for energy production and can replace fossil fuels. Additionally,
the nutrient-rich solids and liquids left after digestion can be used
as fertilizer.
The anaerobic digestion of food waste is a complicated
biochemical process. Based on temperature and input substrate,
different strains of bacteria digest complex chains of carbohydrates,
fats and proteins into their component parts. Anaerobic digestion
occurs in four separate phases: hydrolysis, acidogenesis, aceto-
genesis, and methanogenesis as shown in Fig. 2. The last stage of
the process, methanogenesis, is where the biogas is produced [31].
 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10 3

how all the different species of bacteria interact and further, how
each particular strain of each bacteria interact with each other. At
some points there can be 8 different methanogenic bacteria con-
verting acetic acid into methane. Any attempt at standardizing or
modeling an AD system will be based on a particular waste with
a particular experimental set up. For complicated wastes such as
food waste, the problems are multiplied as the biochemical path-
ways involved in the digestion are estimations and require
a case-by-case interaction to run efficiently. In addition, the
research motivation affects available data on digestion of food
waste. It’s rare to find experiments done from an energy or
economics perspective as the focus of the research is often on the
stability and completion of biochemical reactions and not on
optimum biogas production and high methane content.

2.3. Food waste as anaerobic digestion substrate

Waste analysis is one of the most important steps in the AD

process. Knowing the general composition of the input material to
the system is essential for calculating the amount and composition
Fig. 2. Anaerobic digestion process. of the biogas produced as well as the amount of energy contained in
the biogas.
Anaerobic degradability greatly depends on the composition of
The advantage of using anaerobic digestion in an urban envi- the input material considered. It’s very difficult to estimate or
ronment to treat organic waste as opposed to composting it is that measure the percentage of carbohydrates, lipids, and proteins in
anaerobic digestion produces biogas with a high percentage of a heterogeneous substrate such as food waste due to the dynamic
methane which can be used as fuel whereas composting produces and sensitive nature of the biological process to the input compo-
mostly carbon dioxide which has no energy value [20]. Importantly, sition. There are several techniques available to estimate the
AD also prefers cooked and oily food waste to be digested where amount of biogas contained in a complex substrate such as food
composting does not [31]. In fact, the AD process produces more waste including ultimate analysis, molecular formula (if known),
biogas when used cooking oil and cooked meats are added [20]. computer simulation, and a literature review of experimentally
There are several types of anaerobic digestion, but for food determined biogas yields.
waste, studies have shown that mesophilic (w35  C) is the most
stable [11,21,22] Thermophilic digestion (w55  C) allows for faster 2.4. Ultimate analysis of food waste for biogas prediction
methane extraction and therefore has a lower substrate retention
time and can be considered in situations where size is a restriction, This technique of biogas estimation involves breaking food
but has been shown not to be as stable and more sensitive to input waste down into its elemental composition of carbon, hydrogen,
composition when dealing with food waste [11,21,22]. oxygen, and nitrogen. If 150 tonnes of food waste (dry basis) is
The anaerobic digestion of food waste can be performed in assumed, then the waste breakdown in Table 2 can be assumed.
a single continuously stirred tank or in a two-phase system. The
process can be run “wet” (less than 15% TS) or “dry” (15e40% TS) in Table 2
both single and two-phase systems, each having specific advan- Ultimate analysis of food waste.
tages and disadvantages depending on substrate, system size and % kg/mol
configuration [22]. C 48 5.45
For the case of small-scale systems that operate at a small scale H 6.4 0.46
(1 tpd food waste load or less), dry digestion systems such as the O 37.6 7.26
Dranco system have not been demonstrated to be feasible as the N 2.6 6.35
S 0.4 14.55
equipment needed to pump a high-solids slurry are much more
expensive than readily available wet digestion system components.
The methane in the biogas produced from the AD process can
If 1 mol of N is assumed, then C22H38013N is the chemical
be combusted to produce both heat and electricity using internal
formula for 150 tonnes of food waste (VS).
combustion engines or microturbines and hot water heaters in
Using Buswell’s Equation, the amount of biogas produced can be
a cogeneration arrangement where the heat generated is used to
predicted [23]:
warm the digesters or to heat buildings. Any excess electricity not
needed on site can be sold back to the grid. In the case of excess 
4a b 2c þ 3d
 
4a þ b 2c 3d

gas production or a repair needed in a boiler or genset, the biogas Ca Hb Oc Nd þ H2 O/
4 8
can be stored in tanks, special balloons, or sent to an enclosed flare
4a b þ 2c þ 3d
 
to prevent the release of methane into the atmosphere. In order to  CH4 þ CO2 þ dNH3 ð1Þ
8
configure an AD system in the urban environment, special
consideration must be paid to standing gas, fire, and building
codes as well as health and safety regulations for handling food C22H38O13N þ 6.75H2O / 12.13CH4 þ 9.88CO2 þ 1NH3
waste.
The literature on the anaerobic digestion of solid wastes is The CH4 yield is 55.7 tonnes at 25  C (0.668 kg/m3) or 83,383 m3
difficult to summarize and is often very confusing due to the “black CH4. The CO2 yield is 124.2 tonnes at 25  C (1.842 kg/m3) or
box” aspect of the process. Not everything is fully understood about 67,318 m3 CO2.This provides a total biogas yield of 1005 m3 of

Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
4 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10
biogas/tonne VS. This value assumes that 100% of the food waste
substrate is broken down. Practically, about 40e65% of the organic
material is broken down, scaling the biogas yield down to
402 m3/tVS to 653 m3/tVS [22,24,25]. This range is close to a steady
state experimentally determined biogas yields for mesophilic
digestion of food waste of 367 m3/tonne VS [11].
2.5. Biogas prediction from molecular formula estimation
If estimated from the values presented in [38] for a “mixed
kitchen waste” the carbohydrate, protein, and lipid concentration of
the sampled food waste are 59, 33, and 8 percent respectively.
Using these values for the nutritional breakdown of food waste and
a biogas yield for each category based on theoretical molecular
formulas for carbohydrates, proteins, and lipids as given in [15],
a value for the maximum biogas potential (in Nm3 of CH4/tVS) of
the mixed food waste can be estimated.
Carbohydrates (based on C6H10O5) e 59% at 415 Nm3/tVS
Proteins (based on C5H7NO2) e 33% at 496 Nm3/tVS
Lipids (C57H104O6) e 8% at 1014 Nm3/tVS
This calculation yields a methane production value of
490 Nm3/tVS which is equivalent to 754 m3/tVS of biogas at 65%
methane. As in Section 2.4, this value is assuming that all of the
volatile solids are destroyed. Applying a realistic correction factor of
40e65% volatile solids destruction, the biogas yield using this
method is 302e490 m3 of biogas produced per tonne of volatile
solids. The experimentally determined reference value of
367 m3/tVS is squarely within this range of biogas yields.
The biggest problem with this method is that the percentages of
carbohydrates, proteins, and lipids are not readily available for
a mixed substrate and difficult to measure. In this case, the result
happens to be close to predicted yield but that might not hold true
for other mixed wastes in general. This method works better for
a homogeneous substrate where the carbohydrate, proteins, and
lipid values are obtainable from USDA or similar archives.
2.6. Computer simulation of biogas production
The most advanced model of the biogas production from
anaerobic digestion comes from the Anaerobic Digestion Model 1
(ADM1). This was developed by the International Water Association
in 2002 with the objective of building a mathematical model of
anaerobic digestion based on the interactive, dynamic chemistry of
anaerobic reactors. Currently, it is the research tool which most
completely models the process in which a complex substrate
breaks down in the absence of oxygen [29].
The first step of the ADM1 model converts solids into carbohy-
drates, lipids, proteins and inert material (soluble and particulate
inert). The second step is the hydrolysis process which disintegrates
the products of the first step into sugars, amino acids and long chain
fatty acids (LCFA). Next, the amino acids and sugars are fermented to
produce volatile fatty acids, hydrogen, and carbon dioxide (acido-
genesis). Then the organic acids - LCFA, proprionic acid, butyric acid
and valeric acid - are anaerobically oxidized into acetate, carbon
dioxide and hydrogen (acetogenesis). In the last step, carbon
dioxide and hydrogen and acetate are metabolized into methane by
methanogenic bacteria (methanogenesis) (see Fig. 2) [29].
For wastewater applications, the ADM1 input parameters are
estimated from experimental data as well as expert knowledge
about expected characteristics. For the Differential and Algebraic
Equation version of the ADM1 model (DAE) developed in 2006
there are 26 dynamic state concentration variables, 19 biochemical
kinetic processes, 3 gas/liquid transfer kinetic processes and 8
implicit algebraic variables represented with varying sensitivities
[29]. In order to use this model with complex substrates, the
Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
practical characteristics of solid wastes need to be interpolated into
their carbohydrate, protein and lipid constituents.
For those outside the wastewater engineering field, ADM1 is
extremely complicated to use and requires extensive knowledge of
the biochemistry involved in the anaerobic process as well as
a complete chemical breakdown of the input substrate. This can be
a hindrance for renewable energy researcher as the output of this
model is very useful for energy calculations.
In 2009, a transformer was developed which improves on the
Continuity Based Interfacing Model (CBIM) developed in 2005 [18].
The CBIM balances the macronutrient (CHNOP) elemental continuity
with respect to chemical oxygen demand (COD) balance and charge
balance. The CBIM assumes the input feedstock to have a constant
composition which does not allow for dynamic simulation [39]. The
transformer developed in [40] takes this process a step further by
attempting to maintain the mass balance, COD balance, and charge
balance according to a predefined, ordered, maximization procedure.
This transformer allows ADM1 to be more accurately employed as
a biogas estimation tool. The process is described in the next section.
2.7. Using the transformer as a tool for biogas prediction
Using the transformer developed in [40], the 32 required inputs
for ADM1 can be estimated for complex substrates through the
input of 11 parameters. This development helps to strengthen the
link between ADM1 and commonly measured characteristics of
solid wastes. The transformer was programmed in C and incorpo-
rated into a General Integrated Solid Waste Co-Digestion model
(GISCOD) that runs in Matlab-Simulink [38].
For this method of solid waste digestion, the following practical
characteristics are considered available: total COD (CODt), soluble
COD (CODs), VFAs, total carbon (TC), total inorganic carbon (TIC),
TKN, TAN, total phosphorous (TP), orthophosphate (orthoP), total
alkalinity (Scat), total solids (TS), and total volatile solids (TVS).
From this complete list of characteristics, the 11 inputs to the
transformer can be determined:
1. Particulate COD (CODp) in gCOD/m3(Calculated as (CODt e
CODs) where CODs is split into soluble substrate minus COD of
VFAs (CODs e COD of VFAs) and VFAs)
2. Soluble COD subtracting COD of VFAs (CODs e VFA) in gCOD/m3
3. Volatile Fatty Acids (VFA) in gCOD/m3
4. Total Organic Carbon (TOC) in gC/m3(Total carbon minus total
inorganic carbon)
5. Total Organic Nitrogen (Norg) in gN/m3 (Comes from estima-
tion of Proteins (C6H1203N2))
6. Total Ammonia Nitrogen (TAN) in gN/m3
7. Organic Phosphorous (TP-OrthoP) in gP/m3
8. Ortho-Phosphate (orthoP) in gP/m3
9. Total Inorganic Carbon (TIC) in mol HCO3 /m3
10. Total Alkalininty (Scat) equ/m3(Estimated from total alkalinity
measurements)
11. Fixed Solids (FS) in g/m3 (Total Solids e Total Volatile Solids)
In the case of not being able to measure all of these parameters
directly, literature surveys and specialized archives (USDA, etc) can
be used to determine the necessary input parameters.
Using ADM1 to predict biogas from food waste is simulated by
using transformer inputs for kitchen waste and experimental
hydrolysis rates (5.22, 1.86, and 1.24 d 1 for carbohydrates,
proteins, and lipids respectively) [38]. The disintegration constant
kdis was set at 0.5 for solid waste according to [22]. Using the
transformer and values from [38], the biogas yield was 316 m3/tVS,
which is only a 14% difference compared with experimental yield of
367 m3/tVS [11].
 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10
2.8. Literature review of biogas produced from food waste
For mixed food waste, Table 3 shows biogas yields that have
been reported for mesophilic digestion:
From Table 3, the average value biogas from digestion food
waste at mesophilic temperatures is approximately 370 m3/tVS
which is less than a 1% deviation from the full-scale 100 tpd study
done on the mesophilic digestion of food waste in [11].
Table 3
Experimental biogas yields for food waste.
Source Biogas yield m3/tVS
Discarded Food 355 [34]
Food waste 367 [11]
OFMSW 310e490 [35]
OFMSW 300e400 [26]
OFMSW 255e494 [36]
Food Waste 288 [32]
OFMSW 390 [33]
Food Waste 472 [27]
3. Design of small-scale anaerobic digestion system for food
waste
There are several important factors to consider when discussing
the feasibility of designing an anaerobic digestion system to oper-
ate with food waste as the predominant input substrate. Food waste
is considered a desirable input substrate but it is also prone to over-
acidification and lower pH levels due to the amount of fatty acids
produced. A review of important considerations for digesting food
waste follows.
3.1. Amount of waste available
An estimation of the total tonnage of input substrate is a very
important first step in the design process. Often this amount is not
directly known and needs to be estimated through a waste audit. A
total amount of waste can be estimated from a weekly, monthly, or
annual audit.
3.2. Dryness of input
One of the more important parameters for AD systems is the
dryness of the input material. Before an accurate size of the
digestion tank or a prediction of biogas content can be made,
the amount of dry solids present in the input substrate is
necessary. In agricultural manure waste, there is only a 2e12%
solids content meaning the input slurry is mostly water. The
potential biogas comes from the solids content, so accordingly,
manure has a very low biogas yield per ton. According to
a literature review of waste taken from cafeterias, restaurants,
and markets, food waste is approximately 30% dry material
[11,22,25e28].
As the water content in the substrate decreases, the density of
the substrate decreases. This is an important factor to note when
converting from tons of input material to a volumetric measure-
ment of cubic meters.
The relationship between dryness and volume can be deter-
mined by the following set of inequalities:
D ¼ 1 for b  0:15
Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
5
0:3


D ¼ 1 e b 0:1
for b>0:15 (3)
where: D ¼ Density in dry tons/m3, b ¼ Dryness (%).
A graphical representation of these equations can be seen in
Fig. 3. As dry material in a substrate increases, density decreases. In
the case of food waste, density is 0.78 tons/m3.
Fig. 3. Dryness vs. Density for organic materials.
3.3. Digester sizing considerations
In order to figure out the size of the tank needed to digest the
waste, several related parameters are needed: input flow rate,
dryness, total solids, volatile solids, organic loading rate, and
hydraulic retention time. The size of the reactor can be calculated
by a modified version of the organic loading rate equation:
 3
m kg
 
Flow Rate *Volatile Solids Concentration
day m3
 
Volume m3 ¼
kg
 
Organic Loading Rate =day
m3
(4)
From the estimate of annual tonnage, the amount of dry mate-
rial can be calculated. Once the amount of dry material is known
and the density, then a flow rate can be calculated after the tonnage
is converted to cubic meters and divided by the number of days in
a year.
Next, the amount of volatile solids (VS) needs to be measured or
taken from literature. This can be done in a laboratory by taking
a sample, weighing it, drying it in an oven at 550  C until it
maintains a stable weight (24 h) and then comparing the two
weights. In the case of food waste, literature shows volatile solids
are usually 90e95% of the total solids (dry material) or 28e29% of
the substrate’s wet weight [11,27,28]. Once the VS percentage is
known, the concentration of organic material in kg/m3 can be
obtained by multiplying VS by the weight of the substrate per cubic
meter (derived from density).
From here, it becomes necessary to have a desired organic
loading rate for the substrate. Studies of the mesophilic digestion of
food waste show that the organic loading rate can be much higher
than typical wastewater treatment or farm waste systems while
remaining stable. For wastewater treatment and farm waste, OLRs
can range from 1 to 5 kgVS/m3 [29,30]. The Environmental
(2) Protection Agency’s yearlong study of food waste AD in California
6 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10

ran successfully at an OLR of 7 kgVS/m3 [11]. Other studies have

shown that the OLR for food waste can go as high as 10 kgVS/m3
while remaining stable and producing biogas [21,26,28].
Once flow rate, concentration, and possible OLR ranges have
been determined, a design choice becomes necessary. Due to the
fact that OLR and HRT depend on each other and can both be used
to size the system, one or the other must be fixed in order to
determine the appropriate reactor size. This process should leave
an amount of eligible play for the parameter that has been fixed
once the system has been sized. Once in operation, changing HRT
and OLR have different effects on the system and the possible
consequences need to be considered before sizing can occur. If the
concentration of VS is increased by increasing the OLR while
keeping the HRT constant, then the viscosity of the slurry changes
and subsequently the pumpability of the substrate changes. There
is a usually an upper limit of 15% on pumps used to transport the
slurry, so the system could experience mechanical failure if the OLR
was already close to the upper limit [17]. In contrast, if the OLR is
held constant and the HRT is changed, then the system could
experience undesirable biochemical effects such as the organic
matter not breaking down as much and subsequent methane
production would decrease. Using the OLR as a prominent design
criterion does not appear often in literature, it is the HRT of
a system that is varied.

3.4. Biogas yield

Once the system has been properly sized, hydraulic retention

time and organic loading rate have definite values, then the biogas
production and specific methane production can be calculated.
A comparison of typical reported biogas yields from literature is
seen in Fig. 4 [11,17,31]. Food waste has a higher yield of biogas per
dry tonne than most substrates. Comparing food waste yields with
cow manure for example shows that food waste yields 15 times
more biogas per tonne than farm waste. This is intuitive because
the manure has already been digested by an animal and therefore
large amounts of the energy have already been removed. A litera-
ture survey of biogas yields from food waste was provided in
Section 2.8.

Fig. 5. System design flow chart.

4. Case study of design of food waste to energy system for an

urban building

4.1. Waste analysis

Fig. 4. Biogas yields of various substrates in m3/tonne VS.
3.5. System design overview
A flow chart of the biogas design process described in the
previous sections can be seen in Fig. 5.
The focus of this case study is on a general mixed food waste
that is found in most large kitchens and cafeterias in urban high-
rise buildings. If organic waste is not source separated, it is often
necessary to do a waste audit to accurately determine the
percentage of the total waste that is biodegradable. At Concordia
University, a student-funded organization named Sustainable
Concordia does a yearly month-long waste audit to estimate the

Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10
average composition of the waste that is sent to the landfill. The
most recent waste audit results from 2007 are shown in Fig. 6.
Fig. 6. Waste breakdown for the downtown campus (2007).
According to the audits from 2002 to 2007, the average amount
of digestible waste is 35e43% of the total waste [37]. From the
records of the waste sent to the landfill, a total tonnage of waste can
be calculated as well as a reasonable estimate of how many tonnes
of waste per year are available for anaerobic digestion. From the
landfill records, it’s also possible to determine how much waste
comes from individual buildings and then estimate the amount of
organic waste located therein.
Taking the most recent year’s data as an example as shown in
Fig. 7, Concordia sent 698 metric tonnes of waste to landfill at a cost
of $131.28 per tonne for a total cost of $91,608. Consultation of the
landfill records show that 55% of Concordia’s waste was produced
in the downtown campus totaling 385 metric tonnes. Taking into
account the fact that up to 43% of this waste is biodegradable and
could be used for composting or anaerobic digestion, 165 tonnes of
food waste is available. The amount of waste per month fluctuates
with the season and as a result the resulting energy output is not
constant. The total amount of biogas can be extrapolated and the
monthly energy production can be estimated using the downtown
campus waste fluctuations as shown in Fig. 8.
Fig. 7. Total tonnes of waste sent to landfill annually.
Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
7
Fig. 8. Downtown waste fluctuations over the course of a year.
For mixed food waste biogas yield, the value from the Envi-
ronmental Protection Agency’s yearlong mesophilic, mixed food
waste study is used: an average of 367 m3 of biogas per tonne of VS
added to the digester. For 165 tonnes of food waste produced on the
Concordia downtown campus, 50 tonnes of VS would be produced
(30% VS average). This gives an annual yield of 18,350 m3 of biogas
with a composition of 65% methane and 35% CO2.
For biogas made up of 65% methane, an energy value of
6.25 kWh/m3 is used [12]. Knowing there are 18,350 m3 of biogas
available, the total energy content of the biogas can be calculated as
114,688 kWh annually. If the biogas is sent to a 10 kW water boiler
to offset natural gas costs, 90e95% of the energy or
103e109 MWh/year (Fig. 8.) is available due to boiler efficiency
[31]. If the biogas is sent to an internal combustion engine attached
to a 3kW generator, then the efficiency drops to 35% and there is
only about 40 MWh (Fig. 9) of energy is available [31]. In Con-
cordia’s case, it has been calculated that if all of the biogas was sent
to the water boilers located in the same building, the total energy
costs for heating could be offset by 3%. Annual thermal and elec-
trical energy production is seen below.
Fig. 9. Annual thermal and electrical energy content of biogas.
8 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10
4.2. System design
Implementing an anaerobic digestion system in the urban
environment presents several unique challenges due to the
explosive nature of the gases involved and the paradigm shift
involved with more waste handling and management occurring
inside of city buildings. Additionally, designing a system that is
feasible for the amount of waste produced and the amount of space
available also presents unique challenges.
In the specific case of Concordia University the following system
design (seen in Fig. 10) is proposed and includes a shredder,
hydrolysis/pasteurization stage, digester tank, boiler, screw press to
separate the liquid digestate from the solid, composter to continue
to neutralize any pathogens in the solids, and aeration tank to
neutralize the liquids. The liquid and solid digestate can then be used
as nutrient-rich fertilizers for the building grounds and greenhouse.
The proposed plan involves collecting 100e165 annual tonnes of
organic wastes from various locations on the downtown campus
including one large cafeteria, a smaller kitchen, and several coffee
shops. Organic waste can be collected using 250 l bins and trans-
ported to the greenhouse on the 13th floor of one of the high-rise
buildings. The food waste can be dumped into a 300 l hopper
which feeds into a shredder to produce a particle size of 5mm or
less. The shredder is mounted to a 4 m3 mixing tank and the
shredded product mixed with liquid from the storage tank of
separated water at the end of the process. This first tank is where
the hydrolysis stage of the AD process begins. A side-mounted
mixer ensures a complete mix of the waste as it starts to break-
down. Hydrolysis can produce very unpleasant odors so the tank is
kept under negative pressure and the air is forced through a bio-
filter for deodorization before it is vented outside.
The resulting slurry is then pumped to a 30 m3 polyethylene
tank where the rest of the digestion process occurs. This tank is
modified to accommodate a top-mounted mixer and insulated with
polyurethane and equipped with custom steel cladding. The tank is
protected with over and under pressure vents and an automated,
enclosed biogas flare.
Fig. 10. Proposed AD system for the urban environment.
Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
The top-mounted 735 W mixer has a stainless steel shaft and
paddles and runs at a slow rpm approximately 8 h per day. The tank
temperature (35  C) is maintained by circulating the substrate
(15e20% TS) through a hot-water heat exchanger. This provides
additional mixing for the system.
The digester is fed through the interruption of substrate re-
circulation by using a 3-way valve that diverts the intake of the
pump from the digester to the mixing/hydrolysis tank. Slurry is
then pumped from the mixing tank to the digester at the pre-
determined organic loading rate (4e10 kg/m3/day) and the 3-way
valve is returned to its re-circulation position for heating and
mixing.
In order to combat the amount of the corrosive hydrogen sulfide
(H2S) in the biogas produced by the process, a small quantity of air
is injected into the digester head. This allows the H2S in the biogas
to be converted by bacteria to elemental sulfur that will cling to the
tank roof and be cleaned when necessary.
Generated biogas is stored in a 10 m3 double-membrane
gasholder. This gasholder is sized by taking the predicted average
hourly biogas flow rate and multiplying by a projected amount of
hours that the system would be down for repairs as well as the
space available for storage. In this case, the 10 m3 gasholder size
comes from a 2 m3/h biogas flow rate, 5 h of downtime gas pres-
ervation, and the size of the proposed greenhouse. If the gasholder
is full and the gas utilization system is still down, the gas will be
flared until operation resumes. The biogas line will be equipped
with standard gas safety devices such as flame arrestor, check valve,
and drip trap.
Biogas can also be directed to a 3 kW genset for demonstration
purposes or a 10 kW biogas boiler for hot water generation. Part of
the hot water can be used for heating the digester circulation
system and the excess can be used for heating applications in the
greenhouse or sent directly to the natural gas heaters located a floor
above the proposed AD system. Any biogas overflow is directed to
the flare for combustion.
The digestate at the end of the process is directed to a screw
press for manual dehydration. The solids fall into a container where
 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10 9

they can be added to the composter for further digestion/neutral-

ization. The separated liquid is sent into a storage tank for use in
dilution of incoming food waste in the mixing tank or nutrient-rich
irrigation for the greenhouse. Any remaining water, once aerated
can be sent directly to the sewer. A diagram of the system just
described is seen in Fig. 10.
This general design is flexible and could be sized based on
expected input tonnage and space requirements of different urban
environments. The design process remains the same. The size of the
anaerobic digestion tank for wet digestion can be determined by
process described by the flow chart in Fig. 5.

4.3. Safety considerations

Due to the possibility of explosive gas/air mixtures existing in

biogas facilities, measures that prevent the ignition of a dangerous
explosive atmosphere must be taken. Explosive areas are
3-dimensional spaces that are divided into zones according to the
probability of the development of a dangerous explosive atmo-
sphere due to the percentage of gas present as shown in Fig. 11.
Fig. 13. View of biogas storage tank explosive zone as seen from above.
4.4. System energy usage
This section deals with the energy consumption of the proposed
anaerobic digestion system for the urban building in downtown
Montreal. These figures have been compiled by researching existing
systems, reading the specification sheets on appropriate parts, and
consultations with biogas system operators about how often typical
system parts are energized.

4.4.1. Pre-treatment stage

Fig. 11. Percentages of gas present in air for explosion to occur.
1. Side mounted mixer on the hydrolysis tank e 0.367 kW 5 min./
day - 0.473 kWh/year
These zones can be broken down as follows: 2. Bio filter e Utilizes 100 W continuously - 876 kWh/year
3. Food Grinder e 5 min per day at 2.2 kW - 67.1 kWH/year
Zone 0: Explosive atmosphere that is Continuous, Long-term, or
Often 4.4.2. Digestion
Zone 1: Explosive atmosphere that is Occasional
Zone 2: Explosive atmosphere that is Unlikely and Short-term. 1. Mixer on top of Digester ¼ 0.735 kW, 1/3 of the time -
2178 kWh/year
For Zone 1, The distance of 1 m is valid for “free ventilation”
whereas a distance of 4.5 m has to be respected in closed rooms.
4.4.3. Heating of slurry
For Zone 2, a distance of 1e3 m must be respected and rooms can
be made “smoke proof” by: open face brick, plastered, or concrete
1. Heating load ¼ 0.3 kW during summer/1.5 kW during winter -
walls or walls with cladding made of non-combustible and
7.78 kWh/year
grouted plates, or metal walls. There is no zone around “perma-
2. Hot water pump ¼ 200 W continuously - 1752 kWh/year
nently technically tight” facility parts [17]. Examples of explosive
zones shown for digesters and biogas storage can be seen in
4.4.4. Digestate
Figs. 12 and 13.
1. Screw Press e 2.2 kW, 30 min/day - 408 kWh/year
2. Recirculation pump e 0.735 kW, 8 h/day - 2178 kWh/year
3. Sulfur removal system e 100 W continuously - 876 kWh/year

4.4.5. Process control

1. Utilizes 200 W continuously e 1752 kWh/year

From Section 4.1, the annual amount of energy available from

biogas produced is 144,688 kWh. The total energy utilization of the
system is approximately 10,095 kWh/year or 7% of total energy
produced. This leaves 134.6 MWh per year of surplus energy.

4.5. Cost overview

The estimated cost for this entire system design and its
Fig. 12. View of explosive zones for AD tank as seen from above. components (excluding building modifications) is $229,306.

Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
10 N. Curry, P. Pillay / Renewable Energy xxx (2011) 1e10
Assuming that the maximum of 165 tonnes of food waste is
diverted from landfill each year, a savings of $21,615 per year ($131/
tonne) in transportation costs is gained. Additionally if 18,350 m3 of
biogas is produced and 7% of it is used to heat the system, then
17,066 m3 of biogas is available to send to the building’s natural gas
heating system representing a savings of $8553 at 50 cents/m3 of
natural gas. The total annual savings from this part of the process is
$30,168.
5. Conclusions
This paper argues for the use of small-scale anaerobic digestion
in the urban buildings to deal with the organic waste produced on
site as a way to save on transportation costs and reducing the
amount of waste that is sent to landfills. Several techniques were
presented to estimate biogas production from food waste. A design
technique for sizing an anaerobic digestion system was also pre-
sented. A case study of a system design was presented for dealing
with 165 tonnes of food waste in a building in downtown Montreal,
Canada, including estimated energy consumption and production.
This technology could be integrated into pre-standing urban
buildings or designed and integrated into new buildings for
a paradigm shift toward a more renewable, sustainable future.
References
[1] United Nations Population Funds, Urbanization: a majority in cities. http://
www.unfpa.org/pds/urbanization.htm, (accessed May 2011).
[2] United Nations Population Division, World population to 2300. http://www.
un.org/esa/population/publications/longrange2/
2004worldpop2300reportfinalc.pdf, (accessed May 2011).
[3] United Nations Population Division, World urbanization prospects, the 2009
revision. http://esa.un.org/unpd/wup/Fig_1.htm, (accessed May 2011).
[4] Department of Sanitation NYC, Annual report 2010. http://www.nyc.gov/
html/dsny/html/Pub_Info/annual.shtml, (accessed May 2011).
[5] Reducing, reusing and recycling London’s waste. http://www.london.gov.uk/
priorities/environment/vision-strategy/waste, (accessed May 2011).
[6] City of Montreal, Montreal’s Draft Master Plan for the Management of
Residual Materials for 2008e2012.
[7] City of Toronto: Solid waste management e facts about Toronto’s trash. http://
www.toronto.ca/garbage/facts.htm, (accessed May 2011).
[8] Garbage disaster looms at giant Mexico city dump. http://www.reuters.com/
article/2009/02/12/us-mexico-garbage-idUSTRE51B4XX20090212, (accessed
May 2011).
[9] Ye, Jianhong. Low-carbon eco-city construction in China. www.china.cupa.
pdx.edu/archives/Low-Carbon_Eco-City.pdf, (accessed May 2011).
[10] Food and Agriculture Organization of the United Nations. Global food losses
and food waste. Dusseldorf, Germany: Interpack2011; 2011.
[11] Anaerobic digestion of food waste. East Bay Municipal Utility District; March
2008. Final Report.
[12] Republic of Slovenia Methodology for determining reference costs of elec-
tricity generated from renewable resources. Ministry of Economy, http://
www.mg.gov.si/fileadmin/mg.gov.si/pageuploads/Energetika/Sprejeti_
predpisi/Methodology_RES.pdf; May 14, 2009. accessed June 2011360e381.
[13] International Energy Agency, Key world energy statistics 2010. www.iea.org/
textbase/nppdf/free/2010/key_stats_2010.pdf, (accessed May 2011).
[14] Environmental Protection Agency, Municipal solid waste generation, recy-
cling, and disposal in the United States: facts and figures for 2009. http://
www.epa.gov/osw/nonhaz/municipal/pubs/msw2009-fs.pdf.
[15] Statistics Canada, Disposal and diversion of waste, by province and territory,
2004 and 2006. http://www40.statcan.gc.ca/l01/cst01/envir32a-eng.htm.
[16] Gooch Martin, Felfel Abdel, Marenick Nicole. Food waste in Canada, growing
the margins conference; November, 2010.
Please cite this article in press as: Curry N, Pillay P, Biogas prediction and design of a food waste to energy system for the urban environment,
Renewable Energy (2011), doi:10.1016/j.renene.2011.10.019
[17] Aragundy Jenny, Kottner Michael. IBBK, Fachgruppe biogas Weckelweiler,
"Biogas reference manual; 2008. Weckelweird, Germany.
[18] Hopfner-Sixt Katharina, Amon Thomas. Monitoring of agricultural biogas
plants in Austria e Mixing technology and specific values of essential process
parameters. Vienna: University Of Natural Resources And Applied Life
Sciences, Department Of Sustainable Agricultural Systems, Division Of Agri-
cultural Engineering; 2007. Final Report, Austria.
[19] Raven RPJM, Gregersen KH. Biogas plants in Denmark: successes and setbacks.
Renewable and Sustainable Energy Reviews January 2007;11(Issue 1):
116e32.
[20] Silva MRQ, Naik TR. Review of composting and anaerobic digestion of
municipal solid waste. Wisconsin: University of Milwaukee; 2008.
[21] The University of Southampton and Greenfinch Ltd, Biodigestion of
kitchen waste: a comparative evaluation of mesophilic and thermophilic
biodigestion for the stabilisation and sanitisation of kitchen waste.
Final Report, www.soton.ac.uk/wsunrise/Biodigestion%20final%20report.pdf,
(accessed May 2011).
[22] Mata-Alvarez J. Biomethanization of the organic fraction of municipal solid
wastes. London: IWA Publishing; 2003.
[23] Tchobanoglous George, Theisen Hilary, Vigil Samuel A. Integrated solid waste
management: engineering principles and management issues. 2 edition.
McGraw-Hill Science/Engineering/Math; 1993.
[24] Ontario Ministry of Agriculture. Final report for the study of food-based inputs
for biogas systems in Ontario; May 2008.
[25] Liu Guangqing, Zhang Rulhong, El-Mashad Hamad M, Dong Renje. Effect of
feed to inoculum ratios on biogas yields of food and green wastes. Bioresource
Technology November 2009;100(21):5103e8.
[26] Davidson Asa, Gruvberger Christopher, Christensen Thomas H. Methane yield
in source-sorted organic fraction of municipal solid waste. Waste Manage-
ment 2007;27(3):406e14.
[27] Cho Jae Kyoung, Park Soon Chul. Biochemical methane potential and solid
State anaerobic digestion of Korean food waste. Bioresource Technology;
1995:245e53.
[28] Zhang Ruihong, El-Mashad Hamed M, Hartman Karl, Wang Fengyu. Charac-
terization of food waste as feedstock for anaerobic digestion. Bioresource
Technology; March 2007:929e36.
[29] International Water Association Task Group. Anaerobic digestion model No. 1
(ADM1). IWA Publishing, ISBN 1900222787; 2002.
[30] Polprasert Chongrak, Yang PY, Kongsricharoern Noppadol,
Kanjanaprapin Wilai. Productive utilization of pig farm wastes: a case study
for developing countries. Resources, Conservation and Recycling 30 June
1994;11:245e59. Environmental biotechnology in waste treatment and
recycling.
[31] House D. The complete biogas handbook. House Press, ISBN 0915238470;
2006.
[32] Mohan S, Bindu BK. Effect of phase separation on anaerobic digestion of
kitchen waste. NRC Research Press, jees.nrc.ca; Feb. 19, 2008.
[33] Karnchanawong Seni, Uparawanna Sarawut. Performance of singe-phase
complete-mix anaerobic digestion of organic fraction of municipal solid
waste. Joint International Conference on sustainable energy and environment,
Bangkok, Thailand; November 2006.
[34] Peck Cara, US EPA. Food waste opportunity, http://www.epa.gov/osw/rcc/
resources/meetings/rcc-2008/sessions/msw/food/foodwaste/peck.pdf; 2008
(accessed May 2011).
[35] California Integrated Waste Management Board. Current anaerobic digestion
technologies used for treatment of municipal organic solid waste. Contractor’s
report. Department of Biological and Agricultural Engineering, UC Davis;
March 2008.
[36] Davidson Asa, Jansen Jes la Cour, Appelqvist Bjorn, Gruvberger Christopher,
Hallmer Martin. Anaerobic digestion potential of urban organic waste: a case
study in Malmo. Waste Management Research 2007;25:162e9.
[37] Concordia University. Sustainable concordia waste audits; 2002e2007.
[38] Zaher Usama, Li Rongping, Jeppsson Ulf, Steyer Jean-Philippe, Chen Shulin.
GISCOD: general integrated solid waste co-digestion model. Water Research
June 2009;43(10):2717e27.
[39] Vanrolleghem PA, Rosen C, Zaher U, Copp J, Benedetti L, Ayesa E, et al.
Continuity-based interfacing of models for wastewater systems described by
Petersen matrices. Water Science Technology 2005;52(1e2):493e500.
[40] Zaher U, Buffiere P, STeyer J-P, Chen S. A procedure to estimate proximate
analysis of mixed organic wastes. Water Environment Research 2009;81:
407e15.