Preliminary Report - McGill University

Tales of a Forgotten Bioresource:
The recycling of spent coffee grounds
Design III
BREE 495
Thomas Ammerlaan
Violette Barrière
Pascal Genest-Richard
Sandra Rabow
April 23, 2012
Department of Bioresource Engineering
McGill University
Montréal, Canada
Tales of a Forgotten Bioresource
Table of contents
1. Executive summary
2. Introduction
p. 3
p. 5
a. Project goal
b. Design proposal
3. Analysis and specification
a.
b.
c.
d.
e.
f.
p. 6
Coffee grounds
Drying
Oil extraction
Biodiesel production
Mushroom production
Production management
4. Prototyping, testing, optimization
a.
b.
c.
d.
Drying
Oil extraction
Biodiesel production
Mushroom production
5. Results and conclusions
a.
b.
c.
d.
e.
f.
p. 20
p. 25
Drying
Oil extraction
Mushroom production
Biodiesel
Production management
Facilities layout
6. Acknowledgements
7. References
8. Appendices
p. 36
p. 37
p. 39
a. Examples of calculations
b. Project journal
Thomas Ammerlaan
Violette Barrière
Pascal Genest Richard
Sandra Rabow
April 23, 2012
2
1. Executive summary
Coffee is one of the most popular beverages in the world. Research targeted at the
recycling of spent coffee grounds remains limited. This report documents the analysis,
testing, and final design of the drying and oil extraction of spent coffee grounds as well as
two value added products: biodiesel and edible mushrooms. Upon collection, spent coffee
grounds have a moisture content of 40.5 %. Different drying methods were experimented
with, such as direct sun exposure in aluminum plates, oven drying at around 175 °F, and
room temperature drying in different vessels. The best results were achieved at room
temperature with exposure to direct sunlight and frequent stirring. The drying mechanism
designed for the final facility is a belt dryer, with heating and ventilation provided by a lowtemperature heat pump and sun exposure from windows positioned on top of the band
pass. Shaking rollers are also positioned under the belt at regular intervals to rotate coffee
grounds onto themselves and allow a more uniform convective drying. After drying, coffee
grounds are sent to the oil extraction phase. Three extraction tests were made, yielding
between 8.9 % and 10.8 % oil by weight. The third and last prototype allowed for the most
oil extraction and consisted of a heating plate, an erlenmeyer flask, a soxhlet tube and an
upright water condenser. The three main steps in solvent-based oil extraction are
percolation, desolventizing, and distillation. Belt extractors were found to be the most
efficient for processing large volumes of material. Sizing is based on the quantity needed to
process 100 kg of biodiesel per day. Assuming a working value of 11 % for the oil
extraction yield, this translates into 909 kg of dried coffee grounds to be processed daily,
which is a volume relatively easy to obtain in an urban environment. The extraction rate of
the final model is 617.46 kg of spent coffee grounds per hour, requiring only an hour and a
half of operation per day. Solvent is added at 1:1 weight ratio with coffee and therefore
each pump has an operating capacity of 0.262 L/s. An airtight screw conveyor belt will be
used to feed the coffee grounds into the extractor. As hexane is hazardous to human health,
attention needs to be taken in the design of an airtight extractor. After oil extraction, the
grounds need to be desolventized using a down draft desolventizer, which heats the
medium to 100 °C to evaporate the solvent and redirect it to the belt extractor. The oil,
after being purified through distillation, is filtered and sent to the biodiesel production
facility. Due to limited quantities of oil, only one experiment for biodiesel was conducted.
3
Literature shows that coffee oil extracted using hexane has relatively high levels of free
fatty acids (FFAs) (3.65%), which is above the satisfactory limit of 0.5%-1% for one-step
alkaline transesterification. The design therefore involves a two-step transesterification
process whereby FFAs are initially converted to esters in an acid-catalyzed pre-treatment,
and then converted to biodiesel in an alkali-catalyzed transesterification. Optimized
conditions were taken from literature and determined to be 20:1 molar ratio alcohol to FFA
and 10 wt% H2SO4 with respect to FFAs for the esterification reaction, and 6:1 molar ratio
of alcohol to oil and 1 wt% KOH with respect to oil for the transesterification reaction.
Volumes for the reactors are designed to accommodate a mass inflow of 100 kg of oil, and
follow a 135 % volume container and a height to diameter ratio of 1.5:1. Mushroom
production tests occurred on a small scale. A mother culture was first isolated, and
reproduced into four petri dishes, which were then used for inoculation. The first
inoculation was of pearled barley followed by a variety of proportions of different media
with spent coffee grounds. The experiments showed that indeed, Pleurotus ostreatus
mushrooms could grow on a substrate made of 50 % spent coffee grounds, and probably
more. The final mushroom production phase design involves two main rooms, one for
spawn preparation, where the mother spawn is prepared in order not to be affected by
sterilization conditions in the growth room, and one for the rest of the production, where
sterilization, incubation and growth occur. The growth system is one of shelf cultivation,
where two levels of mushrooms grow on top of the other. The final production facility is
modeled on 100 kg of oil for biodiesel production which corresponds to 1000 kg of dry
coffee grounds at 10% oil extraction.
4
2. Introduction
Coffee trees belong to the family of Rubiaceae, however only two species of the
genus Coffea are economically significant: C. arabica and C. canephora. The stimulating
effects of Coffea plants have been known for centuries. According to available records,
nomadic tribes from Yemen first adopted the coffee culture in the 15th century (Weinberg
& Bealer, 2001). First users chewed the leaves and berries, much like Andean populations
chew coca leaves to relieve themselves from pain, hunger, and altitude sickness. Its
controlled cultivation then spread around the Middle East and Northern Africa regions.
From then on, coffee consumption has grown worldwide to become one of the most
popular beverages in the world. Its production is now at the heart of the economy of over
50 developing countries (FAO, 2006).
The coffee bean is the reproductive unit of the coffee plant. The endosperm is where
the nutrients are kept for the seed to grow into a plant. Raw constituents of the arabica
coffee are approximately 12 % moisture, 50 % carbohydrates, 16 % lipids and 10 %
protein-like material, in addition to 1.2% caffeine (Petracco, 2005). Roasting and brewing
alters the chemical composition of coffee, however the overall content of caffeine and lipids
remain relatively unchanged (Ibid.). Currently, the majority of spent coffee grounds are
sent to landfills. Spent coffee grounds, however, do hold a significant commercial potential,
which have only begun to be explored. This paper will therefore investigate the
optimization of the complete recycling of spent coffee grounds.
2.1 Project goal
The goal of this project is to define and optimize the best system for the recycling of
spent coffee grounds using both existing and novel scientific methods. In order to do so, we
reviewed some fundamental scientific and empirical knowledge associated with a number
of technologies, namely composting, soil amendments, gasification, combustion, biodiesel
production, and mushroom cultivation. This was done with the goal of developing a
technological system that best serves the requirements of the project. A key dictum of the
project is the idea of cradle-to-cradle design, in that case transforming a waste-stream of
coffee grounds into renewable products that have ecological, social and economic value
(McDonough & Braungart, 2002).
5
2.2 Design proposal
The team worked on defining a system that is both sustainable and economically
feasible. Considering factors like market needs, properties of spent coffee grounds,
previous research, and constraints such as economic feasibility, environmental
sustainability, time efficiency and ease of use, the team has narrowed down the potential
solution to a system composed of two value-added products: biodiesel and mushrooms.
The objective is to optimize the production processes of the above products from raw spent
coffee grounds. This report outlines the different technical specifications behind our
assumptions and hypotheses, the testing and prototyping that was conducted, as well as
the details of the final design for all steps of the process: drying of the spent coffee grounds,
extraction of the oil contained in the grounds, conversion of the extracted oil into biodiesel,
and the production of Pleurotus ostreatus mushrooms on the remaining grounds. The final
design is consolidated into a single factory system, to model the complete transformation of
spent coffee grounds into biodiesel and mushrooms.
3. Analysis and specifications
3.1 Spent coffee grounds
Spent coffee grounds have had boiling water – or condensing vapor – run through
them, a process known as brewing but in fact called extraction. The extraction process is a
critical one in the preparation of a coffee beverage, as its purpose is to dissolve the soluble
flavors from the coffee into the water. Different indicators are used to quantify the
extraction process: the brew ratio is the ratio of coffee grounds to water (50-70 g/L); the
strength is a measure of the total dissolved solids in the beverage (1-1.5%); and the
extraction, or yield, is the percentage of the coffee grounds that dissolved into the water
during the extraction process (Lingle, 1995). Under normal conditions, the yield achieved is
between 15% and 25% (Ibid). There was no discrimination as to the provenance of the
coffee grounds, whether they were from espresso, filter, or decaffeinated beverages. In the
context of our experiments, an interrogation remained as to whether the extracted solids
contained oil, so as to cause a reduction in our oil extraction yield. As can be seen in section
3.3, the extraction rate achieved suggests that no significant amount of oil is removed from
the grounds, as the extraction rates achieved approach the ones found in the literature.
6
3.2 Drying
According to Ratti (2009), dehydration is probably the oldest and most frequently
used method of food preservation: “the term dehydration refers to the removal of moisture
from a material with the primary objective of reducing microbial activity and product
deterioration" (p. 16). It is therefore important to design an efficient system for the
immediate drying of the spent coffee grounds upon arrival to the premises, as microbial
activity can degrade the product and moisture can impede the oil extraction process.
Amongst the various dehydration methods available, namely vacuum, superheated
steam, contact, freeze-, osmotic and hot air drying, the latter will be the one chosen as it is
the only one with a drying medium that is air at atmospheric pressure, cheap and readily
available on site. The design style is one of low-speed, continuous, atmospheric, convective
and radiative (solar) single-pass conveyor band dryer with heat pump. This was based by
elimination of other drying technologies based on the product at hand. Certain
disadvantages remain, however, as there are nutritional quality losses reported in the
literature, and the drying time is relatively long (Hawlader et al., 2006; Bimbenet & Lebert,
1992). According to Ratti (2005), hot air drying efficiency depends on the thickness,
geometry, and density of the product. Of these, only the latter is controllable on site. Drying
medium variables include temperature, velocity, and relative humidity, all controllable on
site. The coffee grounds density, as well as the drying air is therefore the factor to optimize
in the case of the dehydration of spent coffee grounds.
The most important variable to start from is the moisture content of the product to
be dried. In the case of conventional spent coffee grounds obtained from filter machines,
the moisture content is around 40 % (see section 4a). This amount of water is highly
manageable, as is it much lower than other conventional fruits or vegetables that are
factory-dried. According to Rizvi (2005), the effect of temperature on water sorption
isotherms can be represented by the Clausius-Clapeyron equation:
[
⁄
]
7
Figure 1: Effect of temperature on the water
sorption isotherm for carrot (adapted from
Kiranoudis et al., 1993)
In the figure above, it is possible to see that temperature is indirectly proportional
to the water activity (aw) in the product. Indeed, temperature affects the mobility of water
molecules and the equilibrium between the vapor and adsorbed phases (binding forces
decrease with increasing temperature) (Kapsalis, 1987). Assuming similar sorption
isotherms as the carrot (Figure 3) for spent coffee grounds as well as a drying room
temperature of around 45 °C, it can be seen that upon entry into the drying room, the water
activity will be of 0.9 for a moisture content of 40 %. Replacing that value in the ClausiusClapeyron equation, we get:
[
⁄
]
The heat of sorption is therefore 39.42 kJ/kg. We can now approximate the energy needed
for drying as follows:
(
)
Converting into kWh yields:
Which is a negligible energy use. Most of the costs related to energy use will be heating and
ventilation. The heat pump used will need to be a low-temperature adjusted heat pump,
such as the Hiseer AW13/F-DPNHE (Hiseer, 2012). The latter is more than sufficient to
8
provide for a 45 °C environment augmented with solar radiation in a cold temperature
climate outside.
For the conveyor belt, the band needs to have holes small enough for grounds not to
go through but air to flow through. A porous and flexible material like hemp fiber can
therefore be used for the conveyor belt. The belt will be sized so as to be able to handle 100
kg of wet spent coffee grounds per day. Assuming a 3-hour drying period, a 1 by 20 meter
belt, the speed of the conveyor will remain around 0.1 m/s.
3.3 Oil extraction
Oil extraction from organic matter has been occurring for centuries, namely to
obtain such staple foods as olive and palm oil. Traditionally, extraction techniques revolved
around mechanical pressing and thermal conditioning. However, the use of organic
solvents for extraction was popularized in Europe in the 1850s with the first continuous
extraction system built in the 1930s in the U.S., which since then has become the industry
standard (Wan, 2007). Several organic solvents can be used to perform oil extraction but
the most effective solvent for spent coffee grounds remains n-hexane. Al-Hamamre et al.
(2012) found a highest yield of 15.3 wt% with hexane for an extraction period of 30
minutes. Special care needs to be taken in the manipulation of n-hexane as it is listed on the
hazardous air pollutant by the EPA, and has been recognized as a health hazard affecting
the central nervous system (Seth et al., 2007).
The principle behind solvent extraction can be summarized as (1) immersion of the
medium in the organic solvent for solid-liquid mass transfer, (2) recuperation of the
miscella (solvent – oil mixture) followed by (3) desolventizing the medium, and (4)
distillation. No single equation can properly model the solid to liquid oil mass transfer
(Carrin & Crapiste, 2007). Several models have been developed for general as well as
specific extraction processes with soybeans, with most recent work having been completed
on the De Smet multistage horizontal extractor with miscella recirculation (Almida et al.,
2010; Carrin et al., 2007; Seth et al., 2007; Karnofsky, 1986; Duggal, 1987; Coats et al.,
1950).
3.3.1 Solid-liquid mass transfer with organic solvents: Mathematical models
reviewed were for soybeans as soybeans and spent coffee grounds have a similar oil
9
content (15 % and 20 %, respectively) and are both judged to be low when compared to
canola and peanut oil, which are 43 % and 50 %, respectively.
Duggal (1987) found the following simplified general equation for soybean
extraction estimating the remaining oil content of the medium:
(2)
Where E is the decimal residual oil in soybean meal after time , Q is the oil
extracted for extraction time with zero moisture,
and
is the total oil content of the sample,
are the model parameters based on experimental data. It has been recognized
that construction of extraction plants remains based on experience and that mathematical
modelling is a more academic exercise (Karnofsky, 1986). However, potential uses of the
model have been recognized for trouble shooting when breakdowns occur (Almeida et al.,
2010).
The multiple equations needed to more accurately model the oil extraction are the
following: (1) oil mass flow from micropores to macropores, (2) oil mass flow in the
macropores and finally (3) oil mass flow rates in the miscella in the different chambers of
the extractor (Ibid.). Almeida et al. (2010) offers an in depth analysis with case study for
the oil extraction process. Since it has already been recognized that the actual system
design is based on experience, further modelling is deemed irrelevant in the present
project. Further design of the extraction plant is therefore based on empirical findings.
Figure 2: Undissolved oil vs. extraction time at
several miscella concentrations ranging from
15.3% oil (curve 1) to 0.3% oil (curve 4)
(Karnofsky, 1986)
10
The rapid decrease of oil mass transfer rate after 10 minutes of being in the organic solvent
can be seen in Figure 4, by which there time only 6% of undissolved soybean oil remains.
As found by Hamamre et al. (2012), no complete oil extraction can be achieved due to the
less solvent soluble oil portions such as phosphatides in spent coffee grounds (Almeida et
al., 2010, Karnofsky 1986).
Industrial technologies available fall under two major categories: rotary extractors
and belt extractors. According to literature, belt extractors require higher capital spending,
but their extraction process is much faster since the flow is continuous instead of batch
type (Almeida et al., 2010).
As can be seen in Figure 3 below, belt extractors consist of a slow moving band
conveyor (5) inside a completely sealed chamber. The band is lined with perforated sheets
and stainless steel cloth unto which the medium is placed at the inlet (1), forming a bed and
exits through the rear (6), where the extracted medium gets redirected to the
desolventizer. While moving through the sealed chamber, the medium gets sprayed with
organic solvent (2, 7), which percolates through the bed to absorb the oil from the medium.
The organic solvent is purest at the end of the chamber (7) to ensure maximal oil uptake,
and the most saturated solvent can be found at the entrance of the medium (1), where the
oil is most readily available.
Figure 3: Belt extractor with recirculating miscella (adapted from Almeida et al., 2010)
11
The De Smet extractor used for modelling by Almeida et al. had a solvent recirculation rate
(Q) of
in each pump, and the material bed measured 1.20 m width (W), 2.10 m
height (H), and 18.5 m length, though its effective length was 15.0 m. Mass flow of soybeans
was regulated to 30,220 kg/h ( ̇ ) and the laminator was adjusted to 0.33 mm thickness (t).
The solid density of the soybeans was measured to be 520
processing volume of 58.12
(ρ). From this, a
h can be found.
̇
/h
However, when computing the belt speed, the answer yielded is unreasonable.
̇
W*t = 58.12
/ (1.20 m*0.00033m) / 3600 s/h = 40.77 m/s
A more reasonable belt speed of 0.0408 m/s is therefore assumed. This is a difference of a
factor of 1000, most likely caused by an unreasonable laminator width (t).
Residence time of the coffee grounds on the belt is calculated to be 367.7 seconds, or
6.13 minutes.
3.3.2 Desolventizing of end medium: Since the extracted material retains the
organic solvent within it, the solvent has to be recuperated for economic and
environmental considerations. The retention varies from 20% to 35% of saturated weight
(Oiltek Canada, 2012). Most popular industrial solvent recuperation techniques involve
direct and indirect heating of the medium well above the boiling temperature of the
solvent, which in the case of hexane is 69°C (Al-Hamamre et al., 2012). Appropriate
desolventizers have been developed by leading oil extraction equipment manufacturers to
ensure minimal leaking of solvent into the atmosphere and recuperation. For instance, the
SRS Engineering Corporation desolventizer heats the medium up to 100°C by jacketed
steam, thus evaporating the solvent. The resulting vapors are then redirected to a dust
catcher, or a scrubber, wherein they are washed with hot water and then led to a
condenser, where the solvent is recuperated (SRS Engineering Corporation, 2012).
3.3.3. Distillation of miscella: Finally, distillation of the miscella remains a
crucial step wherein the oil is separated from the solvent. Novel technologies using nano-
12
sized composite membranes are currently being researched but the industry standard
remains distillation. Distillation consists of heating the miscella under vacuum to a
temperature above the boiling point of the solvent, as with the desolventizing process
(Perry & Green, 1997). The oil is then recuperated, cooled and redirected to a storage tank.
As for the gasified solvent, it is also redirected to another condenser and recuperated for
subsequent use in future extraction cycles (Ibid.). Since the inflow is a two-component
system, the simplest continuous-distillation process is the adiabatic single-stage
equilibrium-flash process (Ibid.).
Figure 4: Simple Adiabatic Flash Distiller
3.4 Biodiesel Production
Biodiesel is a non-toxic, biodegradable, and renewable alternative to petroleum
based diesel fuels (Balat & Balat, 2010). It consists of monoalkyl esters obtained through
the transesterification of vegetable oils and animal fats with alcohol and an acid or base
catalyst (Al-Hamamre et al., 2012).
Coffee oil is a valuable recycled feedstock for the biodiesel industry. Compared to
other waste sources; such as cooking oils, animal fats, and other biomass residues, coffee
has the additional benefits of being less expensive, more stable due to the presence of
antioxidants, and comprising of a nice smell (Al-Hamamre et al., 2012). Similar to other
waste sources, oil extracted from spent coffee grounds have a relatively high level of free
fatty acids (FFAs) (3.65%), which is above the satisfactory limit of 0.5%-1% for one-step
alkaline
transesterification
(Ibid.).
The
design
therefore
involves
a
two-step
transesterification process whereby FFAs are initially converted to esters in an acidcatalyzed pre-treatment, and then converted to biodiesel in an alkali-catalyzed
transesterification reaction.
13
The main factors affecting the transesterification reaction and thereby quality of
biodiesel are the molar ratio of alcohol to triglycerides, type of alcohol, catalyst type and
concentration, reaction conditions (temperature, mixing intensity) and condition of the
vegetable oil (including FFA and water content) (Bart et al., 2010). These parameters were
optimized through the compiling of scientific and empirical knowledge with respect to
biodiesel production.
3.4.1 Pretreatment: Pretreatment involves filtering coffee oil to remove any
residues and being sure the oil is dried to a water content of 0.4 wt% (Santori et al., 2012).
3.4.2 Titration: Dissolve 1 gram of pure potassium hydroxide lye (KOH) in 1 L
of distilled or de-ionized water (0.1 % w/v KOH solution). In a smaller beaker, dissolve 1
mL of oil in 10 mL of pure isopropyl alcohol. Warm the beaker gently. Stir until all the oil
dissolves in the alcohol and the mixture turns clear. Add 2 drops of phenolphthalein
solution. Using a graduated syringe, add the 0.1% KOH solution drop by drop to the oilalcohol-phenolphthalein solution, stirring all the time. Keep on carefully adding the lye
solution until the solution stays pink for 15 seconds (Addison, 2010). FFA % can be
determined using the following equation:
(
(
)
(
)
(
)
(
)
)
Acid-Catalyzed Esterification: In the esterification reaction, which is provided below,
alcohols and carboxylic acids react to produce esters:
RCOOH + R’OH
R-COOR’ + H2O
Oil is mixed with methanol (a molar ratio of alcohol to free fatty acids of 20:1) and
significant quantities of H2SO4 (10 wt% of total fatty acids content). The reactor should be
stirred at about 600 rpm, with a temperature of 60 °C for 2 hours. The reaction is
transferred to the separator where methanol, water and H2SO4 separate from esters and
fatty acids. The oily phase is returned to the reactor, and the methanol/acid/water mixture
is removed and treated for potential reuse (Santori et al., 2012; Knoth et al., 2005).
3.4.3. Alkali-catalyzed Transesterification: The transesterification process
occurs stepwise and given below:
Triglyceride (TG) + R’OH
Diglyceride (DG) + R’COOR1
14
Diglyceride (DG) + R’OH
Monoglyceride (MG) + R’OH
Monoglyceride (MG) + R’COOR 2
Glycerol (GL) + R’COOR 3
Prepare a methoxide solution using a 6:1 alcohol to oil molar ratio and 1 wt% KOH.
Heat the reaction tank to 60 °C and then add the methoxide solution. The reactor should be
stirred at 600 rpm for 2 to 4 hours. Transfer to separator and let the glycerol and methyl
esters separate overnight (Santori et al., 2012; Knoth et al., 2005).
Figure 5: Transesterification of rapeseed
oil with 6:1 molar ratio at 60 °C and 600
rpm (Rashid & Anwar, 2008)
Stirring is one of the most important factors affecting reaction yield (Santori et al.,
2012). The Reynolds number is used to classify and compare turbulence in a vessel:
where:
= density of the fluid (kg/m3)
= rotational speed (rps)
= diameter of the agitator (m)
= dynamic viscosity of the fluid kg/(m s)
Santori et al. (2012) found that mixing the reactants with a mechanical stirrer
between speeds of 300 rpm and 600 rpm (corresponding to Reynolds numbers of 6200 and
12400), yielded higher conversion at higher intensities. Mixing intensities higher than
12400 (600 rpm) did not have any effect on triglyceride conversion (Ibid.).
15
3.4.4. Methanol Removal: The mixture then moves through a methanol
stripper; such as a vacuum flash processor or falling film evaporator (see drying for more
information on flash evaporation) (Ibid.).
3.4.5. Water Washing and Neutralization: The water should be warmed (about
50 °C) and have a slightly acidified pH. Acid is added to the esters in order to neutralize
residual KOH and split soap that formed during the reaction into water soluble salts and
fatty acids, and therefore requires titration to determine exact neutralization quantities.
H3PO4 is recommended since the resulting salt K3PO4 can be used as a fertilizer. Water
separation is best done with a centrifuge. The entire process may be repeated a couple of
times (Ibid.).
3.4.6. Drying: Drying usually occurs in an isothermal flash evaporation unit. A
flash evaporator first heats the liquid at an elevated pressure, then sends it through a flash
valve that decreases pressure, causing the more volatile component of the liquid (water) to
vapourize (Gerpen et al., 2004). Flash evaporation of a multi-component liquid is governed
by the Rachford-Rice equation, which is based on subtracting mole fraction summation
equations from each other (Kooijman et al., 2000):
∑
[
]
where:
= mole fraction of component i in the feed liquid
= mole fraction of feed that is vaporized = V/F
= vapor-liquid equilibrium constant = yi/xi
= molar feed rate with component mole fractions zi
= vapor flow rate
yi = mole fraction of component i in the flashed evaporator
xi = mole fraction of component i in the residual liquid
Given pressure and temperature, K-values can be calculated. Vapour feed- flow ratio
can then be determined from the root of the Rachford-Rice equation (Koojiman et al, 2000).
3.4.7. Glycerol Phase Acidification: The glycerol phase leaving the separator is
approximately 50 % glycerol and contains excess methanol, most of the catalyst, and soap.
Acid is added to convert soap into FFAs and salts. The FFAs are insoluble and can be
separated and removed using a centrifuge, and then recycled (Santori et al., 2012).
16
3.4.8. Methanol Removal of Glycerol: The liquid mixture is heated to a
temperature between 90 and 120 °C and sent through a pressure reducing valve into a
tank. The more volatile part of the liquid evaporates, at which time the glycerol would have
about 85 % purity. Methanol recovered both in this stage and from the esters tends to have
high water content. Water should be removed using a continuous distillation column
before the methanol is recycled (Santori et al., 2012).
Figure 6: Flow diagram for biodiesel process (Adapted from Knoth et al., 2005)
The two main types of reactor vessels used in biodiesel production facilities are
stirred batch reactors (BRs) and continuous stirred tank reactors (CSTRs) (Santori et al.,
2012). BRs have good mixing characteristics and facilitate highly controlled reaction
conditions (Gerpen et al., 2004), which is favourable to smaller production volumes. The
materials of the reactor vessels are another important consideration in that they must be
able to withstand acidic and/or basic conditions during the reaction. Stainless steel is a
good choice for base-catalyzed transesterification; however an acid resistant material, such
as Hastelloy, should be used for the acid-catalyzed esterification reaction and washing tank
(Gerpen et al., 2004).
Pumping is used to transfer liquid from one vessel to the next, and also sometimes
for mixing. Centrifugal and positive displacement pumps, such as gear and lobe pumps, are
17
the two most commonly used. Centrifugal pumps are used when no emulsion will occur
among the compounds, for example with methoxide, and positive displacement pumps are
better suited for viscous fluids, for example with oil, and when careful flow control is
required to guarantee consistent, low-speed and uniform flow for various heads and no
pulsing (Santori et al., 2012).
Figure 7: Sliding vane pump and centrifugal pump (Adapted from Blackmer, 2012)
3.5 Mushroom production
Fungi play a pivotal role in most ecosystems. They serve as one of the forest’s
greatest decomposers and establish beneficial symbiotic relationships with the roots of
many trees and plants. Mushrooms, the fruiting bodies of fungi, also carry important
nutritional benefits for humans. Among these benefits are: i) a protein content averaging
20 % of their dry weight; ii) an assortment of essential amino acids; iii) vitamins such as
thiamine, riboflavin, niacin, biotin, ascorbic acid and Vitamin D; iv) a low fat content; and v)
a variety of medicinal properties (Stamets, 1983). Although there are more than 2000
species of edible mushrooms, white mushrooms, oysters and shiitakes are among the most
cultivated (De Carvalho et al. 2010). Pleurotus ostreatus (Oyster mushrooms) belong to
Pleurotaceae, Agaricales, Basidiomycota. They have been termed lignicolous, due to their
capacity to utilize lignin: a constituent of wood that is also present in plants (Stamets &
Chilton, 1983). It is very rare to find biological processes domesticated that are as efficient
as mushrooms to accomplish that task. For this reason, Pleurotus species are most often
cultivated on agricultural residues.
Pleurotus ostreatus is heavily grown all over the world, and protocols vary. Usually,
tissue culture (mother culture, step 1) is recommended as the start point. This is usually in
the form of a petri dish. From there, the mother spawn (step 2) can be produced by
18
inoculating a growth media. In turn, inoculation of mushroom spawns (step 3) on a
nutrient-rich substrate will produce the final fruiting media. This substrate is a
determining variable in mushroom production. Spent coffee grounds having undergone a
solvent extraction do contain carbon and nitrogen in a 16:1 ratio (Kondamudi et al., 2008).
Table 1: Nutrient contents of different substrates (from Gabriel, 2004)
Table 1 shows the nutrient content of the most frequently used agricultural residues
in mushroom production. One can notice how low in Nitrogen, cellulose and hemicellulose
hardwoods are. However, Pleurotus ostreatus grows on hardwoods in nature, as it has has
the power to degrade lignin.
Table 2: Physico-chemical trend of substrate in each stage (from Gabriel, 2004)
Table 2 shows that the most used nutrients are cellulose and hemicellulose. Growth tests
would help determine if spent coffee grounds are a viable substrate and contains enough of
the required nutrients to sustain Oyster mushrooms.
If that is the case, the design parameters could be the same as for mushroom
production on other agricultural wastes. Different growth methods exist, and need to be
optimized based on the materials available, the climate, as well as the Pleurotus species
chosen.
19
Table 3: Parameters for optimal Pleurotus production (from Stamets, 1993)
Temperature and relative humidity are not too complicated to control. Low light
(the optimal range corresponds to outdoor light on a cloudy day) and CO2 levels can also be
maintained with lights and a proper ventilation system. More complicated parameters to
design for are the need for steam injection, as the substrate needs to be sterilized, ideally
on-site. Steam tubes will have to run along the walls and be externally controlled. This
would allow for the sterilization room and the growth room to be one and the same, which
would ease the process by diminishing the need for labor and space.
4. Prototyping, testing, optimization
4.1 Drying
Coffee grounds were obtained from the households of the design team members, as
well as from the Coopérative du Grand Orme and the McGill Engineering Undergraduate
Society. The grounds were dried as soon as possible so as to diminish the risk of fungal
growth before oil extraction or mushroom colonization (see Appendix, Figure 17).
Different drying methods were experimented with, such as direct sun exposure in
aluminum plates, oven drying at around 175 °F, or simple room temperature drying in
different vessels. It appears like the best results were not achieved in the oven, even
considering the hot air. The lack of air circulation is probably to blame (see section 3b).
Best results were achieved at room temperature with exposure to direct sunlight and
frequent shaking (see Appendix, Figures 18 and 19). A calculation of the moisture content
of the spent coffee grounds (from filter coffee beverage) was undergone through the
following:
20
The water content of wet and fresh spent coffee grounds is therefore around 40 %. This is
the moisture content we assumed for the design of the drying apparatus in the facility.
3.5 Oil extraction
Though two previous oil extractions tests were conducted. The third one is
described as follows, as the first extraction yielded 10.8 % oil recovery, the second failed.
Notes taken on the two previous extractions can be found in the Appendix.
The third prototype was a custom built soxhlet extractor. The set-up consisted of
standard laboratory parts, with the exception of the soxhlet unit, which was ordered from
Pyrex. The parts were the following:
 (1) flat plate heater and magnetic stirrer Cimarec 2
 (1) magnet stirrer
 (1) stainless steel support rod 45 cm long
 (2) clamps
 (1) 250ml Erlenmeyer flask with 24/40 top
 (1) Pyrex 3740-M soxhlet tube with 24/40 bottom, 45/50 top, and 75 ml solvent capacity
 (1) Supelco 6-4186 condeser with 45/40 joint
 (2) Fisher c-219-A tubing for water flow to and from the condenser
 (1) Filtering thimble with 30 mm diameter and 80 mm height
The extraction took place under a fume hood to prevent solvent vapours from
escaping into the room. 20 g of dried spent grounds were loaded into the thimble.
Following this, the Erlenmeyer flask were filled with 175 mL of n-hexane. The extractor
unit is then placed on the heating plate and fastened from the rod to the soxhlet and the
condenser using the clamps. The cold water flow is turned on to a mass flow of 10 L/min,
the element is heated to approximately 150 °C, and the magnetic stirrer is activated to its
lowest setting (approximately 80 RPM).
As the solvent evaporates from the Erlenmeyer flask, it gets redirected to the
condenser where it is liquefied and percolated through the spent coffee grounds. The
thimble slowly fills up and once it reaches a volume of 75 ml, when it syphon-discharges
the miscella to the Erlenmeyer flask. The solvent then goes through the process a second
time. A third repeat is inefficient as the time required to harvest the remaining oil is too
long, as is indicated by the relative clarity of the miscella.
21
On average, the extraction process took 17.2 min for the first run and 9.9 min for the
second run, yielding into an average residence time of the grounds in solvent of 27.1
minutes.
During each batch, there was a 50 mL loss of hexane due to the remaining solvent in
the grounds after extraction and the losses to the atmosphere from the condenser, as the
condenser does not operate at 100% recuperation.
After the batch, it is thus necessary to add 50 mL of solvent to the Erlenmeyer and
load a new thimble with dried coffee grounds. After 15 repeats, the miscella was distilled
using a Buchi R-210 Rotovapor yielding 26 ml. At a density of 892.1 g/L, the result was
23.19 g of oil for an extraction yield of 8.9 %. Grounds were desolventized by leaving them
under the fume hood, thus considering the remaining n-hexane a loss.
3.6 Biodiesel production
Given limitations due to the quantity of oil extracted, only one test batch of biodiesel
was carried out. The experiment was done using optimization conditions derived from
literature (see analysis and specification), and 10 mL of oil. The process and results are
outlined below.
1. Coffee oil was filtered using a coffee filter.
2. Graduated cylinder was weighed followed by 10 mL of coffee oil to determine density of oil:
0.89 kg/L.
3. Based on the literature it was assumed that the oil has an FFA content of 3.65% (AlHamamre et al., 2012). 10 mL of oil was poured into a beaker and heated to 60 °C on a hot
plate. 10 ml of methanol was added to the oil and stirred at 600 rpm for 5 minutes using a
magnetic stirrer. 0.02 ml H2SO4 was then added to the methanol/oil mixture and mixed at
600 rpm, 60 °C for 2 hours. The beaker was covered with paraffin to minimize methanol
escape.
4. Mixture was left to sit overnight. Meanwhile methoxide was prepared by adding 0.09 grams
of KOH to 24 ml of methanol, shaken, covered and left to sit overnight.
5. Half of the methoxide was poured into the unheated mixture and mixed for 5 minutes at 600
rpm. The mixture was then heated to 60 °C and the second half of the methoxide was added.
Mixture was mixed for 4 hours at 600 rpm and 60 °C.
6. Mixture was left to sit overnight.
22
7. Biodiesel and glycerol separated in the beaker. The biodiesel was pipetted into a separate
tube and washed with water, then left to separate. Biodiesel was again pipetted into
separate tube and washed with water, then separated.
The final result was a translucent, orange biodiesel. The yield of biodiesel was 7 mL,
a conversion of 69 wt%. This conversion rate is very low, as literature shows optimized
conversion from 85 wt% to 100 wt%. One seemingly limiting factor that varied throughout
the literature was the amount of time that esterification or transesterification should occur
for, and should therefore be considered in future experiments. Moderately higher
temperatures, such as 65 °C, may also be experimented with. It was evident that significant
amounts of oil and biodiesel were lost when transferring liquids from one container to the
next, which was not accounted for in the conversion rate.
Post washing, the biodiesel would not burn, which was attributed to the lack of time
or equipment available for drying, however, may also be indicative of incomplete
transesterification and therefore more mixing time may have been required. Final
operation conditions are based on optimized parameters from literature, but the group
suggests further experiments to better determine and familiarize a final design.
3.7 Mushroom production
The prototyping conducted to model industrial mushroom production was done at a
much smaller scale in order to allow for trials of manageable size. The goal of this test was
to verify if Pleurotus ostreatus could efficiently colonize and produce fruits from a media of
spent coffee grounds.
Culture: The mother culture used was taken from Fungi Perfecti, LLC, and
reproduced into four (4) petri dishes in which agar (20 g light malt extract, 2 g yeast, 15-20
g agar, 1 L water) had been poured (Appendix, Figure 1). These were left sealed in the dark
for about a month, after which time the cultures had colonized most of the dishes. The two
strongest cultures (no contamination, presence of “growth rings”, not too mush fluffy white
matter) were chosen to inoculate bags of barley from which grains will be taken to
inoculate the final media mixtures.
3.7.1 Inoculation: The first inoculation was of pearled barley that had been
soaked in water for 24 hours prior to inoculation. After soaking, the barley was dried and
23
sent in the autoclave in autoclave-proof bags that have a 20-micron filter. These bags were
then inoculated with small squares of colonized agar under a fume hood, and subsequently
left to rest for about a week in a warm, dark corner. After that time, colonization has
effectively started and the barley can be shaken every day to facilitate mycelium
propagation (Appendix, Figure 6). Even with all precautionary measures taken, a bag was
contaminated in the process (Appendix, Figure 5).
The second inoculation involved different proportions of different media in order to
test the capacity of Pleurotus mycelium to colonize them (Appendix, Thursday March 8th).
These were:






Bags 1 and 2: 100 % barley with a CaCO3 pH buffer (575 g and 825 g)
Bag 3: 100 % poplar shavings (780 g)
Bag 4: 100 % straw (620 g)
Bag 5: 66 % coffee (480 g) – 33 % poplar shavings (240 g)
Bag 6: 50 % coffee (400 g) – 50 % poplar shavings
Bag 7: 50 % coffee (380 g) – 50 % straw
Not long after inoculation, there was contamination of bags 4 and 7. This contamination
must have occurred during inoculation, as the fume hood is in fact pushing the air down
(positive pressure hood) and therefore can allow for contamination on the arms of the
manipulator to fall into the bag. There was a failed trial at fruiting those contaminated bags,
as the team did not soak the cakes prior to exposure to air and light (Appendix, Figures 15
and 16). The two barley bags were contaminated not long after, this time not with
Trichoderma, Penicillum or Aspergillus (Appendix, Monday, March 19th) but with a sort of
yellowish unknown substance.
3.7.2 Fruiting: A terrarium was built for mushroom production with the three
remaining cakes (bags 3, 5 and 6) (Appendix, Figure 20). After a week, it became evident
that bag 5 would not fruit anytime soon, as mycelium colonization was not advanced
enough. The mistake that was made was not to soak the coffee grounds before sterilization
in the autoclave. There was not enough moisture in the bag as it was incubating, and
therefore most of the colonization happened in the week following the release, which
delayed the operation and took up room in the terrarium. After 5 days, the first mushroom
clusters appeared on bag 3; clusters appeared on bag 6 the following day. After 11 days,
some mushrooms appeared to be of harvestable size, and therefore the experiment proved
24
that indeed, Pleurotus ostreatus mushrooms could grow on a substrate made of 50 % spent
coffee grounds. The fruiting bodies of bag 6 appeared to be a little darker, maybe because
of the coffee. That hypothesis, as well as the hypothesis that the fruiting time is slightly
delayed in a substrate containing spent coffee grounds, remains to be verified. It would also
be necessary to test for mushroom production on a 100 % coffee grounds substrate.
3.7.3 Time requirement: Mushroom growth is a lengthy operation. Two
months passed between the first inoculations of a mother spawn to the first harvest. This
can be broken down in the following (after reaching a successful mother culture):
Table 5: Time required for the different steps in mushroom production
Operation
Time needed (days)
Incubating a mother spawn
15
Sterilization
1
Incubating a mushroom spawn
15
Fruiting the mushroom spawns
15
Harvesting
1
Total
47
5. Results and conclusions (Final design specifications, schematics, etc.)
5.1 Drying
Upon arrival to the site, the wet spent coffee grounds will be taken out of their
boxes/containers, and poured into a shaking funnel comprised of a square-shaped funnel
receptacle under which a cylinder with teeth will be rotating in order to break up potential
conglomerates of spent coffee grounds. A second funnel under the cylinder will channel the
grounds to the conveyor belt dryer below. The heat pump drying mechanism will not be
conventional, as part of the heat and drying will be provided by sun radiation through a
ceiling window on top of the band pass.
According to Greensmith (1998), parameters of the drying air for granular materials
can be approximated as being similar to vegetable and fruit drying air. Four fans will be
positioned along the longitudinal axis of the drying room so as to ensure a constant wind
speed. The air temperature will be maintained between 40 and 60 °C, depending on outside
temperatures and sun conditions. The speed of the conveyor belt needs to be adjustable,
and therefore a 2 hp electrical motor with controllable power will be installed at the end of
the belt.
25
Figure 8: Conventional heat pump drying apparatus (Adapted
from Sosle et al., 2003)
The actual design will be a linear, low-speed adaptation of the above drying apparatus, with
a specific start and end point.
Figure 9: Spent coffee grounds drying apparatus
As can be seen on the figure above, a Hiseer AW13/F-DPNHE Low temperature heat
pump will be used to heat up the drying room (Hiseer, 2012). A funnel will be used as
receptacle for the spent coffee grounds containers being emptied as they arrive into the
facility. These will fall on the 1 m by 20 m conveyor belt gradually, as the conveyor belt
speed will fluctuate around 0.1 m/s and larger lumps of coffee grounds would not allow for
good air circulation. The small orange circles under the top portion of the conveyor belt
represent shaking rollers that help rotating the coffee grounds on themselves so as to allow
for better ventilation and therefore convective drying. The ceiling window allows natural
26
light to come in. Prototyping has shown that at room temperature, the difference in drying
speed varies a lot depending on sun exposure. Fans on the right are in fact air outlets to
create a ventilation effect. Three 12-inch fans along the length of the conveyor belt provide
further air circulation. Dry coffee grounds then fall into a large box at the end of the
conveyor belt and are brought into a larger rotating container (similar to a rotating
composting vessel with vanes on the inside of the cylinder) before brought to the oil
extraction phase.
5.2 Oil extraction
As mentioned previously, the three main steps in solvent oil extraction are the
percolation, desolventizing and distillation steps. The process flows are well summarized in
Figure 10 where the path flows of the oil, solvent and water are outline.
Figure 10: Percolation solvent extraction plant (Adapted from Santori et al., 2012)
As mentioned above, belt extractors provide efficiency by processing large volumes
of medium for solvent extraction on a continuous basis. In this operation, sizing is
performed with the production biodiesel in mind. Extraction will yield 100 kg of coffee oil
daily. Although literature reports yields between 11% and 20%, only 8.9% and 10.8% were
found during the team’s testing. Therefore, an oil yield of 11% will be retained to remain
consistent between literature and the team’s results. From the drying operation, 909 kg of
dried grounds will be taken daily.
5.2.1 Extractor: Since manufacturers were unwilling to share technical details
27
with the team, the design will be based according to the De Smet belt extractor reported by
Almeida et al. (2010). To accommodate the small scale nature of the project, a belt
extractor the third of the belt length is designed for space considerations. The total length
(L) is further reduced to 6.00 m for convenience. The belt width (W) is designed to be 0.45
m, a standard size. The material used is perforated stainless steel. A cloth filter on top of the
stainless steel plates will serve as a filter for the coffee grounds, allowing the miscella to
pass through. The corrected laminator width is retained at 0.33m. Maintaining the optimal
residence time of 30 minutes within extractor as found in literature (Karnofsky, 1986) and
in the prototyping, the belt will travel at a speed of 0.0033 m/s. The extraction rate of the
downscaled model is thus 617.46 kg spent coffee grounds/hour, or 67.92 kg oil/hour in
miscella. Thus, the belt extractor will only need to operate 1.47 hours per day (see
Appendix for calculations).
Solvent will be added in at every meter with in a counter flow system meaning that
the most saturated solvent is sprayed first and the purest solvent is sprayed at the last
station along the conveyor belt to ensure maximal oil uptake. Fluid flow will be equal
among the 4 stations. Solvent is typically added in a 1:1 weight ratio with the coffee.
Therefore each pump will have an operating capacity of 0.262 L/s (see Appendix for
calculations).
5.2.2 Screw conveyor feeder: As mentioned previously special heed needs to
be taken in the design of an airtight extractor since hexane is considered as an air pollutant
and is hazardous to human health. Therefore, an airtight screw conveyor belt will feed the
coffee grounds into the extractor.
Requiring an input of 617.46 kg of spent coffee grounds/hour, which given its
density of 350 kg/m3, is equal to 62.1 ft3/hr. Using Screw Conveyor Corporation’s catalogue
and engineering manual, one can find a required 6 in screw diameter operating at 41.68
rpm to give the desired performance. The operation will require a 2.5 hp motor (see
Appendix).
5.2.3 Desolventizer: once the coffee grounds will have passed through the
extractor, residual solvent will be removed using the desolventizer. Leading manufacturers
such as De Smet and Crown Iron Works Company customize desolventizer-toaster units for
large-scale operations. For smaller operations, Crown Iron Works developed a laboratory
28
Down Draft Desolventizer model remaining unknown though illustrated below:
Figure
11:
Down
Draft
Desolventizer
(Adapted from Crown Iron Works, 2006)
These units are designed to process between 50 and 300 kg/h of raw material with
typical solvent content of 15-30 wt% (POS Bio-Sciences, 2010). The belt extractor having a
throughput of 617.46 kg/h, a 1 tonne storage unit is required after the unit since the rate of
desolventizing is slower. Operating at 100 % capacity, the desolventizer will be slower than
the extractor by a rate of 317.46 kg/h. Operating for 1.47 hours, a 1 tonne storage unit will
provide 466 % storage capacity. However, the desolventizer will have to operate at the
same time as the extractor to prevent overflow.
5.2.4 Distillation: Typical miscella contains 30% oil (Berk, 1992). The
distillation rate is the same as the solvent feed to and from the extractor, namely 0.262 L/s
or 943.2 L/h. Such a rate can be achieved by the use of a Desmet Ballestra Hytech
distillation column.
5.3 Biodiesel production
Reactor sizes were determined based on an input of 100 kg of coffee oil and the
process both outlined in analysis and specifications, and used in prototyping. From
prototyping it is known that coffee oil has a density of 0.89 kg/L. Table 6 presents a list of
densities and molecular weights for various substances used in the calculations. Oleic acid
is taken as a standard fatty acid size to conduct calculations, as it is in the mid-range of fatty
acid molecules present in coffee oil (Martin et al., 2001). It has a molecular weight of 282.5
g/mol (Gerpen et al., 2004).
29
Table 6: Densities and molecular weights for various substances (Gerpen et al., 2004)
Substance
Coffee Oil
Methanol
Glycerol
H2SO4
Density
(kg/L)
0.89
0.7914
1.2613
1.84
Molecular Weight
(g/mol)
885.46
32.04
92.10
98.08
5.3.1 Esterification: Al-Hamamre et al., found that oil extracted from spent
coffee grounds had a FFA% of 3.65 (2012). Using this value, calculations follow
esterification guidelines of 20:1 molar ratio of alcohol to FFAs and H2SO4 quantities of 10%
wt of total FFAs (Santori et al., 2012) (see Appendix for calculations).
5.3.2 Transesterification: Calculations follow transesterification guidelines
based on a 6:1 alcohol to oil molar ratio and 1% wt KOH with respect to oil (Santori et al.,
2012), and assumes that the total volume of oil is recovered from the first stage (see
Appendix for calculations).
Figure 12: Typical design for a well stirred batch reactor (Hatzikioseyian et al., 2007)
The volumes of the reactors need to accommodate splashing during the reaction
process. For calculations, the design team adopted a guideline of 135 % volume, and a
height to diameter ratio of 1.5-1.75 (Kac, 2010). The resulting volumes and dimensions
30
(diameter x height) for esterification and transesterification are: 166.2 L and 188.6 L, and
0.5 m x 0.875 m and 0.52 m x 0.91 m, respectively (see Appendix for calculations).
Assuming about half of the excess methanol ends up in biodiesel, and half in glycerol, a flow
of the facility layout is presented below:
Figure 13: Biodiesel facility layout
Originally, oil contains 10.4 % glycerol, whereas standards for biodiesel allow for 0.24 %
total glycerol. Therefore the transesterification is modeled on a completion rate of:
5.4 Mushroom production
The mushroom production process is quite complex and requires different
components. Acknowledging that the growth and preparation parameters can be adjusted
incrementally, a basic room system is designed.
5.4.1 Spawn preparation room: The mother spawn needs to be prepared
outside the growth room in order not to be affected by the sterilization of the growth room
31
prior to inoculation. That room can be much smaller (the size does not matter and can be as
small as a washroom), must not have windows, and can be normally ventilated and heated.
5.4.2 Growth room: Sterilization, incubation and growth will occur in the
same room, once the spawns are ready. The steam delivery system must be able to provide
for temperatures over 60°C over three hours, as well as lukewarm water vapor to maintain
a good moisture level during the fruiting phase. A sterile ventilation system is also needed
to optimize yield and prevent contamination. The growth system will be one of shelfcultivation, where two levels of mushrooms will be allowed to grow on top of each other,
much like bunk beds for kids.
Figure 14: Spawning of shelfcultivated mushrooms (from
Kang et al., 2004)
5.4.3 Material: Quantities of spent coffee grounds needed in the substrate
mixture will vary depending on yields obtained from different substrate mixtures. Tests
have shown that Oyster mushrooms can grow on substrate containing spent coffee grounds
but no substrate consisted exclusively of spent coffee grounds (Section 4d). Assuming the
latter is viable, we will need a 10 cm depth of it for mycelium to colonize it fully. To get
better access for harvesting, the beds will not exceed one meter in width and the top bed
should not exceed 1.6 m in height. A reasonable design will place the two levels at 0.9 m
and 1.6 m, with a width of 1 m and side panels of 15 cm to prevent the substrate from
falling. The plates will be stainless steel to prevent from rust and mold, and ease of
cleaning. A daily input of 100 kg of spent coffee grounds (dry basis) corresponds to a 700
kg weekly input. A 15 % reduction in weight is expected after oil extraction, which leaves
595 kg of substrate. No such thing as a weekly harvest can be achieved, as it would require
32
sterilizing every week, which would kill the neighboring mushrooms in fruiting mode.
Acknowledging that it takes
⁄
to grow a full crop, a total of
of dry spent coffee grounds will be available for mushroom production. As calculated
experimentally, we know that the density of dry spent coffee grounds is 400 kg/m 3.
Converting the available mass into a volume, we get:
This volume can be used completely with 4 rows of 2 superimposed, 13 m-long, 1 m-wide
shelves filled with 10 cm of dry spent coffee grounds. This can be installed in a 15 m-long,
12.5 m-wide, 2.5 m-high room with 1.5 m corridors in between rows of growing
mushrooms.
Figure 15: Shelf-cultivated Oyster mushrooms (from
Kong, 2004)
33
6. Conclusion
Coffee remains to this day a highly popular beverage consumed around the world.
After brewing the prepared coffee beans, the remains are either thrown out to the garbage
or to the compost heap. However, the nutrient contents of the spent coffee grounds remain
rich particularly in oil and nitrogen. This project consisted of creating a technological
system in which this lost resource is reused for the benefit of society. Thus, the high oil
content is extracted and processed into biodiesel, a valuable combustible, and the
remaining nitrogen and other nutrients are used to grow gourmet Pleurotus oyster
mushrooms.
The team researched the feasibility of this idea by conducting small-scale experiments
to validate findings from literature and customize the processes to meet local conditions.
Drying, oil extraction, biodiesel production and mushroom cultivation all proved to be
feasible, and to some extent, a success.
Next came the design of a larger-scale processing factory that would output 100 kg of
biodiesel per day. Technology is certainly available and only slight modifications need to be
made in order to adapt current systems to the use of spent coffee grounds. A challenge
from the oil extraction and biodiesel production sections was the downscaling from large
industrial settings where several tons of raw material are processed per hour. The team
recognizes that the use of certain systems is not optimal since they would only operate for
a fraction of the day, as is the case with the 1.47 hour daily operation of the oil extractor.
Instead of having a plant where all processes occur simultaneously, perhaps having several
plants specializing in one process and shipping the material from one to the next would be
more favorable. In addition, by increasing the size of individual operations, other markets
could be found for the product. For example, the oil extracted from the spent coffee
grounds was found to be very pleasant to the smell and taste, thus the oil could be used as
specialty cooking oil or perhaps as an ingredient in luxury perfumes.
34
7. Acknowledgements
The team would like to thank:
 PhD Candidate François Gagné-Bourque for his invaluable help for almost all
manipulations involving mushrooms;
 MSc Candidate David Bernard Perron for his technical consulting help in mushroom
production;
 Dr. Michael Ngadi for the use of his lab and assistance in designing a larger oil
extractor;
 Dr. Vijaya Raghavan for the purchase of a Soxhlet extraction tube as well as for using
his lab facilities and chemicals;
 PhD Candidate Jamshid Rahimi for his help with the first oil extraction;
 Dr. Mark Lefsrud for his help with the purchase of n-hexane;
 Dr. Grant Clark for punctual redirection of the focus of the project;
 Mrs. Lise Amarasekera and La Coop du Grand Orme for providing us with free
freshly spent coffee grounds;
 Mr. Darwin Lyew for his assistance in the biodiesel production;
 Mr. Yvan Gariépy for his assistance in the oil extraction;
 Mr. Ebrahim Noroozi for providing us with laboratory equipment;
 Mr. John Watson for providing us with free poplar shavings;
 All our roommates and friends for putting up with our tinkering in the house.
35
8. References
Addison, K. 2010. Make your own biodiesel.
Journey to Forever. Accessed 8 April 2012.
Available at:
http://journeytoforever.org/biodiesel_make.ht
ml
Aguilar-Rivera, N., De Jesus-Merales, J. 2010.
Edible
mushroom
Pleurotus
ostreatus
production on cellulosic biomass of sugar cane.
Sugar Tech 12 (2): 176-178.
Al-Hamamre, Z., Foerster, S., Hartmann, F., Kröger,
M., Kaltschmitt, M. 2012. Oil extracted from
spent coffee grounds as a renewable source for
fatty acid methyl ester manufacturing. Fuel. 96,
70-76.
Almeida, R.L., Ravagnani, M.A.S.S., Modenes, A.N.
2010. Soybean oil extraction in belt extractors
with
miscella
recirculation.
Chemical
Engineering and Processing. 49, 996-1005.
Attokaran, Mathew. 2011. Natural Food Flavors
and Colorants. Wiley Online Library. 429
Barkenbus & Zimmerman, 1927. Kentucky coffee
nut tree seed oil. Journal of American Chemical
Society, 49: 2061-2064.
Bart, J. C. J., Palmeri, N., Cavallaro, S. 2010.
Biodiesel science and technology. Florida:
Woodhead Publishing Limited.
Bernard-Perron, David. Working with Agar. PDF
document shared on December 21st, 2011.
Berk, Z., 1992.Technology of production of edible
flours and protein products from soybeans. FAO
Agricultural Services Bulletin No. 97. Available
at:
http://www.fao.org/docrep/t0532E/t0532E0
0.htm
Blackmer, 2012. Experts in mission-critical flow
solutions. Accessed 16 April, 2012. Available
at: http://www.blackmer.com/default.jsp.
Bimbenet, J.J. & Lebert, A., 1992. Food drying and
quality interactions, in Drying ’92 (ed. By
Mujumdar, A.S.), Elsevier Science, Amsterdam,
42-57.
Brookfield Engineering. 2000. Ground Coffee;
Laboratory Viscometer Application Data Sheet.
Consulted on the 18th of April 2012, available
at:
http://www.brookfieldengineering.com/educa
tion/applications/powder-ground-coffee.asp
Cha, D. Y., J. S. Park, C. H. You, G. P. Kim, C. S. Jeon,
and D. W. Lee. 1998. Oyster mushroom
cultivation technology and management. The
Farmers Newspaper (in Korean).
Crown Iron Work Company. 2006. R&D / Test
Facilities. Available at:
http://www.crowniron.com/userimages/LabR
D.pdf
Gabriel, V. 2004. Mushroom substrate: cereal
straw and corncobs. In Mushroom Grower’s
Handbook. MushWorld.
Gerpen, J.V., Shanks, B. Pruszko, R., Clements, D.,
Knothe, G. 2004. Biodiesel Production
Technology August 2002 to January 2004.
National Renewable Energy Laboratory.
Colorado, US.
Gharehaghaji, A.N., Goltapeh, E.M., Masiha, S.,
Gordan, H.R., 2007. Hybrid production of
oyster mushroom Pleurotus ostreatus (Jacq:
Fries) Kummer. Pakistan Journal of Biological
Sciences. 10: 14. 2334-2340.
Greensmith, M. 1998. Practical Dehydration. Boca
Raton, CLC Press.
Hatzikioseyian, A., Remoundaki, E., Tsezos, M.
2007. Typical proportions of a well stirred
batch reactor. Accessed 10 April 2012.
Available at:
http://wiki.biomine.skelleftea.se/wiki/index.p
hp/Well-stirred_batch_reactors.
Hawlader, M.N.A., Perera, C.O., Tian, M., Yeo, K.L.
2006. Drying of guava and papaya: Impact of
drying methods, Dry. Technol. 24, 77-87.
Hiseer, 2012. Accessed on April 1st, 2012. Small
Split low temperature air heat pump high
efficiency. Available online at:
http://www.chineseheatpump.com/chinasmall_split_low_temperature_air_heat_pump_hi
gh_efficiency-169875.html
Kac, A. 2010. The foolproof way to make biodiesel.
Journey to Forever. Accessed 8 April 2012.
Available at:
http://journeytoforever.org/biodiesel_aleksne
w.html.
Kang, S. W., Kwon, H., Kim, B. S., 2004. Illustrated
guide to Oyster mushroom cultivation. In
Mushroom Grower’s Handbook. MushWorld.
Kapsalis, J.G., 1987. Influence of hysteresis and
temperature on moisture sorption isotherms, in
Water activity theory and applications to food
(ed. By Rockland, L.B. and Beuchat, L.R.),
Marcel Dekker, New York, 173-207
36
Knoth, G., Gerpen, J. V., Krahl, J. 2005. The Biodiesel
Handbook. Illinois: AOCS Press.
Kong, W. S., 2004. Description of commerciallyimportant Pleurotus species. In Mushroom
Grower’s Handbook. MushWorld.
Kiranoudis, C.T., Maroulis, E., Tsami, E., MarinosKouris, D., 1993. Equilibrium moisture content
and heat of desorption of some vegetables.
Journal of Food Engineering, 20: 55-74.
Kondamudi, N., Mohapatra, S. K., Misra, M. 2008.
Spent coffee grounds as a versatile source of
green energy. Journal of Agricultural and Food
Chemistry. 56, 11757–11760
Kooijman, H., Taylor, R. 2000. The ChemSep Book,
2nd Edition. Accessed on April 8th, 2012.
Available online at:
http://www.chemsep.com/downloads/docs/b
ook2.pdf
Lingle, T. R. 1995. The Coffee Brewing Handbook.
Long Beach, CA, Specialty Coffee Association of
America.
Martin, M. J., Pablos, F., Gonzalez, A. G.,
Valdenebro, M. S., Leon-Camacho, M. 2001.
Fatty acid profiles as discriminant parameters
for coffee varieties differentiation. Talanta. 54,
291-297.
McDonough, W. and Braungart, M., 2002. Cradleto-Cradle. New York, North Point Press.
Misra, M., Kondamudi, N., Mohapatra, S.K., John,
S.E. 2008. High quality biodiesel from spent
coffee grounds. Clean Technology 2008, 39-42.
Molloy, F. J., Critchley, A. T., Kandjengo, L.,
Mshigeni, K. E., 2003. Ambio, Vol. 32, No. 1 pp.
76-78.
Petracco, M. 2005. Our everyday cup of coffee: the
chemistry behind its magic. Chemistry for
Everyone. Accessed 10 April 2012. Available
at:
http://pubs.acs.org/doi/pdfplus/10.1021/ed0
82p1161.
POS Bio-Sciences, 2010. Plant and Equipment.
Available at:
http://www.rozel38.com/pos/assets/uploads
/pdfPlant__Equipment.pdf
Rashid, U. & Anwar, F. 2008. Production of
biodiesel through optimized alkaline catalyzed
transesterification of rapeseed oil. Fuel. 87, 265273.
Ratti, C. 2009. Advances in food dehydration. Boca
Raton, CRC Press.
Read, J. 2011. Let’s Grow Mushrooms. Videos
accessed online on February 13, 2012.
Available at:
http://www.mushroomvideos.com/
Rizvi, S.S.H., 2005. Thermodynamic properties of
foods in dehydration. In Engineering properties
of foods. 3rd ed. CRC Press, Boca Raton, 127159.
Screw Conveyor Corporation. 2010. Screw
Conveyor Catalogue and Engineering Manual.
Available at:
http://www.screwconveyor.com/SCC%20Eng
Cat10_LR.pdf
Santori, G., Nicola, G. D., Moglie, M., Polonara, F.
2012. A review analyzing the industrial
biodiesel production practice starting from
vegetable oil refining. Applied Energy, 92, 109132.
Sosle, V. Raghavan, G.S.V., Kittler, R., 2003. Lowtemperature drying using a versatile heat pump
dehumidifier. Drying Technology, 21, 539-554.
Stamets, P., Chilton, J. S., 1983. The Mushroom
Cultivator: A Practical Guide to Growing
Mushrooms at Home. Agaricon Press, 415 p.
Stamets, P. 1993. Growing gourmet and medicinal
mushrooms. Random House Digital, Inc. 574 p.
Wan, P.J., 1997. Hydrocarbon Solvents. In
Technology and solvents for extracting
oilseeds and nonpetroleum oils. AOCS Press,
Champaign, IL, 361 p.
37
9. Appendices
Appendix I: Two first oil extraction prototypes
Trial 1: The first prototyping for oil extraction began in Dr. Ngadi’s lab using a Soxhtet
extractor Velp Scientifica model SER148. For the extraction, coffee grounds are placed inside of a
thimble made from thick filter paper, which is then positioned inside of the main chamber of the
machine. The Soxhtet extractor is then placed onto a flask that contains the extraction solvent
(petroleum ether), and subsequently fitted with a condenser.
The solvent is heated to reflux, and the vapour travels up through a distillation arm,
flooding into the chamber housing where the coffee grounds are located. Meanwhile, the condenser
ensures that solvent vapours cool and return to the chamber housing the coffee grounds. Heating
the solvent during the first phase serves mostly as a pump in circulating the solvent trough the
medium at a higher temperature and maintaining its purity. At 60 °C, the oil is more soluble in the
solvent.
Meanwhile, the chamber of coffee grounds slowly fills with hexane. Some oil thus dissolves
in the hexane, and when the Soxhlet chamber is full, it automatically empties by a siphon side arm
with the solvent returning to the distillation flask. After the solvent has circulated several times
through the medium, the solvent is then boiled off to keep only the oil in the bottom flask, yielding a
pure, particle-free product. Extraction rate was 10.8 %, compared to around 15 % in the literature.
Trial 2: The second prototyping for oil extraction involved a paint shaker for solventgrounds mixing and consequent oil extraction. This was carried out at The Hub hardware store in
Beaconsfield and in Dr. Lefsrud’s lab. The hexane mixture used was from Fisher Scientific: UN1208,
H306-4 at a concentration of 95% hexane.
Container 1
Container 2
Batch Extraction Specifications
o 520.70 g coffee grounds
o 800 ml (509.25 g) hexane
o 375.31 g coffee grounds
o 500 ml (320.05 g) hexane
Each container was mixed for a period of approximately 5 minutes. Post-mixing, the
expected oil film was not found on top of the hexane. Also, the attempt at extracting the liquid
portion of the can contents failed with the pipette as the contents of the pipette would leak, causing
mixing. Extraction proceeded by pouring the liquid contents into a 2L round bottom flask, setting
up a Liebig condenser and recuperating the hexane. Evaporation of hexane was not carried out due
to precaution; we did not wish to create a cake of oil and coffee grounds in the bottom of the round
bottom flask, which was foreseen as cleaning nightmare. A thimble should have been used to keep
the round bottom flask clean. The recuperation of hexane was already better in this system when
compared to Dr. Ngadi’s Soxhlet extractor. We achieved a recuperation rate of 45 %, despite not
having boiled off all the hexane in the flask and having left significant quantities in the tin can.
The idea of separating the oil from the coffee grounds by letting the hexane settle did not
function since, unlike oil and water, the oil is dissolved in the solvent, therefore no separation
occurs.
38
Appendix II
Extractor Design:
Operation Time:
0.45 m
0.33 m
Conveyor Feed Cross-Section
Conveyor feed cross-sectional area: a = 0.1485 m2
Conveyor speed: v = 0.0033 m/s
Coffee density: ρc = 350 kg/m3 (Brookefield Engineering, 2010)
Coffee mass flow:
̇
/s or 617 kg/hr
Processing quantity: ̇ = 909 kg/day
Daily processing time: ̇ ̇ = 1.47 hrs/day
Solvent Flow:
1:1 weight ratio solvent:medium
̇ 0.1715 kg/s
Liquid hexane density: ρh = 657 kg/m3 (engineeringtoolbox.com)
Volumetric flow: ̇ / ρh= 0.000262 m3/s or 0.262 L/s hexane
Conveyor design (adapted from Screw Conveyor Corporation, 2010):
Volumetric flow: ̇ /ρc = 1.76m3/hr
1.76 m3/hr*35.288 ft3/m3 = 62.1 ft3/hr
Material classification:
Coffee grounds = 350 kg/m3
/
kg/m3 = 21.85 lb/ft3
*similar to cottonseed flakes according to ρ (20-25 lb/ft3) in Material Table 6
Compound group: 1A-1B
HP Material Factor F: 0.8
Material class: 23CI/235HWY
From Table 7 with 30 % trough loading, non-abrasive materials, a 6 inch diameter screw will be
180 ft3/hr max and 1.49 ft3/hr at 1 RPM.
The screw has a max RPM of 120
Operational RPM = 62.1 ft3/hr / 1.49 ft3/hr at 1 RPM = 41.68 RPM
Power Requirements:
From Table 2 Horsepower Rating:
6 inch conveyor, 2 inch pipe, 1 ½ inch coupling
MAX HP at 50 RPM = 2.5 HP
39
Appendix IV
Calculations for esterification reactant volumes:
Calculations for transesterification reactant volumes:
Calculations for volume and dimensions of cylindrical reactor vessels:
100 kg oil + 21.7 kg methanol + 0.1 kg KOH
112.3 L oil + 27.4 L methanol + 0.1 kg KOH
100.45 kg biodiesel + 10.4 kg glycerol + 10.86 kg methanol
114.9 L biodiesel + 8.24 L glycerol + 13.7 L methanol
Esterification (1.75h:d):
( )
Transesterifcation (1.75h:d):
( )
40
Glycerol Neutralization (1.5h:d):
( )
Centrifuge (larger volume to account for particle separation 1.5h:d):
( )
Methanol Distillation (Glycerol 1.75h:d):
( )
Methanol Distillation (Biodiesel 1.75h:d):
( )
Washing (1.75h:d):
( )
Drying (1.75h:d):
( )
41
Appendix IV: Project journal
Week 0
Summary:
 Set up meeting with David Bernard-Perron for Pleurotus propagation
Tuesday, January 10th
 First day of the Design III course
 Made an appointment with David Bernard-Perron for spore propagation
Week 1
Summary:
 First petri dishes of Pleurotus made with D. Perron
 First team meeting with division of tasks for next days
 Get materials ready for biodiesel production
 Find sources of funding for project
Wednesday, January 11th
 Met up with D. Perron and learned all the steps to successfully propagate Pleurotus
mycelium. These steps included preparation of agar, sterilization using autoclave hardware,
proper specimen handling in vacuum hood for minimal contamination. 5 petri dishes were
prepared with the fungal source.
Tuesday, January 17th
 First team meeting held, as all members were finally back in town.
 This week, we are going to get all the materials ready for the biodiesel production. This will
be in tight collaboration with Dr. Lefsrud as he has the supplies and the equipment for this
procedure.
 Coffee grounds will be collected from the different cafes in town
 A list of possible sponsors/donators/funds will be made and applied to in order to mitigate
the costs associated to the project. Possible sources are
o McGill Sustainability Fund
o BREE department
o Faculty of Ag. Env. Sciences
o ASABE/ CSABE scholarships
o Etc.
Week 2
Summary:
 Mycelium were moved to room temperature, started to grow
 Coffee grounds collected from Twigs
 Preparations for oil extraction with Dr. Lefsrud furthered
42
 Meeting with Dr. Clark held
 Submission of funding request with BEA
Wednesday, January 18th
 Petri dishes were taken out of the fridge and stored in a dark box at room temperature.
Mycelia started growing 2-3 days afterwards and are now successfully increasing in size.
Figure 1: Pleurotus ostreatus culture growing on agar
Monday, January 23rd
 Meeting held with Dr. Lefsrud about price of hexane and soxhlet extraction apparatus.
40$/L of hexane.
 Dr. Ngadi apparently has a Soxhlet extractor
 First draft of budget made for BEA
Tuesday, January 24th
 Team meeting held before class to agree upon discussion topics with prof. Clark.
o Entire system design vs. component design
o What is bioengineering? What focus should be made on biology vs technology?
o Dry run of all the steps?
o Awards that were cancelled
 Discussion with Dr. Clark yielded a legitimacy of performing a dry run of all the steps
involved in our design. Once dry run has been performed, bottlenecks can be identified and
further design and optimization can be targeted.
 Design entire process with CAD drawings and further technical details.
Week 3
Summary:
 Propagate subculture
 Coffee grounds collected from Twigs
 Preparations for oil extraction with Dr. Lefsrud furthered
 Preparation of Gantt chart timeline
43
 Submission of funding request with BEA
Tuesday, January 31st
 Completed Gantt chart (Thomas and Violette).
 Apply to BEA for fundraising (Thomas).
 Divide writing of preliminary report (Sandra and Pascal).
 Calculate quantities of hexane and talk to Mark Lefsrud about purchasing (Pascal).
 Contact Ngadi for Soxhlet (Thomas).
 Divide mycelium subculture into further subcultures for “safe supply” (Sandra).
We plan to procure hexane by the end of the week and begin working on oil extraction as well as
mushroom propagation by weekend/early next week. We will also be looking into further funding
possibilities for hexane as we have noticed this part of the experiment will be quite costly.
Week 4
Summary:
 Inoculation of 2 barley bags
 Coffee grounds collection from the Coop, drying (continuous)
 Preparations for oil extraction with Dr. Lefsrud and Dr. Ngadi furthered
 More precise timeline for biodiesel and mushrooms assigned, along with responsibilities
 Final report outline
 Preliminary report started
Thursday, February 2, 2012
 Preliminary report outline
 Meeting with Dr. Lefsrud and Dr. Ngadi, some academic talk needs to be done in order to
have access to the extraction equipment in one of the BREE laboratories.
Friday, February 3, 2012
 Acquisition of autoclave bags
 Meeting with François Gagné-Bourque (Plant Science PhD candidate) to learn about
autoclave and propagation techniques
 Soaking of barley grounds
 Precise timeline written, with tentative responsibility allocation for team members
Monday, February 6, 2012
 Autoclaving and inoculation of the barley
o 1200 g were soaked in order to activate all fungi and bacteria in the barley, to better
kill everything when autoclaving
o 2 bags of 600 g were autoclaved and inoculated
o Storing in a warm, dark place
o Expecting full colonization in 2-3 weeks
Tuesday, February 7, 2012
 Drying of new coffee grounds from the Coop du Grand Orme
 Meeting with the team to talk about the preliminary report
44

Setting up of a time to undergo a Soxhlet extraction on Thursday, 8 AM, in Dr. Ngadi’s lab.
The supervision of a graduate student will be provided.
Week 5
Summary:
 Mycelium is slowly colonizing the barley
 Coffee grounds collection and drying
 First oil extraction conducted, analysis is undergoing
 Preliminary report written
Thursday, February 9, 2012
 First oil extraction conducted in Dr. Ngadi’s lab.
o Graduate student Gamshi helped us greatly with the Soxtet machine and extraction
process details, as well as provided us with some of the materials
o Quantities to be processed at one time are very limited, mostly for research – not
industrial scale at all
o Extraction rate of 10 % (compared to around 15 % in the literature)
o The team is currently analyzing different possibilities for moving forward
Figure 2: Dr. Ngadi’s Soxhtet extraction machine
Figure 3: Thomas A. pouring coffee oil
45
Figure 4 : Spent coffee grounds from which coffee has been extracted (right) vs. normal spent coffee
grounds (left)
Friday, February 10, 2012
 Preliminary report finished and handed in
 Purchase of CaCO3, an additive (pH buffer) to be mixed with the barley for our next Oyster
colony.
Tuesday, February 14, 2012
 Team meeting
o Discussion about format of our deliverable project and technological aspect
 Factory type
 We will start the report with a datum of 100 L biodiesel required, and work
our way back from there to see what real design constraints come in
between us and success.
Week 6
Summary:
 Mycelium is slowly colonizing the barley
 Second set of bags of barley started
Thursday, February 16, 2012
 Soaking of over 1 kg of barley with CaCO3. Subsequent partial drying to bring back to
saturated state.
Friday, February 17, 2012
 Reproduction of Oyster subculture in 4 new petri dishes to serve as a backup in case
something happens.
 Autoclaving of two bags of barley (soaked with CaCO3) and subsequent cooling and
inoculation.
o A thousand thank you’s to Francois Gagne Bourque for his invaluable help.
 Dr. Ngadi on oil: A fatty acid profile is not necessary, especially since these values can be
found in the literature. One thing that could be interesting is to compare the fatty acid
profile of spent coffee oil with something like olive oil to see if we could cook with it. Also
boiling temp., etc. are interesting parameters to check.
46

Dr. Lefsrud on extraction: He says the way he does extraction is by putting the material with
the solvent in the same container and shaking it for 2 hours. No need to heat up. Then let
settle and skim off the oil layer at the top.
o I was looking at paint shakers on the internet and they're pretty expensive. We
could try doing it by hand one time and see if it works well... If so then we can try a
paint shaker at a store, say we pay them to use their shaker for a while. Or if not we
can design/prescribe a paint shaker or container shaker for our final ''factory"
design.
Week 7
Summary:
 Mycelium is slowly colonizing the barley
 Discussions around narrowing the scope of our project
Wednesday, February 22
 Started the final report with a ‘backwards’ approach: what do we need to produce 100 L of
biodiesel? Then work from there.
Sunday, February 26
 One of the last two bags of barley was found to be contaminated. On the picture below, it is
easy to distinguish the oyster (white) mycelium from the contaminated (blue) mycelium.
Figure 5: Contaminated barley spawn

Two of the other three bags are very strong. Soon we will be able to inoculate a mix of spent
coffee grounds. The hope is that by that time we will have enough coffee grounds from
which we have extracted the spent coffee grounds to be able to compare the growth of
mycelium on the different types of grounds.
Figure 6: Healthy colonization of the barley by Pleurotus mycelium
47
Tuesday, February 28
 Team meeting: assigned responsibilities.
o Go to a hardware store to see if they have empty paint containers we could get to try
to extract the oil with cold n-hexane by shaking it up. (V + P)
o Ensure we have all the necessary materials for conversion of oil into diesel (T)
o Keep going with the final report – structure (P)
o Email Dr. Jabaji to let her know where we are at and ask for advice or insight (P)
o Start drawing the factory layout (S)
o Write a letter to the department head and the dean to ask for funds with a budget
(S)
Week 8
Summary:
 First larger-volume extraction operated
 Barley is fully colonized – ready for inoculation into coffee grounds
 Coffee grounds mixture trials before inoculation
Tuesday, February 28
 Went to hardware store in Beaurepaire to buy clean paint cans and ask for permission to
use their paint shaker for solvent-grounds mixing and consequent oil extraction.
Wednesday, February 29
 Carried out the mixing and settling of the coffee ground and hexane mixture in Dr. Lefsrud’s
Lab in the tin cans. Many thanks to Yvan for his assistance. The hexane mixture used was
from Fisher Scientific: UN1208, H306-4 at a concentration of 95% hexane.

Container 1
o
o
Container 2
o
o
520.70 g coffee grounds
800 ml (509.25 g)
hexane
375.31 g coffee grounds
500 ml (320.05 g)
hexane
Many thanks to The Hub hardware store of Beaconsfield for allowing us to use their paint
shakers to thoroughly mix the contents of the containers. Each container was mixed for a
period of approximately 5 minutes.
Figure 7: Shaking the coffee – hexane mixture in a paint shaker
Friday, March 2nd
 Assessment of our situation vs. the timeline of the semester:
48
o
o
o
Mushroom fruiting will be approximately 2 weeks late, as it took longer than
expected to colonize the barley
Oil extraction is on track
We are very late for the design/report section of our project!
Week 9
Summary:
 First larger-volume extraction assessed: some difficulties,
 Inoculation of different coffee grounds mixtures
Tuesday, March 6th
 Meeting with Dr. Clark
o Discussions about our progresses
o How will we ‘show’ that we have been working consistently on the project? It needs
to be on the final report, as this is how our work will be assessed.
o Some drawings will need to be included in the report: probably the extraction
apparatus and a layout of the factory floor.
Wednesday, March 7th
 Multiple attempts at taking care of the contents of the paint cans (coffee grounds + hexane).
o The expected oil film was not found on top of the hexane. The attempt at extracting
the liquid portion of the can contents failed with the pipette as the contents of the
pipette would leak, causing mixing.
o Extraction proceeded by pouring the liquid contents into a 2L round bottom flask,
setting up a Liebig condenser and recuperating the hexane.
o Evaporation of hexane was not completed out of precaution; we did not wish to
create a cake of oil and coffee grounds in the bottom of the round bottom flask,
which was foreseen as cleaning nightmare. A thimble should have been used to keep
the round bottom flask clean.
o The recuperation of hexane was already better in this system when compared to Dr.
Ngadi’s soxhlet extractor. We achieved a recuperation rate of 45 %, despite not
having boiled off all the hexane in the flask and having left significant quantities in
the tin can.
 Many thanks go to Ebrahim Norooz of the undergraduate food science laboratory for
supplying us with the glassware needed to perform the distillation. Many thanks also to Dr.
Orsat and Dr. Raghavan for providing us with the laboratory flow hood needed to perform
the distillation.
Figure 8: Remnants of hexane, oil and coffee grounds in the round bottom flask
49
Thursday, March 8th
 Collection of autoclave-able bags from François, straw from David Bernard-Perron, and
poplar shavings from John Watson at the Morgan arboretum.
 Autoclaving of all substrate contents, as well as new barley for multiplication of our spawn.
 Inoculation of the following (with the help of François Gagné-Bourque):
o 100 % barley (2 bags, 575 g and 825 g)
o 100 % poplar shavings (780 g) (1 bag)
o 100 % straw (620 g) (1 bag)
o 66 % coffee (480 g) – 33 % poplar shavings (240 g) (1 bag)
o 50 % coffee (400 g) – 50 % poplar shavings (1 bag)
o 50 % coffee (380 g) – 50 % straw (1 bag)
Figure 9: Source bags
Figure 11: Chopped straw
Figure 10: Poplar shavings
Figure 12: Before autoclaving
Figure 13: Inoculation
50

It appears like the assessment tool we will use to measure the efficiency of our different
growing media is the concept of biological efficiency (mushroom fresh weight over
substrate dry weight)
Friday, March 9th
 Outline of final report made according to the following criteria
/15 Analysis and specification: Rigorous use of mathematics and scientific principles to develop the final design of the
chosen solution.
/15 Prototyping: Numerical or physical modeling and simulation of the proposed design
/15 Revision: Students redeveloped and improved the design based on the experience with numerical or physical
prototyping
/15 Testing: rigorous testing and consideration of failure modes, risk analysis, economics, etc.
/15 Professionalism: The project was executed with diligence and professionalism, according to the stated objectives
and methods.
Week 10
Summary:
 Fundamental understanding of oil extraction by solvent achieved with Dr. Ngadi
 Purchase of a soxhlet extractor by Dr. Raghavan’s lab
 Work on the extractor design and final report.
 Research post-oil extraction biodiesel production
Wednesday, March 16th
 While wanting to perform another extraction of oil, we were confronted with the
shortcomings of our current design and wished to find a proper soxhlet tube. We are
thankful to Dr. Ngadi for the time he spent in helping us understand the fundamentals of oil
extraction by means of a solvent.
 Heating the solvent during the first phase serves mostly as a pump in circulating the solvent
trough the medium at a higher temperature and maintaining its purity. At 60 °C, the oil is
more soluble in the solvent. This also explains why the idea of separating the oil from the
coffee grounds by letting the hexane settle did not function since, unlike oil and water, the
oil is dissolved in the solvent, therefore no separation occurs. After the solvent has
circulated several times through the medium, the solvent is then boiled off to keep only the
oil in the bottom flask, yielding a pure, particle-free product.
 With this understanding, we ended up receiving assistance from Dr. Raghavan’s lab in the
purchasing of a soxhlet extractor tube as it already had a proper condenser. The previously
used condenser from Mr. Norooz was unfitting due to the narrowness of its joint. We are
very grateful to Dr. Raghavan in assisting us with financial support and prompt execution.
ETA Tuesday 20.03.2012.
Thursday, March 15th
 Organized collection of spent coffee grounds from engineering café.
 Researched and documented methods for titration and the production of biodiesel.
Monday, March 19th
 Most inoculated bags are almost fully colonized
51
Figure 14: Colonization of coffee grounds media

Two bags were contaminated with Trichoderma, Penicillum or Aspergillus species.
Figure 15: Contaminated bags

Those bags were suspended in indirect light (room facing north) to test if they can still fruit
even with a contaminated part.
Figure 16: Contaminated bags put to fruit
Week 11
Summary:
 Collection of the last coffee grounds needed for the experiment; drying.
 Preparatation for placing of mushrooms into fruit mode (non-contaminated bags)
52
Friday, March 23rd
 The mushrooms put to fruition were drying faster than expected. We will need to vaporize
them with water twice every day, according to mushroom videos.
 Purchase of vermiculite
 Work on the report
 Bags of contaminated mycelium are not fruiting. Disposal.
o Probable cause is no soaking before release; the mycelium did not have enough
water.
Monday, March 26rd
 Call David and watch videos for practical tips on mushroom growing:
o Soak content of bags in previously boiled water for one full day before releasing.
o Use transparent plastic container, drill holes through it, fill bottom with perlite (2
in).
o Drain mycelium “breads” and roll in vermiculite (maintain moisture)
o Put in the container, close lid, vaporize with water as often as 3-4 times a day.
o Container needs to be put in daylight enough to fruit, however not so strong as to
dry the content too quickly.
 Containers of coffee taken from downtown need to be dried as fast as possible otherwise
will get contaminated, as in picture below:
Figure 17: Contaminated spent coffee grounds
Tuesday, March 27rd
 Purchase of two transparent plastic bins for mushroom fruiting
 Purchase of perlite to put at bottom of bins
 Research and design of a drying system to put at the entrance of the factory
 General coordination for the write-up of the report
53
Week 12
Summary:
 Drying of the last coffee grounds collected, measurements taken for moisture content
calculation
 Placement of mushrooms into fruit mode in a terrarium
 Soxhlet extractions conducted
 Drying apparatus specifications found
 Biodiesel production specifications found
Wednesday, March 28th
 Drilling holes in the terrarium, preparation of perlite and vermiculite
 Soaking of the coffee-poplar cakes in preparation for fruiting
 The contaminated mushroom bags were thrown out as they were not fruiting and suspected
to be contaminating our growth chamber with bad spores.
 Continuous, passive drying of the remaining coffee ground in aluminum plates, by exposure
to sun and dry environment.
Figure 18: Pascal removing the coffee filters before drying
Figure 19: Spent coffee grounds in drying mode
54
Thursday, March 29th
 Placement of the mushroom cakes in the terrarium, humidification and put in the light.
 Set up of the Soxhlet extraction apparatus with Yvan Gariépy.
Figure 20: The terrarium
Friday, March 30th
 Soxhlet extraction conducted in Dr. Raghavan’s lab. Early morning for fume hood
availability.
 Drying apparatus specifications finished
Figure 21: The Soxhlet extraction apparatus
Saturday, March 31st
 Other Soxhlet extractions conducted in Dr. Raghavan’s lab. We need a reasonable amount of
oil to be able to go through biodiesel production.
Sunday, April 1st
 Analysis and specifications worked out for the biodiesel production section. We hope to be
able to try it at least once before the end of the semester.
 Work done on the introduction of the report.
 Timeline assessment: everything is on time, except for the biodiesel reaction and the final
report writing. It seems like a lot of focus was put on the practical work side, and now the
focus has to shift into the specifications and design side of things.
55
Monday, April 2nd
 The mushrooms are fruiting! It took 5 days. Many, many clusters of around 20 mushrooms
each are appearing. We are expecting some of these clusters to abort, and some mushrooms
to take the lead in the remaining clusters.
Figure 22: First fruit clusters (see middle of image)
Tuesday, April 3rd
 Biodiesel production through a two-step reaction in the Waste Management lab (Dr.
Barrington’s old lab).
Figure 23: Sandra R. waiting for the oil to react with the methanol and sulphuric acid
Week 13
Summary:
 Mushrooms are still fruiting
 Conversion of coffee oil into biodiesel
 Preparation of final oral presentation
 Work on the final report
Wednesday, April 4th
 Second of the two steps for conversion of oil into biodiesel carried.
56
Figure 24: Biodiesel obtained through the two-step method
Friday, April 6th
 Removal of one of the three mushroom cakes, as it was not fruiting yet and the other two
needed space.
o The problem is that the coffee grounds had not been wet before autoclaving, and
therefore the mycelium had trouble colonizing it all. It started colonizing more in
the terrarium, but for the sake of space we removed it.
Figure 23: Inside the terrarium after 11 days
Sunday, April 8th
 Work on the report.
 Washing of the biodiesel to purify the product
Monday, April 9th
 Final adjustments made to the presentation
 Rehearsal
Tuesday, April 10th
 Final presentation made in front of the class
57
Week 14
Summary:
 Discussions pertaining to the report
 Work on the report
Wednesday, April 11th to Tuesday, April 17th
 Work on the final report
Week 15
Summary:
 Discussions pertaining to the report
 Work on the report
Wednesday, April 18th to Monday, April 23rd
 Work on the final report
58