Bio-Ethanol Productions Plant in Ontario

DESIGN OF MULTI-FEEDSTOCK BIO-ETHANOL
PLANT IN ONTARIO
Winter Term 2015

Department of Chemical Engineering
McMaster University
By
Team:
Vytautas Stasiulevicius, Fahd Ilyas, Carlo Bantug, Danish Fahzal, Leo (Xiau) Zhou
A Project Report
CHE 4W4 Chemical Plant Design and Simulation
4W4 2015 Syngas Fermentation
Executive Summary
This document explores the feasibility of building a multi-feedstock biofuels production
plant in Ontario to produce Ethanol. A basis of producing 100 million litres of Ethanol was
selected for designing and costing the production plant, constrained mainly to meeting the quality
standards outlined in the Canadian General Standards Board. The production of Ethanol through
a thermochemical pathway and a biochemical pathway were studied as technological alternatives.
The thermochemical pathway proceeds via Gasification followed by Fischer-Tropsch synthesis
but was found to be very energy intensive and required a specific gas composition for production.
The alternative, biochemical pathway involves enzymatic hydrolysis followed by fermentation,
but high enzyme costs, cost-intensive pre-treatment, and low feedstock flexibility deterred the
selection of this process.
Instead, a hybrid production pathway was selected, referred to as syngas fermentation,
which combines aspects from both the thermochemical and biochemical processes. Compared to
the alternatives, the hybrid process was selected mainly due to high feedstock and gas
composition flexibilities, allowing nearly any lignocellulosic material to be converted into
Ethanol. Syngas fermentation was also advantageous over alternative processes due to high
Ethanol yield, selectivity, and high resistance to contaminants. The feedstock of interest is first
crushed and dried before it is sent to a fluidized bed gasifier to produce a syngas mixture. The
syngas mixture goes through various cleaning and cooling stages to remove impurities before
being fed into a fermenter containing a specific bacteria (clostridium Ijungdahlii) acting as a
biocatalyst. The bacteria continuously converts syngas to ethanol within the fermenter, while the
broth is continuously extracted and sent to distillation columns to separate out the desired Ethanol
to be used for fuel.
The proposed plant would require an approximate $90 million investment for capital
costs to establish the plant infrastructure, and would cost roughly $110 million per year to operate
and maintain. Though due to high feedstock costs incurred from growing the feedstock the plant
would run a net negative NPV over a 25 year project lifetime unless government subsidies were
provided on the price of ethanol. The overall production process produces approximately 2 kg of
CO2 equivalent emissions, comparable to ~20 kg for crude oil processes. Though it must be
noted that the final Ethanol will likely be blended with gasoline, so final emission reductions will
be in the order of 6-8%, which is a great improvement.
Table of Contents
Executive Summary ........................................................................................................... 2
Table of Contents ............................................................................................................... 3
Table of Figures ............................................................................................................. 5
Table of Tables .............................................................................................................. 6
Table of Reacions ........................................................................................................... 6
Table of Equations.......................................................................................................... 6
1. Project basis................................................................................................................... 7
1.1 Summary ................................................................................................................. 7
1.2 Economic Approximation of Process ........................................................................... 8
1.3 Relevant standards .................................................................................................... 9
2. Existing solutions ......................................................................................................... 10
2.1 1st generation feedstock ........................................................................................... 10
2.2 2nd generation feedstock .......................................................................................... 11
3. Design alternatives ....................................................................................................... 14
3.1 Hybrid process ........................................................................................................ 15
3.2 Biochemical process ................................................................................................ 16
3.3 Thermochemical process .......................................................................................... 18
3.4 Bioethanol Location ................................................................................................ 20
4. Overview of Proposed Process Design ............................................................................ 22
4.1 Process Summary .................................................................................................... 22
4.1.1 Pre-Treatment ................................................................................................... 23
4.1.2 Gasification ...................................................................................................... 24
4.1.3 Gas Cleaning .................................................................................................... 26
4.1.4 Fermentation .................................................................................................... 29
4.1.5 Distillation ....................................................................................................... 32
4.2 Design Basis........................................................................................................... 33
4.3 Product Specifications ............................................................................................. 33
5. Process Behaviour ........................................................................................................ 33
5.1 Normal operation .................................................................................................... 33
5.2 Start-up and shutdown ............................................................................................. 36
3
5.2.1 Start-up ............................................................................................................ 36

5.2.2 Shutdown ......................................................................................................... 38
5.3 Emergency Procedures............................................................................................. 39
6. Overall Material and Energy Balances ............................................................................ 40
6.1 Overall Material Balance:......................................................................................... 40
6.1.1 Process Side ..................................................................................................... 41
6.1.2 Utilities side ................................................................................................... 41
6.2 Overall Energy Balance ........................................................................................... 43
6.2.1 Process Side ..................................................................................................... 43
6.2.2 Utility Side ....................................................................................................... 43
6.3 Stream and Equipment tables .................................................................................... 45
6.3.1 Process Side ..................................................................................................... 45
6.3.2 Utilities Side .................................................................................................... 55
7. Process Control ............................................................................................................ 58
7.1 Control Overview.................................................................................................... 58
7.2 Preliminary P&ID of Process .................................................................................... 73
8. Equipment design, sizing and costing process side ......................................................... 76
8.1 Costing overview .................................................................................................... 76
8.2 Capital costs ........................................................................................................... 77
8.3 Operating costs ....................................................................................................... 82
8.4 NPV ...................................................................................................................... 86
8.5 Sensitivity Analysis ................................................................................................. 88
8.6 Equipment Sizing .................................................................................................... 91
Heat exchanger design ............................................................................................... 93
9. Environmental Impact ................................................................................................... 95
9.1 LCA ...................................................................................................................... 95
9.2 GHG Emissions ...................................................................................................... 97
10. Process safety........................................................................................................... 100
10.1 Hazardous Materials ............................................................................................ 100
10.2 Process Hazards .................................................................................................. 103
11. Risk Assessment ....................................................................................................... 104
4
11. 1 Technical........................................................................................................... 104

11.2 Societal .............................................................................................................. 105
11.3 Economical......................................................................................................... 106
Appendices ................................................................................................................... 108
Appendix 1 Various Lists Relating to Process ................................................................. 108
List of Materials ......................................................................................................... 108
List of Equipment ....................................................................................................... 108
List of Symbols .......................................................................................................... 108
Appendix 2- Detailed Equipment List ............................................................................... 110
Appendix 3 - HAZOP Study ............................................................................................ 111
References .................................................................................................................... 115
Table of Figures
Figure 1. A hydrolysis-based cellulosic ethanol production process
Figure 2. Length of Growing Season in Ontario
Figure 3. Block flow diagram of syngas fermentation process
Figure 4. Typical fluidized bed gasifier configuration
Figure 5. Typical biomass feeding system for fluidized bed gasifier
Figure 6. Typical wet scrubber configuration
Figure 7. Wood-Ljungdahlii biochemical pathway
Figure 8. Typical stirred-tank bioreactor configuration
Figure 9. Ratio control loop design for steam to feed ratio
Figure 10. Ratio control design for Feed to reboiler utility ratio
Figure 11. Ratio control structure between distillate and reflux
Figure 12. Pressure control inside the gasifier unit
Figure 13. Pressure control loop design for the distillation columns
Figure 14. Level control structure for the fermenter
Figure 15. Level control for reflux drum
Figure 16. Cascaded temperature control design around the condenser E-104
Figure 17. pH control loop structure for the fermenter.
Figure 18. Ratio control structure between purge stream and recycle stream
Figure 19. pH control design for unit S-101
Figure 20. Pre-treatment section
Figure 21. Gasification section
Figure 22. Gas cleaning section of the P&ID
Figure 23. Summary of Fermentation section of P&ID
Figure 24. Summary of Distillation section of the process
Figure 25. NPV Analysis
Figure 26. Sensitivity Analysis
13
21
22
25
26
28
30
31
59
61
62
64
65
66
67
69
71
72
73
75
75
76
77
77
88
90
Table of Tables
Table 1. Economic analysis of a cellulosic ethanol plant using the biochemical process
Table 2. Economic analysis of a cellulosic ethanol plant using the thermochemical process
Table 3. Process side material inflows and outflows of the bioethanol plant
Table 4. Utilities side overall material inflows of the bioethanol plant
Table 5. Utilities side overall material outflows of the bioethanol plant
Table 6. Overall energy balance of the process streams
Table 7. Overall energy inflow of the utility streams
Table 8. Overall energy inflow of the utility streams
Table 9. Stream table of process streams entering and exiting the pre-treatment section
Table 10. Material and energy inflow and outflow to the equipment of the pre-treatment
Table 11. Stream table of process streams entering and exiting gasification section
Table 12. Material and energy inflow and outflow to the equipment of the gasification section
Table 13. Stream table of process streams entering and exiting the gas cleaning section
Table 14. Material and energy inflow and outflow to the equipment of the gasification section
Table 15. Stream table of process streams entering and exiting the fermentation section
Table 16. Material and energy inflow and outflow to the equipment of the fermentation section
Table 17. Stream table of process streams entering and exiting the separation section
Table 18. Material and energy inflow and outflow to the equipment of the separation section
Table 19. Stream table of utilities streams entering and exiting the pre-treatment section
Table 20. Utilities side material and energy inflow and outflow to equipment of the gasifier
Table 21. Sum of capital costs for each type of unit and total capital cost.
Table 22. All operating costs for the syngas fermentation plant.
Table 23. Cradle gate GHG emissions of ethanol produced.
Table 24. List of hazardous chemicals used and produced in the biochemical plant
18
20
41
42
43
44
45
46
46
47
48
49
50
51
52
53
54
55
56
57
81
84
99
101
Table of Reacions
Reactions 1-3. Reactions taking place within gasifier
Reactions 4-10. Primary reactions that occur within fermenter, dependant on H2/CO content
Reaction 11. Regeneration reaction of adsorbent bed.
25
30
34
Table of Equations
Equation 1.Price of E100
Equation 2. Incentive calculation
Equation 3. Heat transfer area
Equation 4. Mean Temperature Difference
Equation 5. Log-Mean Temperature Difference
86
88
91
91
91
1. Project basis
1.1 Summary
The inevitable paradigm shift away from fossil-based fuels and products within
the next few decades has necessitated the introduction of alternative fuels and methods of
producing power. Recent volatility in the oil market, general uncertainty in the future
outlook of fossil fuels, rising oil prices and unavoidable concerns for global warming,
with greenhouse gases and volatile organic compounds (VOCs) being released into the
atmosphere, are factors that drive this paradigm shift. Bioethanol or ethanol is one fuel
that has been researched in depth for the last decade and is a promising fuel because of
several key advantages. It presents an alternative that has many similarities to fossil fuels,
especially in terms of the infrastructure and supply chain, but is different in the categories
that make fossil fuels undesirable, such as oil drilling and byproducts. Currently, the
projected potential demand of ethanol in Canada by 2022 is 2 billion liters and the
production capacity is 1.2 billion liters (United Nations, 2009). This 0.8 billion liter
difference is a key financial incentive and makes ethanol a viable fuel to pursue in terms
of research, development and finally implementation. In addition, governments like
Canadas that subsidize ethanol or biofuels in general give another financial incentive to
ethanol fuel startups.
Although ethanol is a viable alternative, in order for it to compete with the oil
market, its use needs to be constrained in several ways to maintain quality, production,
and profit. In the case of quality, the ethanol that is produced from an ethanol plant needs
to be very pure (> 95% purity), with little to no water content and very small traces of
other byproducts from upstream such as ammonia, hydrogen sulfide, methane and acetic
acid. Although acetic acid is considered a byproduct here, there are numerous uses for it
and it can be sold instead of discarded. This may require more investment towards
separation of byproducts but presents a financial incentive to pursue production of ethanol
by means in which acetic acid is also produced. Canadian regulations on gasoline supply
require that ethanol be 5% v/v (volume percent) of the gasoline mixture. Although there
are no explicit environmental regulations on ethanol fuel production and distribution, an
7
estimated 1 Mt CO2 equivalents per year reduction in greenhouse gas (GHG) emissions is
expected on top of existing projected reductions that result from other regulations in
Canada (National Resources Canada, 2013).
In terms of plant and feedstock feasibility, water requirements are the most
important constraints as the plant would not operate without water and the feedstock
would not grow without sufficient water. The location of the facility can be closer to
water sources such as lakes or ponds to meet these requirements relatively easily and
without significant capital investment. Since the feedstock will be grown on arable
farmland, irrigation systems would be necessary and would provide sufficient water.
However, this limits the feedstock location to southern Ontario as there is little rainfall in
Northern Ontario, with icy conditions and heavy snow hindering the growth of feedstock.
There is also the consideration of feedstock availability in Ontario. The feedstock that
will be used in this project is Miscanthus. This feedstock does not grow naturally in
Ontario, which means that there is no feedstock available currently. However, conditions
in southern or southwestern Ontario are highly favorable for the growth of this feedstock.
Introduction of this feedstock in southern Ontario is therefore not expected to be hindered
by adverse climate or weather effects, and the salinity and sand percentage of topsoil in
Ontario should be alright for its growth. The exact amount of feedstock that needs to be
purchased is based on 100 million liter output of ethanol per year.
Another raw material that is needed to run the process apart from the water and
the feedstock is the bacteria required for the fermentation process where ethanol is
produced. This primarily comes from chicken yard waste. Power requirements are
minimal in this plant as heaters, compressors or a large amount of pumps are not required.
Energy requirements to run the unit operations are mostly fulfilled by pressurized steam.
1.2 Economic Approximation of Process

The production cost of ethanol is estimated based on both fixed costs and variable
costs. The fixed costs include installation, labor, maintenance and interest on investment.
The variable costs include feedstock, enzyme production, utilities and waste management.
8
The following cost estimation is based on producing 20 million litres of ethanol per year.
The cost of miscanthus, including planting and harvesting of the crop is estimated to be
about 63-74 $/tDM (Roy, 2014). For every 1 kg of dry miscanthus, an estimated 0.360.39 litres of ethanol are being produced (Roy, 2014). Therefore in order to produce 20
million litres of ethanol, we need 55 million kg of miscanthus on a dry basis. The cost of
buying 55 million kg of miscanthus on a dry basis is 4 million. Ethanol processing plant
construction cost is estimated to be 45 million (Roy, 2014). Selling price of ethanol used
is $3.85 per litre (U.S. Department of Energy, 2015). However this price is subjected to
change depending on the demand of ethanol in the market and other competitors currently
producing ethanol. The total revenue that will be generated from selling 20 million litres
of ethanol is approximated at $77 million.
1.3 Relevant standards

Environmental standards by the government are not used to govern the production
of biofuels as they are an alternative source of energy and not as detrimental to the
environment as fossil fuels. However, the environment including surrounding ecosystems
and bodies of water near either the feedstock location(s) or the plant location need to be
cared for. Waste gases and tailings/byproduct ponds cannot be close to the habitats of
wildlife, and the surrounding ecosystem should not be greatly transformed in order to
introduce feedstock or to build a chemical plant.
Manufacturing standards and constraints are minimal, with distillation columns
having a diameter that will allow them to be transported to the plant and the gasifier being
built with durable walls that can withstand high temperatures and pressures for long
periods of time. General safety standards will be accounted for and the plant site location
will be constrained to locations further from population centers, environmental reserves,
wildlife or public water sources. Proximity to water sources such as ponds or lakes will
have to be optimized in order to not pollute the water while keeping water transportation
costs low. One key safety requirement is that the chimney or release of waste gases has to
be high above ground level in order to keep air pollution near the ground low and to
disperse the waste gases.
9
In this syngas fermentation process, this requirement is very important as syngas

will have small amounts of unwanted or toxic gases such as H2S and methane that will
have to be dispersed. Otherwise, these gases will have to be converted to other byproducts before they are released. IT systems that will be required to maintain safety will
include the basic process control systems such as PID controllers as well as MPC
controllers. Safety IT systems such as SIS interlock systems will also be put in place in
order to lockdown processes such as gasification that can endanger the entire plant if they
are out of control. Lastly, the key safety standards for a plant such as containment and
emergency procedures will have to be detailed and put in place before the plant begins
operating.
2. Existing solutions
2.1 1st generation feedstock
One of the leaders in ethanol production from starch along with sugarcane is corn.
Currently, about 95% of ethanol in the United States comes from corn due to lower cost
and vast research on production (Pimentel & Patzek, 2008). There are two major ways of
processing corn into ethanol, namely the dry milling and wet milling process. The dry
milling process is more common as it requires less capital to build, is more focused on
ethanol production and provides animal feed (dry distillers grain) as co-product. On the
other hand the wet milling method separates the corn for different uses and is able to
produce a variety of product but is also more costly. Over 88% of the ethanol produced in
the United States is produced using the dry milling process and the remaining 12% is
from the wet milling process (Kwiatkowski, 2006) .An overview of the dry milling
process which uses the biochemical process of hydrolysis using enzymes and then
fermentation is as follows:
The corn grain is sent through a series of screen or blowers in order to separate
any foreign object such as rocks or minerals.
The corn then is then crushing and/or grinded and sent to a slurry tank which
contains water, enzymes and pH stabilizing chemicals.
10
The mixture is heated and the enzyme break down the starch polymer into
shorter chains, in a step called liquefaction.
The resulting slurry then undergoes hydrolysis which further breaks down the
glucose chains into glucose units. The glucose is cooled and undergoes
fermentation where it is converted into ethanol with water and carbon dioxide as
by products
The ethanol obtained from fermentation is heated and sent through a degasser
drum to flash off the vapour. The resulting products go through a series of
distillation, stripping column and molecular sieve in order to separate the
ethanol from the rest of the products. The rest of the product separated from
ethanol is dehydrated through series of liquid-liquid separation and liquid
separation such as centrifuge and dryer (Kwiatkowski, 2006) (Wang, 2007)
The capital cost of a corn ethanol production plant with capacity of 400 ML/year
will be $220 million (using CEPCI to find value in present value of 2014) per plant and
the cost of corn will be $725 million and utility costs of $642 million. Through research,
corn ethanol production and use could reduce GHG emissions by 18% of current levels.
This however does not account for the deforestation of land in order to grow more corn
since it is a crop that requires soil with high nutrient concentration.
While corn ethanol is a mature industry, it continues to face issues of minimal
greenhouse gas emission reduction, negative net energy balance and decrease in corn
food supply. Corn is a big part of human food consumption and the use of corn as fuel
often become an ethical issue. As a result, intense research on cellulosic ethanol lead to
the discovery of second generation feedstock which is also known as lignocellulosic
feedstock.
2.2 2nd generation feedstock

Second generation feedstock takes advantage of the abundance of biomass on the
planet. Second generation feedstock uses cellulose and hemicellulose which are complex
11
sugar polymers found in natural biomass such as grass, wood and trees as source of
ethanol. In terms of feedstock, many types of biomass can be used since all of them
contain cellulose. Switchgrass and Miscanthus are mostly studied and used as the
feedstock. Switch grass is a perennial grass native to North America and its abundance
make it appealing to be used as feedstock for ethanol production.. A few examples of
second generation feedstock include Switchgrass and Miscanthus where Miscanthus was
the preferred feedstock in the bioethanol plant design. Unlike corn the use of switchgrass
or miscanthus as feedstock has no impact on food supply and is therefore more appealing.
Furthermore, growth of miscanthus and/or switchgrass requires lands with little to no
fertilizer, pesticides or energy input which is opposite to that of corn.
Preference of using miscanthus as the bioethanol plant feedstock over corn
and other second generation can be seen by looking at greenhouse gas emissions
associated to land conversion for increased corn, switchgrass and miscanthus growth. A
study by Mueller et. al has shown that CO2 equivalent emissions from corn ethanol plant
in the U.S. is rated at 92 g CO2 equivalent per MJ energy provided which is a marginal
benefit to gasolines 96 g CO2 equivalent per MJ energy provided (Dunn, 2013). The
same study also indicate that greenhouse gas emission from land conversion for increased
production of corn ethanol is highest at 7.6 g CO2e/MJ while Miscanthus has the lowest
at -10 g CO2 equivalent per MJ due to its carbon sequestrating ability and high yield
(Dunn, 2013). Furthermore,. Dunn et.al has shown that the average peak annual biomass
of miscanthus is 22 tonnes of biomass per hectare while switch grass only produced
10tonnes of biomass per hectare. The same study also shows that the yield of miscanthus
is less sensitive to the amount of rainfall and fertilizer compared to switch grass (Dunn,
2013). Field trials in three locations the United States have shown that miscanthus yield is
three to four times of that of switchgrass (Liska, 2009). A side by side comparison of
switchgrass and miscanthus greenhouse gas emission reveal that emissions produced by
Miscanthus growth, harvesting and transportation is about 31% lower than that of
switchgrass.
12
In second generation ethanol production, lignocellulosic ethanol has many

advantages over first generation including lower GHG levels and abundant feedstock
supply. However, this technology is not commercialized yet due to high capital and
operating costs on some of the process components as well as the enzymes/bacteria used
can be expensive.
The steps to lignocellulosic ethanol production include pretreatment, hydrolysis
and fermentation. This can be seen in Figure 1.
Figure 1. A hydrolysis-based cellulosic ethanol production process. (Dwivedi et al., 2009)
In the pre-treatment step, the lignin walls of the biomass is broken down or
pushed apart in order to expose the cellulose in order to undergo hydrolysis and
fermentation. This step requires high amounts of energy due to the strength in the walls
and is the most expensive step and the hardest step in the whole process. Pre-treatment
can be done physically, chemically and biologically with the chemical method currently
being the most common. One way of physical pre-treatment is done using liquid hot
water where high temperature and pressure water is used to breakdown the lignin walls.
This method has also shown improvements in the sugar recovery as well as partial
hydroxylation of the cellulose in the biomass. Another chemical treatment method is
13
Ammonia Fiber/Freeze Explosion (AFEX). The biomass is exposed to liquid ammonia at

high temperature and pressure, and then a swift reduction in pressure exposes the
cellulose which can then be processed. Other types of chemical treatment include alkali,
ionic liquid and dilute acid treatments. Out of the three types of treatment, biological
methods are much preferred due to their ability to produce higher yields (theoretically)
while having faster breakdown times and lower emissions. However, biological methods
are some of the most expensive and furthest away from commercializing methods out of
the three types of pretreatment. Biological pretreatment uses enzymes to breakdown the
lignin cell wall to expose the cellulose (Alvira, 2010).
After pretreatment, the cellulose undergoes hydrolysis through a biochemical,
thermochemical or a combination of both biochemical and thermochemical process which
will be referred to as the hybrid process. All three methods will be explained in the
following section of this report.
The estimated capital cost for a cellulosic plant of 400 ML/year was found to be
around $496 million (using 2006 prices and scaled to 2014 present value using CECPI
and assuming linear relationship between cost and production capacity) and operating
cost of $249 million/year which includes raw materials such as feedstock and enzymes
($102 million), utilities such as water, electricity and maintenance ($54 million) as well
as other charges. this estimation is lower than the costs for a corn ethanol plant. It was
also found through research miscanthus is able to reduce GHG emissions by up to 88% of
current biofuel production.
3. Design alternatives
Miscanthus belongs to the second generation feedstock of ethanol known as
lignocellulosic ethanol. Miscanthus can be converted into ethanol through the
biochemical, thermochemical or the hybrid process which is a combination of both
biochemical and thermochemical process. The hybrid process is the recommend process
for the bioethanol plant design due to several advantages over the biochemical process
and thermochemical process which are highlighted below. A brief description of each
14
process is presented and a comparison between each process is made. Finally, proposed
location of the bioethanol plant is also presented at the end of the section
3.1 Hybrid process

The hybrid process combines the thermochemical pathway of gasification of feed
stock into syngas with the biochemical pathway of fermenting syngas into ethanol. A
general step of producing ethanol from Miscanthus using the hybrid process is as follows;
1. Drying and crushing
2. Gasification
3. Gas cleaning and cooling
3. Fermentation
4. Distillation/purification
In the pre-treatment step, raw feedstock is dried to a moisture content of 10%
water. The dried feedstock is crushed and introduced to a gasification reactor where steam
is also introduced. The heat from steam disintegrates the feedstock into its elemental
components. A series of exothermic reactions occur and heat the gasification reactor to
around 850 C (Dwivedi, 2009). The reactions are also responsible for the production of
carbon dioxide, carbon monoxide, hydrogen and trace amounts of hydrogen sulfide,
ammonia and methane - a mixture gases known as syngas. Other products from the
gasification step include solids such as ash and char. Syngas undergo a series of gas
cleaning and gas cooling steps where any impurities like hydrogen sulfide, ammonia and
methane are removed and syngas is cooled to 37C for fermentation.. Equipment used to
remove impurities in syngas may include, adsorption column, scrubbers or cyclone for
solids removal The cleaned syngas is then sent to a fermentation vessel where bacteria
such as Clostridium Ljungdahlii anaerobically digests syngas into acetic acid, ethanol and
water at 37 C and 1 atm (Abubackar, 2011). The product from fermentation is a
combination of ethanol, acetic acid and water. The fermenter product is then sent to a
15
series of separation sections such as distillation column and molecular sieves in order to
purify the ethanol.
3.2 Biochemical process

Production of ethanol from Miscanthus employs the biochemical process of
hydrolysis using enzymes and then fermentation similar along with several pre-treatment
steps. A general step for producing ethanol from Miscanthus is as follows:
1. Pre-treatment
2. Enzymatic hydrolysis
3. Fermentation
4. Distillation/purification
First the raw feed stock of miscanthus or any second generation undergo drying
where moisture content is generally brought down to approximately 10%. The dried
feedstock is then crushed to a size of approximately 3.2 mm. The crushed feedstock goes
through a series of pre-treatment steps as outlined in section 2. Once the cellulose and
hemicellulose are rid of lignin and can be exposed to enzymes, enzymatic hydrolysis
proceeds. Enzymatic hydrolysis is the process where the long polymer sugar chains which
makes up of cellulose are broken down into sugar monomers such as glucose, fructose
and xylose. The resulting monomers are then metabolically digested by bacteria under
anaerobic conditions where alcohols such as ethanol are produced. The products which
consist of several long chained alcohols, acetic acid and water are then sent to a series of
separation steps in order to purify the ethanol.
The biggest difference between the biochemical and hybrid process is the need of
a pre-treatment step in the biochemical process. Pre-treatment is energy intensive and is a
huge drawback to the biochemical process of ethanol production. In contrast, the hybrid
process does not require pre-treatment and is therefore less energy intensive which results
to lower operating costs. It is projected that about 20% of total cost of cellulosic ethanol
16
production is from pre-treatment, a cost that is non-existent in the thermal-biochemical

process (Khanna, 2008). The type of pre-treatment employed is also dependent on the
type of feedstock for cellulosic ethanol. For example, using corn stover as feedstock uses
a different, less expensive pre-treatment process as compared to using switchgrass, which
benefits from ammonia fire explosion pre-treatment. This results to lesser feedstock
flexibility for an ethanol plant using the biochemical process and played an important in
the groups decision of using the hybrid process. Another important factor issue would be
fully breaking down the grass, as the pre-treatment stage is not as effective on grasses
with high lignin contents like Miscanthus (~23% lignin content) (Sanchez, 2008).
Studies also reveal that the choice of pre-treatment has an effect on upstream
processes (i.e. harvesting and storage) since aging of the feed stock during storage can
make it resistant to certain types of pre-treatments. Furthermore, the choice of pretreatment has great effects on the downstream processing. In the thermal-biochemical
hybrid process pre-treatment is completely eliminated and as a result upstream process
such as harvesting and storage has very little effect on downstream processes. Another
advantage of thermal-biochemical process over the biochemical process is the increased
ethanol yield associated with the thermal-biochemical hybrid process. In the biochemical
process lignin is often unused and separated in the pre-treatment process. In contrast, the
hybrid process utilizes the whole biomass including the lignin in the gasification process.
Furthermore, a significant portion of 5-carbon sugars from hemicellulose cannot be
completely converted into alcohol and better enzyme technology is needed (Daniell,
2012). This results to lower ethanol yield per tonne of feedstock using the biochemical
process. Finally, pre-treatment in biochemical process is a relatively new technology and
research is currently ongoing. On the other hand, gasification in the hybrid process is a
much older technology and is used in processes besides ethanol production.
In addition to the pretreatment step, the biochemical process requires enzymatic
hydrolysis which breaks down the network of polymers that make up cellulose and
hemicellulose into sugar monomers. The two main types of hydrolysis are either acid or
enzymatic hydrolysis. The downfall of acid hydrolysis is it produces inhibiting
17
microorganism which lower ethanol yield in the fermentation stage. Acid hydrolysis also
causes corrosion of equipment and the acid needs to be recovered at the end of the
process. In the hybrid process inhibitors are not present since acid hydrolysis does not
occur which allows for a more consistent ethanol yield. Furthermore, acid is not involved
in the hybrid process and therefore lower corrosive material can be used. In enzymatic
hydrolysis, enzymes such as cellulose break down cellulose and hemicellulose into sugar
monomer units. A drawback of enzyme hydrolysis is the cost and need for large scale
production of enzyme. A table of cost of a cellulosic ethanol plant producing 58 M
gallons/year of ethanol using the biochemical process is shown in Table 1.
Table 1. Economic analysis of a cellulosic ethanol plant producing 58 M gallons/year of ethanol using the
biochemical process
Process Section
Cost (Millions $U.S. 2013)
Feedstock handling[1]
14.5
Pretreatment[2]
47.9
Xylose fermentation
12.5
Enzyme production[1]
5.7
Saccharification and fermentation
42.2
Ethanol recovery[3]
8.1
Utilities
102.6
Total
233.5
Table reproduced from Foust,2009 and inflated to 2013 dollars using CEPCI index.
3.3 Thermochemical process

Besides the biochemical and hybrid process, lignocellulosic ethanol is also
produced through thermochemical process which converts syngas produced in the
18
gasification step into ethanol and other linear alcohols using a synthetic catalyst. A
general step of producing ethanol from the thermochemical process is as follows:
1. Drying
2. Gasification
3. Syngas cleaning
4. Catalytic conversion of syngas into ethanol and alcohol
5. Distillation/Purification
First, the feedstock is removed of impurities through washing. The washed feed
stock is then dried and grinded/crushed into smaller pieces. The feed stock is then fed to a
fluidized bed gasifier and can reach high temperatures (800C). Due to high temperature,
the feedstock decomposes to syngas which is made up of carbon monoxide, carbon
dioxide and hydrogen. The syngas is collected from the top of the gasifier and is cooled
through a series of heat exchangers. The cooled gas undergoes water scrubbing steps
where tar and residuals are removed. The gas is compressed to a higher pressure and
impurities such as hydrogen sulphide and carbon dioxide is removed in an amine unit.
The cleaned gas is sent through a bed of fixed bed molybdenum disulphide based catalyst
which produces ethanol along with other linear alcohols. The mixture is sent through a
series of distillation and separation steps where the ethanol is obtained (Yang, 2008).
Compared to the hybrid process, it is evident that the pure thermochemical
process result to several by products such as methanol and other linear alcohols which
require several separation steps. In the hybrid process, the main products are ethanol,
acetone and water (bacteria media) which requires fewer separation units. Another
advantage of the hybrid process is that conversion of syngas to ethanol occurs at low
pressures (1 bar) and low temperatures (37C) which results to lower operating costs. A
study has also shown that the bacteria used in the hybrid process is also able to tolerate
sulfur impurities in the syngas which results to lower energy and cost allocated in the gas
cleaning step (Roy, 2014).
19
In the thermochemical process, sulfur impurities must be eliminated before the

catalytic conversion to ethanol since sulfur irreversibly poisons the catalyst. This also
serves to be potentially cost saving since replacing a poisoned catalyst is not an issue in
the hybrid process. Finally, ethanol yield in the thermochemical process is very sensitive
to hydrogen to carbon dioxide ratio. In order to achieve optimum hydrogen to carbon
dioxide ratio of the syngas, a water-gas shift reaction step is normally employed which
requires the use of another reactor and more steam input (AdvancedBiofuelsUSA, 2011).
In the hybrid process, the hydrogen to carbon dioxide is not needed since
hydrogen to carbon dioxide is less of an issue. This results in lower operating and capital
costs for the hybrid process. Shown in table 2 is an economic analysis of a cellulosic
ethanol plant producing 58 M gallons/year of ethanol using the thermochemical process.
Table 2. Economic analysis of a cellulosic ethanol plant producing 58 M gallons/year of ethanol using the
thermochemical process
Process Section
Cost (in Millions $U.S. 2013)
Feedstock handling
32.1
Catalyst
2.8
Gasification
34.9
Gas cleaning
84.7
Separation
9.28
Utilities
67.5
Total
231.3
Table reproduced from Daniell, 2012 and inflated to 2013 dollars using CEPCI index.
3.4 Bioethanol Location

Due to Ontarios geographic location, all of Ontario experience climate that is
well below the freezing point. While this feature is generally unattractive for crop growth,
some regions of Ontario do enjoy warmer than others. Ontario can be split into 5 different
20
regions which are 1)Eastern (Ottawa), 2)Central (Hamilton/Toronto), 3)Southwestern

(Sarnia), 4)Northeastern (Timmins) and 5)Northwestern (Thunder Bay) . Northeastern
and Northwestern Ontario experience longer and colder climates with annual average
temperature of around 8C.
On the other hand, Southwestern, Eastern and Central Ontario enjoy warmer
climates with average annual temperatures of 12C. This leaves 3 possible regions of
Southwestern, Eastern and Central Ontario (Hamilton/Toronto) as possible location for
the proposed bioethanol plant. Based on Figure 2 we can see that as we move towards
Southern Ontario, the length of growing season increase. This implies that the Sarnia,
Windsor and Hamilton/Toronto area is a more preferred region than Eastern Ontario
where length of growing days of less than 170 days can be observed. It is expected that
the Hamilton and Toronto region is generally unfavourable to the approval of a
bioethanol plant due to highly dense residential area. Finally, several bio refineries such
as Suncor refinery already exists in the Sarnia region which makes it the preferred
location
Figure 2.Length of Growing Season in Ontario.( Agriculture and Agri-Food Canada, 2014).
21
4. Overview of Proposed Process Design

4.1 Process Summary
The proposed process for producing Ethanol is referred to as Syngas Fermentation
and is considered a hybrid approach for converting biomaterials into fuel. The term
hybrid is used because the process incorporates thermochemical aspects such as
gasification with biochemical ones like fermentation. The process uses five main stages
to turn any lignocellulosic biomass (i.e. switchgrass, miscanthus, wood chips) or biowaste (i.e. corn stover) material into ethanol to be used as fuel.
A block flow diagram
shown below in Figure 3 outlines the process, and a full process description follows.
Figure 3. Block flow diagram of Syngas Fermentation process
Fermentation has the advantage of operating at low temperatures (~37C) and

pressures (~1 bar) compared to alternative processes, lowering overall energy costs for
production. Syngas fermentation also has the advantage of high selectivity of ethanol
(leading to increased yields) and good tolerance to typical syngas impurities such as
sulfur, which in turn reduces costs for syngas cleaning (Daniell, 2012). Additionally,
syngas fermentation operability is not impacted greatly by the H2:CO ratio of the syngas,
meaning that the gasification process and the proceeding syngas cleaning steps are
awarded flexibility (Daniell, 2012). Lastly syngas fermentation allows for a very large
variety of feedstock to be used to produce ethanol, capable of converting virtually any
lignocellulosic material into ethanol.
22
4.1.1 Pre-Treatment
The raw or harvested feedstock is first pretreated through washing, drying and
crushing stages to bring the biomass into a desirable pellet form, ideally with diameters in
the range of 3-6 mm (Roy, 2014) (Michel, 2011). The raw biomass will likely be stored
on site in bales (if grassy biomass) or in large storage containers (if woody biomass)
before being dumped/placed onto a conveyer belt which initiates the pre-treatment
process that follows. The washing stage is a precautionary cleaning stage meant to clean
the feedstock of any dirt or lingering chemicals such as pesticides. This pre-treatment
stage can be done through a variety of methods, but a conveyer-belt spraying unit was
selected for this process.
The next pre-treatment step is drying, which is achieved using a belt drying unit,
operated using excess steam or air as the drying force (Li, 2012). The belt dryer is simply
a conveyer belt pushing the biomass through a unit that continuously dries the materials
as they pass.
Drying is a necessary pre-treatment step because the moisture within the
biomass takes away energy from the gasifier which cannot be recovered at an
approximate rate of 2260 kJ lost per kilogram of moisture (Basu, 2013). The total
moisture content of the biomass should be between 10-20% ideally for minimal energy
loss (Roy, 2014) (Basu, 2013).
Next, the dried biomass must go through a size reduction step for ease of loading,
and for optimal performance within the gasifier. A continuously operated hammer mill
crushing device is used in this process, selected for its ease of operation and control of
desired particle size while being able to handle a variety of different feeds (Kratky, 2010).
For optimal operating conditions the biomass moisture content must not exceed 10-15%
(Kratky, 2010). The crushed dried pellets that are left from the pre-treatment process are
sent towards the gasifier via conveyer belt.
23
4.1.2 Gasification
Once pretreated, the biomass is fed into a gasifier unit where it is converted into a
syngas mixture, composed mainly of carbon monoxide (CO), hydrogen (H2), carbon
dioxide (CO2), methane (CH4), and impurities.
Many types of gasifier configurations
may be employed for this task, such as counter/co-current fixed bed, fluidized bed, or
entrained flow gasifiers. For the purpose of biomass gasification a fluidized bed gasifier
was selected, mainly due to its insensitivity to fuel quality, allowing for flexibility in the
biomass feedstock (Basu, 2013).
The fluidized bed gasifier is identified through its use of a bed, which is
essentially a collection of granular solids that are kept suspended via the continuous flow
of gases at specific velocities (Basu, 2013). The bed, selected as quartz sand, provides
excellent solid-gas mixing and a relatively uniform temperature profile within the gasifier
(Basu, 2013). More specifically, the fluidized bed gasifier is a circulating fluidized bed
gasifier, where the bed is recirculated within the gasifier providing longer gas residence
times and allowing for larger units in general (Basu, 2013). An image of a typical
circulating fluidized bed gasifier can be seen below in Figure 4. The gasifier is operated
at low pressures (~1 bar) and high temperatures (~800-1000C) and is naturally an energy
intensive process. Most of the lost energy in operating the gasifier can be recovered
downstream as heat through cooling of the syngas.
24
Figure 4. Typical fluidized bed gasifier configuration (Basu, 2013)
Within the gasifier, the biomass feed undergoes incomplete combustion, in the presence
of either air or trace amounts of oxygen, to produce syngas mainly through the following
reactions.
+ 2 2
[1]
+ 2 2 +
[2]
+ 2 2 + 2
[3]
Reactions 1-3. Reactions taking place within gasifier. Incomplete combustion is achieved with a controlled
amount of oxygen.
The carbon containing biomass is fed into the gasifier with steam and a controlled
amount of oxygen that ensures the biomass undergoes incomplete combustion, starting a
chain of reactions resulting in the final syngas mixture. The hot syngas is continuously
drawn from the gasifier at an approximate temperature of 850C, as is the produced ash
from the bottom of the unit.
25
For the purposes of feeding the biomass to the gasifier there are a variety of
different methods, but a gravity chute was selected for this process due to its simplicity.
After pre-treatment the feed is dropped onto a feed screw which leads to a gravity chute
that feeds the biomass directly into the gasifier (Basu, 2013). A schematic of this feed
system is shown below in Figure 5.
Figure 5. Typical biomass feeding system for fluidized bed gasifier.(Basu,
2013)
The feed screw allows for relatively simple control of feed flow, and the gravity
chute offers a simple method of feeding biomass to the gasifier. The tip of the gravity
chute lies within the gasifier itself and must be properly insulated to withstand the high
temperatures within the unit. For this feed configuration the unit is often operated at
slightly below atmospheric pressure to ensure that the rising gas doesnt travel into the
feed chute (Basu, 2013). A jet vapour stream placed directly under the chute is often
installed to ensure that no gas travels up the chute.
4.1.3 Gas Cleaning

The syngas produced from the gasifier is a gaseous mixture containing H2, CO,
CO2, CH4 and a multitude of impurities. Based on further downstream processes, the
26
ideal syngas composition should be low in impurities and high in H2/CO content for the
purposes of fermentation. Certain impurities such as hydrogen sulfide (H 2S), ammonium
(NH3), and hydrochloric acid (HCl) can cause issues downstream if their levels are too
high. Syngas cleaning is categorized into two types, hot-gas cleaning (HGC) or cold-gas
cleaning (CGC). Attached to the gasifier is a cyclone which quickly removes any solid
particulates or ash within the syngas before it undergoes further stripping. A cyclone is a
simple way to screen out solid impurities and can be operated at temperatures up to
~1000C (Basu, 2013).
The hot syngas mixture leaving the cyclone is next sent into an adsorption column
used primarily for removing H2S which can cause potential issues during downstream
processes and sulfur has been known to corrode metal surfaces (Woolcock, 2013). First
the syngas must be cooled to an approximate temperature of 600C for the adsorbent
within the column, zinc oxide in this case, to work effectively (Woolcock, 2013). During
the cooling of the syngas some of the heat lost in the gasification process may be
recovered as steam, which can be fed to a steam turbine to produce electricity. The
adsorption column is packed with an iron oxide adsorbent which selectively binds with
the sulfur particles to form a metal sulfur compound such as ZnS or FeS (Woolcock,
2013). For this process a Zinc oxide (ZnO) adsorbent was selected mainly due to its low
cost and high availability. The reversible adsorption columns are to be run in parallel,
with one column running at a time. When the sulfur compounds fully bind to the
adsorbent bed the flow is sent to the parallel adsorption column. The fully bound
adsorption column is then fed a stream of oxygen which regenerates the bed by unbinding
the sulfur back into the gaseous stream (Woolcock, 2013). A gas rich in sulfur dioxide
exits the regenerated adsorption column where it is sent to a sulfur recovery unit to obtain
elemental sulfur or sulfuric acid (Woolcock, 2013). The parallel configuration of the
adsorption columns ensures that the process may be run continuously as one bed is being
regenerated while the other one is in operation.
Following the adsorption column the syngas undergoes rigorous cooling stages to
reach an approximate temperature of 45C, all the while recovering significant amounts
27
of energy as heat. Having now transitioned into cold-gas cleaning, a wet scrubber is
selected to remove the remaining particulates. Ammonium and chlorine are both highly
soluble in water making the wet scrubber a great choice in removing these impurities
(Woolcock, 2013). The gas enters a column that is known as a spray tower, which is
essentially a vessel that contains porous pipes that spray the passing gas with water which
collects the impurities. Along with removing ammonium and chlorine the wet scrubber
will also remove any leftover or newly formed solid particulates in the gas. The water is
continuously drained from the bottom the tower and sent to wastewater treatment for
processing. A typical configuration of a wet scrubber is shown below in Figure 6.
Figure6. Typical wet scrubber configuration (Woolcock, 2013).
After leaving the wet scrubber the syngas temperature has fallen to approximately
37C which is the ideal temperature required for the fermentation step that follows (Roy,
2014). The gaseous mixture is also free of the impurities that could cause problems
downstream and is ready to be converted into ethanol via microbial fermentation.
28
4.1.4 Fermentation
The clean syngas is now free of impurities and cooled to a low temperature perfect
for the fermentation process that follows. The syngas is fed to the fermenter where it aids
in the production of ethanol. Essentially the syngas is converted to ethanol via a number
of reaction pathways that are made possible by certain strains of bacteria. To date, the
most relevant family of bacteria utilized in syngas fermentation is the Clostridium family
(Abubackar, 2011) (Daniell, 2012). Of the various strains within the family, Clostridium
Ljungdahlii is the most widely studied and is used in this process primarily for its ethanol
selectivity properties. Isolated primarily from chicken farm waste, the main challenge of
the process would be obtaining the bacteria, as it is not easily isolated (Abubackar, 2011).
Though there are pilot-scale and pre-commercial plants in operation that use these
bacteria, demonstrating the feasibility of the process (Daniell, 2012).
The role of the bacteria in the process is that of a biocatalyst, it enables certain
reactions to occur, while the bacteria itself is hardly consumed (Abubackar, 2011). The
bacteria can then be regenerated or recycled to maximize process efficiency and minimize
bacteria losses. For optimal growth of the bacteria the temperature of the reactor should
be held as close to 37C as possible with the pH maintained at 6, though acceptable
performance can be achieved within a pH range of 4-7 (Roy, 2014) (Abubackar, 2011)
(Daniell, 2012). The bacterium has also been shown to support growth on ethanol,
further improving the overall bacteria efficiency (Daniell 2012). This bacterium and
others of the clostridium family enable the syngas to take the Wood-Ljungdahl
biochemical pathway in order to produce ethanol. A simplified reaction pathway is
shown below in Figure 7.
29
Figure 7. Wood-Ljungdahl biochemical pathway (Abubackar, 2011).
In short, the CO and H2 are utilized as the main reactants for ethanol production
and their amounts in the syngas dictate which reactions are utilized (Daniell, 2012). The
main reactions that take place within the fermenter, made possible through the biocatalyst
are shown in Equations 4-10, forming ethanol (CH3CH2OH) and acetic acid (CH3COOH)
as the main products (Daniell, 2012).
6 + 32 3 2 + 42
[4]
4 + 22 3 + 22
[5]
3 + 32 3 2 + 2
[6]
2 + 22 3
[7]
2 + 42 3 2 + 2
[8]
22 + 62 3 2 + 32
[9]
22 + 42 3 + 22
[10]
Reactions 4-10.Primary reactions that occur within fermenter, dependant on H2/CO content(Daniell, 2012)
30
Initially acetic acid will be the favoured product but once the fermenter is run with
recycle for several hours ethanol production will be favoured, reaching an approximate,
steady ethanol: alcohol ratio of 2 (Abubackar, 2011). It should be noted that since the
bacteria is anaerobic the reactor must be kept free of any oxygen or the bacteria will die
(Abubackar, 2011).
As with gasification, many fermenter types may be incorporated for the purposes
of syngas fermentation and the process can be run in either batch, semi-continuous, or
continuous modes of operation dependent on which fermenter is used. The most studied
and widely employed reactor configuration for the purposes of syngas fermentation is a
stirred-tank bioreactor (STB) and was selected for this process under continuous
operation (Abubackar, 2011). A typical configuration of an STB is shown below in
Figure 8 .
Figure 8. Typical stirred-tank bioreactor configuration. l- gas sparger; i- gas feed; ii- medium feed; iiipump; iv- liquid outlet; v- gaseous outlet (Abubackar, 2011).
31
The gaseous feed enters the fermenter at the bottom of the reactor where most of
the syngas breaks into smaller bubbles, well dispersed by the continuous mixing of the
tank. Syngas retention times vary but can be expected to be approximately 1 minute
(Abubackar, 2011). The liquid broth from the reactor is continuously drawn as fresh
medium is being pumped into the bioreactor. Ideally the syngas will have a carbon
conversion efficiency that can reach up to 80%, where the unconverted gas is also
removed from the bioreactor continuously (Daniell, 2012).
The unconverted syngas, now high in CO2 content, can be combusted to recover
even more of the energy that was used up in the gasification step. The fermenter also
contains a fermentation medium, which varies greatly in composition, but is largely made
up of acidic water. The medium also includes the bacteria, nutrients, vitamins, minerals,
salts, yeast extracts and/or other additives that are required for ethanol production. The
liquid medium extracted from the fermenter is usually immediately filtered to remove the
bacteria, which is recycled, as it would die during the distillation stage.
Some of the ongoing challenges with this process are limited mass transfer rates,
which can be improved through modification of the bacteria or incorporating a 2-stage
system, and limited ethanol concentrations in the fermentation broth (Abubackar, 2011).
Typically, the broth cant contain much more than 5% ethanol before it impacts the
bacteria and causes problems within the unit, so the resulting ethanol yield is ~3-6% of
the concentration within the broth (Abubackar, 2011) (Daniell, 2012). Ideally the yield
from the bioreactor is 0.3-0.4 L ethanol/kg-dry feedstock (Roy, 2014).
4.1.5 Distillation
The final broth that is pulled from the fermenter contains mainly the fermentation
medium, with ethanol (~3-6%) and acetic acid (~3-6%) in low concentrations. This broth
is then sent to a series of distillation columns followed by extractive dewatering to reach
final product purity.
The ethanol is to be separated from the acetic acid and the
fermentation medium using successive distillation columns in series. The bottoms from
the first distillation columns can be recycled back to the bioreactor since it will mainly
32
contain the fermentation medium, though a purge stream is necessary to prevent

accumulation of acetic acid. All of the energy recovered from cooling the syngas and
combusting the unused syngas can now be used to provide energy for the operation of the
distillation columns. Upon successive separation of ethanol, oftentimes the ethanol is
sent to further separation processes such as dewatering to ensure high product purity to
meet quality specifications.
4.2 Design Basis

The design basis that was selected to base the sizing, costing and economic
analysis for this process was to produce 100 million liters of ethanol per year. Meeting
this production rate must be done while maintaining high product purity (>99%) and
ensuring safe process operation.
4.3 Product Specifications

The final ethanol blend must meet the standards set by the Canadian General
Standards Board and the American Society for Testing Materials (ASTM) to ensure safe
operation within a motorized vehicle, and other standards outlined further in the report.
5. Process Behaviour
5.1 Normal operation
The feedstock (miscanthus) which contains 12-20% moisture content is fed to a
continuous belt dryer drier at atmospheric pressure and room temperature where the dried
it is dried to a 10% moisture content. Medium pressure steam which enters at a
temperature and pressure of 162C and 7.8 atm respectively and exits at 120C and 2 atm
is used as the heating media. The dried feedstock is then sent into a hammer mill which
reduces the feedstock into a target particle size of 3.2 mm. Once the feedstock are milled
into the appropriate particle size it is sent into the gasifier unit which operates at 850C
33
and 1 bar. Steam is also introduced into the gasifier and provides heat to disintegrate the
feedstock into its elements.
The reactions presented in chapter 4 occur inside the gasifier and the resulting
products consist of solids such as char and ash along with the major product that is
syngas. Other major by products include ammonia, hydrogen sulphide and hydrogen
chloride. The gasifier products are sent to an H2S adsorber column where H2S is removed
by adsorbing on to a zinc oxide chemical adsorbent. The zinc oxide adsorbent eventually
needs to be regenerated after normal operation, so after about 15 days (based on volume
of adsorbent and its capacity), the feed going into the adsorber is sent to a secondary
adsorber that operates in the same way. During this time, the used zinc sulfide being
regenerated is contacted with oxygen in air to convert it back to zinc oxide through the
Reaction 11:
+
3
+ 2
2 2
[11]
Reaction 11. Regeneration of adsorbent bed via oxygen.
The sulfur dioxide produced from this reaction is then sent to a sulfur plant to
produce a sulfur compound or for other processing. The sulfur-free syngas is then cooled
from its temperature of 550C to 37C and is sent into the wet scrubber part of the gas
cleaning section, also known as cool gas cleaning. The wet scrubber is simply a vessel
where the cool syngas is contacted with water from a water spray to removeCH3, HCl and
CO2. Once the impurities are removed, the syngas is continuously sent into the anaerobic
(closed-roof) fermentation vessel through an entrance from the top. A recycle stream
which contains bacteria and nutrients from a storage vessel is also introduced into the
fermentation vessel as a mixed liquid broth at 37C and 1 bar. At the same time, a
continuous feed of fresh bacteria broth which contains nutrients essential for bacterial life
is mixed with the recycle broth from the bottom stream of the first distillation column..
The mixed recycle and fresh broth mix to a temperature of 73C and passes through a
shell and tube heat exchanger where the mixture is cooled to 37C. The pH or acidity of
the broth mixture introduced into the fermentation vessel is kept at an optimal pH of 6 by
34
controlling both the amount of bacteria broth recycled into the storage vessel and the
amount of fresh bacteria broth.
The resulting products from fermentation are ethanol, acetic acid and water which
represent bacteria broth at a temperature of 73C and pressure of 1 bar. The fermenter
liquid effluent is then fed into the first distillation column which operates at a condenser
pressure of 1 bar and a reflux to distillate ratio of 0.1. The bottom stream of the
distillation column has a molar fraction of 95.69% water, 4.30% acetic acid and .0042%
ethanol at 100.174C and 1 atm. The distillation column reboiler uses low pressure steam
at inlet conditions of 135C and 3 atm and outlet conditions of saturated liquid (vapour
fraction = 0) and a pressure of 2.7 atm. The bottom stream which is now composed of
mostly water and hence bacteria is mixed with fresh bacteria broth. Meanwhile, the
distillate stream exits the top of the distillation column at a temperature of 86.95C and
pressure of 1 atm with molar composition of 89.5% water, 1.40% acetic acid and 9.05%
ethanol. Cooling water with inlet conditions of 32C and 1 atm and outlet conditions of
48C and 1 atm is used to condense the vapour from the top of the distillation column.
The distillate stream is then fed into a second distillation column which concentrates the
ethanol.
The second distillation column operates at a condenser pressure of 1 atm and a
reflux to distillate ratio of 0.79 moles. The bottom stream of the second distillation
column exits at a temperature of 97.8C and 1 atm with molar fractions of 97.7% water,
0.73% ethanol and 1.6% acetic acid. The bottoms stream of the second distillation column
also uses low pressure steam at inlet conditions of 135C and 3 atm and outlet conditions
of saturated liquid (vapour fraction = 0) and a pressure loss of 0.25 atm. Meanwhile, the
distillate stream of the second distillation column exits at a temperature of 79C and 1
atm with molar fractions of 21.5% water, 78.4% ethanol and less than 0.1% acetic acid.
Cooling water which enters at 32C and 1 atm and exits at 48C and 0.9 atm is also used
to condense the vapour stream of the second distillation column.
The distillate stream of the second distillation column is sent to a dehydration
process where ethanol with a purity of 99.9% is obtained. Although these conditions are
35
based off of a simulation, the only realistic expected change is a slight increase in the
acetic acid concentration in the final ethanol stream leaving the plant.
5.2 Start-up and shutdown

Clarity and unambiguity in startup and shutdown procedures is necessary in order
for plant operators, engineers and technicians to operate the plant with little variability
and a constant throughput. However, to appropriately address this essential part of
running a chemical plant, each startup and shutdown should begin with awareness and
preparation by operators. In addition, all plant personnel and employees should
familiarize themselves with the correct operation and maintenance procedures. There
should be a checklist for the startup and shutdown procedures for the whole plant as well
as for each individual unit and section of the plant (e.g. for a compressor or cyclone). The
operators who will be starting the procedures should be well trained and confident in their
ability to handle unexpected circumstances. Inter-personnel communication between
plant operators, engineers, technicians and managers is also necessary, and everyone
should work cohesively and with specific tasks and objectives in mind, which should be
decided upon before the plant starts up or shuts down. Lastly, the working materials such
as catalysts, refrigerants or adsorbents and utilities like cooling water or steam should be
readily available for the process as needed. The following is only a general operating
procedure for the miscanthus syngas fermentation plant, with startup and shutdown
sections.
5.2.1 Start-up
1. Run hot pressurized air through the gasifier and it should exit through the flash
column which releases used syngas out of the fermentation tank. This ensures that
the syngas exit out of the gasifier is unhindered and all residual gases/materials
present in the gasifier and the adsorption column, scrubber and fermentation tank
are removed. It also ensures that the valves are working properly and the
equipment, pipes and valves are not blocked. Slowly, the miscanthus feed and
high pressure steam (used to heat the gasifier for pyrolysis to begin) can be added
36
simultaneously, and can be brought to steady state by changing the throughput

from upstream (pretreatment).
2. Startup the hammer mill by first starting up any necessary heaters or burners,
bringing these up to the desired operating temperature, and once these are at set
point, begin introducing feed slowly, and gradually increase the throughput until
the desired steady throughput is achieved.
3. Drying can be started up by running air through the process before bringing in the
feed.
4. First, run a test batch of miscanthus through the hammer mill and drying processes
to see if the miscanthus water content and size is according to specifications that
ensure maximum efficiency through the pyrolysis process in the gasifier.
5. The product from the gasifier exit is not yet run through the gas cleaning
processes, so it exits through a flare side stream.
6. For the heat exchangers, open shell side vent valve to release air or gases, slowly
introduce cooling fluid until shell side is flooded with cooling fluid, shut the shell
side vent valve, open the tube side vent valve to release air or gases, slowly
introduce the syngas until all tube (passes) are filled, close the tube side vent
valve, slowly increase syngas flow rate up to operating conditions. Any utility
processes including condensers, reboilers and heaters should run this process.
7. Fermentation can begin by first running it as a batch process separately before the
syngas is produced and sent through. This batch process is required in order to
facilitate the conditions necessary for the bacteria to survive, namely a
temperature of 37C and around 5-6 pH. Once the bacteria has been introduced
and the conditions are satisfactory, the process can be made continuous by the
introduction of syngas and the release of some broth along with the products
(including ethanol) and the used syngas.
8. For the pump (which is used to pump broth and water for fermentation), make
sure all connections are in place. Close the discharge valve and open the suction
valve. Slowly introduce the water/broth to the pump until the pump suction line
37
fills. Open the discharge valve and start the pump. Once the pump reaches the
desired speed, open the discharge valve to a setting that gives the best efficiency
point.
9. When the gas cleaning, heat exchanger and fermentation processes are ready, the
syngas flow rate to the flare is slowly reduced and instead this flow begins
running through the downstream processes. Steady state is achieved when almost
no syngas exits through the flare.
10. Operation of the distillation column begins by removing undesirable materials in
the column using air or inert gases. The next step is to slowly increase the pressure
inside the column using one of the components of the feed or an inert gas. A small
amount of the feed is then introduced to the distillation column and the column is
ran at total reflux and utilities are turned on. Gradually bring the column into
normal operating conditions.
5.2.2 Shutdown
1. The pretreatment processes including the hammer mill and the drying need to be
shut down simultaneously by first reducing the air into the dryer and subsequently
lowering the feed rate.
2. Let the remaining pyrolysis reactions happening in the gasifier finish after there is
no feedstock entering the gasifier. Gradually reduce the steam flow rate into the
gasifier. Shut down the gasifier after all the miscanthus has burned and the char
and particles at the bottom of the gasifier are cleaned out.
3. Shut down the cyclone, adsorber and wet scrubber when there is no feed.
4. Shut down the water inlet into the fermentation tank when the last syngas has
reacted in the tank. Simultaneously slow down the steam inlet and cooling water
inlet into both distillation columns reboilers and condensers. Slowly, the feed will
exit through the bottoms of the second distillation column. The recycle stream will
still operate until all the broth and water has exited from the second distillation
column. This entire step ensures that the ethanol product stream does not receive
any of the other components.
38
5. Drain the entire system of broth, wastewater and ethanol.

6. Clean all units and removal residual materials.
5.3 Emergency Procedures

1. In the event of an emergency, response teams as well as authorities must be
contacted immediately. This will include on site response as well as off site (ie,
911).
2. Evacuate the site of emergency to ensure minimum amount of damage and
potential casualties.
3. In the event of temperature emergencies, the equipment would be shut down
immediately and the energy cooling system would started in order to contain the
heat.
a. If the heat issue continue, shut down the whole plant to make sure heating
is not coming from another part of the plant.
b. If a quench system is available on plant, initiate as soon as equipment is
shut down.
4. In high pressure emergency situations, shut down the equipment in question and
sections prior in attempt to reduce the pressure.
a. Open all of the relief valves close to the high pressure area
b. If the pressure continue to increase on larger units pass critical point, alert
all parties in and around the plant and evacuate as quickly as possible
c. In case of pressure being too low in a part of the plant to the point of
causing a vacuum, the same procedures as the high pressure emergency
situation can be used.
5. Due to the high temperature nature of the gasifier and the streams/units
afterwards, any leaks in that part of the plant may result in large amount of
damages.
a. Shut down the gasifier to stop the production of high temperature syngas.
b. Initiate quench system and/or emergency cooling system if available.
c. Evacuate the plant floor
39
d. In case of fire in the plant, activate the fire alarm and alert
authorities.
d. Shut down the plant and evacuate.
e. Allow authorities and emergency response teams to resolve the issue.
6. In case of spilling/leaking from pipes or units (liquid or vapour), the spill/leak
must be located and shut down or the process must be re-routed.
a. Evacuate areas of spill.
b. Any spills must be cleaned up as soon as possible to prevent any potential
chemical damage or fires
c. Clean up spills using spill pillows or equivalent substance which can
absorb the spill
d. In case of gas leak from the process, make sure the plant is fully sealed in
order to ensure the gas does not escape to the atmosphere
e. Check for toxicity before returning to the plant after spill incident
f. If contacted with liquid or vapour substance, use proper treatment methods
to disinfect contacted area.
6. Overall Material and Energy Balances

6.1 Overall Material Balance:
Table 3 displays the overall material balance of the process streams of the
bioethanol plant starting from the pre-treatment section which includes drying and
milling, to the gasifying unit, gas cleaning section, fermentation unit and the ethanol
separation section. Table 3 includes process stream names and the total mass flows in
kg/hr for each stream entering and exiting the bioethanol plant.
40
6.1.1 Process Side

Table 3. Process side material inflows and outflows of the bioethanol plant
PROCESS SIDE OVERALL MATERIAL BALANCE
INFLOW
Component
Raw
Steam
Mat
Total
Total mass
flow
122,728
93,992
247,760
OUTFLOW
Fresh
O2-
NH3
Broth
H2O
Removal
31,040
70,846
735
Waste
WWT
DC-2
Ethanol
Bottom
127,650
952
38,406
8,819
Evaporated
Cyclone
H2S
Water
underflow
removed
17
218
247,762
6.1.2 Utilities side

Table 4 and Table 5 displays the overall material balance of the utilities used throughout the bioethanol plant starting
from the pre-treatment section, to the gasifying unit, gas cleaning section, fermentation unit and ethanol separation section.
Table 4 includes the utility material inflows while Table 5 includes the utility material outflows. All numbers have units of
kg/hr. Utilities used throughout the bioethanol plant include medium pressure steam (DRYER-STEAM) utilized in the dryer
with inlet conditions of 163C and 7.8 atm and outlet conditions of 120C and 2 atm. Meanwhile, cooling water (CW1, CW2,
CW3 ,CW4) are used for all heat exchangers with inlet conditions of 37C and 1 atm. Low pressure steam (stream name of
BOTTOMS1 and BOTTOMS2) with inlet conditions of 135C and 3 atm is used for the reboiler of the distillation columns
while cooling water (DIST1-CW and DIST2-CW) at inlet conditions of 37C and 1 atm is used for the total condenser of the
distillation columns. Cooling water (CW1 and CW2) exit heat exchangers E-101 and E-102 as low pressure steam (stream
119
name of LPS1 and LPS2 respectively) at conditions of 135C and 3 atm. On the other hand, streams CW3 and CW4 exit heat
exchangers E-103 and fermenter cooling jacket respectively at conditions of 49C and 1 atm. Low pressure stream
(BOTTOMS1-LPS and BOTTOMS2-LPS) exit as saturated liquid (vapour fraction of zero) and a pressure drop of 0.3 atm was
assumed.
Table 4. Utilities side overall material inflows of the bioethanol plant
UTILITIES SIDE OVERALL MATERIAL INFLOWS
Component
Total
DRYERSTEAM
69,945
CW1
CW2
CW3
153,766
89,667
205,641
DIST1-CW
CW4
6,469,180
Total mass
1,490,110
DIST2CW
827,897
BOTTOMS1LPS
60,645
BOTTOMS2-LPS
27,262
9,520,617
flow
Table 5. Utilities side overall material outflows of the bioethanol plant

UTILITIES SIDE- OVERALL MATERIAL OUTFLOWS
DRYERComponent
STEAM
LPS1
LPS2
CWR3
CWR4
EXIT
Total
69,945
153,766
89,667
205,641
6,469,180
DIST1-CW
DIST2-CW
RETURN
RETURN
1,490,110
9,520,617
Total mass flow
42
827,897
BOTTOMS1
-LPS
RETURN
60,645
BOTTOMS2-LPS
RETURN
27,262
6.2 Overall Energy Balance

Table 6 displays the overall energy balance of the process streams of the bioethanol plant starting from the pretreatment section which includes drying and milling, to the gasifying unit, gas cleaning section, fermentation unit and the
ethanol separation section. Table 6 includes process stream names and the total enthalpy flow in kg/hr for each stream entering
and exiting the bioethanol plant.
6.2.1 Process Side
Table 6. Overall energy balance of the process streams
PROCESS SIDE OVERALL ENERGY BALANCE
INFLOW
Energy
Raw Mat
Steam
Total Fluid
-356,727
-1,099,100
OUTFLOW
Fresh
Broth
-495,460
O2-H2O
-16,262
NH3
Waste
Removal
-1,963
-61,705
WWT
-15,014
DC-2
Bottom
-575,230
Ethanol
-52,071
Evaporated
Cyclone
H2S
Water
underflow
removed
-217,910
-1.7
-10,382
Enthalpy Flow
6.2.2 Utility Side

Table 7and Table 8 displays the overall energy balance of the utility streams of the bioethanol plant starting from the
pre-treatment section which includes drying and milling, to the gasifying unit, gas cleaning section, fermentation unit and the
43
ethanol separation section. Table 8 includes the incoming utility stream names while Table 8 includes the outgoing utility
stream names and the total enthalpy flow in MJ/hr for each stream entering and exiting the bioethanol plant.
Table 7. Overall energy inflow of the utility streams . All values are in MJ/hr, except temperature which is in units of C and pressure in units of atm
UTILITIES SIDE: ENERGY INFLOW
Energy
Total Fluid Enthalpy
Flow
CW1
CW2
CW3
CW4
DRYER-STEAM
-2,451,400
-1,427,900
-3,278,400
-103,130,000
73,485
Energy
DIST1-CW
DIST2-CW
DIST1-LPS
DIST2-LPS
Fluid Enthalpy Flow
-23,737,000
-13,188,000
-937,570
-421,480
Electricity
Fluid Enthalpy Flow
ELEC
.0025
Electricity
Energy
MILLING-
P-101
P-102
ELECTRICITY
ELECTRICITY
8x10-4
1.3x10-3
P-103
P-104
P-105
P-106
P-107
ELECTRICITY
ELECTRICITY
ELECTRICITY
ELECTRICITY
ELECTRICITY
1.3x10-3
1.3x10-3
1.3x10-3
1.3x10-3
1.3x10-3
44
Table 8. Overall energy inflow of the utility streams . All values are in MJ/hr, except temperature which is in units of C and pressure in units of atm
UTILITIES SIDE: ENERGY OUTFLOW
DRYEREnergy
LPS1
LPS2
CWR3
CWR4
STEAM
EXIT
Fluid Enthalpy Flow
-2,356,900
-1,384,700
-3,260,300
-102,570,000
Energy
DIST1-CW RETURN
DIST2-CW RETURN
DIST1-LPS RETURN
DIST2- LPS RETURN
Fluid Enthalpy Flow
-23,633,731
-13,130,424
-1,069,866
-480,871
69,302
6.3 Stream and Equipment tables

6.3.1 Process Side
Table 9 and Table 10 show the stream tables and equipment table respectively for the pre-treatment section of the
bioethanol plant. All mass flows are per mass basis with units of kg/hr while total enthalpy flows are in units of MJ/hr.
Pre-treatment section stream table
45
Table 9. Stream table of process streams entering and exiting the pre-treatment section of the bioethanol plant.
Raw Mat.
Evaporated Water
Dried Feed
Miller Feed
Milled Feed
55,689
55,689
55,689
55,689
H2
523
523
523
523
N2
604
604
604
604
O2
64,776
64,776
64,776
64,776
H2O
28
17
11
11
11
CL2
789
789
789
789
101
101
101
101
Solids
218
218
218
218
122,728
17
122,711
122,711
122,711
-356,727
-10,382
-350,498
-350,498
-350,498
Component
Total Flow
Total Enthalpy
Flow
46
Table 10. Material and energy inflow and outflow to the equipment of the pre-treatment section of the bioethanol plant
Component
PT-101 [BELT DRYER]

IN
OUT
PT-102 [HAMMER MILL]

IN
OUT
55689
55689
55689
55689
H2
523
523
523
523
N2
604
604
604
604
O2
64776
64776
64776
64776
H2O
28
17
11
11
11
CL2
789
789
789
789
101
101
101
101
Solids
218
218
218
218
Total Flow
122,728
17
122,711
122,711
122,711
Total Enthalpy Flow
-356,727
-10,382
-350,498
-35,0498
-35,0498
Table 11 and Table 12 show the stream tables and equipment table for the gasification section of the bioethanol plant.
Component and total flows are per mass basis with units of kg/hr while total enthalpy flows are in units of MJ/hr.
Gasification section stream table
47
Table 11. Stream table of process streams entering and exiting gasification section of the bioethanol plant.
STEAM
MILLED FEED
SYNGAS
O2-H2O
55,689
H2
523
10112
N2
604
O2
64,776
64776
CH4
355
CO
121799
CO2
11709
H2O
93,992
11
6070
NH3
735
HCL
812
H2S
119
CL2
789
101
Solids
218
218
93,992
122,711
145,859
70846
-1,099,100
350,498
-342,170
-16,262
Component
Total Flow
Total Enthalpy Flow
48
Table 12. Material and energy inflow and outflow to the equipment of the gasification section of the bioethanol plant.
R-101 [GASIFIER]
Component
IN
OUT
55,689
H2
523
10,112
N2
604
O2
64,776
64,776
CH4
355
CO
121,799
CO2
11,709
H2O
93,992
11
6,070
NH3
735
HCL
812
H2S
119
CL2
789
101
Solids
218
218
93,992
122,711
145,859
70,846
-1,099,100
350,498
-342,170
-16,262
Total Flow
Total Enthalpy Flow
Table 13 and Table 14 show the stream tables and equipment table for the gas cleaning section of the bioethanol plant.
All flows are per mass basis with units of kg/hr while total enthalpy flows are in units of MJ/hr.
49
Table 13. Stream table of process streams entering and exiting the gas cleaning section of the bioethanol plant.
Component
SYNGAS
CYCLONE
CYCLONE
OVERFLOW
UNDERFLOW
PRECOOLED
SYNGAS
CLEAN
H2S
COOLED
FERMENTER
NH3
SYNGAS
REMOVED
SYNGAS
FEED
REMOVAL
H2
10,112
10,112
10,112
10,112
10,112
10,112
CH4
355
355
355
355
355
355
CO
121,799
121,799
121,799
121,799
121,799
121,799
CO2
11,709
11,709
11,709
11,709
11,709
11,709
NH3
735
735
735
735
735
735
HCL
812
812
812
812
812
812
H2S
119
119
119
119
119
Solids
218
218
Total Flow
145,859
145,641
218
145,641
145,641
119
145,522
144,906
735
-342,170
-560,080
217,910
-654,220
-436,300
-1.7
-587,620
-585,660
-1,963
Total Enthalpy
Flow
50
Table 14. Material and energy inflow and outflow to the equipment of the gasification section of the bioethanol plant.
GC-101 [CYCLONE]
Component
H2
IN
OUT
E-101
GC-102 A/B
E-102
GC-103
[Heat exchanger]
[Adsorption column]
[Heat exchanger]
[Ammonia scrubber]
IN
OUT
IN
OUT
OUT
IN
OUT
IN
OUT
OUT
10,112
10,112
10,112
10,112
10,112
10,112
10,112
10,112
10,112
10,112
CH4
355
355
355
355
355
355
355
355
355
355
CO
121,799
121,799
121,799
121,799
121,799
121,799
121,799
121,799
121,799
121,799
CO2
11,709
11,709
11,709
11,709
11,709
11,709
11,709
11,709
11,709
11,709
NH3
735
735
735
735
735
735
735
735
735
735
HCL
812
812
812
812
812
812
812
812
812
812
H2S
119
119
119
119
119
119
145,859
145,641
218
145,641
145,641
145,641
119
145,522
145,522
145,522
145,522
144,787
-336,600
-560,080
-217,910
-560,080
-645,220
-645,220
-1.7
77,605
436,300
587,620
587,620
585,660
Total Flow
Total Enthalpy
Flow
Table 15 and Table 16 shows the stream table and equipment table for the fermentation section of the bioethanol plant.
51
-1,963
Table 15. Stream table of process streams entering and exiting the fermentation section of the bioethanol plant.
COOLED
PUMP
MIXED
MIXED
BROTH
BROTH
11,709
H2O
NH3
HCL
FERMENTER
FRESH
RECYLED
MIXED
FEED
BROTH
BROTH
BROTH
10,112
CH4
355
CO
121,799
CO2
Component
H2
FERMENTER
DC-1
PRODUCT
FEED
315
355
6188
31,040
56,531
87,571
87,571
87,571
93,274
93,274
308
812
812
H2S
WASTE
Ethanol
9,509
9,509
93
Acetic Acid
8,467
8,467
8,467
8,467
10,398
10,398
119,579
Total Flow
144,787
31,040
65,003
96,043
96,043
96,043
113,181
113,181
127,650
-585,660
-495,460
-942,330
-1,437,800
-1,455,800
-1,455,800
-1,614,500
-1,614,500
Total Enthalpy
Flow
52
-61,705
Table 16. Material and energy inflow and outflow to the equipment of the fermentation section of the bioethanol plant.
R-102 [Fermentation vessel]
Component
IN
E-103 [Heat exchanger]
OUT
IN
P-101 A/B [PUMP]
OUT
IN
P-102 A/B [PUMP]
OUT
IN
OUT
H2 [kg/hr]
10,112
315
CH4
355
355
CO
121,799
CO2
11,709
6,188
H2O
87,571
93,274
308
87,571
87,571
87,571
87,571
93,274
93,274
NH3
HCL
812
812
H2S
Ethanol
9,509
93
9,509
9,509
Acetic Acid
8,467
10,398
119,579
8,467
8,467
8,467
8,467
10,398
10,398
Total Flow [kg/hr]
144,787
96,043
113,181
127,650
96,043
96,043
96,043
96,043
113,181
113,181
Total Enthalpy
Flow [MJ/hr]
585,660
-1,437,800
-1,614,4500
-61,705
-1,437,800
-1,455,800
-1,455,800
-1,455,800
-1,614,500
-1,614,500
Table 17 and Table 18 shows the stream table and equipment table for the separation section of the bioethanol plant.
53
Table 17. Stream table of process streams entering and exiting the separation section of the bioethanol plant.
DC-1
DC-1
FEED
BOTTOM
H2O
93,274
113,062
Ethanol
9,509
Acetic Acid
Total Flow
Component
Total Enthalpy
Flow
Component
DC-1
RECYCLED
Boilup
BROTH
113,062
56,531
56,531
40,417
40,417
40,417
3674.3
36,743
`12
`12
10,452
10,452
10,452
950.2
9,502
10,398
16,934
16,934
8,467
8,467
2,125
2,125
2,125
193.2
1,932
113,181
130,008
130,008
65,003
65,003
52,994
52,994
52,994
4817.7
48,177
-942,330
-942,330
-64,372
-643,720
HEATED
BOTTOM
-1,614,500
DC-2
BOTTOM
DC-2
HEATED
BOTTOM
DC-2
DC-2
Boilup
BOTTOM
DC-2 TOP
DC-1 TOP
DC-2 TOP
CONDENS
DC-1 TOP
DC-1
DC-1
CONDENS
DC-2
VESSEL
EXIT
VESSEL
EXIT
DC-1
REFLUX
DC-2
DC-2
REFLUX
DISTILLATE
DC-1 DISTILLATE
ETHANOL
WWT
H2O
60,844
60,844
25,053
35,791
1,736
1,736
1,736
784
952
952
Ethanol
1,161
1,161
478
683
15,752
15,752
15,752
6933
8,819
8,819
Acetic Acid
3,284
3,284
1,352
1,932
1.6
1.6
1.6
.07
.09
0.09
Total Flow
65,289
65,289
26,884
38,406
17,405
17,405
17,405
7,634
9,771
8,819
952
-281,863
-402,661
-52,412
-66,673
-52,071
Total Enthalpy
Flow
54
15,014
Table 18. Material and energy inflow and outflow to the equipment of the separation section of the bioethanol plant.
T-101
Component
IN
E-104
OUT
P-104
E-105
P-105
IN
OUT
IN
OUT
IN
OUT
IN
OUT
H2O
93,274
56,531
36,743
113,062
113,062
113,062
113,062
40,417
40,417
40,417
40,417
Ethanol
9,509
9,502
`12
`12
`12
`12
10,452
10,452
10,452
10,452
Acetic Acid
10,398
8,467
1,932
16,934
16,934
16,934
16,934
2,125
2,125
2,125
2,125
Total Flow
113,181
65,003
48,177
130,008
130,008
130,008
130,008
52,994
52,994
52,994
52,994
-1,614,500
-942,330
-643,720
Total Enthalpy Flow
T-102
Component
IN
E-106
OUT
P-106
E-107
P-107
IN
OUT
IN
OUT
IN
OUT
IN
OUT
H2O
36,743
35,791
952
60,844
60,844
60,844
60,844
1,736
1,736
1,736
1,736
Ethanol
9,502
683
8,819
1,161
1,161
1,161
1,161
15,752
15,752
15,752
15,752
Acetic Acid
1,932
1,932
.09
3,284
3,284
3,284
3,284
1.6
1.6
1.6
1.6
Total Flow
48,177
38,406
9,771
65,289
65,289
65,289
65,289
17,405
17,405
17,405
17,405
-643,720
-575,230
-66,673
Total Enthalpy Flow
6.3.2 Utilities Side

Table 19 and Table 20 show the stream table and equipment table respectively for the utility streams of the entire
bioethanol plant section of. Component and total flows are per mass basis with units of kg/hr while total enthalpy flows are in
units of MJ/hr.
55
Table 19. Stream table of utilities used in the entire bioethanol plant.
Component
H20
DRYER-
DRYER-
CW1
LPS1
CW2
LPS2
CW3
CW3 RETURN
STEAM
STEAM EXIT
69,945
69,945
153,766
153,766
89,667
89,667
205,641
205,641
-73,485
-69,302
-2,451,400
-2,356,900
-1,427,900
-1,384,700
-3,278,400
-3,260,300
CW4
CWR4
DIST1-CW
DIST1-CW
DIST2-CW
DIST2-CW
Total
Enthalpy
Flow
Component
RETURN
H20
DIST1-LPS
RETURN
DIST1-LPS
DIST2-
RETURN
LPS
6,469,180
6,469,180
1,490,110
1,490,110
827,897
827,897
60,645
60,645
27,262
103,130,000
102,570,000
23,737,000
23,633,731
13,188,000
13,130,424
937,570
1,069,866
421,480
DIST2-
P-101
P-102
P-104
P-105
P-106
P-107
MILLING
LPS
ELECTRICITY
ELECTRICITY
ELECTRICITY
ELECTRICITY
ELECTRICITY
ELECTRICTY
ELECTRICITY
8x10-4
1.3x10-3
1.3x10-3
1.3x10-3
1.3x10-3
1.3x10-3
.0253
Total
Enthalpy
Flow
Component
RETURN
H20
827,897
Total
Enthalpy
Flow
Electricity
-480,871
56
Table 20. Utilities side material and energy inflow and outflow to the equipment of the entire bioethanol plant
Component
H20
PT-101
PT-102
E-101
E-102
E-103
69,945
69,945
153,766
153,766
89,666
89,666
205,641
205,641
73,485
69,302
-2,451,400
-2,356,900
-1,427,900
-1,384,700
-3,278,400
-3,260,300
Total
Enthalpy
Flow
Electricity
.0253
Component
H20
E-104
E-105
E-106
E-107
P-101
56,531
56,531
1,490,110
1,490,110
27,262
27,262
827,897
827,897
-937,570
-1,069,866
-23,737,000
-23,633,731
-421,480
-480,871
-13,188,000
-13,130,424
Total
Enthalpy
Flow
8x10-4
Electricity
Component
Electricity
P-102
P-103
P-104
P-105
P-106
P-107
1.3x10-3
1.3x10-3
1.3x10-3
1.3x10-3
1.3x10-3
1.3x10-3
57
7. Process Control
7.1 Control Overview
This section of the report covers all the controls that are added to P&ID. The
controls are arranged in the order in which they are presented in the P&ID. There are
controls on the same type used more than once in the system and will be described only
once here.
Starting from the gasifier unit (R-101), we have steam and feedstock coming in at
a specified ratio. Now we know that flow of steam and air can fluctuate and therefore
introduce a disturbance into our system. Therefore we have implemented a ratio control
for this section, in order to keep the ratio of feed to steam constant, entering the gasifier
unit. Figure 1.0 displays how this ratio control is applied to the system.
For many processes, a key objective is to
maintain the flow rates of two process steams in some
proportion to one another. In such cases, ratio control is
applied. When ratio control is applied, one process
input, the dependent input, is proportioned to the other
process input, known as the independent input. The
independent input may be a process measurement or its
set point. The proportion that needs to be maintained is
between the two inputs is known as the ratio. In Figure
9, the independent input measurement is the flow rate of
feed coming into the gasifier. The ratio controller sets
the set points of the flow controller rather than the valve
position, as illustrated in Figure 9.
Figure 9: Ratio control loop design for steam to feed ratio
Thus any nonlinearity installed characteristics associated with the valves is

addressed by the flow controllers and has no impact on the ratio controller being able to
maintain the ratio set point. The input to the ratio control is the measurement or set point
of the independent flow which is the flow of feed coming in to the gasifier. The ratio
controller, multiplies this measurement to the ratio, to determine the set point of the
dependent flow which is steam in our case. The set point of dependent flow will be sent to
the flow controller that will manipulate valve v-6 shown in Figure 9 in order to keep the
ratio constant.
The measurement of both flows must be done as close to the gasifier unit to avoid
any time delay in the response. Therefore as shown in Figure 9 the measurement of flow
is taken right before the streams enter the gasifier unit. Controller type PI will be used in
this situation because they eliminate forced oscillations and steady error resulting in
operation of on-off controller and P controller respectively. However, introducing integral
mode has a negative effect on speed of the response and overall stability of the system.
Since in our case we are using a flow controller to manipulate the valve position, we
know that this type of control has a very fast response time, therefore we dont have to
worry about the delay in response introduced by the integral mode. Furthermore, integral
mode will eliminate any off-set that is present in the system.
There are other section of the P&ID where ratio control is implemented. This
included controlling the ratio between Feed entering each distillation column and the
reboiler utility. Sometimes in the process we might decide to increase our production, so
once we do that we need to change the reboiler duty as well. Therefore instead of a person
going and manually adjusting the utility flow of the reboiler, we will use a ratio control to
keep the flow of steam constant with the coming feed in the column. Furthermore a
disturbance can occur and alter the flow of steam by either decreasing it or increasing it.
This disturbance will also be eliminated by the ratio control strategy shown in Figure 10.
Note that we have two distillation columns in our process and only the first one has this
type of control strategy. This is because we are already using the feed entering the second
distillation column to control the level of the reflux drum of the first distillation column.
59
Therefore we cant have two controllers fighting and didnt apply this strategy to the
second column.
Figure 10: Ratio control design for Feed to reboiler utility ratio
The idea is similar to what was described in the previous ratio controlled system.
In this scenario, the independent variable for ratio control is the measurement of the feed
entering the column. This measurement is sent to the ratio control that multiplies it to the
ratio and sends an output to the flow control which adjusts the position of valve V-496, as
shown in Figure 10. The dependent variable in this case is the flow of LPS entering the
reboiler.
Similar to the previous case, flow measurements are taken right before the streams
enter their desired unit. Measurement of feed flow is taken before it enters the distillation
column and measurement of LPS flow is taken before it enters the reboiler. All this is
done to avoid any type of time delay in the response of the control system. A PI controller
type is suitable for this control system because as mentioned earlier they will eliminate
forced oscillations and steady state error resulting in operation of on-off controller and P
controller selectively. A key point to note which wasnt mentioned earlier is that PI
60
controller does not increase the speed of response. It can be expected since PI controller
does not have any means to predict what will happen with the error in the near future.
This problem can be solved by introducing the derivative mode which has ability to
predict what will happen with the error in the near future and thus to decrease a reaction
time of the controller. From our prior knowledge of control theory, flow controllers have
a very fast response time and therefore we have concluded just to use a PI controller for
the ratio control applied to this distillation column.
Temperature inside a distillation column is one of the variables that needs to be
controlled. This is because distillation is temperature dependent; any variation in
temperature will cause the purity of the product stream to decrease. A ratio control
structure for this system is shown in Figure 11.
Figure 11: Ratio control structure between distillate and reflux to maintain a steady temperature inside the
column
The independent measurement for the ratio control is the flow of distillate leaving
the reflux drum. This measurement is sent to a ratio control that multiplies it to the ratio.
However another independent measurement is sent to the ratio control and this is the
61
temperature inside the column. The ratio control compares the set point of temperature to
the set point stored by the operator and takes action by manipulating valve V-105 which
either decreases or increases the reflux flow depending on what is the situation. The flow
of reflux is measured right before it enters the distillation column and the flow of
distillate is measured right after the stream leaving reflux drum (V-102) splits. Since this
temperature control will affect the purity of the product stream, we want the temperature
measurement to be taken from the top trays of the distillation column. A similar approach
is used to determine the location of temperature measurement for the 2nd distillation
column, it will be measured from one of the trays at the top.
A PID controller is used for this control structure. They have all the necessary
dynamics including fast reaction on change on controller input (D mode), increase in
control signal to lead error towards zero (I mode) and suitable action inside control error
are to eliminate oscillations (P mode). The reason behind using a PID controller is that
this is the most important area in our system and it effects the purity of our desired
product, therefore we want the control system to be perfect in all aspects mentioned
above. The derivative mode improves the stability of the system and enables increase in
gain K and decrease in integral time constant Ti, which increases speed of the controller
response. From our prior knowledge or process control, we know that temperature control
in systems have a slower response time, therefore we need to a controller type that is fast
and will not change the temperature of the system and therefore keep the purity constant.
Note that this type of control structure is applied to the second distillation column as well
and therefore we havent shown it again as all the parts are similar to what is described
above.
Some of the units in the process require pressure control inside. This is because
high pressure can lead to explosion of the unit and therefore cause damage to the
surrounding units and also might kill workers around that area. Gasifier unit is one of the
reactors that has a pressure control used in it. Figure 12 shows the control loop structure
for pressure control in the gasifier unit (R-101)
62
Figure 12: Pressure control inside the gasifier unit
This is a regular single loop feedback control system which has a control variable
and a manipulated variable. The controlled variable is the pressure inside the unit and the
manipulated variable is the flow of the product stream leaving the gasifier unit. When the
pressure inside the unit is too high, valve V-2 will open to push more vapour out the unit
in order to decrease the pressure. When the pressure is too low, the valve will close
slightly to keep the vapour inside the unit so that the pressure can reach its desired set
point. Since gas is being formed in this reactor, the pressure measurement can be taken
anywhere on the top reactor. However the measurement needs to be taken away from the
outlet stream in order to avoid any errors in the reading.
A PI controller will be a suitable type of controller for this system. Since pressure
control by themselves have a fast reaction time, therefore we dont need a derivative
mode in this situation. We still require an integral mode to remove the offset and a
proportional mode to eliminate any oscillations.
Another section of the system where pressure control is used are two distillation
columns. Both the columns have a similar structure of the pressure control system and
therefore only one is explained in detail here. Figure 13 shows the structure of pressure
control loop designed on the distillation column.
63
Figure 13: Pressure control loop design for the distillation columns
This type of pressure control strategy is similar to that which was used in the
Gasifier unit (R-101). However this one has a different manipulated variable. In order to
control the pressure inside the column, we are manipulating the cooling water flow
entering the condenser. The idea is to reduce vapour accumulation at the top of the
distillation column during high pressure scenarios. In a high pressure scenario, the
position of valve V-83 in Figure 13 will open more to let more cooling water enter the
condenser. This eventually will liquefy more vapour and therefore will decrease the
pressure inside the column. The pressure sensor must be placed somewhere on top of the
column away from the exit location to avoid any errors in the measurement.
A PID controller type will be suitable for this system. Even though we have
mentioned earlier that pressure control is fast in terms of dynamics but in this scenario we
time delay. When the cooling water flow will increase, it will take time for the vapour to
condense in the condenser and therefore it will take time for the pressure reading to
64
change inside the column. So we want a controller that has fast response and can predict
the future error using the derivative mode.
Sometimes when we are dealing with liquid systems, we have a problem of
flooding in the vessel. In order to avoid this situation, level control is applied to different
vessels that are used in the system. One of this includes level control inside the fermenter
unit (R-102). Since the product is liquid leaving the fermenter, we do not want the liquid
to fill up the fermenter and therefore flood it. A simple level control structure for this
system is shown in Figure 14. Level control is also a single loop feedback control that
uses a control variable and a manipulated variable. The control variable in this case is the
level of liquid inside the fermenter and the manipulated variable is the valve position of
V-550, which changes the flow of product stream leaving the fermenter.
Figure14: Level control structure for the fermenter
This type of level sensor calculated the hydrostatic pressure inside the unit at two
different heights and the difference gives us the level of liquid in the tank. The location of
these hydrostatic sensors is determined by what is the maximum allowable level that can
lead to safe operation. The minimum level is determined by how fast we are pumping the
65
liquid out. Since pumps cannot run dry, some level of liquid needs to be present all the
time in order to avoid dry operation of the pump. If the level inside is high, the level
controller will send a signal to valve V-550 which will open more in order to reduce the
level inside the fermenter and bring it back to the desired set point given by the operator.
A PI control type is suitable for this situation. Since level controls have a fast
response time and the hydrostatic measurement itself is automated, therefore applying a
derivate mode here wont make a difference in terms of improving the response of the
system.
Reflux drum vessels used to store the liquid after the condenser in a distillation
column also requires level control to avoid flooding of the vessel. Figure 15 shows the
control loop structure design of the level control used for V-101. Note that the reflux
drum for the second distillation column has a similar control strategy being applied and
therefore is not mentioned in detail.
Figure 15: Level control for reflux drum
66
The level control here uses a similar strategy as the previous level control system.
Two hydrostatic pressure measurements are taken from the vessel and the corresponding
output is sent to a level controller which manipulates the position of valve V-546. This
type of level control uses a single feedback control loop designed with a PI type
controller. As mentioned above that we dont need a derivative mode since we are not
looking for an improvement in the control system, its already operating at optimum
conditions. From prior knowledge we can conclude that this type of response behaviour
between level and flow is stable and therefore further stability in the system is not
required by the input of derivative mode.
When a heat exchanger is used to heat up a desired stream or cool it down.
Temperature of the product stream leaving the heat exchanger must be controlled. This is
because any fluctuations in the flow or temperature of the utility stream can cause
deviation in the temperature of the product stream exiting the heat exchanger and we
might not get the desired temperature output that we are looking for in the product stream.
Therefore a cascaded temperature control strategy is applied to all the heat exchangers
and condenser used in the process and only one of them is explained in detail here. Figure
16 shows how one these heat exchangers have a cascaded temperature control being
implemented.
67
Figure 16: Cascaded temperature control design around the condenser E-104
A certain degree of flexibility needs to be added around the heat exchanger and
condensers, since the flow rate of the CW coming in might fluctuate and act as a
disturbance to the temperature of the product stream leaving the exchanger. To avoid this
disturbance, a cascade loop is implemented around the condenser as shown in Figure 16.
The inner loop of this cascade control is measuring the flow of the CW and controlling it
by manipulating the pneumatic valve V-68. This valve is also labelled fail closed because
during a failure, if the valve is in the closed position the utility is not wasted. The outer
cascade loop measures the temperature of the product stream exiting the condenser, and
this is the set point for the inner cascade loop. Therefore 2 controllers are required for
keeping the temperature of the product stream constant when a disturbance occurs in the
flow of CW. The good thing about having a cascade control is that the inner loop will
have a much faster dynamic response than the outer loop, therefore a disturbance will
have a minimal effect on the temperature of the product stream. The temperature sensor
which measures the temperature of the product stream exiting the heat exchanger is
68
located right after the stream exits the heat exchanger to avoid the disturbance in reading
from any temperature losses along the pipe.
The flow sensor for the measurement of CW flow is located before the pneumatic
valve V-68. The sensor could also be located after the pneumatic valve but since the
pressure in the change will change on both sides of the valve, locating the flow sensor on
either side will not have a drastic effect on the control strategy being applied. In other
words it doesnt matter where we put the flow sensor, we are going to get the same
control behaviour. However we cannot have the sensor far away from the heat exchanger,
since there will be error in the reading due to pressure losses along the pipe.
A PID type controller is best suitable for this control scheme. This is because
temperature change will take time once the flow rate of the utility is increased therefore
the system dynamics will be slow giving rise to a very high time delay in the control
response. The derivative mode will decrease the integral time constant Ti as mentioned
earlier, and will therefore increase the speed of the controller response. A key point to
note here about derivative mode that wasnt mentioned in the previous sections, is that
its not taken from the error signal but rather from the system output variable. This is
done to avoid effects of the sudden change in the value of error signal. Sudden change in
error signal will cause sudden change in control output. To avoid that it is suitable to
design D mode to be proportional to the change of the output variable.
pH is another parameter that needs to be controlled in a system where there is a
pH sensitive medium. In our case, the bacteria in the fermenter works best at a pH of 6,
therefore it is our goal to keep the pH of the fermenter constant at 6 for optimum
conditions. A cascade control structure is applied for this pH control, Figure 17 shows
how the control loop is designed around the fermenter.
69
Figure 17: pH control loop structure for the fermenter.
The reason behind using a cascade control for this system is that pH control itself
is very slow in terms of dynamics of the system. We are using a single pH probe to
measure the pH of a huge fermenter, it will take a lot of time for the pH of the fermenter
to change, and therefore we need a very fast control system for this case. The pH
measurement is sent to a flow controller which manipulates the position of valve V-78.
The inner loop which controls the flow has a much faster response than the outer loop
which controls the pH and therefore the inner loop will run much faster than the outer
loop, keeping the pH constant at the desired set point.
This type of cascade control will involve using a PID control. Since we already
have a slow response time in the system, we need to have a PID type controller so that the
D mode improves the stability and increases the speed of the controller response. The pH
measurement needs to be taken at the bottom of the liquid level, away from any mixer or
inlet or exit. I have previously done a pH control lab and the problem in that was the
location at which pH measurement was taken. Therefore its very important that the pH
probe is located in the area where there are no fluctuations in flow and we have a steady
flow profile.
Since we are recycling our broth back into the fermenter, we need to purge some
of it out of the system to avoid any accumulation inside our units. The purge stream needs
70
to be in a certain ratio with the recycle stream, this is achieved by having a ratio control
as shown in Figure 18. Since the flow of recycle stream can change and act as a
disturbance, we want to fix our flow of purge so that we dont remove extra stuff from
our system and save cost.
Figure 18: Ratio control structure between purge stream and recycle stream
As mentioned in the earlier sections, ratio control involves independent and

dependent variables. The independent variable is the flow of recycle stream and the
dependent variable is the flow of the purge stream. The independent measurement of the
recycle flow is sent to the ratio controller that multiplies it by the ratio and manipulates
the position of valve V-562 accordingly. The flow sensor of the recycle stream is attached
right before the split is made to avoid any error is measurement that can be caused by
pressure losses in the pipe. Similarly, the flow measurement for purge stream is done
right after the spilt is made to avoid any errors due to pressure losses.
Using a PI controller alone without the derivative mode will be suitable for this
case since we are only doing a ratio control with flow which has fast control dynamics.
Lastly we need to control the pH inside the unit S-101. Since everything exiting S-101
enters our fermenter, we dont wait to disturb the pH of the fermenter and therefore we
71
need to keep S-101 at out desired pH level of 6. Figure 19 demonstrates how this type of
control structure is designed around the unit S-101. This type of control design also
involves a ratio control between recycle stream and the fresh medium stream entering the
unit S-101. The independent measurement will be the flow rate of recycle stream and the
dependent measurement will be the flow rate of the fresh medium stream. Since the flow
rate and pH of the fresh medium stream can change and therefore can act as a source of
disturbance in our system, it needs to be controlled.
Figure 19: pH control design for unit S-101
The pH measurement from the unit S-101 is sent to the ratio controller which
compares it to the set point of pH that we have defined already in the system, along with
72
this input the ratio control also gets an input from flow of the recycle and accordingly
adjusts the flow of the fresh medium entering the unit. The location of the flow sensor for
the recycle is done right before the mixing point so that we can avoid errors due to
pressure drop in the pipes. Similarly, the location of the flow sensor for the fresh medium
is right before the mixing point to avoid any errors in flow measurement due to pressure
losses in pipe that can lead to drop in the flow across the pipe. In order to get an accurate
pH measurement inside the unit S-101, the probe needs to be fully emerged in the liquid
medium all the way till the bottom, away from any inlet and outlet, to avoid any error in
pH measurement due to flow fluctuations.
We need to use a PID controller for this section because as mentioned earlier pH
system dynamics are very slow and we need a derivative mode for fast response of the
controller. The derivative mode will look at the slope of the error and decide what action
to take. In other words we will have a feedforward control strategy being applied by
looking into the future of our error and predicting what its going to be so that we can
take the appropriate action in the present.
This brings us to the end of all the control loops that are used in the system. Now
the final copy of all the sections of the P&ID will be presented that consist of all the
control loops that were mentioned above. It also contains those control loops that were
mentioned earlier but not explained in detail because they had the same control design as
those which were mentioned.
7.2 Preliminary P&ID of Process

The process is split into different sections so that the P&ID can fit into this
document. The first section which doesnt require any controlling is the pre-treatment
section shown in Figure 20.
73
Figure 20: Pre-treatment section
The output of the pre-treatment goes into the gasification unit which is described
in the next section. The second section of the P&ID consists of the gasifier unit and all
the control loops around it demonstrated by Figure 21.
Figure 21: Gasification section
74
The output of the Gasification unit goes into a series of gas cleaning steps which are
shown in Figure 22.
Figure 22: Gas cleaning section of the P&ID
The next section of the P&ID displays the fermenter unit and the controls that are
applied and the recycle storage unit S-101 with the pH control and other controls are also
shown.
Figure 23: Fermenter section of the P&ID
75
The final section of the P&ID consists of the distillation and is shown in Figure 16.
Figure 24: Summary of the Distillation section of the process
8. Equipment design, sizing and costing process side

8.1 Costing overview
Moving on from the P&ID and process design discussion, this section aims to
present a study estimate of the costs of the plant, though the costs presented are not bare
module costs so they do not include installation factors. Because this is a study estimate,
there are large margins of error associated with each capital cost, some of which can be
up to 200%. Additionally, both capital costs and operating costs use results from the
Aspen Plus simulation, which were presented previously in stream tables. These results
are not exact and may contain error due to the difficulty in accurately modeling the
gasification and continuous fermentation processes as well as other small differences
between the simulation or model and reality. The margin of error for all costs is assumed
to be 50%.
76
Capital costs were calculated using cost correlations in (Seider et al., 2009)
(Towler and Sinnott, 2012) and (Woods, 1983). The costing method used for both capital
costs and operating costs are explained in their respective sections. Additional important
details regarding the economics of the plant are listed here:
The MARR was chosen as 10%.
The plant lifetime was chosen as 25 years.
Nominally, the plant runs for 24 hours and 330 days a year but since this is
unrealistic due to extra time required on some days for maintenance, the hours lost
from not running the plant can be made up by adding more days of operation per
year.
The first year is used to buy, setup, test and run the plant at lower capacity in
order to troubleshoot any problems that come up during initial operation. The
plant starts operating in the second year.
The tax rate is 25%.
Plant economics are initially analyzed without an ethanol subsidy or a carbon tax.
Transportation costs are ignored completely for this analysis due to the inability to
accurately document which population centers are using the fuel, what their
demand is and what the travel pathways are in order to reach these destinations.
The following USD to CAD conversion rate was used throughout: 1 USD = 1.25
CAD.
8.2 Capital costs

Capital costs were calculated for the following components: dryer, crusher
(hammer mill), gasifier, cyclone, heat exchangers, fermenter, vessels, pumps, distillation
columns and storage tanks. In addition, the only component which was cost as working
capital was zinc oxide, which is the adsorbent used in the H2S gas cleaning adsorber
column. All costs were converted to 2013 dollars using CEPCI inflation factors. The final
costs were converted to Canadian dollars from U.S. dollars, so the entire economics
analysis is done in Canadian dollars. While Woods textbook uses cost information and
correlations from the 1970s, these correlations were only used for vessels, pumps and
77
distillation column costs. However, the fact that distillation columns are a significant
portion of the final capital cost makes this final estimate a weak one as cost correlations
from the 1970s cannot be used with confidence in modern times. The other two textbooks
use cost information and correlations from 2006 so they are far more accurate. During
cost estimation, some of the cost correlation factors exceeded their given bounds.
Although this means that the associated capital costs are not as accurate, these deviations
are not significant relative to the existing error associated with each cost (approximated as
50%), especially since the bounds are not exceeded by much.
The dryer was cost as a spray dryer because it uses the evaporation rate in the
correlation, which is one of the design factors chosen during the dryer design. The crusher
was cost as a hammer mill as this is the type of crusher chosen in the plant design. Since
there is no direct gasifier cost correlation in the books used, the gasifier was cost as a
pyrolysis furnace. This models the gasifier accurately because one of the key steps in the
gasification process is pyrolysis and the gasifier essentially takes on the form of a furnace
as miscanthus is fed and burned. For gas cleaning, the cyclone was cost using the gas
flow rate correlation, and the H2S adsorber and wet scrubber were cost as vessels using a
volume correlation. The volume for the adsorption vessel was calculated by equating it to
the volume of the adsorbent during operation, which used the vessels adsorption capacity
and density and an operation time of 15 days. An operation time of 15 days was picked to
limit the size of the adsorption column.
Heat exchangers were cost using the standard heat exchanger area correlation.
This heat exchanger area was calculated during heat exchanger design. Specifically, the
closest overall heat transfer coefficient was chosen to suit the heat exchangers shell and
tube species based on knowledge and experience, then heat exchanger duty was found
from the Aspen Plus simulation using the HEATER model and finally the corresponding
area was calculated.
Since there was only one fermenter, there was some freedom regarding the design
specifications and type of tank to use. The final chosen design was a jacketed, closed-lid
stirred tank design. As a result, the cost correlation used fit perfectly here as it was for a
78
jacketed, agitated design. However, the fermenter was far too big and the cost factor
exceeded its bounds by far too much. Therefore, the fermenter was cost as 10 fermenters
and these costs were added together. There were two distillation column condenser
vessels and both were cost using a Woods cost correlation involving the length and
diameter.
Pumps were cost using the standard power/electricity usage correlation. However,
since the Aspen Plus simulation did not initially include pumps (as there was no pressure
loss), several pumps were duplicated. These were duplicated from two pumps that were
added to specific parts of the simulation or flow sheet where the flows differed, namely
the recycle part and the fermentation outflow part. In the end, the combined capital cost
of pumps was far lower than the other capital costs so this duplication of pumps did not
greatly affect the final capital cost. The effect was considered negligible.
Distillation columns were cost using a correlation in Woods based on diameter
and tower height. The tower height was simply calculated by multiplying the tray spacing
found from the Aspen Plus simulation sizing analyzer by the number of trays. However,
the sizing analyzer gave a diameter far larger than the length, which does not resemble the
shape of an actual distillation column and cannot be transported by truck. Instead, a 5 m
diameter was chosen so that it could be transported but the volume required (based on
feed flow rate) raised the height to >400 m. Since this is not feasible either, the distillation
column in reality would have to be divided up into multiple distillation columns of height
of around 50-80 m, as shown in the P&ID.
Storage containers had their own correlation. The mixing tank that is before the
fermenter in the P&ID needs to be a closed-lid tank that is kept at fermenter operating
conditions so it was cost in the same way the fermenter was cost. As a result, its cost
ended up being the same as that of the fermenter. Total capital costs for each component
of the plant and the total capital cost for the entire plant is shown in Table 21.
79
Table 21. Sum of capital costs for each type of unit and total capital cost.
Expense
CAD Cost
Pretreatment
2,104,000
Gasifier
8,182,000
Cyclone
729,000
Gas Cleaning
1,208,000
Heat Exchangers
1,598,000
Fermenter
$ 20,465,000
Vessels
516,000
Pumps
83,000
Distillation Columns $ 27,919,000

Storage
$ 21,953,000
Zinc oxide
TOTAL
2,374,000
$ 87,129,000
Table 21 shows that the largest costs are for the gasifier, fermenter, distillation
columns and storage containers. Storage costs are inflated due to the pre-fermenter
mixing tank being cost as a reactor instead of a storage tank. This was discussed in the
previous paragraph. When comparing the largest costs, it is clear that the highest cost is
that of the distillation columns. These distillation columns are unusually large in size due
to the high feed flow rate. This is because there is no real parallelism in the process, as the
fermenter exit contents are not split into multiple streams and instead go through two
distillation columns in series. Although this was remedied by splitting the feed flow into
multiple distillation columns, this does not lower the high cost of the columns. In fact,
since multiple distillation columns require more material to be produced than a single
long distillation column, splitting the feed as stated would actually increase the cost even
more, possibly to around $30 million. However, this nonlinearity was not considered
80
when adding more columns and it was assumed that splitting the feed would not change
the cost and not have an effect on the separation performance.
The next largest cost, ignoring storage costs, is that of the fermenter. The
fermenter, unlike the distillation columns, does not require feed splitting (even though it
was cost as multiple fermentation tanks) because it can be extended in width and height
without too much trouble. The only thing to consider here would be the gas-liquid mass
transfer rate, contact area or collision rate, all referring to how much the gas and liquid
mix in the tank. However, the syngas only takes a few minutes to contact and react with
the contents in the fermenter and since the gas moves quickly, it is assumed that the edges
of the tank are reached fairly easily by the syngas. Therefore, it is not as critical to split
the feed into multiple fermenters to reduce the size. In addition, unlike the distillation
columns, running the fermentation process is a lot more difficult in parallel due to the
requirement of each fermenter to maintain operating conditions like pH, temperature,
pressure and ethanol levels under strict ranges. This would also require additional safety
systems, more operators, splitting the recycle stream running through the plant and more
mixing tanks or storage containers.
The third largest cost is the gasifier cost. It was initially assumed that this would
be the largest cost due to its position in the flow sheet at the beginning where it is one of
the first units to be in contact with the miscanthus feedstock. However, since the capital
cost is simply cost as a pyrolysis furnace, the cost is low because the expected extra costs
in maintenance of the gasifier and energy management are not taken into account here.
Pretreatment, gas cleaning and heat exchangers make up the next three biggest capital
costs. These costs are relatively proportional to their role in the syngas fermentation
process. Gas cleaning costs (including the cyclone) add up to almost $2 million, which is
ironically smaller than the cost of the zinc oxide adsorbent. Zinc oxide was cost by
obtaining its volume through its adsorption capacity and density as stated in the previous
paragraph (in the adsorption section).
The pump capital cost is relatively the smallest cost in Table 21. This shows that
having two pumps side by side (one as a backup) is not only feasible but is the best option
81
to take when building a syngas fermentation plant. For this reason, only two pumps out of
the seven pumps were single (without a backup).
8.3 Operating costs

Operating costs were calculated on an annual basis and like all other costs, they
are in Canadian dollars. Utilities make up the largest portion of the operating
requirements in terms of the role they play in running the plant and their constant need in
order to keep the plant safe. The electricity price is the on-peak electricity price in
Ontario: 14 cents per kilowatt-hour. This may not be reasonable for the current time but it
is there to account for changes (most likely increases) in the electricity price over the
plant lifetime, which will last multiple decades. In other words, it is there for contingency.
All other utility prices were taken from (Seider et al., 2009) presented in the costing
overview section. Utility operating costs were calculated by simply multiplying the cost
factor by the utility flow over the annual duration of plant operation (24 hours for 330
days).
The three other important operating costs are the costs of the feedstock, broth and
wet scrubber operation. The feedstock cost is based on a cost factor estimate of
$71.50/tonne for growing and harvesting the miscanthus in Canada (Roy, 2014). This
includes costs to run an irrigation system, to properly plant and maintain the crops and the
cost of the fertilizer. The wet scrubber annual cost is based on the flow of the waste
stream according to a cost factor by the Environmental Protection Agency (EPA) in their
Air Pollution Control Technology Fact Sheet (U.S. Environmental Protection Agency,
n.d.) . The waste stream carries the separated impurities ammonia, hydrogen chloride and
carbon dioxide at conditions close to standard temperature and pressure.
The broth is cost as a fermentation medium known as corn steep liquor. Various
journals, such as Maddipati, 2011 and Saxena, 2012, have recognized its low cost and
effectiveness as a medium for the bacteria to produce ethanol in comparison to yeast
extract, the traditional fermentation medium. It provides the necessary nutrients and
conditions for the bacteria to thrive. The cost was calculated using by using a factor of
$0.31/L of corn steep liquor but the entire stream of new broth was not cost in order to
82
counteract the overestimation of new broth required according to the Aspen Plus
simulation. Other operating costs include the pretreatment steps, which require large
amounts of energy input in the form of electricity for the crushing step and medium
pressure steam for the drying step. These were added to the total operating cost.
The final additions to the annual operating cost were the labour and infrastructure
costs. Labour costs include the salaries and work related costs of operators, supervisors,
engineers, maintenance personnel, management, and lab technicians. It was assumed that
10 operator posts would be needed to run the 10 primary processes, and that 4.4 people
would be needed per post to run the plant 24/7 while changing shifts. Other infrastructure
costs include overhead, maintenance materials, insurance, property tax, laboratories and
operating overhead. Carbon tax was included but is only used when performing the
sensitivity analysis. The complete operating costs are shown in Table 22.
Table 22. All operating costs for the syngas fermentation plant.
Expense
Factor
Cost/Unit
Annual Cost [CAD]
Feedstock
$71.50/tonne
86,873,000
Wet Scrubber Annual Cost
$64000/waste stream
104,000
Raw Material (Broth)
$0.31/L
5,612,000
Electricity (P-101 A/B)
$0.14/kW-hr
1,811
Electricity (P-102 A)
$0.14/kW-hr
1,811
$0.14/kW-hr
1,171
Electricity (P-104 A)
$0.14/kW-hr
1,171
$0.14/kW-hr
1,171
$0.14/kW-hr
1,171
$0.14/kW-hr
1,171
Electricity (M-102)
$0.14/kW-hr
3,511,000
MPS to M-101
$10.5/1000 kg
7,271,000
LPS to E-103
$6.6/1000 kg
-$
6,085,000
LPS to E-106
$6.6/1000 kg
-$
4,078,000
$0.020/m
CW to E-101
83
36,000
CW to E-102
$0.020/m3
21,000
CW to E-104
$0.020/m3
48,000
CW to E-105
$0.020/m3
295,000
CW to E-107
164,000
$0.020/m
CW to FE-101
$0.020/m
1,514,000
(A) Operators 1 post - 4.4 people
$70,000
3,080,000
(B) Supervision & Engineering
0.25*(A)
$100,000
770,000
(C) Maintenance Personnel
0.03*(fixed cost)
$75,000
2,091,000
(D) Engineering & Management
0.5*(A)
$100,000
1,540,000
Overhead
0.4*(A+B+C+D)
2,992,000
Maintenance Materials
0.03*(fixed cost)
2,091,000
Insurance
0.01*(fixed cost)
697,000
Property Tax
0.02*(fixed cost)
1,394,000
Laboratories
0.15*(A+B+C)
891,000
$
Carbon Tax
Operating overhead
0.25*(A+B+C+D)
TOTAL
1,870,000
$ 112,713,000
Table 22 clearly lists the most costly operations. Disregarding labour, the largest
operating costs, ones that exceed $1 million annually, are for feedstock, adding fresh
broth to the fermenter, electricity and medium pressure steam for pretreatment, and
cooling water for the fermenter. The feedstock cost exceeds all other operating costs by
far but is not unusual by itself. It is large due to the various costs associated with growing
the feedstock and maintaining it, which essentially requires as much work as it does to
start and maintain a massive food crop year round.
The two costs for the fermenter are expected since the fermentation process is at
the heart of this process and is one of two big processes in the plant, the other being
gasification. To maintain the health of the bacteria and the resultant ethanol production
yield, high quality broth will always be required. As well, temperature control of the
fermenter using cooling water is essential to maintaining the bacterias ability to produce
84
ethanol, as fermentation is exothermic and can inhibit the process. To summarize, both
the cooling water and broth provide the ideal conditions to make this process work.
Therefore, cost cuts to this operation are not recommended.
The pretreatment or milling processes are both large costs, with the medium
pressure steam going to the drying process being the most costly. In fact, this cost is the
largest utility cost. The use of medium pressure steam instead of low pressure steam
should be re-evaluated as this drives up the cost greatly. Electricity to the crushing
process is needed, but other types of crushing processes (other than hammer mills) should
be investigated to see if they can use less electricity or energy.
The two operating costs shown for the low pressure steam going to the two
distillation column reboilers are negative because they subtract the profit made from
selling low pressure steam made in the heat exchangers. In reality, part of the low
pressure steam made in the HXs goes to the reboilers while the rest is sold but this still
gives the same negative costs. Cost savings can be made on these two reboiler costs as
well, which can be accomplished by reducing the boilup ratio of the bottoms exiting the
distillation columns. However, this may reduce the purity of the final ethanol product,
resulting in a less efficient process. Instead, it is recommended that the distillation
columns should be further optimized or other changes should be made like feeding
vapour and liquid from one stage to another, so that the reboilers are not used as much. In
this way, the costs of the condenser utilities are brought down as well. As seen in Table
22, labour and infrastructure costs make up a large part of the operating costs and these
costs are essential to running and maintaining the plant. However, it should be noted that
these costs are based on fundamental assumptions like the amount of operators needed
and rely on multiplication factors that can only give rough estimates.
Compared to a total capital cost of almost $90 million, the total annual operating
cost is relatively large as it is on another order of magnitude. Though this may indicate
that the operating cost is too large, it is in fact the capital cost that is very low due to the
simplicity and linear nature of the syngas fermentation process. Still, operating costs can
85
and should be brought down by saving on feedstock costs and making the process more
efficient, especially by making it energy efficient.
8.4 NPV
NPV calculations were performed using the previous total capital cost and total
annual operating cost values. The costing overview section gives some of the key
information used in these NPV calculations like the tax rate and the MARR. It is also
important to note that the rate of depreciation on the equipment bought using the initial
capital is 30% annually according to Class 43 of Canadas CCA (capital cost allowance)
law (Canada Revenue Agency, 2015). Since no other byproducts of the process are being
sold except for excess steam, the only revenue that is made comes from selling the
ethanol fuel.
In Canada, ethanol can be used in multiple blends. These are E10, a 10% v/v
(volume %) ethanol blend, or E85, an 85% v/v ethanol blend. These blends are only made
by volume, and the rest of the blends contain gasoline. The sale of both these blends is
not currently widespread in Canada and E85 is currently only commercially available in a
select few areas. For this reason, US prices of ethanol were used as it is more widespread
and the national average is more indicative of the price of ethanol. The E85 price was
used from the US Department of Energys Alternative Fuels Data Center (U.S.
Department of Energy, 2015). Since we know the blend according to volume percentage
and prices are given by volume, the following formula (Equation 1) involving the price of
E85 and gasoline was used to calculate the price of pure ethanol (E100).
$2.21
$2.30
0.15
85

100 = (
0.85
86
1.25
1
$2.74
(1)
The process of syngas fermentation discussed in this report produces 99% ethanol
so this price is appropriate given the purity. NPV calculations are shown in Figure 25
Period
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
All Income
$
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
$
90,000,207
All eligible expenses

$
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
$
(112,713,359)
All ineligible expenses

$
(87,129,306)
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
-
Book value
$ 87,129,306
$ 74,059,910
$ 51,841,937
$ 36,289,356
$ 25,402,549
$ 17,781,784
$ 12,447,249
$
8,713,074
$
6,099,152
$
4,269,406
$
2,988,585
$
2,092,009
$
1,464,406
$
1,025,084
$
717,559
$
502,291
$
351,604
$
246,123
$
172,286
$
120,600
$
84,420
$
59,094
$
41,366
$
28,956
$
20,269
Depreciation
$ 13,069,396
$ 22,217,973
$ 15,552,581
$ 10,886,807
$
7,620,765
$
5,334,535
$
3,734,175
$
2,613,922
$
1,829,746
$
1,280,822
$
896,575
$
627,603
$
439,322
$
307,525
$
215,268
$
150,687
$
105,481
$
73,837
$
51,686
$
36,180
$
25,326
$
17,728
$
12,410
$
8,687
$
6,081
Taxable income
$ (13,069,396)
$ (44,931,125)
$ (38,265,733)
$ (33,599,958)
$ (30,333,916)
$ (28,047,687)
$ (26,447,326)
$ (25,327,074)
$ (24,542,897)
$ (23,993,974)
$ (23,609,727)
$ (23,340,754)
$ (23,152,474)
$ (23,020,677)
$ (22,928,419)
$ (22,863,839)
$ (22,818,633)
$ (22,786,988)
$ (22,764,837)
$ (22,749,332)
$ (22,738,478)
$ (22,730,880)
$ (22,725,561)
$ (22,721,838)
$ (22,719,232)
Tax paid
$ (3,267,349)
$ (11,232,781)
$ (9,566,433)
$ (8,399,990)
$ (7,583,479)
$ (7,011,922)
$ (6,611,832)
$ (6,331,768)
$ (6,135,724)
$ (5,998,493)
$ (5,902,432)
$ (5,835,189)
$ (5,788,118)
$ (5,755,169)
$ (5,732,105)
$ (5,715,960)
$ (5,704,658)
$ (5,696,747)
$ (5,691,209)
$ (5,687,333)
$ (5,684,619)
$ (5,682,720)
$ (5,681,390)
$ (5,680,460)
$ (5,679,808)
Net cash flow in period (TVM)

$
(83,861,957)
$
(10,436,700)
$
(10,865,057)
$
(10,753,690)
$
(10,333,770)
$
(9,749,228)
$
(9,088,775)
$
(8,406,240)
$
(7,733,492)
$
(7,088,647)
$
(6,481,260)
$
(5,915,623)
$
(5,392,837)
$
(4,912,123)
$
(4,471,640)
$
(4,068,993)
$
(3,701,544)
$
(3,366,605)
$
(3,061,546)
$
(2,783,857)
$
(2,531,183)
$
(2,301,332)
$
(2,092,283)
$
(1,902,180)
$
(1,729,320)
Figure 25. NPV Analysis
Figure 25 shows unusual results as ethanol revenue exceeds the total capital but
the yearly operating costs exceed both by a significant amount. Over the plants 25 year
lifetime, NPV is negative and amounts to a loss of $223 million. This includes tax and
depreciation as well as the yearly MARR. Therefore, the rate of return generated from
this project does not exceed the MARR of 10% and makes this project not worthy of
investment. It is also worth noting that a negative NPV of almost a quarter billion dollars
is not much lower than the baseline of $0 NPV given the 25 year lifetime. (An NPV of $0
is the cut-off for project investment.) This means that the confidence level in rejecting
investment into this project should not be high. Judging by the NPV values in Figure 25
across all the years, the NPV and the rate of return are reduced significantly by the high
87
NPV
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
(83,861,957)
(94,298,658)
(105,163,714)
(115,917,405)
(126,251,175)
(136,000,403)
(145,089,178)
(153,495,418)
(161,228,910)
(168,317,557)
(174,798,817)
(180,714,440)
(186,107,278)
(191,019,401)
(195,491,041)
(199,560,034)
(203,261,578)
(206,628,183)
(209,689,729)
(212,473,586)
(215,004,769)
(217,306,101)
(219,398,384)
(221,300,564)
(223,029,884)
operating costs, which take away from the revenues generated. This is primarily because
of the high feedstock cost, as mentioned previously. It is therefore critical to lower
operating costs by coming up with better ways grow feedstock while still maintaining the
efficiency and yield from the process.
8.5 Sensitivity Analysis

The sensitivity analysis was done as part of the previous NPV analysis. The following
factors were varied to see possible changes in the NPV in different circumstances.
Ethanol price
Feedstock cost (based on feedstock price)
Plant lifetime
Carbon tax
Ethanol price was varied instead of any kind of subsidy because a subsidy program
which supports biofuel production already exists in Canada. It is known as The
ecoENERGY for Biofuels Program and lasts until 2017 (National Resources Canada,
2014). Its incentive is based on the following formula (Equation 2) and the exact
incentive amount is calculated further in the section.
$
Incentive = Eligible Sales (L) Incentive Rate (L)
[2]
Eligible Sales are assumed to be equal to the production capacity of ethanol from
the plant. The Incentive Rate is a given rate that is fixed by the government of Canada for
every year until 2017. Therefore, the ethanol price sensitivity shown here does not
represent a subsidy but rather the normal variations in market price of ethanol. Although
these normal variations change the NPV in a linear fashion in the sensitivity analysis,
changes in the market demand due to price are also expected in reality. In addition,
significant shifts in demand are expected due to the lower energy content of ethanol
blends when compared to pure gasoline. Specifically, one gallon of E100 has around 73%
of the energy content of gasoline (U.S. Department of Energy, 2014), so ethanol fuel will
only be competitive in the market once the price of the ethanol is around 70% of the price
of gasoline. For this to happen, the price would have to reduce drastically from current
88
levels. Therefore, a key assumption in this sensitivity analysis is that all the ethanol that is
produced is sold at market price. In other words, there is market demand but it does not
govern supply from this plant.
The final part of the sensitivity analysis is for carbon tax. In British Columbia,
Canada, the carbon tax for ethanol is equal to that for gasoline, which amounts to 6.67
cents per liter of fuel burned (B.C. Ministry of Finance, n.d.). Although the plant is being
built in Ontario, carbon tax only exists for gasoline (not ethanol) in Ontario, which means
that the baseline NPV does not include a carbon tax. However, it is possible that a carbon
tax for ethanol equivalent to that for gasoline can be imposed in Ontarios future, similar
to B.C.s carbon tax. This tax would then be 4.7 cents per liter of ethanol (Ontario
Ministry of Finance, 2015). Figure 26 shows the sensitivity of the four factors mentioned.
Energy Price
NPV [CAD]
Millions
Ethanol Price
-100
-50
Plant Lifetime
Carbon Tax
$300.00
$200.00
$100.00
$$(100.00) 0
50
100
$(200.00)
$(300.00)
$(400.00)
$(500.00)
$(600.00)
$(700.00)
% Change
Figure 26. A +20% sensitivity in ethanol price is an increase of 0.1 cents per liter. A +20% sensitivity in
feedstock price is an increase of 5 dollars per tonne. A +20% sensitivity in plant lifetime is an increase of 5
years (ranging from -40% to 40%). Carbon tax is a binary sensitivity with 0% representing NPV w/o
carbon tax and 100% representing NPV w/ carbon tax.
The baseline NPV shown in Figure 26 is -$223 million for an ethanol price of 0.72
cents per liter, a feedstock cost of $71.5 per tonne of miscanthus, a 25 year lifetime and
no carbon tax. Reiterating the point made in previous discussions, the primary reason why
NPV is negative is due to the high feedstock cost, as the ethanol revenue generated
89
annually barely exceeds the annual feedstock operating cost ($90 million to $87 million).
Figure 26 clearly shows that the only way for NPV to be positive is that the price of
ethanol exceed about 1 dollar per liter. This represents 28 cents per liter or 1.06 dollars
per gallon increase in the price. This is not only unlikely under current market conditions
and variations but such a high price will turn off customers from buying the ethanol due
to the lower energy content of ethanol blends when compared to pure gasoline. Therefore,
realistically NPV can only become positive if the gasoline price goes back to normal
levels from its current state, as it went through a downward spike around a year ago due
to oversupply of crude oil.
From the illustration in Figure 26, it can also be seen that a -100% sensitivity in
feedstock price representing 46.5 dollars per tonne still cannot make the NPV positive.
This shows how the plant can almost never be profitable, no matter how low the
feedstock cost goes. It signifies the risk of introducing miscanthus to Ontario, with the
associated risks of being an invasive species and requiring vast amounts of land to grow.
The plant lifetime sensitivity in Figure 26 does not result in a positive NPV either.
However, it is nonlinear and shows that a smaller plant lifetime can make the NPV
significantly larger. Finally, the carbon tax sensitivity causes a further decrease in NPV as
the tax takes away from the ethanol sales revenue. This is not important as the NPV is
already negative and is not affected significantly by an added tax.
The calculated incentive amount that was mentioned in Equation ### is exactly $7
million dollars. Though this is helpful in raising NPV, it is only a small amount and only
useful when NPV is already above $0. The final NPV including the incentive over 2016
and 2017 is negative $217 million. This is still far below $0. Therefore, investment into
this project is not recommended.
90
8.6 Equipment Sizing

A list of each type of unit and how it was sized is shown below. The Aspen Plus
simulation was used to size only a few units as the given sizes were unrealistic in most
cases.
Pretreatment The belt dryer size was chosen based on relevant literature. It was 4-5 m
wide and 30 m long. It supported a maximum throughput of 500 tonnes/hr so it was
enough to allow our process to run with no issues. The crusher size was based on the
dryer, which meant that it would have to be around 4-5 m wide as well. As for the total
volume of the crusher, the exact speed of crushing and the velocity of the belt dyers
speed would have to be known in order to estimate. Additionally, we would have to make
sure the two processes remain continuous when making adjustments to these
specifications.
Gasifier The gasifier size was initially found using the Aspen Plus simulation, which
gave a 4.5 feet diameter as one of the parameters. Since this did not accurately fit the
incoming feedstock size and amount, another method was used. In this method, the gas
velocity out of the gasifier was decided as 5 m/s (typical) and the volumetric flow rate out
of it was used to find the cross sectional area, giving a 20 m2 value. The diameter was
then calculated as 5 meters. Usually, the height of the gasifier is longer than the diameter
to account for the char and ash base, the incoming feedstock, the pyrolysis section and the
fluidized bed. A reasonable height that was decided upon was 20 m high.
Cyclone It was assumed that an industrial cyclone with a comparable height to that of
the gasifier would have to be used since the gasifiers outlet is fed into the cyclone.
However, this did not have to be as high as the fluidized bed because a fluidized bed
serves other functions except for moving/separating gas. Therefore, a 15 m height was
chosen for the cyclone. The rest of the dimensions were proportional to this height.
Heat exchangers Heat exchangers were sized using the heat transfer area. To get a
detailed estimate, a diameter and length would have to be chosen for each pass of a shell
and tube heat exchanger. The heat transfer area would then have to be divided by the
surface area of a single pass to give the amount of passes required. Based on this, a rough
91
estimate of the total volume would be found. However, this detailed approach was not
used as the pipe diameters and velocities inside and outside the HXs would have to be
known or decided upon.
E-101 = 620 m2
E-102 = 7543 m2 (this would have to be divided up into two or more HXs)
E-103 = 529 m2
E-104 = 1224 m2
E-105 = 450 m2
E-106 = 463 m2
E-107 = 530 m2
Fermenter A volume was picked based on literature (6000 m3), as explained previously.
Vessels Vessels (condensers) were sized using a length and diameter from the Aspen
Plus simulation.
V-101 = 5.5 m length, 1.8 m diameter
V-102 = 5.9 m length, 2 m diameter
Pumps Pump sizes had to be based on piping diameters. However, these were not set. In
addition, the volumes had to use the incoming volumetric flow amounts but the exact
correlation was not known. It was therefore assumed that the pipe diameters were around
1 foot and that the pump sizes were proportional to this diameter.
Distillation columns Distillation columns were initially sized using Aspen Plus.
However, a 5 m diameter had to be set by us to account for transportation constraints on
moving the columns, so the height of the columns were increased (using the same volume
from Aspen Plus), which gave ~6 distillation columns for each single distillation column
block, measuring 50-80 m each. These heights were slightly unreasonable but the values
matched our smaller ethanol yield compared to theoretical values.
T-101 = 405 m length (6 distillation columns / 67.5 m length each), 5 m diameter
T-102 = 456 m length (6 distillation columns / 76 m length each), 5 m diameter
Storage containers Storage container volumes were based on a residence time of 3 days
and the given incoming flow rate. The storage containers were floating roof types. One
92
storage tank was actually the pre-fermentation mixer. Therefore, it had to be close to the
fermenter size to maintain similar conditions.
S-101 = 21,000 m3
S-102 = 6000 m3
Heat exchanger design
Shell and tube countercurrent heat exchangers were simulated in Aspen plus in
order to determine the heat duty required to cool/heat a process stream to a specified
outlet stream temperature using heat transfer coefficients from tabulated sources. Heat
exchangers E-101, E-102 and E-103 were designed using equations and tabulated overall
heat transfer coefficients from the textbook chemical engineering design: principles
practice and economic design by Towler and Sinnott. The area of heat transfer was
calculated using Equation 3. Meanwhile, the required heat transfer area for heat
exchangers E-104, E-105, E-106 and E-107 were obtained using Aspen Plus Process
Economic Analyzer.
=
[3]
Where,
Q = heat transferred per unit time [W]
U = the overall heat transfer coefficient, [W/m2]
A = heat transfer area [m2 ]
Tm = the mean temperature difference [C]
The mean temperature difference was calculated using Equation 4;
=
[4]
Where ,
Tlm = log mean temperature difference [C] and calculated using Equation 5;
Ft = correction factor which depends on the heat exchanger geometry and obtained from
(Towler and Sinnott, 2012).
=
(1 2 ) (2 1 )
( 2 )
ln 1
(2 1 )
Where,
93
[5]
T1 = hot fluid inlet temperature [C]

T2 = hot fluid outlet temperature [C]
t1 = cold fluid inlet temperature [C]
t2 = cold fluid outlet temperature [C]
Heat Exchanger (E-101)

T1 = 850C
T2 = 550C
t1 = 37C
t2 = 134.976C
Q = 26,201,374.9 J/sec, obtained from Aspen Plus
U = 70 W/m2 C for a liquid water as cooling fluid in the shell and gas at 1 bar flowing
inside the tubes.
=
(850 C 134.976 C) (550 C 37 C)

(850 C 134.976 C)
ln
(850 C 134.976 C)
= 612.66C
For a one shell and even multiple of tube passes, the correction factor, Ft, equals .9857.
Therefore Tm is
= .9857 612.66
= 603.9C
Rearranging, Equation 3 for A, the total heat transfer area required is then;
=
=
26,201,374.9
70 W/m2 C 612.66C
=
26,201,374.9
W
70
C 612.66C
m2
= 619.8 2
94
Specifications of heat exchangers E-101, E-103, E-104, E-105, E-106 and E-107 are
summarized in Table 23.
Table 23. A summary of parameters used to design heat exchangers E-101, E-102 and E-103.
Equipment ID
E-101
E-102
E-103
Heat exchanger
Countercurrent Shell
type
and Tube
and Tube
and Tube
Q [J/sec]
26,201,374.9
12,000,822.4
5006737.3
U [W/m2 C]
70.0
70.0
800
T1 [C]
850
550
73.2
T2 [C]
550
187
37
t1[C]
29.4
29.4
29.4
t2 [C]
134.9
134.9
134.9
Tlm [C]
612.6
23.0
14.3
Ft
.9857
0.8
0.825
Tm [C]
603.9
18.4
11.8
A [m2]
619.8
9,299
529.3
The heat transfer area for heat exchangers E-104, E-105, E-106, E-107 were obtained
using Aspen Plus Process Economic Analyzer. The heat transfer area for heat exchangers
E-104, E-105, E-106 and E-107 are 1224.177 m2, 450.2378 m2, 463.421 and 530.3198 m2
respectively.
9. Environmental Impact
9.1 LCA
A cradle to gate LCA was decided in order to show the environmental impact
from the various stages of the gasification-fermentation ethanol production process. This
can be seen in Figure 27 below.
95
Figure 27. Cradle to gate LCA for the plant.
Starting from the production of the feedstock, the seeds and fertilizer are planted
and the plant is grown over time in the presence of air, water and fertilizer before it is
collected. During the growth of the feedstock, emissions are absorbed by the plant,
however the equipment used on the farms in order to grow the feedstock are mostly
powered with diesel from the refinery will have various emissions which are labeled as
CO2, H2O, NOx, Sox (to air) on the LCA. Electricity from the grid will also contribute to
the emissions on the farm. This results in the farm having a positive GHG value.
Transportation emissions are due to using trucks to transport the feedstock from the farm
to the plant and the trucks returning.
For this project trucks of 44 tonne capacity are used as the vehicle since emission
values are found in literature with 44 tonne trucks as basis (Bonitta and Whittaker, 2009).
This contribute a large amount of emission since combustion of diesel results in high
amounts of waste gas into the air. The diesel used by transportation and the farm comes
from crude oil. The crude oil is taken out of the ground via drilling and is sent to the
refinery to be refined and finished into usable fuel (gasoline, diesel, etc).
96
During the drilling phase, non-usable components of the oil is often combusted.
Along with the production of the pipes, energy use from machines in drilling and refining,
the process of diesel refining also contribute a positive GHG level. Another source of
emission is from electricity generation. Since there are many forms of power plants, some
contribute a much higher percentage of emissions than others. For example, gas power
plants and coal power plants are older technologies which are banned from certain places
due to their large environmental impact. Renewable energy sources such as wind and
hydro have constantly been under research and development with the hope of replacing
coal and gas one day. In terms of nuclear power plant, it is the cleanest reliable source of
power today which resulted in nuclear power being the highest electricity production
method to the grid (nuclear energy GHG emission: 16-55g CO2 equivalent/kWh, Gas
energy GHG emissions: 700-1000g CO2 equivalent/kWh) (Fthenakis and Kim, 2007).
Looking at the plant itself, there are emissions from the electricity use and fuel
that may be consumed on site. The emissions from the plant can be found in the following
section in Table 25. One reason the syngas fermentation method is said to emit less GHG
is because the reactions themselves only produce two types of greenhouse gases in carbon
dioxide and methane.
9.2 GHG Emissions

When calculating the total GHG used in the syngas fermentation ethanol
production process, the GHG emission from various steps in the LCA is calculated and
added together. The LCA can be found in the previous section. The first portion of the
LCA shows the farms that grows the feedstock used in the plant. For this project,
Miscanthus is used as a representation of the feedstock however the process can be used
with almost all types of biomass. Through research, it was found that for every kilogram
of miscanthus grown, the total GHG emission is equivalent to 51 grams of CO2. (Maxime
et al., 2013) Therefore since this process uses 122,728kg of miscanthus per hour,
6,259,128 g or 6,259kg of CO2 is released into the air per hour.
Another source of GHG emission from the LCA is the transportation of feedstock
from the farm to the plant. Assuming the trucks used for transportation are large diesel
97
trucks of 44 tonne capacity, meeting the usage demand of the plant would require up to
67 trucks every day to meet the required feedstock for the plant. From research articles, it
was found these trucks emit equivalent of 120g CO2 per every kilometer traveled (Bonitta
and Whittaker, 2009). Assuming the feedstock is available in a radius of 100km from the
plant as well as the return trip for the trucks, the total GHG emissions from transportation
would be equivalent to 1,608,000g or 1,608kg of CO2 per day for the trucks to go from
the farms to the plant and then back. Since diesel is used for the truck transportation, the
amount of energy used by each truck per trip is 356 MJ. Using correlation from research,
19.224 kg of CO2 equivalent is released per truck for each trip (Hsu, 2011). Taking
account of the 67 trucks needed, the total CO2 equivalent comes out to be 1,288kg per
day.
The next source of emissions is from the generation of electricity. Since there has
been insufficient research done on the farms supplying the feedstock thus far, only the
electricity used for the production of ethanol in the plant will be included in the
calculations. Since nuclear energy makes up of around two thirds of the or the electricity
going to the grid, it is assumed the plant takes electricity purely from nuclear source since
the supply of power through different methods (i.e. wind, hydro and solar) all vary
depending on the time and weather of the day. Natural gas electricity accounts for a small
portion of the power grid therefore it is neglected in this analysis. The high amount of
nuclear energy is due to nuclear power plants are the most powerful and at the same time
has the lowest GHG emissions from any of the non-renewable/non-natural methods of
generating electricity (coal, natural gas). It is also reliable unlike methods such as wind,
hydro and solar. Through research, it was found for every kilowatt/hour of electricity
used, the total equivalent of CO2 was found to be between 16-55g (Fthenakis and Kim,
2007) . Assuming the Ontario nuclear power plants emit at the middle of the range, the
total GHG was found to be 90,164.18g CO2 per day, or 90.16kg CO2 per day.
Finally, in terms of the emissions from the plant itself, the environmental
advantage of syngas fermentation over other types of ethanol production is showed by the
difference in the GHG emissions. Looking at the results from the plant process, it was
98
found only two types of greenhouse gas was produced in the process of fermentation
(CO2 and CH4). Through simulations, the amount of CO2 in the products was found to be
6,188.09kg per hour and CH4 was produced at 354.57 kg per hour. The CO2 equivalent of
CH4 produced was 12,055.4kg per hour was found using correlations found from research
(Intergovernmental Panel on Climate Change ,2013). Due to inability to find related
research, the emissions from diesel production from crude was neglected in the
calculations of the emissions. Further research can be done in order to determine more
detailed GHG emission and other methods of diesel production that may have lower
emissions. Table 24 shows the CO2 equivalent of each source of GHG per unit capacity of
the plant. The capacity of this plant is 303,031L of ethanol per day.
Table 24. Cradle gate GHG emissions of ethanol produced.
CO2 Equivalent
Source
[kg/L-ethanol]
Feedstock Farm
0.496
Diesel Production
0.00531
for Transportation
Transportation
0.00425
Electricity
0.000317
CO2 Produced by Plant
0.490
CH4 Produced by Plant
0.955
Total
1.945
As shown in Table 24, the highest amount of emissions from this process is the
methane produced since it is assumed the methane would be flared into the atmosphere.
However since methane is a usable resource, it could be separate out and used as a utility
99
or sold to customers if needed. It would increase the profitability of the plant greatly and
will allow the plant to have lower emissions based on a cradle to gate analysis.
In terms of the environmental benefits of bioethanol to gasoline, a LCA analysis
for a pyrolysis gasoline production resulted in GHG emission of 22.34 kg CO2 Equivalent
per liter of gasoline (Hsu, 2011) which is more than 10 times the amount found through a
biomass syngas fermentation process shown in Table 24 above. Research showed
potential of 88% decrease in GHG emission from using miscanthus as a biofuel feedstock
(Huang et al., 2009) and this result agrees with that. This shows the improvement of a
second generation ethanol production plant from a first generation plant in terms of
emissions. GHG emission from burning bioethanol is much lower than burning gasoline.
Using statistics from the United States Environmental Protection Agency, Motor
gasoline had a total GHG emission of 2.329kg CO2 equivalent and the ethanol obtained
total GHG emission of 1.52kg CO2 equivalent per liter. From those figures, just using
ethanol instead of gasoline decreased the GHG by 35%. (EPA). Adding up the emissions
from production and use, the total emission level from gasoline is 24.669 kg CO2 per liter
and 3.465 kg CO2 per liter for bioethanol produced. This means an overall GHG
reduction of around 85% by using bioethanol produced by Miscanthus syngas
fermentation which comes close to values found from research. Just looking at those
figures, replacing gasoline with ethanol will improve the environment greatly.
10. Process safety

10.1 Hazardous Materials
Several chemicals are used as raw materials, produced in different sections of the
plant or used as the ultimate products in the biochemical plant. The following chemicals
listed in Table 25 pose safety hazards to both the equipment and personnel in the
bioethanol plant.
100
Table 25. List of hazardous chemicals used and produced in the biochemical plant along with a description
of hazards they pose.
Material
Material
Description
Incompatible materials
-Can cause asphyxiation by

displacing oxygen
-Exposure may cause
headaches, drowsiness,
vomiting etc.
-LC50 of hydrogen is
>15000 ppm/1h
-Can form explosive mixture with

air and result to fire/explosion
-Reacts violently with oxidants
which leads to fire
-Lower flammability limit : 4%
-Upper flammability limit: 76%
-Auto ignition temperature: 500
to 571C
-Reactive with oxidizing

materials
-Odourless
-Colourless
-Highly toxic gas.

-Causes severe asphyxiation
and is toxic when inhaled
-Over exposure may lead to
loss of consciousness and
muscle weakness
-PEL of carbon monoxide is
50 ppm
-LD50 of carbon monoxide
is 1807 mg/kg
-Extremely flammable.
-Must be isolated from sparks and
source of heat
-Lower flammable limit: 10.9%
-Upper flammable limit of 79.2%
-Auto ignition temperature of
607C.
Reactive with oxidizing

materials
-Odourless
-Colourless
-May displace oxygen and

cause suffocation at high
concentration
-PEL of carbon monoxide is
50000 ppm
-Odourless (<
5000 ppm)
-Colourless
TLV of 1000 ppm
Gas
Carbon
Monoxide
Dioxide
Methane
Combustibility
-Odourless
-Colourless
Hydrogen
Carbon
Flammability and
Toxicity
-Incompatible/Reactive
with Magnesium,
Titanium, Aluminum
-Ignites in presence of heat,

sparks, open flames and hot
surfaces
-Decomposes to hazardous
chemicals such as carbon dioxide
and carbon monoxide.
-Accidental release poses a
serious fire or explosive hazard
-Lower flammable limit of 1.8%
an upper explosive limit of 8.4%
287C
101
Reactive with oxidizing

materials
Ammonia
-Colourless
-Strong
(pungent)
-Unpleasant
odour
-Causes extreme poisoning

when inhaled
-Causes serious eye damage
and skin burns
-Irritating and corrosive to
the respiratory system
-PEL of ammonia is 50 ppm
-LC50 of 7338 ppm over an
hour exposure
-Causes severe burns to
exposed skin
-Extremely toxic to
respiratory tract system
-TLV of 2 ppm
-LD50 of 4701 mg/kg.
-Lower flammable limit of 15.4%

-Upper explosive limit of 25%
-Auto ignition temperature
651C.
-Ignites in presence of oxidizing
materials
-Greenishyellow colour
-Pungent odour
-Nose and throat irritant

-Chlorine is converted to
hydrochloric acid in the
respiratory system
-Causes pneumonia and
pulmonary enema.
-Chlorine gas has a
threshold limit value of 0.5
ppm
-Flammable gases will form

explosive mixtures with chlorine
-Incompatible with
several compounds
including alcohols
-Colourless
-Rotten eggs
odour
-LC50 of 3124 ppm (1

Hour)
-Extremely toxic and fatal
when inhaled
-TLV of 1 ppm
-LC50 of 712 ppm over 1
hour
-Inhalation or contact with
ethanol can cause skin and
eye irritation
-PEL of ethanol is 1000
ppm
-LD50 of ethanol is
3450(oral, mouse).
-Corrosive
-Causes severe irritation to
the eyes and severe skin
burns
-Extremely flammable in
presence of sparks, heat source
-Lower flammable limit: 4.3%
-Upper flammable limit: 45%
-Auto ignition temperature:
270C
-Incompatible with
oxidants such as oxygen
difluoride
-Highly flammable and may form

explosive mixtures with air
-Lower flammable limit: of 3.3%
-Upper flammable limit:19%
423C

material
-PEL of acetic acid is 10

ppm
-Sulphur dust suspended in air is

readily ignited by flame
-Readily ignites with air
-Colorless
-Acidic odour
Hydrogen
Chloride
Chlorine
Gas
Hydrogen
Sulphide
Ethanol
Acetic acid
Sulphur
-Yellow color

materials
-Incompatible with
halogens and acids
-Reactive with acetic

anhydride, propylene
oxide, sodium
hydroxide
102
LC50/LD50 = Lethal concentration/Lethal dose at which 50% of animal (tests) were

killed
TLV = Threshold limit value
PEL = Permissible exposure limit
10.2 Process Hazards

An advantage of producing ethanol through the hybrid process is the generally mild
operating conditions. Fermentation requires atmospheric conditions and a temperature of
37C which does not pose any extreme effects. On the other hand, gasification reaction
operates at a high temperature of 850C. This results to extreme temperatures that can
melt tubes and pipes as well as cause ignition and start fire. In order to mitigate this risk,
temperature and pressure control sensors and process control loops were put in place in
order to counteract any sudden disturbances to temperature and pressure. Furthermore,
the flow rate of the feed to the gasification is controlled in order to prevent runaway
reactions.
The preferred material of construction was high grade stainless steel.
In addition to the gasifier reactor, safety control systems were integrated in to the
distillation columns. Distillation columns pose the risk of high vapour flow rates which
can lead to over pressurised distillation column and pressure vessels. This may cause
explosions which can damage equipment and more importantly personnel. Temperature,
pressure and flow rate control systems were put in place in order to mitigate these
hazards. Furthermore, pressure relief systems were integrated into the pressure vessels
which will release vapour to relieve high pressures. Duplicated temperature, pressure and
flow rate sensors were also included in the design in order to safely and accurately
measure any deviations. Automatic safety shutdown systems such as the use of solenoid
valves which halts feed to equipment in case of unsafe conditions were also included in
the plant design.
Other equipment such as pumps, heat exchangers, valves are also integrated with
control systems which counteracts any disturbances (which may pose safety hazards) to
103
flow rates, temperature and/or pressure. A more detailed safety analysis of the distillation
columns is included in the HAZOP analysis in Appendix 3.
Bypass streams were designed to handle increased flow rates, pipe leaks and
prevent pipes from bursting caused by blockages. Storage vessels were also placed in key
areas of the plant such as prior to the distillation column and fermentation vessel to not
only increase reliability but to also store materials in case feed to equipment must be
stopped to prevent safety hazards such as high level of material in the equipment. Other
operating conditions include flushing equipment with an inert gas in order to prevent
unwanted reactions caused by foreign materials. Finally, equipment are released of any
gas by opening vent valves before commissioning and decommissioning.
11. Risk Assessment

11. 1 Technical
One of the largest risks regarded with the syngas fermentation process is the
bacteria used in fermentation. Since the bacteria is not yet commercialized or even used
on a regular basis, the bacteria is completely unavailable for use on an industrial scale and
is extremely expensive to purchase a culture from suppliers. As previously said in this
report, one culture of these bacteria will cost over 70 dollars. In order to make this
process feasible, growing the bacteria would be the most realistic option in order to
provide the required amount of bacteria. Since the said bacteria do not have any history
related to growing, various tests has to be done in order to see if it can be grown in large
quantities for the use in the industry. If the bacteria cannot be grown, then this process is
not a viable method of ethanol production.
Another large risk technologically is the lack of knowledge and experience in the
field of cellulosic bioethanol or even second generation bioethanol in general. As of now,
Ontario does not have a full operating second generation bioethanol plant in operation
(Decker, 2009). If this purposed plant were somehow to get approval, this would be the
first plant ever. Because of the lack of industrial size plants, no information is known on
the subjects of these plants in this region. For example, questions such as feedstock
growth, temperature effects, demands and prices are all undetermined and since there are
104
no historical trends, establishing a knowledge base is extremely difficult which means a

very extreme risk. Since there are no industrial scale syngas fermentation unit, creating a
plant like this would require the designing of a unit without any prior knowledge which
leads to very large uncertainties which may make it difficult to develop.
There are also technological risks when growing the feedstock such as
Miscanthus. From research, it may take up to five years in order to introduce the crop
properly and to have desired growth in a new environment (Roy, 2014). Therefore if
decided to start with the growth of Miscanthus, a solution that will speed up the growing
process greatly would need to be find in order for this method to be feasible. The amount
of feedstock needed for this plant is also very large and since there are always risks with
crop growing in terms of climate, securing the required amount of feedstock every year
can be difficult. This risk is not of large concern for this plant however since the syngas
fermentation process is able to use almost every type of biomass available to produce
ethanol. If not enough miscanthus can be grown or bought, other feedstock such as corn
stover can be used in its place.
11.2 Societal
Since this plant does not currently exist in Ontario, the effects of opening a new
plant are not available to see. However, one of the risks of opening a new plant could be
the potential impact on the workforce in the community. If the plant was opened in an
area of low population, there would not be enough workers to supply the demand of the
plant. Without workers, the plant would not produce anything, therefore the location of
the plant is extremely important. Building a plant too close to large community may also
have risks due to the response of the residents. Having a plant nearby may result in higher
noise levels, higher emissions in the air and even potential pollution in the area. If these
issues are not managed correctly, the residents could oppose the opening of the plant or
the operation once built. If opposition were to occur, then it would lead to huge loss in the
economics. This will be discussed further in the next section.
Since the work force for each community is limited, adding further demand for
workers in the community may take away jobs from other sectors in the community that
105
may result in lack of workers in those sectors. For example, workers from service sectors
could see a job at the plant as a better choice and relocate themselves in the plant instead.
By doing that, the workers in the service sector are greatly reduced, it may even cause a
shortage of workers in service. This would lead to longer hours as well as more
responsibilities than before. Unless the worker wages are increased, then the workers will
be unsatisfied and important services such as caring for children and elders may take a hit
in the total availability of workers. With the need to grow the feedstock, more land will be
required and in worst case scenarios it may require residents to relocate in order to free up
the land needed which would cause opposition from residents. If relocation of residents is
unavoidable, there would be additional costs on the start-up of the plant since there would
need to be compensation for the relocated.
11.3 Economical
Some of the economic risks can be seen in the sensitivity analysis. Since the most
important part of economics is the profit, the goal must always be to maximize the profit
of the plant. Looking at the sensitivity analysis, the largest factor that will affect
profitability is the ethanol sales price. Since the price of ethanol fluctuates and
introducing more ethanol into the market will likely to decrease the price of ethanol at the
beginning, the operation is at risk of losing profits especially at the start of operation. The
second most significant factor explored on the sensitivity analysis is the Energy price. In
this case, the energy price is directly related to the electricity cost of running the plant.
Since profit is relatively sensitive to the energy price, any increase in the energy price will
result in loss of profits.
Looking at the recent trends of energy prices in Ontario, the price has been
increasing steadily over the past years (stats Canada) therefore assuming the energy prices
will continue in that trend, there is a risk of profit much lower than anticipated once the
plant is in production. Another concern economically would be the demand of ethanol. In
estimations, the market for ethanol is assumed to be open where there are no barriers to
entry. In this market, any cowpony that chooses to produce ethanol will be able to enter
the market without any restrictions which means the market could potentially become
106
oversaturated with supply and not enough demand to keep the price at an acceptable rate.
A market such as this has the potential to drive the market price below the profitable level
for every supplier in the market which will lead stop in operation for many plants.
While the government can demand a production limit to every plant in order to
prevent the market from deteriorating, larger plants will require a certain level of
production in order to stay profitable and may simply decide to not operate under the
constraints while smaller companies will not be able to match the total demand by
themselves. This would then raise the price of ethanol and also raise the unemployment in
the area due to plant shutdowns (Vazirani, 2007). Since the gasoline industry is so large
and has large amounts of influence in the world, it will not be easy to even secure a
market in order to become profitable in ethanol production. Even now, there are many
uses of gasoline that cannot be replaced with ethanol and at best, most of the machines
require a blend of ethanol and gasoline in order to function. In order for the production
process to be profitable and ensure sufficient demand, further ethanol uses will need to be
developed and ethanol will need to be integrated as a viable source of energy along with
gasoline. The government will likely have to intervene in order to establish the base for
ethanol to become successful economically.
107
Appendices
Appendix 1 Various Lists Relating to Process
List of Materials
Adsorbent (Zinc Oxide [ZnO])

Bacteria (Clostridium Ijungdahlii)
Corn Steep Liquor (Fermentation Medium)
Feedstock (i.e. Miscanthus, Switchgrass, Corn Stover etc.)
Natural Gas
Nutrients
Oxygen
Water
List of Equipment
Adsorption Vessels
Conveyer Belts
Cyclone
Distillation Columns
Fluidized Bed Gasifier
Gravity Chute
Heat Exchangers
Pumps
Screw Belts
Sensors
Stirred-Tank Bioreactor
Storage Tank (Distillation Column Feed)
Storage Tank (Feedstock)
Storage Tank (Fermentation Medium)
Valves (i.e. Manual, Automatic)
Wet Scrubber (Spray Tower)
List of Symbols
Q Heat transferred/time [W]

U Overall heat transfer coefficient [W/m2]
108
A Heat transfer area [m2]

Tm Mean temperature difference [C]
Tlm Log-mean temperature difference [C]
Ft Correction factor
T1 Hot fluid inlet temperature
T2 Hot fluid outlet temperature
t1 Cold fluid inlet temperature
t2 Cold fluid outlet temperature
109
Appendix 2- Detailed Equipment List

Equipment List
Displayed Text
Type
Specific
B-101
B-102
B-103
Conveyer Belt
Electricity
Screw Belt
Chute
E-101
Gravity Chute
Insulation tip
Cooling syngas from R-101
E-102
Heat Exchanger
Cooling recycle from T-101 before S-101

Reboiler for T-101
None
Cooling Water
Cooling Water
Cooling of syngas during gas cleaning
E-103
E-105
Rubber
Screw Belt
B-105
E-104
Utility
Conveyer Belt
Belt
B-104
C-101
Extra
Conveyer Belt
Cooling Water
Shell & Tube
Cooling Water
Low Pressure Steam
E-106
Condenser for T-101
Cooling Water
E-107
Reboiler for T-102
Low Pressure Steam
E-108
Condenser for T-102
GC-101
Cyclone
None
None
GC-102 A
Adsorption Column
Adsorbent (ZnO)
Oxygen
GC-102 B
Gas Cleaning
Cooling Water
Adsorption Column
Adsorbent (ZnO)
Oxygen
GC-103
Wet Scrubber
Spray Nozzle
Cooling Water
P-102 A
Pump broth from R-102 to T-101
None
Electricity
Hammer Mill
None
Electricity
Spray Dryer
Conveyer Belt
Steam
Gasifier
Fluidized Bed (Quartz-Sand)
Oxygen/Natural Gas
Fermenter
Bacteria (Clostridium ljungdahlii)

None
Electricity/Water
Medium Storage
Column feed storage
None
None
Distillation Column
Trays
None
Distillation Column
Trays
None
Condenser vessel for T-101
None
None
Condenser vessel for T-102
None
None
P-101 A
Pump new medium into R-102
P-101 B
P-102 B
Pump broth from R-102 to T-101
P-103
Pump from S-102 to T-101
P-104 A
P-104 B
Pump
P-105 A
Pump liquid from V-101
P-105 B
P-106 A
Pump bottoms from T-102
P-106 B
P-107 A
Pump liquid from V-102
P-107 B
PT-101
PT-102
R-101
R-102
S-101
S-102
T-101
T-102
V-101
V-102
Pump bottoms from T-101
Pre-Treatment
Reactor
Storage
Tower
Vessel
Figure A2.1. Detailed equipment list.
110
None
Appendix 3 - HAZOP Study

Node T-101 Distillation Column
Parameters Flow, Temperature, Pressure
Guide Words Low, High, No (if applicable)
Parameter: Flow
Guide Word
Deviation
Low liquid column flow

Low
Low vapour column flow
High liquid column flow
Cause
Consequence
- Fouling in the pipe

- Faulty level control loop from reactor,
causing feed flow to decrease
- Reboiler duty too high
- Condenser duty too low
- Pump breakdown on reflux loop
- Damage the pumps

surrounding the column
- Poor separation quality
- Not meeting production
requirements
- Column level control loop using distillate

flow
- Reflux and reboiler flow controls can also
help raise liquid flow
- Flow indicators placed at inlets and outlets
of column
- Reflux flow too high due to temperature

control loop, or condenser vessel level
control loop
- Reboiler duty too low, or boiler feed
water flow/temperature too low
- Faulty level control on fermenter, feeding
excess liquid from reactor
- Reflux too high from temperature control
or condenser level control
- Can lead to weeping and

even dumping of the trays
- Reduces separation
efficiency
- Control loop within reboiler controlling

utility flow in ratio to the feed flow
- Reboiler flow also in ratio control with
bottoms flow to minimum vapour flow met
- Reduce separation quality

- Can lead to flooding of
condenser vessel
- Can damage pumps
- Reboiler flow too high from column level

control
- Reflux too low
- Condenser duty too low or cooling water
- Can knock trays out of

place
- Can cause flooding within
the column
- Column level alarms

- Reflux flow control loop
- If feed flow too much for column to handle,
send feed to storage tanks (done using
control loop)
- Level alarms placed on the column
- Reflux control loop as preliminary control
to prevent rising vapour flow
High
High vapour column flow
Action
flow too low
No liquid column flow
No
No vapour column flow
- Leads to entrainment
- Tray efficiency reduced
- Increase pressure within
column
- Breakdown of pumps
surrounding column
- Column will not operate,
no separation
- Broken valve/piping
- Reactor failure/shutdown
- Fermenter flash vessel rupture
- Reboiler broken or fouled

- Broken pipe in reboiler loop
- Boiler water issues (temperature too low,
or control causes low flow)
- Reboiler ratio error causing no flow to go
to reboiler
- Poor separation, reduced

product quality, poor column
performance
- Can lead to weeping or
dumping of trays
- Feed sent to storage tanks until distillation

column is back and running properly
- Parallel pumps used around column with
alternate motors in case one pump runs dry
or gets damaged
- Control loop placed on the reboiler loop
controlling utility flow in ratio to the feed
flow entering the column
Parameter: Temperature
Guide Word
Deviation
Low
Low column temperature
High
High column temperature
Cause
- High liquid flow through the column
- Low vapour flow through the column
- Feed flow is too large
- Reboiler duty too low
- Reflux flow too high
- High vapour flow within the column, or
low liquid flow
- Vapour condenser not working properly
112
Consequence
- Poor column operation
and product purity directly
influenced by column
temperature
- Can lead to unsafe
conditions and poor column
operation causing product
purity to suffer
Action
- Temperature control loop added using
reflux flow rate to control column
temperature and distillate flow rate
- Two alternate temperature sensors were
used sending the average value to the
controller
- Temperature indicators were placed at the
top and bottom of the column, as well as
near the feed tray.
Parameter: Pressure
Guide Word
Deviation
Low
Low column pressure
High
High column pressure
Cause
- Reboiler duty is too low
- Feed flow is too low
- Reflux flow too high
Consequence
- Can lead to weeping or
dumping of trays
- Column needs to be
restarted if dumping occurs
- Decrease in separation
efficiency
- If vapour flow is too high into the column

from reboiler loop
- If condenser duty is too low
- Outlet valve is broken or malfunctioning
- Low/no liquid flow into column
- Can cause entrainment,

reduce tray efficiency
- Can eventually lead to
flooding in the column
- Significant decrease in
separation efficiency
113
Action
- Pressure within column is controlled via
utility flow into the condenser
- Two alternate pressure sensors used to
increase reliability, where the average
reading between the two sensors is sent to
the controller
- Low pressure alarm placed near top of
column
- Pressure relief is achieved via a safety
relief valve near the top of the column in
case of pressure buildup
- High pressure alarm placed near the top of
the column
- Pressure indicators placed throughout the
column
Figure. A3.1 HAZOP changes reflected on a P&ID.
References
Abubackar, H., & Veiga, M. et. al., (2011). Biological conversion of carbon monoxide:
Rich syngas or waste gases to bioethanol. Biofuels, Bioproducts and Biorefining, 5, 93114. doi: 10.1002/bbb.256
Advanced Biofuels USA . (2011). Syngas Fermentation, the third pathway to cellulosic
ethanol. Retrieved from http://advancedbiofuelsusa.info/wp-content/uploads/2011/06/3rdPathway-Revision-FINAL-6-17-11.pdf
Agriculture and Agri-Food Canada. (2014). Length of Growing Season in Ontario.
Retrieved From:
http://www.agr.gc.ca/eng/science-and-innovation/agricultural-practices/climate/futureoutlook/climate-change-scenarios/length-of-growing-season-inontario/?id=1363033977515
Alvira, P., Toms-Pej, E., et. al., (2010). Pretreatment technologies for an efficient
bioethanol production process based on enzymatic hydrolysis: A review, Bioresour.
Technol., 101(13), pp. 48514861.
B.C. Ministry of Finance. (n.d.). How the Carbon Tax Works. Retrieved from:
http://www.fin.gov.bc.ca/tbs/tp/climate/A4.htm
Basu, P. (2013). Biomass Gasification, Pyrolysis, and Torrefaction: Practical Design and
Theory (2nd ed.). Amsterdam: Elsevier; Academic press.
Canada Revenue Agency. (2015) Classes of Depreciable Property. Retrieved from:
http://www.cra-arc.gc.ca/tx/bsnss/tpcs/slprtnr/rprtng/cptl/dprcbl-eng.html#Class_43
Daniell, J., & Kpke, M. et. al., (2012). Commercial Biomass Syngas Fermentation.
Energies, 5, 5372-5417. doi: 10.3390/en5125372
Dunn, J.B., Mueller, S., et. al., (2013) Land-use change and greenhouse gas emissions
from corn and cellulosic ethanol.
Environment Canada. (2015). 1981-2010 Climate Normals & Averages. Retrieved from:
http://climate.weather.gc.ca/climate_normals/index_e.html
Foust, T.D., Aden, A., Dutta, A., and Phillips, S. (2009). An economic and
environmental comparison of a biochemical and a thermochemical lignocellulosic ethanol
conversion processes, Cellulose, 16(4), pp. 547565.
Intergovernmental Panel on Climate Change. (2013). Climate Change 2013: The Physical
Science Basis. Retrieved from: https://www.ipcc.ch/report/ar5/wg1/
Khanna, M., Dhungana, B., and Clifton-Brown, J. (2008) Costs of producing miscanthus
and switchgrass for bioenergy in Illinois, Biomass Bioenergy, 32(6), pp. 482493.
doi: 10.1016/j.biombioe.2008.08.008
Kratky, L., & Jirout, T. (2010). Biomass Size Reduction Machines for Enhancing Biogas
Production. Chemical Engineering & Technology, 391-399. doi:
10.1002/ceat.201000357
Kwiatkowski, J.R., McAloon, A.J., et. al., (2006). Modeling the process and costs of fuel
ethanol production by the corn dry-grind process, Ind. Crops Prod., 23(3), pp. 288296,
Li, H., & Chen, Q. (2012.). Evaluation of a biomass drying process using waste heat
from process industries: A case study. Applied Thermal Engineering, 71-80.
doi: 10.1016/j.applthermaleng.2011.10.009
Liska, A. J. H. S. Yang, et. al., (2009) Improvements in Life Cycle Energy Efficiency
and Greenhouse Gas Emissions of Corn-Ethanol, J. Ind. Ecol., 13(1), pp. 5874. doi:
10.1111/j.1530-9290.2008.105.
Maddipati, P., & Atiyeh, H. et al., (2011). Ethanol production from syngas by
Clostridium strain P11 using corn steep liquor as a nutrient replacement to yeast extract.
Bioresource Technology, 102(11), 6494-6501. doi: 10.1016/j.biortech.2011.03.047
Michel, R., and Rapagn, S., et. al., (2011). Catalytic steam gasification of Miscanthus X
giganteus in fluidised bed reactor on olivine based catalysts. Biomass and Bioenergy, 35,
2650-2658. doi: 10.1016/j.biombioe.2011.02.054
National Resources Canada. (2014). ecoENERGY for Biofuels Program: About the
Incentive. Retrieved from:
http://www.nrcan.gc.ca/energy/alternative-fuels/programs/ecoenergy-biofuels/3591
National Resources Canada. (2013). Ethanol Benefits. Retrieved from:
https://www.nrcan.gc.ca/energy/alternative-fuels/fuel-facts/ethanol/3499
116
Ontario Ministry of Finance. (2015). Gasoline Tax. Retrieved from:

http://www.fin.gov.on.ca/en/tax/gt/
Phillips, S. (2007). Technoeconomic Analysis of a Lignocellulosic Biomass Indirect
Gasification Process To Make Ethanol via Mixed Alcohols Synthesis. Industrial &
Engineering Chemistry Research, 46(26), 8887-8897. doi: 10.1021/ie071224u
Piccolo, C., & Bezzo, F. (2009). A techno-economic comparison between two
technologies for bioethanol production from lignocellulose. Biomass and Bioenergy,
33(3), 478-491.
Pimentel, D., and Patzek, T., (2008). Ethanol Production Using Corn, Switchgrass and
Wood; Biodiesel Production Using Soybean, in Biofuels, Solar and Wind as Renewable
Energy Systems. Ed. Springer Netherlands, pp. 373394.
Roy, P. (2014). Life Cycle Assessment of Ethanol produced from Lignocellulosic
Biomass: Techno-economic and Environmental Evaluation. Ph.D. Thesis. The
University of Guelph: Canada
Snchez, O.J., and Cardona, C.A. (2008). Trends in biotechnological production of fuel
ethanol from different feedstocks, Bioresour. Technol.,99(13), pp. 52705295.
Saxena, J., & Tanner, R. (2012). Optimization of a corn steep medium for production of
ethanol from synthesis gas fermentation by Clostridium ragsdalei. World Journal of
Microbiology and Biotechnology, 28(4), 1553-1561. doi: 10.1007/s11274-011-0959-0
Seider, W., & Seader, J. (2004). Product and process design principles: Synthesis,
analysis, and evaluation (2nd ed.). New York: Wiley.
Solomon, B. D., Barnes, J.R, and Halvorsen, K.E, . (2007). Grain and cellulosic ethanol:
History, economics, and energy policy, Biomass Bioenergy, 31(6), pp. 416425.
doi: 10.1016/j.biombioe.2007.01.023
Tanner, R., & Miller, L. et. al., (1993). Clostridium ljungdahlii sp. nov., an Acetogenic
Species in Clostridial rRNA Homology Group I. International Journal of Systematic
Bacteriology, 43(2), 232-236.doi: 0020-7713/93/020232-05
Towler, G., & Sinnott, R. (2013). Chemical engineering design principles, practice and
economics of plant and process design, second edition (2nd ed.). Kidlington, Oxford,
U.K.: Butterworth-Heinemann.
117
United Nations Conference on Trade and Development. (2009). The Biofuels Market:
Current Situation and Alternative Scenarios. Retrieved from:
http://unctad.org/en/Docs/ditcbcc20091_en.pdf
U.S Department of Energy. (2014). Alternative Fuels Data Center Fuel Properties
Comparison. Retrieved from:
http://www.afdc.energy.gov/fuels/fuel_comparison_chart.pdf
U.S Department of Energy. (2015). Alternative Fuel Price Report. Retrieved from:
http://www.afdc.energy.gov/fuels/prices.html
U.S. Environmental Protection Agency. (n.d.) Air Pollution Control Technology Fact
Sheet. Retrieved from http://www.epa.gov/ttncatc1/dir1/fsprytwr.pdf
Wang, M., Wu, M., and Huo, H. (2007). Life-cycle energy and greenhouse gas emission
impacts of different corn ethanol plant types, Environ. Res. Lett., 2(2), p. 024001.
Woods, D.R. (1983). Cost Estimation for the Process Industries. McMaster University,
Hamilton: Canada.
Woolcock, P., & Brown, R. (2013). A review of cleaning technologies for biomassderived syngas. Biomass and Bioenergy, 52, 54-84.
Yang, B., and Wyman C.E . (2008). Pretreatment: the key to unlocking low-cost
cellulosic ethanol, Biofuels Bioprod. Biorefining, 2(1), pp. 2640. doi: 10.1002/bbb.49
118

Bio-Ethanol Productions Plant in Ontario

Uploaded by

Copyright:

Available Formats

Bio-Ethanol Productions Plant in Ontario

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bio-Ethanol Productions Plant in Ontario

Uploaded by

Copyright:

Available Formats

What feedstock and process were selected for the proposed bioethanol plant?

What feedstock and process were selected for the proposed bioethanol plant?

What are the main steps in the syngas fermentation process?

What are the main steps in the syngas fermentation process?

DESIGN OF MULTI-FEEDSTOCK BIO-ETHANOL

Winter Term 2015

4W4 2015 Syngas Fermentation

4W4 2015 Syngas Fermentation

4W4 2015 Syngas Fermentation

5.2.1 Start-up ............................................................................................................ 36

4W4 2015 Syngas Fermentation

11. 1 Technical........................................................................................................... 104

4W4 2015 Syngas Fermentation

4W4 2015 Syngas Fermentation

4W4 2015 Syngas Fermentation

1.2 Economic Approximation of Process

4W4 2015 Syngas Fermentation

1.3 Relevant standards

4W4 2015 Syngas Fermentation

In this syngas fermentation process, this requirement is very important as syngas

4W4 2015 Syngas Fermentation

2.2 2nd generation feedstock

4W4 2015 Syngas Fermentation

4W4 2015 Syngas Fermentation

In second generation ethanol production, lignocellulosic ethanol has many

Figure 1. A hydrolysis-based cellulosic ethanol production process. (Dwivedi et al., 2009)

4W4 2015 Syngas Fermentation

Ammonia Fiber/Freeze Explosion (AFEX). The biomass is exposed to liquid ammonia at

4W4 2015 Syngas Fermentation

3.1 Hybrid process

4W4 2015 Syngas Fermentation

3.2 Biochemical process

4W4 2015 Syngas Fermentation

production is from pre-treatment, a cost that is non-existent in the thermal-biochemical

4W4 2015 Syngas Fermentation

Cost (Millions $U.S. 2013)

Saccharification and fermentation

3.3 Thermochemical process

4W4 2015 Syngas Fermentation

4W4 2015 Syngas Fermentation

In the thermochemical process, sulfur impurities must be eliminated before the

Cost (in Millions $U.S. 2013)

3.4 Bioethanol Location

4W4 2015 Syngas Fermentation

regions which are 1)Eastern (Ottawa), 2)Central (Hamilton/Toronto), 3)Southwestern

4W4 2015 Syngas Fermentation

4. Overview of Proposed Process Design

A block flow diagram

Figure 3. Block flow diagram of Syngas Fermentation process

Fermentation has the advantage of operating at low temperatures (~37C) and

4W4 2015 Syngas Fermentation

Drying is a necessary pre-treatment step because the moisture within the

4W4 2015 Syngas Fermentation

Many types of gasifier configurations

4W4 2015 Syngas Fermentation

Figure 4. Typical fluidized bed gasifier configuration (Basu, 2013)

4W4 2015 Syngas Fermentation

Figure 5. Typical biomass feeding system for fluidized bed gasifier.(Basu,

4.1.3 Gas Cleaning

4W4 2015 Syngas Fermentation