Abstract - Sustainable Transportation Energy Pathways Program ...

anticipatedbananaBiotechnology

Feb 20, 2013 (4 years and 8 months ago)

362 views

Modeling biofuel

production

from
wastewater
-
grown microalgae

Research Proposal

Brendan Higgins


Abstract


Biofuel

production from wastewater
-
grown algae offers the potential to produce renewable
transportation fuels in a cost
-
effective and environmentally sound manner.
A
lgae minimize
eutrophication by removing inorganic nutrients from the wastewater while pro
ducing

b
iomass

as
a co
-
product
. This research seeks to
quantify the amount of

algal

biofuel

that

can
be produced
from wastewater streams in California by

conducting a resource assessment and developing

a
supply chain

model. Specifically, the model will rely on spa
tial data and mixed
-
integer linear
programming to determine
the optimal size and location for
cultivation and biorefinery facilities
.
Preferred regions will be indentified as well as total

expected

production

of biofuel and co
-
products.

The results can
inf
orm
managers of agricultural and municipal wastewater facilities as
they make decisions about

whether to include algal technology in future

upgrades and expansion.
The model could also aid state and federal agencies as they determine the role of algal fuel
s in
meeting sustainability goals such as the Low Carbon Fuel Standard and Renewable Fuel
Standard.


Introduction


Algal biofuels have been touted as a way to reduce foreign oil dependence, greenhouse gases,
and the land requirements of biofuel production
(Chisti 2007)
.
The US Department of Energy’s
Aquatic Species Program conducted eighteen years of research on algal biofuel
production with
a particular focus on biodiesel

from open pond systems
. The program ended in 1996
with the
conclusion

that oil from microalgae could not compete with
the low cost of petroleum in the

1990s

(Sheehan, Dunahay et al. 1998)
. A cost
estimate

found that oil from

pond
-
grown

microalgae
would cost $56 per barrel
(1994$)
under fairly optimistic production conditions or as
little as $38 per barrel under very optimistic production conditions

(Benemann and Oswald
1996)
. Using the Chemical Engineering Plant Cost Index and the Construction Cost Index to
adjust appropriate
cost

factors

to 2009 dollars, these figures are $9
3

and $6
5

per barrel
respectively.


Assumed Productivity
(g

m
-
2

d
-
1
)

Lipid Content

Cost per Barrel
($1994
)

Cost per Barrel
($2009)

30

50%

$56

$9
3

60

50%

$38

$6
5

15

20%


~ $
4
50


The growth rates and lipid productivities assumed in
the 1996

cost estimate

have never been
achieved sustainably in outdoor cultures. Sustainable year
-
round productivities of 16
-
20 g m
-
2

day
-
1

with lipid contents of 20
-
28%
were

achieved
under the Aquatic S
pecies Program

(SERI
1984)
.
Development of a

conservative scenario with productivity of 15 g m
-
2

day
-
1

and

20% lipid
content
resulted

in a much higher cost estimate

of $4
5
0 per barrel

of oil
.
This presents a
slightly
more conser
vative production level than that

observed
during the Aquatic Species Program.


The findings of the Aquatic Species Program suggest that without a major technolo
gical
breakthrough, alga
l

biofuel production is not cost effective unless significant co
-
product value
can be derived. As a result, there is considerable interest in producing biofuels from wastewater
grown algae.
Stri
n
gent wastewater regulation
s ensure

a
market for wastewater treatment and
algae bio
fuel

could become a valuable co
-
product.

In addition, Green (1996) contends that algae
wastewater treatment facilities have lower capital and operating costs than equivalent activated
sludge plants.


Algae based

systems for the treatment of municipal and agricultural wastewater have been
employed successfully to remove excess nutrients
(Green, Bernstone et al. 1996)
.
Under proper
management
,

algae
increase
the dissolved oxygen

(DO)
level

in the water
which can be utilized
by BOD consuming bacteria. Algae systems can also be combined with anaerobic digestion of
primary
wastewater

where an algae pond receives the digester effluent. This system results in
methane gas
production fr
om the digester

along with
algal

biomass

production from the pond

(Green, Bernstone et al. 1996; Woertz, Feffer et al. 2009)
. The microalgae consume fixed
nitrogen and phosph
o
rous
to synthesi
ze protein biomass
(Kebede
-
Westhead, Pizarro et al. 2003;
Woertz, Feffer et al. 2009)
. If released into natural water bodies, these inorganic nutrients
contribute

to eutrophication and ecosystem disruption
(Willemsen 1980; Oswald 2003)
.



Wastewater from both municipal sources and dairies contain lower C:N and C:P ratios than algae
biomass
(Woertz, Feffer et al. 2009)
. Thus, carbon is a limited resource and needs to be provided
in order for the algae to fully utilize the nitrogen and phosphorous. Carbon dioxide has
traditional
ly

been the carbon source supplied to algae mass cultures
(Sheehan, Dunahay et al.
1998)

and has the b
enefit of consuming a known greenhouse gas

(IPCC 2006)
.


Wastewater grown algae can be used as fertilizers
or

animal feed
(Kebede
-
Westhead, Pizarro et
al. 2004)

but also can be used to make biofuel. Woertz (2009) found that wastewater grown algae
can have lipid contents as high

as 14
-
29% depending on the wastewater reactor
;

lipids

can be
used to make biodiesel
and other renewable fuels
.

Proposal Concept


To date, there has been little analysis of the potential for scaled up algal biofuel production from
wastewater. Part of the
reason for the limited research has been due to a lack of

algae

productivity data
associated with different types of wastewater

systems
. Such data has become
more widely available recently as pilot scale studies
have been

conducted.


This research
builds o
n previous biofuel modeling research at UC Davis by quantifying the
amount

of biofuel that ca
n be produced from wastewater
-
grown algae in California.
A

profit
maximizing mixed
-
integer linear programming model will be developed that will determine the
optim
al size and location of algae
cultivation facilities
. The model will also determine locations
for biorefineries and storage facilities.
Production sites will be selected based on
feasibility

criteria

and the highway

and rail

network
s

will serve as

potentia
l

transportation links

both for
off
-
site
feedstock delivery

as well as finished fuel transport
.


Municipal wastewater and dairy wastewater will be examined.
The nutrient contents (particularly
nitrogen and phosphorous) of diff
erent wastewater streams will
be

estimated

from database data
.
County level data on animal wastes is available from the California Biomass Collaborative

(Williams 2007)

and

w
astewater treatment influent d
ata
by plant is available from
US EPA
databases
.


Two primary methods have been employed to treat wastewater using algae: high rate ponds
(HRP) and algal turf scrubbers (ATS).

The high rate pond is typically 10
-
20 cm deep and mixed
by paddlewheel

(Sheehan, Dunahay et al. 1998)
. The algal turf
scrubber is an inclined flow
-
way
over which wastewater flows

in repetitive bursts
.
A

mixture of algae species attach to the flow
-
way

of the ATS

and absorb the wastewater nutrients

(Craggs, Adey et al. 1996)
. Both methods
will be inc
luded in the analysis and data on system productivity and costs will be obtained from
literature.

Limited literature data is available on the biomass composition

of algae grown using
these two methods so sensitivity analysis will be performed.


Both
the HRP and ATS

can be fed
aerobically treated

or anaerobically digested wastewater.
Aerobic treatment methods convert BOD (biological oxygen demand) into CO
2

through the
action of bacteria and are most appropriate for wastewaters with low COD (chemical oxygen
demand) loading rates
(
Malina and Pohland 1992)
.
Anaerobic digestion

is more appropriate for
wastewaters with moderate to high COD loading

and

removes a portion of the
COD (chemical
oxygen demand)

in the form of methane and carbon dioxide

(Malina and Pohland 1992)
.

The
methane can be sold directly or combusted for heat and electricity while the CO
2

can be supplied
to the algae.
However, the CO
2

a
vailable from anaerobic digestion
may not

cover
the full carbon
requirement of the algae depending on the C:N and C:P ratio of the wastewater
.

Thus make
-
up
CO
2

may be required.


Make
-
up c
arbon dioxide will be supplied to the algae growth facilities in one
of two forms:
industrial grade CO
2

or as flue gas.
Most pilot scale cultivation facilities use industrial grade CO
2

but a life cycle assessment has shown that using power plant flue gas results in lower greenhouse
gas emissions
(Clarens, Resurreccion et al. 2010)
.
Still, industrial grade CO
2

will be included as
an option in the model since it has the advantage of lower capital costs

(Sheehan, Dunahay et al.
1998)
.
Existing power plants, ethanol plants, ammonia plants, and other
facilities

that

could
provide flue gas will be incorporated into the model

provided spatial data is available
.
This gas
will be transported by pipeline to the cultivation site.


The Aquatic Species Program found that flocculation and settling followed by c
entrifugation was
the most cost effective harvesting method from high rate ponds

(Sheehan, Dunahay et al. 1998)
.
After centrifugation, the algal biomass has a water content of approximately
78
-
92
%

wet basis
(Molina Grima, Belarbi et al. 2003; Johnson and Wen 2009)
.
This ha
rvesting method will be
included if high rate ponds are chosen by the model.
Algae attached to the algal turf scrubber can
be harvested by mechanical scraping
(Kebede
-
Westhead, Pizarro et al. 2
003)

or a vacuum
system
(Craggs, Adey et al. 1996)
.

Water content of the harvested biomass was found to be

about
9
4
%

wet basis

(Johnson and Wen 2
009)
.

A screw press may be used
to dewater the
harvested biomass, reducing
the water content to 80% wet basis
(Pizarro, Mulbry et al. 2006)
.


Most
fuel conversion processes

are well established

for biomass

but
algae sp
ecific data is limited
in the literature.
Some of the conversion technologies considered
require

nearly

dry biomass and
drying cost
s

will be
included for

these cases
.

Biomchemical c
onversion technologies that will be
considered include:



Lipid extraction and transesterification (biodiesel)



Lipid extraction and hydrotreatment (renewable diesel
)



Anaerobic d
igestion (methane)



Alcohol fermentation (ethanol)

Thermochemical conversion technologies that will be considered include:



Wet
gasification (methane)



Gasification with Fischer Tropsch (gasoline, diesel)



Thermochemical liquefaction (
gasoline, diesel
)



Pyrolysis (gasoline, diesel)

Processing costs, conversion efficiencies, and any necessary biomass drying will be incorporated
into th
e model.
Cost data for a variety of thermochemical processes

and alcohol fermentation

will be obtained from work done by Antares Group
(2009)
.

Other

cost data on con
version
processes will be
estimated from

literature

data
.


Emission factor data will be incorporated into the model.
The production effects of pricing
greenhouse gases will also be explored.

Model Results



Model outputs include size and siting decisions f
or algae
cultivation

facilities, biorefineries, and
storage facilities
. Maps will show the spatial arrangement. The model will also provide

estimates
of

industry profit for given price points for fuel and co
-
products

under

a variety of scenarios
.
Multiple model runs at varying prices will allow for the development of a supply curve for algal
biofuel from wastewater.

Research Impact


The model

and its results will prove use
ful to a number of stakeholders including

dairy and
feedlot operators, munic
ipal treatment plant operators,
and state and federal agencies
. Biofuels
from wastewater
-
grown algae hold the promise of enhanced water treatment, reduced greenhouse
gas emissions, improved energy security, and profit for agricultural operations.


The
supp
ly chain model

can serve as a strategic planning tool for government and industry
stakeholders as the algal fuel industry grows. For a given fuel price, agricultural managers can
determine if their facility is in a position to take advantage of algal techn
ology.

Municipalities
considering renovations or upgrades to existing wastewater facilities could also determine if they
are in a good location to pursue algal treatment technology.

Government agencies
can use the
model results to determine the expected co
ntribution of wastewater
-
grown algal fuels toward the

Low Carbon Fuel Standard,

Renewable Fuel Standard
,

and other sustainability goals.

Limitations


Growth rates and biomass compositions in the model are average

values

or
values obtained from
regression
models
. Therefore, actual yields will vary from site to site.

On
-
site testing should take
place prior to investment in algal production infrastructure.
The model results can serve as a
guide to sites that show promise.


Sensitivity analysis will be perform
ed on the fraction of

municipal, d
ai
ry, and feedlot
wastewaters

that

are available for algal facilities.
It is possible that

a portion of a facility’s
wastewater cannot be diverted to algae

facilities. In many cases, an algae facility could be added
to the

existing
treatment infrastructure

or could be implemented when upgrades are made to
existing plants.

Future Work


-

Life cycle analysis of fuel production pathways being considered in the model

-

Predictions of spatial algae biomass outputs should be combined

with the existing
biorefinery siting model de
veloped by UC Davis researchers

-

The model should be expanded to include the entire continental US



Statement of Work


An overview of the modeling process is illustrated in the graphic below:








Task 1: Develop a Secondary Wastewater Treatment Model


Two methods of secondary wastewater treatment will be considered:



Activated Sludge (aerobic)



Anaerobic Diges
tion

(anaerobic)


Activated sludge will be used for wastewaters with less than 2,000 mg/ L COD whereas
anaerobic digestion will be used for wastewaters with more than 2,000 mg/ L COD

(Malina and
Pohland 1992)
.


Each system will be characterized in the treatment model to determine the required inputs for
each cubic meter of wastewater treated. These inputs include:



Ammortized

capital expenditures



Electricity consumption



Heating requirements (anaerobic digestion)

Site specific
characteristics (weather,
wastewater)

Algae yield from Algae
Cultivation Model

Harvested algae yield
from Harvesting Model

Fuel yield from Fuel
Conversion Model

Linear Programming
Model

Site Selection

CO2
Requirement,
CO2 availability

Site
characteristics and resource proximity database


Output characteristics will include the composition of the treated water with a focus on COD,
nitrogen, and phosphorous. For anaerobic digestion, outputs include quan
tities of CO
2

and CH
4

produced per cubic meter of water treated.


For sites in which an anaerobic digester is used, a biogas generator will also be included.
Electricity will be sold as a co
-
product and heat will be captured and used on
-
site if needed eith
er
to heat the digester influent or to dry biomass.


Outputs of the secondary wastewater treatment model include:
































































The treatment system in turn is a function of the
wastewater characteristics (COD) as discussed
above. Hence the cost of treatment is completely defined a priori by the conditions at a given
site.

One critical assumption that is being made at this step is that completely new facilities will
be constructed
. It is possible that existing facilities could be modified, resulting in lower costs.
This issue should be addressed in future work and would require an in depth survey of existing
treatment operations.


Some estimates are available for the value of remov
ing

nitrogen and phosphorous

nutrients from
wastewater and sensitivity analysis will be used to determine the co
-
product value of treated
wastewater
(Pizarro, Mulbry et al. 2006)
.


For anaerobic digestion, the following e
quations will be used
(Malina and Pohland 1992)










































































Task
2
: Develop an Algae Cultivation Model


Literature data on productivity and growth conditions of wastewater
cultivated algae will be
compiled into a spreadsheet model.

A model will be developed for the two processes discussed
previously:



H
igh rate algae pond

(HRP)



Algal

turf scrubber

(ATS)


Data has been compiled on the nitrogen and phosphorous uptake rate by algae
for each system
using

a variety of wastewater sources. The available nitrogen and phosphorous will be the
limiting factor that determines algal biomass production levels for a give
n site.


Regression analysis will be used to determine the

expected

productivity

for a given site based on
key

environmental factors
including
:



Inorganic nutrient concentrations in water



Solar intensity



Air temperature (minimum, average, and maximum)

The f
actors were chosen because they are site specific environmental factors. Given a site
location, expected productivity can then be predicted and the quantity of land required to
produce a
given quantity of

dry biomass can be calculated.


The spreadsheet mod
el will also include the resources consumed for e
ach kilogram of dry weight
algal

biomass produced. These r
esources include

e
lectricity

and c
arbon
d
ioxide
.
The cost of
electricity will be based on
the average rate for

California industry. The cost of carbo
n dioxide
will be determined specifically for each site using a mixed
-
integer linear programming model
discussed later.

Carbon dioxide supplied from onsite processes such as anaerobic digestion and
biogas combustion is assumed to be free.


The empirical fo
rmula for typical
algae biomass is assumed to be C
106
H
181
O
45
N
16
P and thus the
molecular weight is 2428 g/ mole
(Green, Bernstone et al. 1996)
. The mass breakdown of the
biomass
theoretical nutrient requirements are give
n below
:


Element

% of Biomass

Source of
Element

Required Mass of Source Material

(g/
kg biomass)

Carbon

52.4%

CO
2

1921

Hydrogen

7.5%

H
2
O

1342

Oxygen

29.7%

CO
2

and H
2
O

Not limiting

Nitrogen

9.2%

Total N

92

Phosphorous

1.3%

Total P

13


Carbon dioxide

is not perfectly utilized

due to losses from outgassing
.
Actual CO
2

supply to a
raceway pond was found to be
2.2

k
g per

kg biomass
(Weissman and Goebel 1987)
.

For an
inclined flow
-
way reactor with a carbonation sump, 4.4 kg CO
2

per

kg biomass was required
since more outgassi
ng can occur in a thin layer of moving
water
than in a pond

(Doucha, Straka
et al. 2005)
. This system has similar physical properties to an algal turf scrubber (which also
experiences a thin layer of water moving down a flow
-
way).


Functions will be developed to relate production capacity and algae
cultivation costs to a number
of site
-
specific parameters. The cost function includes

the cost of carbon dioxide
(
which will be
taken as an output from a separate model
)
.





































(






)




























The same treatment co
-
product values will be assigned to algal treatment as were used with
secondary treatment.

Task
3
:
Develop

Carbon Dioxide Procurement Model


A
GIS database of carbon dioxide sour
ces will be compiled to include those listed in the table
provided that data can be obtained for ammonia and ethanol plants. Data for power plants is
available from NATCARB Atlas.


Source

Production

Emission Output

Power plants

kW
-
hr

Flue gas
/ day

Ammonia plants

Kg ammonia

CO
2
/ day

Ethanol Plants

Kg ethanol

CO
2
/ day


A mixed integer linear program will be developed to match each potential cultivation site with a
CO
2

source. The model will be a cost minimizing model that determines the optimal pairing
between sources and cultivation sites. This model is independent of decisions to construct a
cultivation facility at a given site. Hence, the output of the CO
2

procureme
nt model can be
treated as a characteristic of a given potential algae cultivation site.


Not all power plants operate continuously (particularly natural gas plants).

Likewise,
other
sources operate continuously but algae can only utilize CO
2

during the da
y
.

Thus an ability to
store CO
2

must be factored into this

model
.


Make
-
up CO
2

will
be compressed and
transported by pipeline. A cost per kg delivered will
developed as a function of transport distance and volume.
Kadam
(1997)

found that it is more
economical to extract CO
2

from flue gas prior to

dry
ing
, compress
ion
, and transport by pipeline
than it was to dry, compress, and transport flue gas by pipeline. Monoethanolamine (MEA)
extraction was used to isolate the CO
2
.

Ammonia and ethanol plants produce nearly pure CO
2
,

removing the necessity to perform MEA extraction.


The output of the procurement model will be a
cost per kilogram

CO
2

delivered to each potential
cultivation site.




















Task 4:
Develop a Model of Harvesting and Dewatering Technology


Harvesting and dewatering are not site
-
specific processes and therefore, costs can be determined
directly as a function of biomass input.
The biomass input will come from the Algae Cultivation
Model.


For the high rate pond cultivation system, flocculation and settling will be used for bulk
harvesting. This technique reduces the water content of the algae slurry to 99% wet basis
(DOE
2009)
. This slurry can be used directly for some of the
fuel conversion technologies:



Wet gasification



Fermentation



Anaerobic Digestion

Additional dewatering of the slurry via centrifugation

(78
-
92% water content wet basis)

is
required for:



Lipid extraction



Thermochemical liquefaction

(Minowa, Yokoyama et al. 1995)

Biomass drying either by natural gas or solar methods

is required for:



Gasification and Fischer Tropsch



Pyrolysis


The algal turf scrubber system yields biomass with a water content of 94% meaning that water
must be add
ed to perform wet gasification, fermentation, or anaerobic digestion.
ATS harvesting
will be done by
scraping based on cost estimates by Pizarro
(2006)
.
A screw press will be used to
reduce
biomass
water content to 80% wet basis for lipid extraction and thermochemical
liquefaction.


Drying will be performed by natural gas powered drum dryers

or solar dryers, whichever is more
cost effective for a given unit of productivity

for a given site

(dollars p
er kg per day).
Solar
insolation, tem
perature, and relative humidity are important factors for solar drying

(Schirmer,
Janjai et al. 1996)

and a simple model will be developed to correlate drying pe
rformance to
expected weather conditions.



Capital and operating costs for each
harvesting and dewatering step will be compiled along with
the harvest yield, and
the wet weight of the biomass
.

Flocculation and settling ponds
will include
land costs
.

Yield

and cost data will be compiled for
four

harvesting/ dewatering pathways for
each cultivation method

(eight total pathways)
:



Harvested biomass with 99% water content wet basis



Harvested biomass with ~90% water content wet basis



Harvested biomass with ~80%
water content wet basis



Harvested biomass that is nearly dry (~10% water content wet basis)


The output

will be eight

cost function
s of the form
:





















w
here







is the

combination capital and O&M costs for a given

harvesting/
dewatering pathway.

The
dry and wet
biomass yield will also be reported for each pathway.































































Wet weight wil
l be used later when determining transportation costs.


Task 5: Develop a Model of Conversion Technology


Each conversion technology must be matched to biomass with appropriate water content
.
Therefore, the conversion technology dictates the harvesting and

dewatering requirements

and
thus influences the cultivation decision and corresponding harvesting pathway.

Biomass that
contains

less water than is

required by a conversion technology

may be used (possibly to save
money on biomass transportation)

but fres
hwater must be added prior to conversion
,

potentially
adding to the cost.

The conversion technology is the deciding factor in what types of fuels can be
produced from the algal bio
mass. The linear programming model will provide the user with the
option to
specify that a certain fraction of total fuel produced be of a particular type.


The conversion technology model will compile capital and O&M costs for the different
technologies
. Capital costs will be determined as a function of facility size (quantity o
f a given
fuel produced) by using a linearized economy of scale function.

In addition, fuel yields and co
-
product yields will be
determined for each technology and biomass cultivation method. The
biomass produced by the HRP and ATS have different compositi
on, particularly with regard to
lipid content. HRP algae have lipid contents in the range of 7
-
29%
(Woertz, Feffer et al. 2009)

while ATS grown algae have lipid content in the range of 5
-
9%
(Mulbry, Kondrad et al. 2008)
.


If the model chooses to perform anaerobic digestion on the algal biomass, conversion will occur
at the cultivation site. The existing anaer
obic digester used to treat the wastewater will be
expanded to handle the extra capacity and the yield will be decreased to reflect the performance
of algae digestion which is about

0.3 g CH
4
/ g VS
(Salerno, Nurdogan et al. 2009)
.

Gas
combustion can take place on
-
site if a generator is present, otherwise, clean
-
up

may take place
on or offsite at a “biorefinery” before entering the natural gas stream.


T
h
e output of this model will be a series of cost and production functions:




























where






is the combination of yield, capital, and O&M costs.

Biomass
type differentiates between HRP
-
grown biomass and ATS
-
grown biomass.



















































Yields of fuel and co
-
products will also be model outputs.

Task
6
: Determine

Cultivation

Site Feasibility and
Develop

Cultivation

Site
Characteristics

Database


This task creates a database of feasible algae cultivation sites
.

Site speci
fic weather and land
slope

data

will be compiled. Only areas with land slopes of less than 10% will be inc
luded in the
feasible set

consistent with work performed under the Aquatic Species Program
(Sheehan,
Dunahay et al. 1998)
. The

results from the

secondary wastewater treatment model (task 1)
, algae
cultivation model (task 2), CO2 procurement model (task 3), and harvesting and dewatering
model (task 4)

will be
applied to the feasible cultivation sites in the database. Hence, each site
will have eight algal cultivation pathways from which to ch
oose:


Cultivation

Harvesting

Dewatering

Drying Method

ATS

Scraping

None

None

ATS

Scraping

Screw Press

None

ATS

Scraping

Screw Press

NG Drying

ATS

Scraping

Screw Press

Solar Drying

HRP

Settling

None

None

HRP

Settling

Centrifuge

None

HRP

Settling

Centrifuge

NG Drying

HRP

Settling

Centrifuge

Solar Drying


Each pathway results in a unique “product” output with regard to biomass composition and water
content. The quantity of each possible product will be determined for each site based on site
-
specif
ic data. This information will feed into the supply chain model.

Task 7: Determine Site Feasibility for Biorefineries


Access to transportation, labor markets, and utilities will be used as criteria to develop a feasible
set of biorefinery locations.


In
addition, the location of existing petroleum terminals will be added to the spatial database.
Finished liquid fuels will b
e delivered to these locations for blending into gasoline or diesel.

For
methane generation, cleaned
-
up gas will have to be piped to t
he nearest pipeline.

Task
8
: Extract Transportation Network Data


Transportation network data will be obtained from ESRI Network Analyst. This data will feed
link travel costs into the linear programming model.

Both the truck and rail network will be
incl
uded in the analysis. Costs per ton mile will be determined for each mode and applied to the
model.


Transportation services will be employed to move algal biomass from the cultivation location to
the biorefinery
. Finished fuel will be transported from the biorefinery to petroleum terminals
or
natural gas pipelines.

Task
9
: Develop a Mixed
-
Integer Linear Programming Model


A profit maximizing mixed
-
integer linear program will be developed
.
































The cost function will be based on the compilation of data from the

Cultivation

Site
Characteristics Database developed in Task 6.


Decision variables:

1.

Build/
don’t
build algae cultivation facility at node

for each of the eight production
pathways

(
8
binary

variables for each feasible site
)

2.

Build/
don’t

build biorefinery

at a node

fo
r each type of biorefinery

(
7
binary

variables
for each feasible site
)

3.

Choose quantity of fuel produced at each biorefinery (1 continuous variable for each
feasible site)

4.

Choose transportation routing for biomass


quantity

of each biomass type

along links

(
8
continuous

variables per link
)

5.

Routing decision
for finished fuel



quantity

along links (
7
contin
u
ous

variables per link
)


Appropriate constrain
t

equations will be developed

to maintain mass balances
.

Only one
cultivation pathway and one type of conve
rsion technology will be chosen for each site.

Task 10: Conduct a Regional Scale Study: The Imperial Valley


Developing a model that covers the entire state of California is a significant undertaking and
involves assumptions about expected productivity lev
els in different parts of the state. In order to
test the model prior to scaling up to the state level, a regional model run will be conducted for the
Imperial Valley in Southern California. This will result in a smaller model which can be tested
and debug
ged.


Task 11: Expand Model to California


After a successful regional model is run, the full California database will be loaded into the
model and run.

The final deliverables will be the model and a report.

Budget


Wage
Personnel

Position

Duties

Length of

Time

Hours

Cost

Graduate Student
Researcher (half
time)

Completion of
tasks 1
-
11

1 year

1040

$17,680

Wage: $17/ hr


Salaried Personnel

Position

Duties

Length of
Time

% Effort

Cost

Programmer

Salary: $/ yr

Assistance in
completion of
tasks
3
,
7
-
8

1
year

25%




Supplies

Item

Quantity

Cost

Server time

Rate: ($40/ run
)

50 runs

$2000


Travel



References


Antares (2009). National Biorefinery Siting Model Draft Final Report
-

Task 2. Landover, MD,
Antares Group Incorporated.

Benemann,

J. and W. J. Oswald (1996). Systems and Economic Analysis of Microalgae Ponds
for Conversion of CO2 to Biomass. Berkeley, CA, University of California Berkeley.

Chisti, Y. (2007). "Biodiesel from microalgae."
Biotechnology Advances

25
: 294
-
306.

Clarens, A
. F., E. P. Resurreccion, et al. (2010). "Environmental Life Cycle Comparison of
Algae to Other Bioenergy Feedstocks."
Environmental Science and Technology

44
:
1813
-
1819.

Craggs, R. J., W. H. Adey, et al. (1996). "Phosphorus Removal from Wastewater Using a
n Algal
Turf Scrubber."
Water Science and Technology

33
(7): 191
-
198.

DOE (2009). National Algal Biofuels Technology Roadmap (Draft), US Department of Energy.

Doucha, J., F. Straka, et al. (2005). "Utilization of flue gas for cultivation of microalgae
(Chlo
rella sp.) in an outdoor open thin
-
layer photobioreactor."
Journal of Applied
Phycology

17
: 403
-
412.

Green, F. B., L. S. Bernstone, et al. (1996). "Advanced Integrated Wastewater Pond System for
Nitrogen Removal."
Water Science and Technology

33
(7): 207
-
21
7.

IPCC (2006). Introduction to the 2006 Guidelines K. Rypdal, N. Paciornik, S. Egglestonet al.
Bern, Switzerland, IPCC.
2006 IPCC Guidelines for National Greenhouse Gas
Inventories:
75.

Johnson, M. B. and Z. Wen (2009). "Development of an attached microalgal growth system for
biofuel production."
Applied Microbiology and Biotechnology

85
: 525
-
234.

Kadam, K. L. (1997). "Power Plant Flue Gas as a Source of CO2 for Microalgae Cultivation:
Ec
onomic Impact of Different Process Options."
Energy Conserv. Mgmt
38
: S505
-
S510.

Kebede
-
Westhead, E., C. Pizarro, et al. (2003). "Production and Nutrient Removal by Periphyton
Grown Under Different Loading Rates of Anaerobically Digested Flushed Dairy
Manu
re."
Journal of Phycology

39
: 1275
-
1282.

Kebede
-
Westhead, E., C. Pizarro, et al. (2004). "Treatment of Dairy Manure Effluent Using
Freshwater Algae: Elemental Composition of Algal Biomass at Different Manure
Loading Rates."
Journal of Agricultural and Food

Chemistry

52
: 7293
-
7296.

Malina, J. and F. G. Pohland (1992).
Design of Anaerobic Processes for Treatment of Industrial
and Muncipal Waste
. Boca Raton, FL, CRC Press.

Minowa, T., S.
-
y. Yokoyama, et al. (1995). "Oil production from algal cells of Dunaliell
a
tertiolecta by direct thermochemical liquefaction."
Fuel

74
(12): 1735
-
1738.

Molina Grima, E., E.
-
H. Belarbi, et al. (2003). "Recovery of microalgal biomass and metabolites:
process options and economics."
Biotechnology Advances

20
: 491
-
515.

Mulbry, W., S
. Kondrad, et al. (2008). "Treatment of dairy and swine manure effluents using
freshwater algae: fatty acid content and composition of algal biomass at different manure
loading rates."
Journal of Applied Phycology

20
: 1079
-
1085.

Oswald, W. J. (2003). "My s
ixty years in applied algology."
Journal of Applied Phycology

15
:
99
-
106.

Pizarro, C., W. Mulbry, et al. (2006). "An economic assessment of algal turf scrubber technology
for treatment of dairy manure effluent."
Ecological Engineering

26
: 321
-
327.

Salerno,

M., Y. Nurdogan, et al. (2009). Biogas Production from Algae Biomass Harvested at
Wastewater Treatment pOnds.
2009 Bioenergy Engineering Conference
. Seattle, WA.

Schirmer, P., S. Janjai, et al. (1996). "Experimental investigation of the performance of the

solar
tunnel dryer for drying bananas."
Renewable Energy

7
(2): 119
-
129.

SERI (1984). SERI Aquatic Species Program 1983 Annual Report. Golden, CO, Solar Energy
Research Institute.

Sheehan, J., T. Dunahay, et al. (1998). A Look Back at the U.S. Department o
f Energy's Aquatic
Species Program
--
Biodiesel from Algae. Golden, CO, National Renewable Energy
Laboratory.

Weissman, J. C. and R. P. Goebel (1987). Design and Analysis of Microalgal Open Pond
Systems for the Purpose of Producing Fuels. Golden, CO, Solar E
nergy Research
Institute.

Willemsen, J. (1980). "Fishery
-
aspects of eutrophication."
Aquatic Ecology

14
(1
-
2): 12
-
21.

Williams, R. B. (2007). An Assessment of Biomass Resources in California, 2007. Davis, CA,
University of California Davis.

Woertz, I., A. F
effer, et al. (2009). "Algae grown on diary and municipal wastewater for
simultaneous nutrient removal and lipid production for biofuel feedstock."
Journal of
Environmental Engineering
: 1115
-
1122.