-
extracted text
-
AN ASSESSMENT OF THE CAP A CITY AND COSTS OF
ELECTROLYTIC HYDROGEN PRODUCTION FROM SURPLUS
HYDROELECTRICTY IN THE PACIFIC NORTHWEST
by
Zachary A. Maskin
A Thesis
Submitted in Partial Fulfillment
of the Requirements for the Degree
Master of Environmental Studies
The Evergreen State College
June 2012
©20 12 by Zachary A. Maskin. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Zachary A. Maskin
has been approved for
The Evergreen State College
~Y~v
al
Murph Ph. .
Member of the Faculty
~ ~~ ccJ/2oate
ii
ABSTRACT
An Assessment of the Capacity and Costs of Electrolytic Hydrogen Production
from Surplus Hydroelectricity in the Pacific Northwest
Zachary A. Maskin
Transitioning to the hydrogen economy is inhibited by the inability to produce
electrolytic hydrogen at a competitive costs and substantial volumes. The Pacific
Northwest's capacity to generate large amounts of low cost surplus hydroelectricity
provides an opportunity to produce hydrogen gas at competitive costs through
forecourt scale electrolysis. This research analyzes Pacific Northwest surplus
hydroelectric capacity and models the production of electrolytic hydrogen from a
single Norsk Hydro Atmospheric Type 5040 (5150 Amp DC) electrolyzer unit.
Modeling projects production of more than 300,000 kilograms of electrolytic
hydrogen gas annually, at approximately $3.88 per kilogram. The results of this
study suggest that hydroelectricity utilities in the Pacific Northwest have the capacity
to produce substantial amounts of electrolytic hydrogen at costs competitive with
conventional hydrogen production.
iii
Table of Contents
Section I: Introduction.........................................................................
Problem Statement...............................................................................
Hypothesis and Rationale........................................................................
Thesis Organization..............................................................................
1
1
3
4
Section II: Study Scope, Parameters and Design.........................................
Selection of Pacific Northwest for Hydroelectricity to Hydrogen.........................
Selection of Electrolysis Production Scale...................................................
Selection of Electrolyzer Unit..................................................................
Study Assumptions..............................................................................
5
5
8
9
11
Section III: Foundations of Hydrogen Energy and Hydrogen Production.........
The Hydrogen Economy........................................................................
Hydrogen's Benefits Over Petroleum.........................................................
Conventional Hydrogen Production...........................................................
Electrolytic Hydrogen Production.............................................................
Electrolysis Opportunities for Power Utilities................................................
14
14
17
20
22
26
Section IV: Background Hydroelectricity to Hydrogen Studies......................
Renewables to Hydrogen and Hydroelectricity's Advantage..............................
Hydroelectricity to Hydrogen Studies.........................................................
30
30
32
Section V: Quantification Surplus Hydroelectricity Capacity, Fixed and
Variable Costs of Electrolysis................................................................
Surplus Hydroelectricity Availability.........................................................
Feedstock Electricity Costs.....................................................................
Electrolysis Capital Costs.......................................................................
Electrolysis Operation and Maintenance Costs..............................................
39
39
44
46
48
Section VI: Calculation of Electrolytic Hydrogen from Hydroelectricity Total
Production Costs................................................................................
50
Section VII: Conclusions......................................................................
56
Section VIII: Suggestions for Further Research..........................................
61
Appendix A: Projected Monthly Surplus Capacity........................................
63
Bibliography of Sources........................................................................
64
iv
Directory of Tables and Figures
Table 1: Projected Pacific Northwest Load Resources for the Years 2012 - 2012
Under Variable Water Year Conditions in Average Megawatts..........................
7
Table 2: Scales of Electrolyzer Production...................................................
9
Table 3: Forecasted Surplus Electricity Generation in Average Megawatts Under
Variable Water Year Conditions...............................................................
41
Table 4: BPA Priority Firm Rates..............................................................
45
Table 5: Costs Per Kilogram of Electrolytic Hydrogen with Variable Feedstok
Electricty Rates...................................................................................
51
Table 6: Annual Electrolytic Costs and Production Capacity at lndustirial Firm and
Prirority Firm Electricity Rates.................................................................
53
Figure 1: Monthly Power Load Demand in the Pacific Northwest........................
6
Figure 2: Projected Pacific Northwest Electricity Surplus Under Critical Water Year
Conditions.........................................................................................
8
Figure 3: Total Cost Forecast For Forecourt Electrolytic Hydrogen Production with
Variable Feedstock Electricity Costs..........................................................
52
v
Acronyms:
$/kg: Dollars per kilogram
aMW: Average Megawatts
BPA: Bonneville Power Administration
HLH: Heavy Load Hours
HHV: Higher Heating Value
lR: Industrial Firm Rates
kWh: Kilowatt Hours
MMBtu : Million British Thermal Unites
PF: Priority Firms
WY: Water Year
Helpful Conversions:
1 MMBtu = 293.1 kWh
1 kWh = 0.003412 MMBtu
1 kg = 1 I .0921 m 3
lmill= 0.1¢ /kWh
MW= 1,000kW
vi
Acknowledgements:
Completion of this thesis would not have been possible without the gracious help from
many individuals. A substantial thank you is indebted to Dr. Ralph Murphy of The
Evergreen State College who served as reader of this thesis . Dr. Murphy contributed in
the guidance and review of this research, and I commend him for his tireless work and
commitment. Additional thanks need to be given to Alan Hardcastle of the Washington
State University Extension Program, and Ken Dragoon of the Northwest Power and
Conservation Council. Both gentlemen were able to direct this study to important data
sources. A final thank you is owed to Buck Rogers, 2012 graduate of The Evergreen
State College Master of Environmental Studies program, who helped with initial
conceptualization and framing of this thesis.
vii
/
Section 1: Introduction
A critical challenge to building a clean energy future has been the inability to produce
substantial quantities of clean burning fuel derived from clean energy sources, at a cost
that is competitive with dirtier conventional fossil fuels. For the past several decades,
hydrogen has been targeted as potential energy game-changer that could drastically
reduce dependence on fossil fuels. ln its H 2 fom1 , hydrogen has been praised for its high
energy density, versatility, storability, lack of pollutants, absence of greenhouse gas
emissions, and ability to be produced from renewable energy sources. Although it is the
most abundant element in the universe, hydrogen rarely naturally exists in its H 2 form.
Without natural H 2 deposits, H 2 must be produced by separating it from hydrogen
compounds. water molecules can be split into hydrogen and oxygen through the simple
process of electrolysis. Requiring substantial amounts of feedstock electricity,
electrolysis can be a very expensive process, largely giving fossil fuel production
economic advantage over electrolytic hydrogen. Low electricity prices, such as those of
Pacific Northwest hydroelectricity, create an opportunity to produce electrolytic
hydrogen at reduced costs, that could potentially elevate the region as a leader in clean
and efficient hydrogen production.
Problem Statement:
Producing a universal energy carrier from renewable and cleaner sources could alleviate
economic, social and environmental burdens engendered from fossil fuel production and
consumption. Hydrogen, in its H 2 form, has the ability to perform all the duties of
conventional fossil fuels and has several additional advantages over fossil fuels
1
including: higher energy efficiency, lack of emissions and ability to be produced through
renewable energy sources. Marred by a large demand for costly feedstock electricity
inputs, electrolytic hydrogen as an energy carrier and transportation fuel has been largely
shelved until there is production efficiency and capacity improvement and/or significant
increase in the cost of petroleum products.
Maturation of electrolytic technologies have allowed for more efficient large scale
electrolysis, potentially allowing the opportunity to produce economically viable
electrolytic H 2 . With constant production, today's largest electrolyzer units can produce
more than I ,000 kg of hydrogen a day. This high level of production comes at a high
electricity consumption, demanding more than two megawatts (MW) of constant
electricity. That is enough energy to power roughly 1,600 homes.
This thesis intends to determine whether a tipping point has been reached that would
allow Pacific Northwest power utilities to use inexpensive surplus hydroelectric resources
to produce economically viable electrolytic hydrogen on a large (forecourt) scale that is
cost competitive with conventional hydrogen production and gasoline.The economic
viability of electrolytic hydrogen production will be assessed through an analysis of
regional surplus hydroelectricity availability, establishment of fixed production costs of
electrolytic hydrogen, and calculation of electrolysis feedstock electricity demand and its
associated variable costs. With establishment and validation of total production costs, an
assessment of the price per kilogram of raw electrolytic hydrogen gas can be made . This
research represents a pilot study for the Pacific Northwest's capacity to produce a clean,
2
storable energy carrier that can serve as back up energy supply or be sold as a
merchantable commodity or fuel.
Hypothesis and Rationale:
The following subsection exhibits the working research hypothesis and supporting
rationale for this thesis. The null hypothesis is represented asH 0 and the alternative
hypothesis is represented as H 0 .
H0 :
Despite the abundance of low cost surplus hydroelectricity in the Pacific
Northwest region, maturation of electrolysis technology and escalation of fossil
fuel costs, large scale production of electrolytic hydrogen is not an economically
viable use of hydropower resources.
Ha:
Because of the abundance of low cost surplus hydroelectricity in the Pacific
Northwest region, the maturation of electrolysis technology, and the escalation
of fossil fuels costs, large scale production of electrolytic hydrogen is an
economically viable use of hydropower resources.
Rationale for utilization of hydroelectric resources to produce electrolytic hydrogen stems
from three advantages of hydroelectricity:
(I)
Hydropower has ahigh capacity to produce large volumes of electricity,
allowing for high production electrolyzers to produce hydrogen at
economies of scale.
(2)
Hydropower resources have the greatest ability to produce large amounts
of low cost surplus electricity.
(3)
Hydropower is an existing installed energy resource, very little
infrastructure would have to be built and carbon costs of hydroelectric
plants have been largely mitigated over their longoperating lifetime.
3
Thesis Organization
Prior to an assessment of the availability, capacity and costs of electrolytic hydrogen
production utilizing surplus hydropower in the Pacific Northwest, validation of study
background and parameters first must be established. Once study foundations are
established then research transitions to an analysis of electrolysis cost variables after
which, overall production capacity and costs can be calculated.
Proceeding from this point, this thesis is organized into several larger sections. First is the
articulation of the study scope and design. Then the thesis progresses to a foundational
discussion of hydrogen energy and hydrogen production . The next portion reports on the
preexisting hydroelectricity-to-hydrogen studies which influence, support and guide this
research. After the review of the preexisting studies, this research transitions to establish
Pacific Northwest surplus hydroelectric availability, and validate fixed and variable
production costs. Finally, analysis oftotal production cost is made, overarching
conclusions are drawn and opportunities for further research are suggested.
4
Section II: Study Scope, Parameters and Design
Assessment of electrolytic hydrogen production from surplus Pacific Northwest
hydropower requires establishment of foundational study parameters. This section
addresses the selection of the Pacific Northwest for the study site, sets appropriate scale
of electrolysis production, addresses the selection of the appropriate electrolyzer unit and
details necessary foundational study assumptions .
Selection of Pacific Northwest for Hydroelectricity to Hydrogen Production:
Natural resources have finite quantities and extraction locations. Since natural resources
have limits, ultimately the goal is to more efficiently and sustainably use these resources.
Hydropower represents the most advantageous methods of producing high capacity, nonintermittent electricity with only marginal greenhouse gas emissions. Single hydro dams
can constantly generate several thousands of megawatts of power, but have maximum
capacities and finite locations that are technologically and economically viable for energy
extraction. Nearly all of these locations have been exploited in North America.
(Altinbilek, Seelos and Taylor 2005) The Pacific Northwest is fortunate to have the
highesthydroelectric capacity in the United States, but as we have installed the
practicable the maximum of hydroelectric capacity, it becomes increasingly important to
most efficiently use this valuable resource.
To sustainability manage power resources, it is essential to forecast energy consumption
demand. Bonneville Power Administration forecasts demand loads ten years in advance
to ensure it has the capacity supply the demanded energy resources Figure 1 displays
5
anticipated regional monthly load demand for 2011-2012, 2015-2016, and 2020-2021 ,
illustrating the regional energy supply requirements each month for an entire year.
Figure 1: Monthly Power Load Demand in the Pacific Northwest
30.000
29,000
28.000
27.000
~ 26,000
;
i.
lr . .....
.
I
g' 24.000
ill
>
<
>- 23.000
0
22.000
-- .....
''\
~
21. 000
20.000
19.000
1-Aug 1 6- Aug
I
I.
25,000
.
...
'
Sep
\
I
Ill ~
'
. . ,,-I
• /1 i
'
""~
l'bv
I- -
Doc
Jan
_,·
'\
:..
\
\
'
·;
Oct
..
.... -- ---- "
Fell
''\
...
~
11
/
'
'
__·---__ ,..., ,
~
M:lr
,
I
I
I
'
......___
·;
I
)
/
"'-./
1-Apr 16-Apr
OY 2012 - - - - OY 2016 . .. . ... OY 2021
I
/
May
Jun
Jul
I
Figure l displays projected electricity load demand in megawatts for the Pacific Northwest region in the
20 12, 20 16, and 2021 operating years.
Source: Bonneville Power Administration 2011 Pacific North west Loads and Resource Study p.64
Figure 1 displays the minimum energy suppiy necessary to fill the electricity demands of
the Pacific Northwest region . Figure I demonstrates the cyclical annual energy demand
in the Pacific Northwest region, depicting highest demand in the coolest months .
Anticipated regional growth generates sizeable energy demand increase over the next ten
years. Whereas energy demand can be forecasted with reasonable accuracy, energy
supply is widely variable. The Pacific Northwest's energy supply is dominated by
hydroelectricity and hydropower's production capacity is directly dictated by water year
conditions. Water year conditions are historic records of river levels resulting from
annual precipitation and snow melt levels. Table I displays the wide range in Pacific
6
Northwest electricity resource in average megawatts for years 2012-2021 under variable
water year conditions.
Table 1: Projected Pacific Northwest Load Resources for the Years 2012-2012
Under Variable Water Year Conditions in Average Megawatts
2013
2014
2015
2016
2017
2018
2019
2020
2021
Operating 2012
Year
33 , 121 33,064 33,061 32,979 33,054 32,850 33 , I 05 32,857 33 ,003 32,272
Top
10%
Water
Year
29,785 29,778 29,819 29,755 29,825 29,660 29,885 29,667 29,774 29,082
Middle
80%
Water
Year
26,434 26,509 26,588 26,537 26,603 26,435 26,669 26,442 26,555 25,857
Bottom
10%
Water
Year
Table 1 displays the forecasted electricity supply in average megawatts (aMW) for the Pacific Northwest
under variable water year conditions
Source: Bonneville Power Administration 2011 Pacific Northwest Loads and Resource Study p.67
Furthermore, Table I demonstrates the potentially wide range in electricity supply water
year conditions generate. The difference in electricity supply between a Top I 0% and
Bottom I 0% water year is greater than 6000 aMW, enough electricity to power roughly
4.8million homes.Comparison of Figure I and Table 1 reveals that even in critical water
years, there is typically ample electricity supply to meet regional demand. Power
generated above regional demand is considered surplus and is eligible to fill supplemental
regional energy demands or is exported out of region. Figure 2, demonstrates projected
surplus through 2021 under critical water year conditions, representing the most
conservative annual average energy surplus.
7
Figure 2: Projected Pacific Northwest Electricity Surplus Under Critical Water
Year Conditions
5.000
4000
3.000
.,"
~
:r."'
.I;"'
>
,..
~
'
.....
..
2.000
-...
i
. __ -- -....
~
1.000
<(
I
·--
-------
..
0
0
~
•
· 1.000
t-- - - - - - - - - - - - - - - -- - - - -
·2.000 +---~----,.--.,---,--~c----,--------r--,----~
2012
2013
2014
2015
2016
2011
2018
2019
2020
2021
I-
2011 W!llQ Book Sludy - - - - 2010 lll'hlQ Book Study · · · • - · · 2009 WlliiQ Book Study
I
Figure 2 displays projected electricity surplus produced by Pacific Northwest energy firms for the
years 20 12 through 2021 under critical water year conditions.
Source: Bonneville Power Administration 2011 Pacific Northwest Loads and Resource Study p.70
Cursory examination of Figure 2 suggests that projected electricity surplus derived from
the Pacific Northwest's large hydroelectricity capacity is ample electricity supply to
utilize surplus electricity to produce electrolytic hydrogen . Even under the lowest of
water years, as seen in Figure 2, annual surplus appears ample to support electrolytic
hydrogen production, exposing the Pacific Northwest's suitability for this study. Further
analysis of surplus electricity resources are discussed later in Section V, but initial
outlooksuggests Pacific Northwest hydroelectric dominated power resources make the
region attractive for electrolytic hydrogen production.
Selection of Electrolysis Production Scale:
There are numerous companies currently producing commercially available electrolyzer
units. Available in differing sizes, designs and production capacities, there are primarily
five scales of electrolytic hydrogen production capacity. Although there is variance in the
8
design of electrolyzer technology, all electrolyzers fundamentally function the same way,
with the application of direct electric current to a hydrogen compound and then the
collection of separated gases. Table 2 displays the five scales of electrolytic production
and the corresponding number of cars such could potentially fuel.
Scale
Home
Table 2: Scales of Electrolyzer Production
Kg. of H 2 Produced
Number of vehicles served
Annually
200 - I ,000
I- 5
Small Neighborhood
1,000- 10,000
5-50
Neighborhood
I 0,000 - 30,000
50 - 150
30,000 - 100,000
150-500
Small Forecourt
Forecourt
~
100,000
~
500
Table 2 displays the five scales of hydrogen production from commercially available electrolyzers
Source: (Ivy, 2004)
In order to make the most sizable impact in clean hydrogen production, this study
selectsthe largest scale electrolytic production, known as forecourt electrolysis. A
forecourt hydrogen production plant has the capacity to produce greater than I OO,OOOkg
of hydrogen annually. (Ivy 2004)A plant of this capacity requires substantial feedstock
electricity input, and has a sizeable physical footprint. Selection offorecourt scale is
paramount to economic feasibility. The large capital cost of electrolysis requires the
selection of forecourt scale so that economies of scale are reached, resulting in the least
expensive price per kilogram of H 2 produced.
Selection of Electrolyzer Unit:
Selection of an electrolyzer unit is closely married to desired production scale. For the
scale considered in this analysis, only a forecourt scale electrolyzer is considered because
9
of its production output. Ideally electrolyzers would operate with a level of input power
yielding the highest production rate. But, as availability of surplus hydroelectricity can
change rapidly, the electrolyzer must have the ability to quickly adapt to varying current
loads. (Ouellette, Rogner and Scott 1997)
The electrolyzer selected for this study is the Norsk Atmospheric Type No. 5040 (5150
Amp DC). It is the largest commercially available electrolyzer; a single unit has the
production capacity of 1,046 kg of H 2 daily, enough to supply fuel for 1909 cars
annually, on the assumption of approximately 200 kg of H 2 annually at 60 miles per kg
ofhydrogen. (Ivy 2004)The Norsk Atmospheric Type No. 5040 (5150 Amp DC) is a
bipolar electrolyzer, employing a filter press and alkaline electrolyte. This design clamps
stacks alternating layers of electrodes, separated by support diaphragms. The stacked
electrode cells are connected in series, generating higher voltage. Bipolar electrolyzers
have the advantage of a smaller footprint, higher current density, and the ability to
produce higher pressure gas . (Kroposki, et al. 2006) The Norsk Atmospheric Type No.
5040 (5150 Amp DC) is amongst the most efficient electricity-to-hydrogen electrolyzers
current available. Requiring 2.328 MW to operate at optimal production levels, it
produces a single kilogram of hydrogen for every 53.5kW input, generating roughly 43.6
kilograms an hour. Its overall electricity to hydrogen efficiency is 73 % and has the ability
to compress gas to 435psi . (Ivy 2004) Further discussion and explanation of the
electrolysis process can be found below in Section Ill.
10
Study Assumptions:
This study asserts several key assumptions in order to produce consistent results. Any
given year' s specific water year (WY) condition dictates surplus power availability.
Reflecting the recorded historical 70 year water levels, water year conditions are the
strongest detem1inant of how much surplus hydropower can be generated and how a
power utility would consider using its water resources. Along with determining
availability, specific water year conditions dictate the price surplus hydropower
commands. The unpredictable nature of water year conditions and conesponding
unpredictable energy supply drives the following assumptions, which are made to
provide predictable study parameters and a reduction of exogenous variables.
Today, the electrical grid in the United States is constructed as a patchwork of smaller
regional electricity grids. Electricity is sold as a commodity and is transmitted from
region to region in times of power deficit and surplus. Electricity transmission is
governed by Regional Transmission Organizations (RTO) and Independent System
Operators (JSO) .Cunently, the majority of surplus electricity generated in the Pacific
Northwest is sold to the California ISO and British Columbia. This study assumes that
any existing contracts and agreements that may bind Pacific Northwest hydroelectric
utilities to sell to inter-regional energy exchanges are disregarded. This assumption
enables any regionally generated hydropower above the load demand to be consumed by
electrolytic hydrogen production. Participation in inter-regional energy exchanges is
voluntary, and the quantity ofsurplus electricity sold on the exchanges is variable and
difficult to ascertain. This assumption does not preclude surplus hydropower to be
11
transmitted out of region, it is simply a caveat giving electrolytic hydrogen production the
first priority to surplus power. Furthermore, Bonneville Power Administration's Pacific
Northwest surplus electricity forecasts through the year 2021 anticipate only extreme
occurrences, consisting of the poorest of water years and during the coldest months,
which surplus power is so marginal that there would be insufficient surplus for both
electrolytic hydrogen production and out transmission out of region. Additional
discussion of Pacific Northwest surplus electricity capacity is addressed in Section V.
The Pacific Northwest has a substantial amount of Independent Power Producers (IPPs)
that contribute 3,287 aMW to the power grid, enough electricity to power more than
roughly 2.4 million homes. IPPs constitute roughly 10% of BPA total generating
capacity. Generating power by wind, biomass, natural gas, and coal the IPPs assist BPA
filling base-load power requirement. Independent Power Producers have the option of
selling electricity out of region to ISOs and RTOs, but this study assumes that alllPP
electricity would stay within the Pacific Northwest Region. Regardless of water year
conditions, this study assumes that all 3,287 aMW are available to the Pacific Northwest
base-load demand. Delivering all IPP electricity to the regional grid allows for a more
consistent measure of base-load power resources. Counting all 3.287 aMW nonhydropower resources as delivering regional base-load power, it can then be assumed that
all electricity generated beyond base-load demand is surplus generated through
hydroelectricity.
Determination of the price of surplus electricity is difficult because of large arnot,mts
12
ofdaily, monthly and annual variance. Surplus electricity prices and availability reflect
seasonal river conditions and overall specific water year conditions. For this study the
price of surplus power is assumed to be equal to or below the lowest contracted
electricity rates offered by Bonneville Power Administration. Currently, the lowest
contracted electricity rates are paid by Priority Firms at prices that vary monthly,
reflecting historical seasonal river flow rates. Calculations in this study that employ
Priority Firm Rates, will represent conservative surplus electricity rates .
Climate change poses a potential challenges to hydroelectricity generating utilities.
Changes in precipitation rates, seasonal conditions, stream lengthand annual freeze and
melt events represent potential disruption of traditional power generation timing and
capacity. Although potentially having substantial impact on surplus availability and
timing, this study does not take in to consideration impacts of climate change on water
year conditions, population change, precipitation, river flow change, and changes
inenergy demand. Climate change represents an intriguing consideration for future
hydroelectricity producers, which has the potential to increase, decrease or cause no
change to availability of surplus electricity, but this consideration is beyond the scope of
this study.
The several assumptions addressed above are established to provide a standard and
predictable baseline of power resources in the Pacific Northwest, and reduce ambiguity
of what power is available and its dispersion . As available power is consistently in flux in
this region, it is important to account for all the consistently predictable power resources.
13
These assumptions allow more predictable allocation of regional electricity resources,
which will allow a more accurate forecast of regionally available surplus hydropower and
pnce.
14
Section III: Foundations of Hydrogen Energy and Hydrogen Production
The most basic and lightest of all elements, hydrogen has a host of uses which makes
hydrogen production enticing. Beyond its potential role in the energy field, hydrogen is
used in a multitude of manufacturing processes, and is an essential part of chemical ,
metal and glass production. Hydrogen ' s versatility keeps demand for production high
and as hydrogen energy continues to gain interest, there is increasingly more demand to
produce larger quantities. This section highlights hydrogen energy and discusses the two
common methods of hydrogen production.
The Hydrogen Economy:
The term 'hydrogen economy ' arose amidst the energy crisis of the 1970s and constitutes
the replacement of the petroleum-based transportation and energy infrastructure with
hydrogen produced from non-fossil fuel based sources. (Balat and Kirtay 201 0)
Essentially transitioning to the hydrogen economy entails three steps: (I) large scale
production of hydrogen fuel ; (2) storage, transportation, and distribution of hydrogen
fuel ; and lastly (3) wide-scale implementation and utilization of hydrogen fuel. (Tomczyk
2009) Certainly, the hydrogen economy is a drastic overhaul of our existing energy and
transportation infrastructure that would generate a complex array of transitional strife to a
host of interconnected industries. A large scale of energy transition would potentially
render currently crucial products and services obsolete, but a transition to a hydrogen
economy has the widely appealing possibility of energy independence, virtually devoid of
greenhouse emissions.
15
Transitioning to the hydrogen economy requires the clearance of substantial obstacles.
The United States consumes roughly 140 billion gallons of gasoline annually, which
would need to be replaced with roughly the same amount of kilograms of hydrogen.
(Kroposki, et al. 2006) Today ' s hydrogen production does not yet approach 140 billion
kg. 1n 2009 world hydrogen production accounted for roughly 45 million kg (500 billion
m 3) and nearly 96% of all hydrogen was produced with fossil fuel based feedstocks .
(Tomczyk 2009) This hydrogen would be inappropriate forbuilding thehydrogen
economy, as producing hydrogen from fossil fuel based feedstock contradicts the
foundation ofthehydrogen economy' s fossil fuel-free energy and transportation fuel. The
roughly 1.8 million kg of hydrogen not produced with fossil fuel feedstocks are generated
through electrolysis. Utilizing electrolysis to produce hydrogen is the ideal method of
producing hydrogen for transition to the hydrogen economy and detailed explanation of
the electrolytic process can be found in the "Electrolytic Hydrogen Production"
subsection below.
The complete transition to the hydrogen economy would require electrolytic hydrogen
production to exponentially increase. To generate enough electrolytic hydrogen to replace
fossil fuels would require the US to " ... double its current electricity capacity."
(Kroposki, et al. 2006, p. 20) Not only would electricity capacity need to double, but the
new electricity would have to come from non-fossil fuel energy sources such as wind,
solar, hydro, biomass, nuclear and other renewables. Although a full transition to the
hydrogen economy requires a dramatically large increase in electricity capacity,
incremental steps towards the hydrogen economy can be made with more efficient and
16
sustainable management of the electric resources already in place.
This pilot study, which analyzes electrolytic hydrogen production potential from surplus
hydroelectricity resources, is an attempt to make more efficient and sustainable
management of currently available resources. If there is ample surplus capacity and low
enough production costs, electrolytic hydrogen from surplus hydropower could represent
a model for potential energy conservation through hydrogen production.
Hydrogen's Benefits Over Petroleum:
The consequences of modem dependence upon fossi I fuels have far reaching negative
economic, social and environmental impacts. As conveniently accessible fossil fuels
continue to diminish, we are faced with exerting increased energy inputs in order meet
our energy demand. The increased exertion of input energyto fill the same energy
demand depicts the growing inefficiency of fossil fuel based energy. Unless substantial
new deposits are discovered, fossil fuels' life-cycle efficiency will continue to decline,
generating increasingly higher costs to the consumer. This does not account for the host
environmentally harmful by-products producing and burning fossil fuels generates, nor
the social strife generated as a cost of conducting business in the fossil fuel industry.
Hydrogen is a proven and viable fuel, directly comparable with gasoline. A single
kilogram ofhydrogen contains the energy capacity of33999.6 kWh (122398.56 MJ or
116MMBtu) which is approximately the same energy capacity as a gallon of gasoline
31676.1-36368.9 kWh (108-124MMBtu or 11403.4 - 130928 MJ). (Levene, Kroposki
17
and Sverdrup 2006) Although a gallon of gasoline and a kilogram of hydrogen have the
same potential energy capacity, when consumed hydrogen fuel delivers superior energy
performance, losing far less potential energy to heat than gasoline.
Half the global oil production is consumed by motor vehicles. Buming fossil fuels in
motor vehicles accounts for more than 70% carbon monoxide (CO) emissions, 17% of
carbon dioxide (C0 2 ) emissions, and a host of other pollutants including: nitrogen
oxides, hydrocarbons, lead and particulates. (Balat and Kirtay 201 0) As fossi I fuel
energy sources are rapidly becoming scarcer with increased global petroleum demand,
there is the need to procure a storable, transportable energy supply derived from nonfossil fuel sources. Hydrogen has received significant endorsements as the fuel of the
future. When bumt, hydrogen emits pure water and only a marginal amount of nitrous
oxide (approximately 11200 of diesel's N0 3 by-product) (Balat and Kirtay 2010)
Hydrogen is more efficient energy carrier than petroleum products, storing more than 2.5
times the energy per unit of mass than gasoline. Hydrogen ' s better efficiency
performance equates to a vehicle traveling further on the same mass unit of fuel and less
energy lost as heat. Hydrogen bums at a rapid rate and at high octane. Hydrogen has
more versatile flammability in air 4%- 75% by volume opposed to gasoline's 1 - 7.6%.
Hydrogen has a very low minimum ignition energy of .02 mJ and is easy to ignite at low
temperatures. (Balat and Kirtay 201 0) Hydrogen can be compressed, stored and
transported efficiently and inexpensively (Rifken, 2002). Much of the infrastructure built
for petroleum transport, and delivery can be used with hydrogen fuel.
18
Hydrogen outperforms petroleum in energy efficiency, is cleaner for the environment and
has potentially limitless production capacity from renewable electricity. Petroleum's
advantage over hydrogen is chiefly economical. The long term investments made to
access reserves, allows fossil fuels to be less expensive to produce on a large scale. As
the cost to produce fossil fuels increases, we must consider whether continuing to make
large investments in fuels that are increasing less economical and exacerbate negative
environmental and social conditions are a better use of our resources than beginning to
make investments into hydrogen and start a transition to an energy carrier that offers
more long term sustainability than petroleum.
Hydrogen can be produced from domestic energy sources. Production is well suited for
alternative and renewable energy resources. The intermittence of renewables can cause
hard to predict spikes and declines in electricity generation. Sudden changes in energy
output from renewables generate issues balancing regional energy supply and demand.
The potential to take surplus electricity and store it for use in times of increased demand
constitutes a major leap forward in energy conservation and efficiency. Once stored,
hydrogen has the versatility to be used as either a transportation fuel or electrical power
source.
Hydrogen has a host of social, environmental, and fuel performance advantages over
fossil fuels, but the problem that persists is the unavailability of a large source of cheaply
produced hydrogen. Until large quantities of hydrogen can be produced at prices
competitive with petroleum, the economics of petroleum will continue outcompete
19
hydrogen.
Conventional Hydrogen Production:
As aforementioned, hydrogen is the most abundant element in the universe, but rarely
exists naturally in its H 2 state. The H 2 compound is the form required forhydrogen 's use
as an energy carrier, meaning H 2 has to be produced through separating hydrogen
fromother elements withina compound. H 2 is an 'energy carrier,' rather than an energy
source. This is an important distinction as it implies that hydrogen (H 2 ) is not a natural
harvestable energy resource. Energy inputs are required to generate the H 2 , which has the
capacity to store a substantial portion of imputed energy, and then can be consumed as
fuel.
The conventional method for producing hydrogen is a process called stream reformation,
in which stream reacts with natural gas, liquefied coal or another fossil fuel in a catalytic
convertor. The reaction between steam and fossil fuel strips away H 2 from the steam
molecules, which then can be collected. Steam reformation production costs range from
$1.00-5.00 per kg of hydrogen and is the current least expansive method ofH 2
production. Fossil fuel feedstock price volatility contributes to this wide ranging cost
scale. (Kroposki, et al. 2006) The process of steam reformation has the unfortunate
byproduct of large amounts of greenhouse gasemissions. Fossil fuel based hydrogen
generation, " ... produces at least the same amount of C0 2 as the direct combustion of the
fossil fuels." (Balat and Kirtay 20 l 0, p. 865) Additionally, steam reformation hydrogen
commonly has high levels of impurities. (Kroposki, et al. 2006) Hydrogen that has higher
20
rates of impurities reduces fuel cell performance. Despite this, steam reformation is still
the hydrogen production method of choice, representing 96% of all hydrogen production
as of 201 0 (Balat and Kirtay 201 0)
Steam reformation has several advantages that make it the current hydrogen production
method of preference. Primarily, steam reformation has economic advantage over nonfossil fuel based hydrogen production. Steam reformation is versatile. Any fos sil fuel can
be used as a feedstock in the process. Steam reformation has the ability to produce
substantial amounts of hydrogen on a smaller physical footprint and has a better input
energy to hydrogen conversion efficiency. Opposed to electrolysis, steam reformation has
the ability to produce economical hydrogen at smaller scale production output.
The advantages of steam reformation production are reduced with consideration of steam
reformation's long-term sustainability, and potential role in the hydrogen economy. As
global fossil fuel deposits decline, steam reformation's hydrocarbon based feedstock
costs will continue to increase. Geopolitical strife is perpetuated through the reliance on
steam reformation, as it continues the need to conduct business in volatile fossil fuel
producing nations. Steam reformation necessary use of fossil fuel feedstocks , releases
greenhouse gases and other pollutants, ruling out steam reformation hydrogen fuel in the
potential hydrogen economy. The hydrogen economy fundamentally prohibits hydrogen
produced through fossil fuel based energy. Steam reformation ' s associated greenhouse
gas emissions could be mitigated with carbon capture systems, but the current application
of carbon sequestration technology to conventional hydrogen production has not proven
21
itself economical. (Balat and Kirtay 201 0) Steam reformation will likely continue to
serve the industrial needs of hydrogen, as the method performs efficiently producing
economical supplies of hydrogen, but its poor sustainability and dependence on fossil
fuel feedstocks, does not make the steam reformation a long term-viable solution towards
advancing to the hydrogen economy.
Electrolytic Hydrogen Production:
It is well established that there is another hydrogen production methods without the
undesirable consequences of the steam reformation process. Electrolysis, discovered
more than two centuries ago by William Nicholson and Sir Anthony Carlisle, is a rather
simple process in which the application of direct current to water splits the atoms into its
two basic elements, Hydrogen (H 2 ) and oxygen (0 2 ). Electrolysis generates these gases
as electricity is introduced to an electrolyzer, which consists of four basic components: an
electrolyte, electrodes, a separator, and container. The electrolyte is a highly conductive
solution or polymer, most commonly an alkali, which supplies atoms to be exchanged.
The electrodes are the actual interaction point between electric current and electrolyte.
Consisting of a positive charged anode and negative charged cathode, electrodes are
highly conductive metals which facilitate the exchange of atoms. Oxidation occurs at the
anode site, which entails stripping away electrons. Reduction, or the gain of electrons,
occurs at the cathode. The anode facilitates the generation of oxygen, while the cathode
generates hydrogen. The electrode is the most variant component between electrolyzer
designs and facilitates different levels of efficiency and productivity. A separator is the
corrosive resistant buffer between anode and cathode which prevents the mixture of gases
22
within the electrolytic cell. Lastly, the container is simply the vessel which holds the
electrolyte and allows for flow of current. (Pratt, et al. 1984) Regardless of individual
electrolyzer unit, the same fundamental reaction occurs with the application of direct
current to water. The process of electrolysis can be expressed by the following reaction
equation, representing the process facilitated within the electrolyzer unit:
H 2 0 + 237.2 kJ /mole electricity 0+ 48.6 kJ/mole heat- H 2 + 12 0 2
The heat required in the electrolysis process is generated from reaction within the
electrolytic cell. As electric and ionic currentsflow through the electrolytic cell they
encounter internal resistance, generating heat. Heat generated in the electrolytic process
is the direct a result of input electricity, so in actuality it takes 285.8kJ of input electricity
complete electrolysis, rather than, aforementioned 237.2kJ. (Harrison, et al. 2010) The
loss of 48.6kJ of electricity to heat in the process of electrolysis is inevitable, constituting
the impossibility of 100% efficient electricity to hydrogen conversion. Actual perfect
efficiency of electrolysis is 84.5% electricity to hydrogen conversion. (Ivy 2004) This
illustrates the fundamental limitation of electrolytic hydrogen production, in the best case
scenario 15.5% of input electricity will be lost to heat in conversation to hydrogen.
Under optimal conditions of electrolysis, 39 kWh of electricity and 8.9 liters of water
would produce a single kg of H 2 at 25 °C and 1 atmosphere of pressure. This represents
the ideal 84.5% electricity to hydrogen conversion efficiency, but current commercially
available electrolyzers do not have the ability operate at such high efficiency rates.
Today's commercially available electrolyzer units operate at 56-73% efficiency equating
to a range of 70.1-53.4kWh of input feedstock electricity per kg of hydrogen produced.
23
(Kroposki, eta!. 2006) How efficiently an electrolyzer can convert electricity into
hydrogen is a vital determinant of economic viability of a large scale electrolysis project.
In the electrolysis process there are several key factors which determine how efficiently
electricity is converted to hydrogen, predominantly: cell size, voltage, conductivity,
current density and process temperature. (Kruger 2000) Setting an electrolyzer to its
highest efficiency rating is not always the best economic decision in an electrolysis
project. Efficiency changes with variance in electricity load, (current density). Literature
states, that amidst electrolysis as current density drops " ... the specific capital costs of
electrolyzer ($/kg) increases rapidly as capacity decreases." (Ouellette, Rogner and Scott
1997, p.399) This suggests that economic vitality is reliant on large scale electrolysis
employing as high a current density as possible. Higher versus lower current density,
generates a tradeoffbetween production level and feedstock costs. High hydrogen gas
production levels yield from higher current density,and constitute better return on capital
investment. In response, high production levels decrease electrolysis efficiency,
increasing electricity costs per unit of hydrogen produced. (Ouellette, Rogner and Scott
1997) ln a project such as this thesis, with the high capital costs inherent in large scale
electrolysis and low costs of surplus feedstock hydroelectricity, it is the optimal choice
to compromise efficiency for a higher hydrogen production rate.
With improvement of technology, a realistic efficiency goal for future electrolyzers is
roughly 50kWh/kg, or 78%, but this also includes compression of gas to 6000 psi.
However, current electrolyzer technology with the ability to compress to 6000 psi only
operates at the 60kWh/kg range, does not have the production capacity of forecourt
24
scale .. (Ivy 2004) Hydrogen gas has a very low density, without compression, hydrogen
requires a large volume of storage space. Compressing raw hydrogen gas is considered a
finishing stage for retail sale. Some electrolyzers have the capacity to compress gas
through the electrolysis process. When this is the case, electrolysis reduces the energy
input of hydrogen storage, transportation and delivery. This study only considers costs of
producing raw hydrogen gas: storage, transportation and delivery are beyond the scope of
research.
Electrolysis of water has secondary advantages over conventional hydrogen production
and petroleum products. The process of electrolysis generates very pure gas, which
generates superior performance of hydrogen fuel. The simplicity of basic water
electrolysis produces H 2 and 0 2 to purities up to 99.9995%. (Kroposki, et al. 2006)
Additionally, electrolysis produces two other merchantable products besides hydrogen:
oxygen (0 2 ) and heavy water (Deuterium monoxide). Oxygen and heavy water account
for up to 45% of the total products generated in the electrolysis process, 30% oxygen and
15% heavy water. (Ouellette, Rogner and Scott 1997) These by-products represent
additional incentive for a large scale electrolysis project. But the literature warns,
" ... these [oxygen and heavy water] benefits should not be the determining factor for the
project feasibility." (Ouellette, Rogner and Scott 1997, p.399) Although an electrolysis
plant at the scale of this study would produce substantial quantities of oxygen and heavy
water, the value of these by-products is not considered in this study.
Hydrogen has had a reputation problem as a potentially dangerous substance, largely
25
steaming out of popular culture. Ease of hydrogen's flammability was demonstrated in
the 1937 Hindenburg disaster. Hydrogen has endured bad sentiment in its association
with the Cold War hydrogen bomb. (Hydrogen Society 20 12) In actuality, hydrogen
currently plays a large role in manufacturing and industry. The safety concerns of
hydrogen ease of flammability, and potential volatility are mitigated by the assertion that
successful testing, implementation, and practice of safety procedures are already in place
in storage, transportation and consumption of hydrogen. (Tomczyk 2009) Hydrogen
should be handled as a potentially hazardous, flammable product, but is considered no
more dangerous than the multitude of fossil fuels widely present in everyday life.
(Rodgers, et al. 201 0)
Electrolytic hydrogen's reputation for its high level of purity, process simplicity and ecofriendliness, (Wang, Wang and Guo 2010) sets it apart from conventional hydrogen
production. Although electrolysis is limited by an unobtainable 100% feedstock
electricity-to-hydrogen conversion efficiency, technology has maturated to the extent that
high production capacity, of greater than 1,000 kg a day, is possible at efficiency rates of
73% . (Ivy 2004) Whether a conversion efficiency of 73% is sufficient enough to make
electrolytic hydrogen a competitive viable opportunity for the Pacific Northwest will be
developed in Section VI.
Electrolysis Opportunities for Power Utilities:
With high water year conditions yielding large hydroelectricity supplies, Pacific
Northwest, utilities most contend with the generation of a surplus energy supply. The
26
power grid does not have the ability to store electricity and as a consequence, utilities
need to find consumers for all power generated above regional demand. The necessity to
align supply with demand is called 'load balancing,' and has engendered a complex
energy exchange system of Independent Service Operators (ISOs) and Regional
Transmission Organizations (RTOs), where electricity is traded from one region to
another. Even with energy exchanges, on occasion regional and inter-regional demand is
so insufficient that hydroelectric dams have had to be spilled without power generation
and wind turbines are shut down. (Sickinger 2011) This illustrates a significant problem
with the current electric grid: there is no capacity to store power; all electricity needs a
consumer, even to the point of economic detriment to the energy supplier. Electrolytic
hydrogen presents an opportunity for power utilities to have a consistently available
consumer that can store a substantial portion of the input energy for later use.
The current electricity grid is burdened with large amounts of energy transmitting
constantly within and outside utility regions. Especially over long distances, transmission
congestion and bottlenecking causes substantial losses intransmitted energy. Nationally,
problems with transmission and distribution results in the losses of 6-8%of annual total
generated electricity.(Fesmire 2007) Regionally, each year several hundreds of
megawatts of electricity are lost in long distance transmission and bottlenecking.
(Bonneville PowerAdministration 20 II.) Further inefficiencies are created by interregional transmission services. Congestion and regional price discrepancy generates
revenue opportunities for transmission line owner. "When facing unregulated pricing of
transmission services, an owner of a transmission network may not have economic
27
incentive to efficiently mitigate transmission congestion." (Kieit and Reitzes 2008, p. 6)
Electrolytic hydrogen production presents utilities a chance to reduce transmission
congestion. Rather than transmitting all surplus electricity out of region, producing
hydrogen from a proportion of surplus can help balance regional electricity supply and
demand and reduce electricity loses in transmission bottlenecking and congestion.
The dominance of hydroelectricity in the Pacific Northwest constitutes high power
production in spring and summer months because of snow-pack melt and reduced
capacity amidst the fall and winter. This presents a local problem as high production and
high demand occur during opposite seasons. Fall and winter seasons require the highest
energy demand, whereas spring and summer energy demand is reduced. (Abraham 2002)
Generating hydrogen with electricity above that demand load during melt months could
be stored and used during times of potential energy deficits in the colder months. Utilities
could also produce hydrogen at times of off peak demand, periods where power costs are
significantly lower. The hydrogen is compressed or liquefied and stored to be used in a
fuel cell to provide supplemental electricity during times of high demand. (Flour Daniel,
Inc. 1991)
Electrolytic hydrogen represents a resource management opportunity for hydroelectricity
producers. The inability to store electricity means supply must always be balanced with
consumption demand. Electricity to hydrogen conversion represents an opportunity to
have consistently available energy consumer to help utilities balance energy demand with
supply. Beyond load balance, converting electricity into hydrogen offers long term
28
storage of a high percentage of the input electricity's energy. Energy stored in hydrogen
can be converted back into grid electricity by fuel cell or in combustion turbine, or the
raw gas can be sold as commodity. Hydrogen production from Pacific Northwest
hydroelectric resources represents opportunity for more efficient and sustainable natural
resource conservation.
29
Section IV: Background Hydroelectricity to Electrolytic Hydrogen
Studies
This study is not pioneering research in hydrogen production from non-fossil fuel
sources. It is rather a Pacific Northwest regional consideration to the capacity and cost to
convert regionally generated surplus hydroelectricity into electrolytic hydrogen. This
research builds off major findings of other alternative energy to electrolytic feasibility
studies and models and applies research findings as foundational study design . The
following subsections review significant background study findings which directed this
research's scope.
Renewables to Hydrogen and Hydroelectricity' s Advantage:
With the maturation of renewable energy technology, electrolytic hydrogen produced
from alternative energy sources has recently received significant research activity. Much
of the current research has focused on wind and solar power as feedstock electricity.
Studies continually reiterate the key factor of the economic competitiveness of
electrolytic hydrogen is the cost of input electricity. Typically 40% of the total cost to
produce raw hydrogen through electrolysis is electricity.(Ouellette, Rogner and Scott
1997) The current high price of installed solar and wind energy limits the ability to
produce electrolytic hydrogen economically competitive with conventional fossil fuel
based hydrogen production in the near tem1 .
Studies have found that intermittency of solar and wind also generates potential
impediments to electrolytic hydrogen production. Variable current density
30
changeselectrolyzer efficiency, and production capacity, as optimal power input may not
be reached. Intermittency continues to harm efficiency with the possibility of
electrolyzers not reaching necessary operation temperature. These conditions generate
uneconomical consumption of input electricity and hydrogen production output. Safety is
also a concern with solar and wind electrolysis. Operation at low capacity can cause
gases to permeate through the electrolyte and come into contact with each other,
potentially causing dangerous inflammable consequences inside the electrolyzer unit.
(Bartels, Pate and Olson 2010)
Results of solar and wind to hydrogen production studies have found a wide variance in
prices of electrolytic hydrogen. Wind power resulted in range of prices from $2.27- 6.03
per kg of H 2 , but the lowest prices reflect long-term production with considerable
electricity subsidies. Solar based electrolysis prices ranged from $5.10- 23.27, but prices
reflected a wide range in variables and production conditions. (Bartels, Pate and Olson
2010) Studies reflect there are still significant barriers impeding economic
competitiveness of wind and solar electrolysis, mainly the large capital investment in
wind and solar power generation in addition to electrolyzer capital investment.
Hydropower has a significant advantage over solar and wind to hydrogen, as the capital
costs of hydroelectricity have largely been mitigated over many decades of production
since dam installation. Although hydropower and renewable energy could work in
concert, renewable electricity's cost per kWh needs to be reduced before challenging
conventional hydrogen and fossil fuel prices.
31
Hydroelectricity to Hydrogen Studies:
Utilizing hydroelectricity to produce economically viable hydrogen through electrolysis
has received significant previous research. Though the majority of recent studies have
focused on the alternative energy sources of solar, wind and nuclear, electrolytic
hydrogen production plants have been operating near hydroelectric dams since the early
111
20 Century. Several regional feasibility studies have been conducted since the early
1980s suggesting favorable conditions for producing economically viable electrolytic
hydrogen. The studies also acknowledge significant obstacles that could impede
economic competitiveness of electrolytic hydrogen produced via hydroelectricity. As
noted, feedstock electricity costs are continually acknowledged as the prominent factor in
producing low costs hydrogen through electrolysis.
Hydrogen production facilities have been operating at hydroelectric plants for some time.
D.S. Tarkay highlights electrolysis plants located at hydropower dams and outlines
opportunities for hydroelectric utilities in his 1985 article Hydrogen Production at
Hydro-Power Plants. Labeling electrolytic hydrogen from hydroelectricity, "technically
and economically feasible," Tarkay concludes that, "If the professional and business
community will recognize the potential of proposed hydrogen production .. . it can open
the door for a new hydrogen era" (Tarkay, 1985 p.583)
Tarkay's research contends that in 1985, hydroelectric utilities were producing
economically viable electrolytic hydrogen, bolstering the rationale of this thesis. lf outof-date technology can produce volumes of electrolytic hydrogen competitively, it is an
32
encouraging prospect that hydrogen could be produce economically, considering this
study models production from a more advanced electrolyzer unit. Despite being an early
and vague feasibility analysis, Takay's study supports the parameters of this thesis.
Regional feasibility studies analyzing the use hydroelectricity to produce electrolytic
hydrogen in the Pacific Northwest first appeared in 1984. As a response to the Pacific
Northwest Electric Power Planning and Conservation Act's (PL96-501) directive to
consider renewable energy opportunities, the Bonneville Power Administration produced
Feasibility Assessment of Electrolytic Hydrogen in the Pacific Northwest(l984). This
study researched the potential of using 'state-of-the-art' technologies to produce
electrolytic hydrogen from surplus and purchased hydropower. The preliminary technical
and economic feasibility analysis concluded that production of electrolytic hydrogen
was, "attractive" with use of surplus and purchased hydropower. The study concluded
that because only approximately 3.5 months produced substantial amounts of surplus
power, a hydrogen plant would need to purchase additional electricity. Although, surplus
power alone could not produce economical electrolytic hydrogen, "Nevertheless, the cost
of hydrogen generated by a mix of unused surplus hydropower and purchased power
proved to be very attractive." (Pratt, eta!. 1984, p. 8-2) Furthermore, the study concluded
that, electrolytic hydrogen produced in the Pacific Northwest could be accomplished
" ... at less cost than it can be produced by steam reformation of natural gas in most other
parts of the country."(Pratt, et al. 1984, p. 8-2)
BPA's Feasibility Assessment of Electrolytic Hydrogen in the Pacific Northwest (1984)
33
supports several components of this thesis ' hypothesis and rationale. Pacific Northwest
hydroelectricity's low cost gives regional advantage in electrolytic hydrogen production.
The study indicates there is opportunity to use surplus hydropower to produce electrolytic
hydrogen. Finally, the research supports that electrolytic hydrogen produced in the
Pacific Northwest can be competitive with conventional hydrogen produced elsewhere.
Although this study suggests electrolytic hydrogen produced with surplus electricity is
only feasible amidst peak melt months, the study models electrolyzer technology nearly
30 years old. This thesis models the use of electrolyzer technology with greatly improved
efficiency and production capacity, potentially supporting electrolytic hydrogen
production from surplus hydropower for additional months.
ln 1991 Bonneville Power Administration continued to assess feasibility of
hydroelectricity to electrolytic hydrogen production in: Pacific Northwest Hydrogen
Feasibility Study (1991), produced by Flour Daniels Inc. This study modeled I 00 MW of
electrolytic hydrogen production to be used astransportation fuel for consumption in
hydrogen internal combustion engine (ICE). The study modeled feedstock electricity at
the lowest BPA guaranteed costs of 1.5 cents (1991 dollars) a kWh. (Flour Daniel, Inc.
1991) Using 1991 technology and a feedstock electricity cost of 1.5cents a kWh, the
study concludedthat Pacific Northwest produced electrolytic hydrogen fuel would cost
motorists 3.5 cents ( 1991 dollars) per mile. Even accounting for a 10 - 30% hydrogen
engine efficiency advantage for hydrogen fuel, the study concluded that producing
hydrogen for transportation could not be competitive with gasoline, which reported a cost
of only 1.5 cents a mile. (Flour Daniel, Inc. 1991) At the time of this study BPA
34
references wholesale gasoline prices as $0.60.gallon (1991 dollars) and concludes that
unless gasoline surpassed the wholesale cost of$1.41 (1991 dollars), gasoline would
continue to economically outcompete hydrogen based fuel. (Flour Daniel, lnc. 1991)
Now more than twenty years later, the Pac(fic Northwest Hydrogen Feasibility Study
( 1991) offers support for electrolytic hydrogen based fuel to be cost competitive with
gasoline. BPA 's proposed benchmark of $1.41 ( 1991) a gallon wholesale represents the
potential tipping point for hydrogen fuel to be cost competitive with gasoline. This point
appears to have been reached. Updated for inflation $1.41 in 1991 amounts to $2 .38 in
2012 dollars. (United States Bureau Of Labor and Statistics 2012) Over the past year
(May 2011 -May 2012), wholesale gasoline prices have had monthly averages ranging
from $2.61 - $3.17 (United States Energy Information Administration 2012) Wholesale
gasoline has seen large price volatility over the preceding twelve months, but remains
above BPA 's 1991 assessment for hydrogen fuel to reach cost competitiveness. The cost
of feedstock electricity in the 1991 study was $0.15/kWh ( 1991 dollars), and adjusted for
inflation equates to roughly $0.03/kWh (2012 dollars) which is very close to BPA's
current guaranteed lowest payer rate. Electrolyzer technology improvement witnessed in
the two decades since this study promises better efficiency and production rates than
electrolysis employed in 1991 study. Preliminary assessment of the Pacific Northwest
Hydrogen Feasibility Study ( 1991 ),indicates producing electrolytic hydrogen fuel in the
Pacific Northwest can be cost competitive with conventional gasoline.
The 1997, Ouellette, Rogner and Scott, study: Hydrogen-Based Industry from Remote
35
Excess Hydroelectricity, investigates the economics of producing electrolytic hydrogen
from surplus hydroelectricity in the remote Northwest Territories, Canada (NWT). The
Taltson Dam, NWT produced a year-round minimum surplus of 8.5 MW at High Load
Hours of operation and 15 MW at Light Load Hours of operation. The study assumed a I
cent/kWh (1993 Canadian Dollars) rate for feedstock electricity. Ouellette, Rogner and
Scott saw opportunity to produce electrolytic hydrogen because of such low demand and
an inability to transmit hydropower outside of the dam region. The study concluded that
compressed and liquefied electrolytic, which is considered hydrogen's retail product
state, could be produced via excess hydroelectricity more economically than hydrogen
produced through steam reformation. (Ouellette, Rogner and Scott 1997) Although local
conditions allow for a more easily predictable surplus loads, the Ouellette, Rogner and
Scott study validates that under the circumstances of low-cost input electricity,
electrolytic hydrogen derived from surplus hydroelectricity can outcompete conventional
hydrogen production methods.
Contreras, Posso and Nejat Veziroglu's 2007 study: Modeling and Simulation of the
Production of Hydrogen Using Hydroelectricity in Venezuela, models production costs
of using hydroelectric resources to produce substantial quantities electrolytic hydrogen in
Venezuela. This research emphasizes the advantage ofVenezuela's low cost ofelectricty
enabled generation of electrolytic hydrogen at a fraction of the cost of electrolysis cost
studies with higher electricty tarrifs. (Contreras, Posso and Nejat Veziroglu 2007) The
low cost feedstock electricity allowed Venezulean production cost models to outcompete
studies that produed more favoriable ecnomies of scale. The study concluded that using
36
Venezuela's hydropower resources for producing electrolytic hydrogen would be "highly
advantageous." (Contreras, Posso and Nejat Veziroglu 2007)
Contreras, Posso and Nejat Veziroglu provides this thesis with the foundational cost
equation for total electrolytic hydrogen production costs.
Cost of energy consumption (annual)+ Cost of Investment (annual)+ Cost of 0 &M (annual)
--------~------------~~~--~----~~~--~~------~----~--~1
Total hydrogen kg produced (annaul)
Calculation of the total cost of electrolytic hydrogenproduction is a simple equation
entailing the summation of the three numerator variables: total cost of feedstock
electricity consumption, total cost of capital, and total cost of operation and
maintenance; divided by denominator variable: total kilograms of electrolytic hydrogen
produced. The resulting metric represents the overall cost of per kilogram of electrolytic
hydrogen. This cost equation will be employed in Section VI to calculate total production
costs of electrolytic hydrogen produced with Pacific Northwest surplus hydropower.
The review ofbackground electrolytic hydrogen production studies provides this thesis
foundational support in the feasibility of cost competitive electrolytic hydrogen
production. Background studies establish potential economic viability of regional
electrolytic hydrogen production and the potential to produce electrolytic hydrogen at
costs competitive with conventional hydrogen and gasoline. Previous studies address the
vitality of low cost feedstock electricity and suggest management strategies in balancing
efficiency with production. Finally, background research furnishes this thesis with a
1
Contreras, Posso and Nejat Veziroglu. 2007. p.l222
37
validated equation for producing an overall hydrogen cost per kilogram metric. Having
addressed the foundational studies guiding this thesis, this study now progresses to
quantify surplus electricity capacity and production costs.
38
Section V: Quantification of Surplus Hydroelectricity Capacity, and
Fixed and Variable Electrolysis Production Costs
In order to draw conclusions on the viability of utilizing Pacific Northwest surplus
hydroelectric resources to produce electrolytic hydrogen, it is necessary to quantify
surplus hydropower capacity, and associated production costs of a large scale electrolysis
project. Surplus hydropower capacity will be quantified from Bonneville Power
Administration's Pacific Northwest Loads and Resources Study 20I I, also known as
the20I I White Book. This annually produced document, forecasts energy loads and
resources based upon historic water year conditions. Fixed production costs will be
established through National Renewable Energy Labs literature that reviewed
commercially available electrolyzers and associated costs. Variable electrolysis costs will
be quantified from Bonneville Power Administration's 20I 2 Power Rate Schedule and
General Rate Schedule Provisions. With the quantification all production variables, an
analysis of capacity and productions costs of electrolytic hydrogen will be made in
Section VI.
Hydroelectricity Availability:
The availability of a substantial supply of low cost electricity is vital in producing
electrolytic hydrogen at competitive costs. In order to offset the high capital costs of
large electrolyzers, electrolysis should be performed nearly around the clock at optimal
production capacity, to deliver lowest capital costs. Power intermittency needs to be
marginalized, as aforementioned, changes in current density resulting from intermittency
generates reductions in electrolysis efficiency and increases thecosts of capital per unit.
39
Therefore, economic competitiveness hinges on availability of a constant flow of
electricity at the electrolyzers maximum capacity rating. The aforementioned Norsk
Hydro Atmosphere 5040 (5150 Amp DC)'s optimal production capacity requires2.328
MW of feedstock electricityand without a consistent delivery of 2.328 MW, producing
hydrogen at costs competitive with conventional hydrogen production and gasoline may
not attainable.
BPA is the marketing authority of all federal hydroelectric power generation projects in
the Pacific Northwest. One the largest energy marketing agencies in the Department of
Energy, BP A oversees the United States ' largest hydroelectric resource. BP A alsohas
partial marketing governance over additional regional electricity resources, including
nuclear and renewables . BPA produces power from a wide variety of sources, yet even
under the worst of water year conditions, hydropower accounts for the highest proportion
of power generation, providing about half of the total regional energy resources. BP A has
a maximum hydroelectric capacity of 20,594 MW , (enough to power more than 16.4
million homes) and an annual average generation of 6,845 MW. (Bonneville
PowerAdministration 2011.) Only in times of critical water conditions, and during winter
months, BPA risks producing no monthly surplus.Critical water levels account for only
10% of all water years, so for inthe vast majority of water years there is no risk of
monthly power deficit.
Because water year conditions are directly responsible for a very large range inelectricity
generation capacity, it is essential for BPA to forecast its potential energy production.
40
Annually, BPA produces energy capacity reports displaying estimated retail loads, load
capacity, surplus, and deficit. The reports projects 10 years in advance, estimating
generation capacity under each of the 70 water year conditions. The report compares
estimated capacity under each water year to projected retail load, the anticipated
electricity consumption. BPA then produces anticipated surplus/deficits under each of the
corresponding 70 year water conditions. All surplus power is a direct result of
hydropower resources. SPA's other electricity resources, nuclear, coal, combustion
turbines, cogeneration units, etc., provide base-load power to the region with fixed
annual generation capacity. These sources are contractually guaranteed to connect to the
grid and always provide the same constant electricity input, but are insufficient to cover
the entire retail load. Additional power generated above these fixed amounts must be
derived from hydropower. Table 3, displays Bonneville Power Administration's
anticipated average annual regional surplus power, in average megawatts (aMW), for the
next ten years under variable water year conditions:
Table 3: Forecasted Surplus Electricity Generation in Average Megawatts Under
Variable Water Year Conditions
Operating
Year
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021*
Critical
Water Level
3,972
3,761
3,419
3,008
2,748
2,274
2,257
1,758
1,594
663
Bottom 10%
Water Years
3,945
3,728
3,383
2,973
2,714
2,241
2,222
1,725
I, 560
630
Middle SO%
Water Years
7,296
6,997
,614
6,192
5,936
5,466
5,439
4,950
4,780
3,855
Top 10%
Water Years
10,632
10,283
9,856
9,415
9,165
8,656
8,659
8, 140
8,009
7,045
Table 3 displays average megawatts of forecasted surplus hydropower generation in the Pacific Northwest
for the years 2012-2021 under Critical , Bottom I 0%, Middle 80% and Top 10% water conditions
Source: Bonneville Power Authority: 201 I Pacific Northwest Loads and Resources Study. May 2011 p 71.
*2021 surplus decline attributed to scheduled Centralia coal plant retirement.
41
The data displayed in Table 3, shows the large range in surplus availability associated
with variable water year conditions, with a minimum surplus of 630 aMW to a maximum
surplus of 10,638 aMW. This report depicts very favorable conditions for electrolytic
hydrogen production for surplus hydropower resources. Furthermore, this surplus data
does not include all regional hydroelectricity production. It reflects only electricity
generated by BPA and delivery of electricity from Independent Power Producers, who
historically contribute to BPA base-load power supply. Significant amounts of
hydroelectricity is generated by non-federal firms including, but not limited to: Seattle
City Light, Tacoma Power, and Puget Sound Energy, all of which have the capacity to
generate surplus power, depending on the water year. Since data in Table 3 does not
represent all potential surplus hydroelectric capacity in the Pacific Northwest, it
represents conservative surplus estimates of surplus capacity. As seen in Table 3,
indicates strong support for sufficient availability of surplus of hydroelectricity to employ
at least one Norsk Hydro Atmosphere Type 5040 (5150 Amp DC.). On an annual scale,
even under critical water conditions, the average MW surplus far exceeds the 2.33 MW
necessary minimum to consistently run the selected electrolyzer. Annual data indicates a
large capacity of surplus power for the proceeding ten years, producing ample supply to
operate at least one electrolyzer unit full time.
Annual surplus numbers suggest ample capacity for electrolytic hydrogen production
from surplus power, but hydroelectric outputs vary drastically depending on the month.
This makes it is important to consider projected monthly surplus conditions. Appendix A
displays a table for the years 2012-2021 projected monthly electricity surplus in average
42
megawatts under critical, bottom 10%, middle 80% and top 1O%water year conditions.
Under cortical water conditions, over the next 120 months, only four months are
projected operation in a deficit. All four months of deficit occur in 2021 when the
Centralia Coal Plants are scheduled for retirement. The remaining 116 months all provide
ample surplus to operate at least one forecourt electrolyzer. Even if the roughly 1,000
aMW of the Centralia Coal Plant are not replaced, data suggests there is ample surplus
electricity capacity for electrolytic hydrogen to be produced in the eight remaining
months, even under critical water conditions. Critical water conditions represent only
roughly 10% of all water years, giving deficit months only a marginal percentage of
occurrences. Consequently, electrolytic hydrogen represents an opportunity to mitigate
times of deficit, as hydrogen could be used to generate grid power, potentially alleviating
rare occurrences of energy deficit.
Analysis of regional surplus hydroelectricity data suggests the Pacific Northwest has the
capacity to utilize its surplus electricity resources to realistically produce constant
electrolytic hydrogen. The vast majority of months for the next ten years are projected to
produce far more surplus than the minimum 2.33 MW required to operate the Norsk
Hydro Atmospheric ' s optimal production state. With surplus electricity capacity data
suggesting favorable resources for electrolysis, the study can proceed to quantify
production cost variables.
43
Feedstock Electricity Costs:
As mentioned earlier, a single forecourt electrolyzer unit'sconstant consumption of 2.33
MW of electricity is enough to power approximately 1850 homes. This large of an energy
demand means fractions of cents per kWh can make the difference between project
economic feasibility orfailure. This high consumption demand facilitates the need to
acquire the least expensive feedstock electricity as possible. The Pacific Northwest 's
substantial hydroelectric capacity helps deliver electricity at the lowest rates in the entire
United States. The high capacity and low electricity tariffs make the Pacific Northwest a
very attractive location for this pilot electrolytic hydrogen production study. Feedstock
electricity is the only variable cost in total electrolytic hydrogen production costs. In the
Pacific Northwest electricity prices per kilowatt hour can vary annually dependent on
operating water year conditions; monthly, dependent on seasonal river conditions; and
daily, dependent on Heavy Load Hours versus Low Load Hours. With forecourt
electrolysis' high energy demand, these feedstock price variations have drastic impact on
costs per kg produced.
It is difficult to forecast prices for surplus electricity. Surplus availability is dependent
onwater year conditions, and availability dictates surplus price. Water year conditions
remain unknown until operation within said calendar year. High water years produce a
greater surplus and a low price for surplus electricity, whereas lower water years
engender a reduced surplus, which subsequently demands a higher price per kilowatt
hour. Water year conditions have less influence over contractual payer rates, as they are
set prior to the beginning of the water year calendar. Regardless of water year, this study
44
assumes that price of surplus electricity would be at least as low as the lowest
contractually guaranteed rate paid by energy consumers. Bonneville Power's lowest rate
is paid by Priority Fim1s and defines the energy as," ... electric power ... continuously
available for direct consumption or resale by public bodies, cooperatives, and Federal
agencies." (B01meville Power Administration 2007, p.95) Priority Firms rates vary with
seasons and load operating hours, and have a publicly listed 2012 average rate of$30.17
per MW consumed. (Bonneville Power Administration 2012) Table 4 displays monthly
rates for Priority Fim1s during High Load Hours (HLH) and Low Load Hours (LLH),
which for this study will employ to represent surplus electricity rates.
Month
October
November
December
January
February
March
April
May
June
July
August
September
Average
Table 4 BPA Priority Firm Rates
Heavy Load Hour Rate*
Light Load Hour Rate*
31.04
24.38
31.55
24.58
34.28
26.57
33.21
24.88
34.11
26.35
32.75
25.51
30.71
23.59
28.24
17.58
29.15
16.2
35.25
23.09
37.53
25.33
36.63
26.77
32.871
23.736
*Rate in mills/kWh
Table 4 displays BPA monthly Priority Firm Rates for electricity consumption during High Load Hours and
Low Load Hours of operation
Source: Bonneville Power Administration : 2012 Power Rate Schedule and General Rate Schedule
Provisions
www.b pa. gov/corporate/ratecase/20 12/docs/ FinaiPo\\ erRateSchedulesGRSPs Upload 01-17-20 12. pdf
Table 4 displays BPA 's lowest contractually guaranteed rates. The payer rates vary
month to monthdue to river conditions and daily operating load hours. For this study,
these Priority Firm Rates will represent the highest surplus electricity prices and will
reflect the feedstock electricitycosts per kWh for the electrolytic hydrogen production
45
from surplus hydroelectricity. Heavy Load Hours (HLH) reflect energy consumption
Monday through Saturday 06:00 -21 :00, and Light Load Hours (LLH) constitute energy
consumption Monday through Saturday 21 :00 - 06:00, all day Sunday, and six additional
federal holidays . LLH amass to 3,576 hours annually or nearly 41 % of total load hours.
HLH make up the remaining 5,184 hours, roughly 59% of the time annually. Priority
firms pay an average rate of 32.871 mills/kWh
mills/kWh
(~2.4 ¢ /kWh)
(~3.3 ¢ /kWh)
during HLH, and 23 .736
during LLH.
Priority Firm Rates seen in Table 4 are selected as study parameters for electricity
feedstock prices because of their annual predictability regardless of operating water
year.Setting Priority Firm Rates as feedstock electricity price represents a high surplus
price estimate, enabling their projected hydrogen production rates to be more
conservative than assigning arbitrary surplus electricity prices, which would be hard to
substantiate because of inability to project water year conditions.
Capital Costs of Electrolysis:
Operating an electrolysis plant entails two categorical fixed investment costs: Capital
costs and operation and maintenance (O&M) Costs. Each of these input expenses make
up a substantial proportion of the price per kilogram of hydrogen produced. Fixed costs
are static, and remain constant through the entire life of an electrolysis project, assuming
electrolysis is continually operating at the current density for optimal production.
Capital costs represent the most burdensome of electrolysis ' two fixed production costs.
46
Although electrolysis is fundamentally a simple process, it requires largest startup capital
expenses, including but not limited to : procurement of the electrolyzer unit, physical
plant of a production facility, construction, engineering costs, and substantial contingency
capital. (Ivy 2004) Even with capital costs spread out over the expected 40 year
electrolysis plant life, capital costs represent the most substantial proportion of
electrolytic hydrogen's production costs.
Capital costs ' impact on and electrolysis projectcan be reduced with economies of scale.
Compared to smaller levels of electrolytic hydrogen production, forecourt electrolysi s'
high capacitygenerates greatly reduced capital investment per kg of H 2 produced.
Electrolysis systems that produce around I 00 kg daily accrue capital costs amounting to
roughly 55 % of the total cost of production. Scaling down to about 20 kg a day capital
costs rise to more than 70% of all production costs. The substantial capital costs currently
rule out economically competitive small scale electrolytic hydrogen production, as it
generates hydrogen at the $8-19 a kg price level. (Ivy 2004) Literature determined the
large production volume of forecourt electrolysis can reduce capital costs to below onethirdof total hydrogen production costs, which is the only production scale able to
mitigate capital costs to that low of a proportion of total costs. (Ivy 2004)
According to literature, the Norsk Hydro Atmosphere type 5040 (5150 Amp DC)
produces raw gas with capital costs of $1.32 (2000 dollars) per kg. This cost expects a 40
year plant life with electrolyzer stack replacement at the I Ot11 , 20th and 30th year of
operation. Unlike feedstock costs which fluctuate with season and time of day, capital
47
costs remain constant, as long as electrolysis is produced at optimal production capacity.
The burden of capital costs of electrolysis practically mandates forecourt scale production
to attain economically competitive electrolytic hydrogen production.
Electrolysis Operation and Maintenance Costs:
Like capital costs, operation and maintenance (O&M) costs are fixed. Literature defines
primary O&M expenditures of electrolysis plant as labor and overhead expenses as well
as operation expenditures of insurance and taxation. O&M contribute to roughly 10% of
the cost to produce electrolytic hydrogen, amounting to $0.37 (2000 dollars) per kg
produced. (Ivy 2004) O&M costs remain proportional to scale in electrolysis, amounting
to about 10% of production costs whether producing 20 or 1000 kg daily.
Literature contends there are secondary O&M costs. Other raw materials and
miscellaneous O&M are less scheduled expenses throughout the life of an electrolysis
project. Other raw materials and miscellaneous O&M primarily are used for electrolyzer
calibration and efficiency optimization and include:replacement KOH electrolyte, demineralized water, cooling water, and inert gas. (Ivy 2004) Combined secondary O&M
costs account for about $0.05 (2000 dollars) per kg ofH 2 .
Capital and O&M fixed costs reflect the unavoidable costs of electrolytic hydrogen
production. At forecourt scale the combined capital and O&M costs amount to $1.74
(2000 dollars) per kg. Adjusting for inflation, capital and O&M expenses at $1.74 in
2000 dollars amasses $2.32 in 2012 dollars. (United States Bureau Of Labor and
48
Statistics 20 12) This represents the bare minimum cost of electrolytic hydrogen
production for the Norsk Hydro Atmosphere 5040 (5150 Amp DC). If the third variable
of production costs, feedstock electricity, was free of charge, raw hydrogen gas could be
produced at the $2.32 per kg, a highly competitive price for fuel. Although free electricity
is not expected, years of high surplus hydroelectricity can engender very inexpensive
feedstock energy.
With the establishment and validation for all the input costs variables of electrolytic
hydrogen production, analysis of total costs of electrolytic hydrogen from surplus
hydroelectricity can now be made. The following section will combine all independent
cost variables and assess the potential price per kilogram of electrolytic hydrogen
produced from surplus hydroelectricity in the Pacific Northwest.
49
Section VI: Cost Calculation of Electrolytic Hydrogen form Surplus
Hydroelectricity
With the validation of sufficient surplus hydroelectric capacity to support at least one
forecourt electrolyzer and the quantification of feedstock electricity, capital and operation
and management costs, the total production costs of raw electrolytic hydrogen gas can be
detem1ined. As mentioned earlier on page 37, this study utilizes the Total Production
Costs equation as employed by Contreras, Posso and Nejat Veziroglu (2007).
The following table, Table 5, Costs per Kilogram of Electrolytic Hydrogen with Variant
Feedstock Electricity Rates, displays a multiple feedstock electricity rates:feedstock [1] , a
rate of $0.048 kWh , represents the national average of industrial electricity rates. [2] , [3] ,
and [4] represent round benchmark feedstock electricity rates, reflecting potential low
electricity prices under high surplus conditions.[5] displays BPA ' s mean Industrial Firm
Rate under High Load Hours (HLH) operation. [6] displays BP A ' s mean Industrial Firm
Rate under Low Load Hours (LLH) operation. [7] demonstrates BPA's mean Priority
Firm Rate under HLH conditions. Finally, [8] demonstrates BPA's mean Priority Firm
Rate under LLH conditions.
50
[1]
Table 5: Costs Per Kilogram of Electrolytic Hydrogen with
Feedstok Electrict~ Rates
$/kWh
Electrcity
Electricty $/kg
Capital
&OM
Costs Hourly
($/43.6kg/hr)
($/kg)
(kWh X 2328)
$2.32
$0.048
$111.744
$2.563
Variant
Capital &
OM$/hr
Total
$/kg
$101.152
4.883
[2]
$0.03
$69.84
$1.602
$2.32
$101.152
3.922
[3]
$0.02
$46.56
$1.068
$2.32
$101.152
3.388
[4]
$0.01
$23.28
$0.534
$2.32
$101.152
2.854
[5]
$0.040381 *
$94.006968
2.156
$2.32
$ 101.152
4.476
[6]
$0.031206**
72.647568
1.666
$2.32
$101.152
3.986
17]
$.03287083***
76.52329224
1.7551
$2.32
$101.152
4.075
[8]
$.02373583****
55.25701224
1.2674
$2.32
$101.152
3.587
Table 5 displays the effect of varaible feedstock electricty prices on in the cost per kilogram of
electrolytic hydrogen production.
Table Notes: *BPA average HLH Industrial Rate **BPA average LLH Industrial Rate 2
*** BPA average HLH PF ****BPA average LLH PF rate
Table 5 demonstrates electrolysis' total production costs dependence on the price of
feedstock and reiterates literatures' assessment of feedstock electricity's importance to
competitive electrolytic hydrogen costs. Results of Table 5 calculations and all feedstock
electricity rates from $0.00-$0.58 kWh , are represented in Figure 3 by the linear
expression: y($/kg)= 53 .3936x ($/kWh)+ 2.31997.
2
Payer rate information courtes y of BPA.20 I 2 Po wer Rate Schedules and General Rate Schedule
Provisionsp.21
http: //www.bpa.gov/corporatc/ratecasc/20 12/docs/FinalPowcrRateSchcdulcsGRSPs Upload 01-17-20 12.pdf
51
Figure 3: Total Cost Forecast For Forecourt Electrolytic Hydrogen
Production with Variable Feedstock Electricity Costs
- 6
-r- -5
l
4
____-r--2
---- ......
--
~-
~~
~
-----
-
I--
-
1
0
$0.00
$0.01
$0.02
$0.03
$0.04
$0.05
$0.06
$0.07
$/kWh
Figure 3 displays projected total production costs of electrolytic hydrogen per kilogram
produced with the Norsk Hydro Atmosphere 5040 (5150 DC) electrolyzer under variable
feedstock electricity costs.
Linear Equation: y=53.3936x +2.31997
Figure 3 also displays the potential economic viability offorecourt electrolysis. Under
conditions of feedstock electricity rates averaging at or below $0.0315 electrolytic
hydrogen can be produced at $4 per kilogram. To reach the National Renewable Energy
Labs goal of $3 per kilogram hydrogen feedstock electricity would have to have an
average cost just over $0.0127 per kilowatt hour. These prices only reflect hourly
production under variable circumstances. Table 6 displays anticipated annual production
rates and costs comparing annual electrolytic production costs of Industrial payer
electricity rates and Priority Firm Rates, which represent a conservative surplus
electricity pricing. Production capacity and costs are calculated for each payer' s mean
HLH and LLH operation. Each payer's HLH and LLH costs and production outputs are
combined to display the mean costs per kilogram of electrolytic hydrogen over an entire
year of production.
52
Table 6: Annual Electrolytic Costs and Production Capacity at lndustirial Firm and
Priroritl: Firm Electricitl: Rates
Feedstock
Rate
IR
Annual
Operating
Time
(hr.)
Annual
Production
Capacity
(kg.)
Annual
Electrcity
Consumption
($)
4.476
5184
226022.4
3.986
3576
4.075
3.587
Equivilent
Production
Costs
($/kg)
Annual
Capital
Costs
($)
Annual
O&M Costs
($)
Annual
Total Costs
($/yr.)
HLH and
LLH
Combined
$/kg
487332.122
397799.424
126572.544
1011676.26
- 4.276
155913.6
259787.703
274407.936
87311.616
621471.61
5184
226022.4
396696.747
397799.424
126572.544
921041.28
3576
155913.6
274407.936
87311.616
559261.01
HLH
IR
LLH
PF
- 3.876
HLH
PF
197599.076
LHL
Table 6 displays electrolytic hydrogen production capacity and production costs on an annual
scale at BPA 's Industrial Rates (IR) and Priority Firm Rates (PF). Production capacity and costs
are presented under High Load Hours (HLH) and Low Load Hours (LLH) and load operating
hours costs and production are combined to demonstrate an overall mean cost per kg of
electrolytic hydrogen . PF Rates represent conservative surplus electricity rates .
The cost of production electrolytic hydrogen under Priority Firm and Industrial Firm rates
as presented in Table 6 is calculated by employing Contreras, Posso and Nejat Veziroglu
2007 Total Production Costs of Electrolysis equation. Displayed below is the Priority
Firm data presented in Contreras, Passo, and Nejat Veziroglu total cost equation:
$/kg=
Cost of energy consumption (annual) + Cost of investment (annual) + Cost of O&M (annual)
Total hydrogen kg produced (annual)
$/kg=
($594295.823) + ($672207 .36) + ($213884.16)
381936 kg.
1
$/kg=$3 .876
Table 6bestarticulates the opportunity for electrolytic hydrogen production in the Pacific
Northwest region, as long as there is a consistent delivery of 2.33MW of electricity, a
single forecourt electrolyzer can produce a maximum of 3 81,936 kg. of electrolytic
hydrogen annually. Regardless of input electricity costs, there are unavoidable fixed costs
3
Contreras, Posso and Nejat Veziroglu. 2007. p.l222
53
totaling $886,091.52 to operate a forecourt electrolyzer annually. The high fixed costs of
a forecourt electrolysis project preventthe ability to produce electrolytic hydrogen below
$2.32 per kg, and this requires zero cost feedstock electricity. Table 6 demonstrates
production costs of both Industrial Firms and Priority Firms for two reasons: Primarily,
Priority Firms is used to represent a potential conservative surplus electricity price rate
and providing Industrial Firm production costs allows for comparison of surplus
feedstock electricity to a contractually guaranteed payer rate. Secondly, using both IR and
PF displays the electrolytic production at the two lowest contractually guaranteed
electricity rates by Bonneville Power Administration, and demonstrates the opportunity
to contract feedstock electricity rates and perform electrolysis, which would eliminate the
potential risk of relying on higher water year condition to generate ample power surplus.
Results of the cost analysis of producing electrolytic hydrogen utilizing projected surplus
Pacific Northwest hydroelectricity indicate quite attractive hydrogen production
opportunities. Analysis of availability of surplus hydroelectricity appears sound
regardless of water year conditions until 2021. The anticipated 2021 closure of the
Centralia Coal Plant, reduces regional base-load electricity capacity by approximately
1,OOOaMW, requiring a substantial quantity of surplus hydropower to replace the
diminished base-load capacity. This presents a long-term potential obstacle for surplus
power electrolytic hydrogen production, but only in critical water year conditions and
only in the four low river flow months. Perceivably, the lost capacity from Centralia Coal
Plant closures could be mitigated by increased renewable projects planning to connect to
the grid in the near future.
54
Overall, results suggest there is capacity to generate electrolytic hydrogen through
surplus hydroelectricity in the Pacific Northwest at competitive costs. The projected
production cost of approximately $3.88 per kilogram, puts electrolytic hydrogen
produced from surplus hydropower in the $1-$5 per kilogram price range of conventional
hydrogen production as reported by literature. Preliminary comparison of electrolytic
hydrogen to gasoline suggests gasoline still has an economic advantage. Raw electrolytic
hydrogen produced at a cost of$3.88 per kg. is above the $2.61 - $3.17 per gallon
wholesale price range of gasoline witnessed over the past twelve months. Hydrogen
outperforms gasoline significantly in efficiency, and prior to sale to direct consumers
both raw hydrogen and wholesale gasoline entail additional costs including:
transportation, delivery and taxation, so further study needs to be conducted in final
hydrogen and gasoline cost comparisons. Nevertheless, results suggest electrolytic
hydrogen production utilizing surplus hydroelectricity is a feasible project for the Pacific
Northwest and generates hydrogen gas at a cost within the range of competitiveness to
conventional hydrogen production.
55
Section VII: Conclusions
Electrolytic hydrogen represents an opportunity to fundamentally change the landscape of
energy resources. Clean, efficient and simple to produce, electrolytic hydrogen can be
produced from renewable, domestic power resources. Electrolytic hydrogen retains a
substantial portion of the input electricity, presenting an energy storage opportunity that
our electrical gird currently lacks. Hydrogen has the versatility to generate grid electricity
or can be consumed as transportation fuel. Hydrogen can perform all the duties of fossil
fuels without many of the undesirable environmental and social consequences. When
consumed hydrogen only emits pure water and a minute amount ofN0 3 . When produced
from renewable energy sources,electrolytic hydrogen has marginal life-cycle greenhouse
gas emissions. Transitioning to a hydrogen based transportation system would alleviate
substantial greenhouse gas emissions and decline the necessity to conduct business in
socially turbulent petroleum producing nations. What suppresses transition to the
hydrogen economy is an inability to produce large volumes of electrolytic hydrogen at
prices competitive with fossil fuels.
This study considered economic competiveness of electrolytic hydrogen produced with
surplus hydroelectric resources forcomparison to gasoline and conventional hydrogen
production. The study analyzed the Bonneville Power Administrations projected surplus
electricity availability for the next I 0 years, established a maximum price for surplus
electricity and establishedcosts of capital and operation and maintenance of the largest
commercially available electrolyzer.Analysis of projected surplus hydroelectricity
suggests there is ample capacity for electrolytic hydrogen production to constantlyoperate
56
at least one forecourt electrolyzer even at critical water year conditions, meaning under
the worst of water year conditions, there is the capacity to produce more than 300,000 kg.
of electrolytic hydrogen annually. This study considered electrolytic hydrogen production
under very conservative parameters of critical water year conditions and surplus
feedstock electricity at conservative rates. More favorable water year conditions would
allow for a great deal more electrolytic opportunities at lower costs. Literature supplied
the equation for calculation of total production costs : the summation of the cost of total
electricity consumption, total capital cost and total O&M costs all divided by total
hydrogen production. The resulting metric represents the overall cost per kilogram of the
electrolysis project. Literature supplied electrolysis fixed cost data and production
capacityfor the Norsk Hydro Atmospheric electrolyzer. The variable costs of feedstock
electricity were supplied from BPA's Priority Firm Rates. The study resulted in
electrolytic hydrogen produced at an average cost of about $3 .88 per kilogram. Priority
Firm Rates and the associated costs of $3.88 per kilogram represents a conservative total
production costs for electrolytic hydrogen utilizing surplus hydropower in the Pacific
Northwest.
Although above the targeted $3/kg goal for electrolytic hydrogen set by National
Renewable Energy Labs, electrolytic hydrogen at $3 .88 per kg. is in the range of $1-5 per
kg of conventional hydrogen costs cited by literature. This suggests that electrolytic
hydrogen produced with Pacific Northwest hydroelectricity can be competitive with
conventional hydrogen production. Moreover, results suggest electrolytic hydrogen can
be produced in the $1-5 per kg range with feedstock electricity rates up to $0.05 per kWh.
57
BPA has several contractually guaranteed payer rates at costs below $0.05 per kWh,
suggesting because of the low cost of feedstock electricity in the Pacific Northwest
region, electrolytic hydrogen can be produce at costs competitive with conventional
hydrogen without utilizing surplus electricity.
Producing raw hydrogen at costs of $3.88 per kilogram is above the twelve month price
range of wholesale gasoline of $2.61 - $3.17 per gallon. This only reflects a cursory
comparison of electrolytic hydrogen and gasoline. Both fuels entail further finishing costs
before retail sale, and although gasoline per gallon and hydrogen per kg contain the same
energy capacity, hydrogen gains up a 2.5 times greater energy efficiency. These factors
require significant additional research before a complete comparison of electrolytic
hydrogen to gasoline may be concluded on.
Transition to the hydrogen economy on the national scale is stymied by lack of feedstock
electricity. Utilizing the largest electrolyzer only produces approximately 1000 kg a day
and consumes 2.33aMW. Converting our entire transportation fleet to hydrogen would
require nearly double the entire United States electricity capacity. Total electricity
capacity could be increased by continual generation at maximum sustainable yield,
enabling generation to remain at its most efficient output all the time, but increased
capacity is still necessary. Outside of the Pacific Northwest the substantial proportion of
electricity is produced with less environmentally friendly energy sources. There is little
value in producing electrolytic hydrogen from fossil fuel-based electricity sources.
58
Nevertheless, results of this study indicate there is significant opportunity to the produce
electrolytic hydrogen through the Pacific Northwest's surplus hydroelectricity capacity.
The Norsk Hydro Atmosphere has a large energy demand at 2.33aMW, but for the
majority of the water year conditions this is a marginal proportion of anticipated surplus
electricity. For the next nine years thereis sufficient surplus hydroelectricity capacity to
utilize additional electrolyzer units even under critical water year conditions. This
suggests there is great opportunity for the Pacific Northwest utilities to invest in multiple
forecourt electrolyzers and produce large quantities of electrolytic hydrogen for surplus
hydropower. There would be limited risk in such a venture considering 90% of water
year conditions produce a minimum of 600 aMW surplus, which is enough surplus
energy to run more than 250 of the largest electrolyzer, producing more than 75 million
kg of hydrogen annually. This level of electrolytic production would be a significant step
in the transition towards hydrogen energy. After the scheduled 2021 closure of the
Centralia Coal Plant, the lowered electricity capacity in the Pacific Northwest creates
increased risk in utilizing surplus energy for hydrogen production, but years with water
years above critical conditions should have ample capacity to run far more electrolyzers
than just single unit modeled in this pilot study.
This study finds evidence that highly suggests that the Pacific Northwest has the capacity
to generate electrolytic hydrogen that can compete with conventional hydrogen and
potentially competitive with gasoline. Results support that there is hydroelectric capacity
to generate a forecourt scale electrolysis project, presenting the region with an
opportunity to produce hydrogen to help balance electricity load, store a backup energy
59
supply, alleviate transmission congestion, and generate merchantable gases. Electrolytic
hydrogen from surplus hydroelectricity represents aviableopportunity for the
sustainability of the Pacific Northwest energy resources and the movement towards a
clean energy future.
Section VII: Suggestions for Further Research:
Researching the opportunity to produce electrolytic hydrogen from surplus
hydroelectricity generated additional avenues for further research that fell beyond the
scope of this thesis . This study was able to conclude that there is capacity to produce
electrolytic hydrogen utilizing surplus hydropower at economically competitive costs to
conventional hydrogen production, but there is room for further assessment of overall
capacity and evaluation of conditions and opportunities which could alter production
costs.
Climate Change poses major concerns for hydroelectricity producers. Forecasted surplus
electricity data did not reflect Climate Change ' s impending impact on precipitation rates,
stream length, peak flow, and power generation timing. There is substantial opportunity
to assess Climate Change ' s influence on electricity capacity in the Pacific Northwest
region. A change in electricity capacity could mean substantial change in feedstock
electricity prices, which is the only variable costs in the electrolysis process.
Water year conditions play the dominant role in determining surplus electricity capacity.
A thorough risk assessment of potential months, conditions and times of electricity deficit
could give a clearer image of long term viability of continual electrolytic hydrogen
production. Although review of monthly data suggested only 4 months out 120 operating
at deficit under critical water year conditions, additional review of low flow months
should be studied.
61
This study did not consider costs and methods of hydrogen compression, transport and
delivery. If a study determines these additional costs, a more accurate assessment of retail
hydrogen gas could be made. There are many methods of compression, transportation and
delivery, so it is vital to study which is the most cost effective and efficient. These
conditions play key roles in transition to the hydrogen economy.
There are emerging technologies which support more efficient hydrogen production.
High gravity, high temperature and new polymer electrodes have great potential in
lowering production costs of electrolysis. An assessment of how these emerging
technologies could reduce feedstock cost and produce cheaper hydrogen gas could
improve electrolysis cost effectiveness.
Electrolysis produces two additional merchantable commodities: oxygen and heavy
water. This study did not assess the potential value of these byproducts. An assessment of
how oxygen and heavy water production could reduce capital costs of electrolysis could
offer more incentive for electrolytic hydrogen production.
This study analyzes only a single electrolyzer unit. A maximum electrolysis capacity
study would give a clearer image of a realistic target for high volume hydrogen
production utilizing multiple electrolyzer plant. There are additional areas of study as
new technology continually emerges. This study by no means covers the gamut of
electrolytic hydrogen production, but hopefully provides jumping off points for
additional research.
62
Appendix A: Pacific Northwest Monthly Surplus Electricty Data
Projected Monthly Surplus at Top 10%, Middle 80% and Bottom 10% water Year Conditions in
Averagre Megawatts
Year
WY
Top
Aug!
7356
Aug16
4803
Sept
20112012
4319
Oct
6475
Mod
5730
3645
4016
Bot
3853
2112
2896
20122013
20132014
20142015
2015 2016
2016 2017
2017 2018
20182019
20192020
20202021
Dec
10921
Jan
13732
Feb
13517
5096
5623
5573
7547
7122
7130
10963
10790
3992
4131
3092
2868
2718
2590
4890
4902
Nov
Mar
13164
Aprl
15012
8709
Ap16
14584
Jun
13395
July
10159
11942
7778
7285
6020
6396
4662
3934
May
11831
10711
Aug
10621
Top
7540
5629
4623
6382
8621
10684
13582
13282
13157
13398
13336
11025
12206
10014
10273
Mid
5672
4218
4571
5002
5541
5359
7462
6934
71 12
9425
9621
9421
10922
7151
6986
Bot
4077
2627
3760
3899
4049
2879
2768
2529
2621
3370
3751
5349
5731
4093
3718
Top
7056
5161
4331
6034
8198
10174
12850
12592
12244
13752
12846
10799
12411
9373
9846
Mid
5196
3748
427 1
4653
5123
4882
6771
6276
6287
9795
9187
9353
11118
6655
6603
Bot
3608
2153
3466
3551
3632
2403
2075
1878
1810
3748
3342
5557
5965
3685
3372
Top
6563
4819
3930
5641
7843
9732
12603
12342
11753
13696
13175
10639
10430
9067
9404
Mid
4705
3403
3873
4253
4774
4452
6468
6015
6041
9858
9610
9107
9274
6231
6181
Bot
3126
1816
3061
3147
3270
1970
1775
1611
1587
3865
3789
5119
4518
3173
2962
Top
6264
4506
3585
5384
7488
9383
12118
12027
11656
11882
11181
11083
11479
8864
9154
Mod
4395
3089
3526
3997
4418
4103
5988
5728
ssn
8057
7629
9518
10331
6082
5925
Bot
2809
1500
2716
2890
2914
1621
1269
1328
1489
2061
1804
5508
5563
3056
2703
Top
5986
4094
3334
5148
7154
9043
11703
11460
10879
12227
11547
9674
10309
8357
8645
Mod
4124
2672
3272
3759
4094
3785
5626
5215
5177
8382
7985
8340
9137
5574
5456
Bot
2549
1091
2465
2655
2591
1303
915
822
759
2385
2166
4515
4098
2513
2231
Top
5637
3767
3059
4902
6933
8759
11482
11244
11041
13054
12112
10525
10594
8198
8648
Mid
3787
2348
2993
3515
3864
3478
5353
4944
5296
9229
8559
8960
9446
5417
5428
Bot
2226
768
2190
2409
2360
997
633
545
874
3233
2735
4950
4678
2391
2211
Top
5325
3433
2732
4634
6634
8423
11122
10889
10641
11714
11224
9165
9893
7813
8129
Mod
3464
2012
2670
3245
3574
3165
5045
4644
4939
7869
7662
7832
8721
5030
4940
Bot
1888
430
1863
2140
2071
683
334
251
521
1871
1843
4006
3682
1969
1715
Top
4971
3112
2490
4372
6340
8151
10886
10856
10461
11837
10823
9718
9973
7518
7998
Mid
3120
1693
2424
2985
3270
2871
4756
4557
4716
8012
7271
8153
8825
4737
4769
Bot
1560
112
1621
1878
1766
389
37
157
295
2016
1447
4144
4057
1711
1550
Top
4596
2873
2216
4123
6104
7240
9523
9328
8994
11231
10277
8517
8047
6050
7034
Mid
2734
1451
2154
2734
3044
1982
3446
3083
3292
7386
6714
7184
6875
3268
3845
Bot
1159
-130
1347
1629
· 1541
-500
- 1265
- 1310
-1125
1389
896
3358
1836
206
619
Appendix A di splays month hydropower surpl us averages for the years 2012-2021 under vari able water year conditions. Top
represents top 10% of Wate Years, Mid repreents middle 80% of water years, and Bot represents botoom I 0% ofWate Years.
Source: http: "" w.bpa.gO\ 12owcr 12gp whitebook 20 I I WhiteBook2011
*Data is in average megawatts
TcchnicalAnllcndi~
Vol 0 o201 Final.12df
63
Bibliography of Sources:
Abraham, Spencer. The National Gid Transmission Study. U.S. Department of Energy, 2002.
Altinbi lek, Dogan, Karin Seelos, and Richard Taylor. "Hydropower's Role in Delivering
Sustainability." Energy & Environment (Multi Science Publishing) 16, no. 5 (September
2005): 815-824.
Balat, M., and E. Kirtay. "Major Technical Barriers to a "Hydrogen Economy"." Energy Sources
(Taylor and Francis Group, LLC.) A, no. 32 (20 I 0): 863-876.
Bartels, Jeffery R, Michael B. Pate, and Norman K. Olson. "An economic survey of hydrogen
production from conventional and alternative energy sources." International Journal of
Hydrogen Energy, Aug2010: 8371-8384.
Bonneville Power Administration. 2007 Wholesale Power Rate Schedule. Portland, OR:
Bonneville Power Administration, 2007.
Bonneville Power Administration.2012 EPA Power Rates. Jan 24, 2012.
http://www.bpa.gov/power/psp/rates/current.shtml#footnote2 (accessed April 20, 20 12).
Bonneville PowerAdministration. "2011 Loads and Resources Study." Bonneville Power
Administration, Depatment of Energy, 2011.
Contreras, Alfonso, Fausto Posso, and T Nejat Veziroglu. "Modeling and simulating of the
production of hydrogen using hydroelectricity in Venezuela." Journal of Hydrogen
Energy 32, no. 9 (June 2007): 1219-1224.
Fesmire, Bob. "Energy Efficiency in the Power Grid." Renewable Energy World, 2007:
http: //www.renewab leenergyworld.com/rea/news/ article/2007 /07I energy-efficiency-inthe-power-grid-49238.
Flour Daniel, Inc. Pacific Northwest Hydrogen Feasibility Study. Pacific orthwest Hydrogen
Feasibility Study, US Department of Energy Bonneville Power Administration, Irvine,
CA: Flour Daniels, Inc., 1991.
Gutierrez-Martin, F., J.M. Garcia-De Maria, A. Bai"ri, and N. Laraqi. "Management Strategies for
Surplus Electricity Loads Using Electrolytic Hydrogen." International Journal of
Hydrogen Energy 39 (Aug 2009): 8468-8475.
Harrison, K.W., R. Remick, G.D. Martin, and A. Hoskin. Hydrogen Production: Fundamentals
and Case Study Summaries. Conference Paper, US Department of Energy, National
Renewable Energy laboratory, Golden, CO: National Renewable Energy laboratory,
2010.
Hydrogen Society. Achieving Safe Hydrogen Systems. 2012.
http://www.hydrogensociety.net/hydrogen_safety_issues.htm (accessed 4 25, 2012).
Ivy, Johanna. Summary of Electrolytic Hydrogen Production: Milestone Completion Report.
National Rewnewable Energy Laboratory, US Department of Energy, Golden, CO:
Midwest Research Institute, 2004.
64
Kleit, Andrew, and James Reitzes. "The effectiveness ofFERC's transmission policy: is
transmission used efficiently and when is it scarce?" Journal of Regulatory Economics
(Springer Science+Business Media, LLC) 34, no. 1 (Aug 2008): 1-26.
Kroposki, B., J. Levene, K. Harrison, P.K. Sen, and F. Novachek. Electrolysis: Information and
Opportunities for Electric Power Utilities. Technical Rep011, National Renewable Energy
Laboratory, U.S. Department of Energy, Golden, CO: National Renewable Energy
Laboratory, 2006.
Kruger, Paul. "Electric Power Requirement in the United States for Large-sclae Production of
Hydrogen fuel." International Journal of Hydrogen Energy 25, no. 11 (2000): 10231033.
Levene, J. I., B. Kroposki, and G. Sverdrup. Wind Energy and the Production of Hydrogen and
Electricity- Opportunities for Renewable Hydrogen. Conference Paper, National
Renewable Energy Laboratory, U.S. Deparment of Energy, Golden, CO: National
Renewable Energy Laboratory, 2006.
Ouellette, N, H. H. Rogner, and D. S. Scott. "Hydrogen-based industry from remote excess
hydroelectricity." International Journal of Hydrogen Energy 22, no. 4 (April 1997): 397403.
Pratt, J ., W. Butcher, L. Steinmann, C. Blazek, and E. Daniels. Feasibility Assessment of
Electrrolytic Hydrogen Production in the Pacific Northwest. Portland, OR: Bonneville
Power Administration, 1984.
Rodgers, Steven F, Shravan K Vudumu, Scott E. Grasman, Susan L Mun-y, and Umit 0 Koylu.
"Hydrogen Safety: A Focus on Power Generating Applications." Professional Safety 55,
no. 9 (Sept 201 0): 39-44.
Sickinger, Ted. "BPA curtails wind farm electricity generation." The Oregonian, May 18, 2011.
Tarkay, D.S. "Hydrogen Production at Hydro-Power Plants." The International Journal of
Hydrogen Energy 10, no. 9 (Feb 1985): 577-584.
Tomczyk, Piotr. "Fundamental Aspects of the Hydrogen Economy." World Futures (Taylor &
Francis Group, LLC.) 65 (2009): 427-435.
United States Bureau Of Labor and Statistics. CPI Inflation Calculator. 2012.
http: //www. bls.gov/data/inflation_ ca1culator.htm.
United States Energy lnfonnation Administration. Refiner Gasoline Prices by Grade and Sales
Type. 5 15,2012. bttp://205.254.135.7/dnav/pet/pet_pri_refmg_dcu_nus_m.htm.
Wang, Mingyong, Zhi Wang, and Zhancheng Guo. "Water electrolysis enhanced by super gravity
field for hydrogen production." International Journal of Hydrogen Energy 35 (20 I 0):
3198-3205.
65