Making Hydrogen with Nuclear Energy for Liquid Fuels
C. O. Bolthrunis, R. W. Kuhr, R. Hannink, R. Bhoomi,1 J. McKinnell2
«Le mieux est l'ennemi du bien.» -- Voltaire (La Bégueule, 1772)
1. Introduction
Although we may all agree that a carbon-free hydrogen economy is much to be
desired and may be necessary sooner than we think, we do not have the luxury of
allowing the perfect to be the enemy of the good. Like it or not, carbon-based
fuels will be with us for a long time. The huge world deposits of coal and tar
sands will not remain in the ground. It is up to us therefore, to reduce the carbon
footprint of carbon-based fuels as much as possible. This means reducing the
carbon emissions in fuel processing and adding as much carbon-free hydrogen to
those fuels as possible. The most likely source of carbon-free, high temperature
process heat and electricity necessary for hydrogen production in large-scale,
centralized installations required for fuel processing is nuclear energy, specifically
from high-temperature gas-cooled reactors (HTGRs). In this paper we will
explore at a high level how such high-temperature process heat and electricity
might be used in making gasoline from coal and other carbon-intensive fuels. We
will also explore the environmental impact and especially the economics of these
technologies in the light of the run-up in crude prices and a potential carbon tax.
2. Enabling Technologies
To be able to make liquid hydrocarbon fuels such as gasoline from coal using
hydrogen derived from nuclear energy, several enabling technologies are
required. An HTGR that is safe, licensable and can provide process heat at about
900ºC is needed. Just such a reactor, the Pebble Bed Modular Reactor (PBMR) is
approaching commercialization. The first commercial size PBMR is expected to
be constructed in Koeberg, South Africa and be ready for fuel loading by about
2014. Figure 1 shows the PBMR configured to provide process heat. Highpressure helium is circulated through the reactor to deliver heat to the
intermediate heat exchangers (IHX). A second helium loop supplies heat to the
process. The PBMR is passively safe because of the design of the fuel itself.
Reactor containment is unnecessary. The fuel pebbles are continuously removed
from the reactor, automatically checked for integrity and burn-up and returned to
the reactor. Refueling is done on line. A diagram of the fuel is shown in Figure 2.
1
Shaw Energy & Chemicals, One Main Street, Cambridge, MA 02142;
[email protected]
2
Pebble Bed Modular Reactor (Pty) Ltd., Pebble House, 1279 Mike Crawford Avenue, P.O. Box
9396, Centurion, 0046, South Africa
1
Figure 1 The PBMR configured to deliver high temperature process heat
Figure 2 Structure of the PBMR pebble fuel
2
One must also make carbon-free hydrogen with this energy. For that, three
leading water-splitting technologies are currently being developed under the
auspices of the United States Department of Energy’s Nuclear Hydrogen Initiative
(NHI). The technologies include high-temperature steam electrolysis, the sulfur
iodine cycle and the hybrid sulfur cycle. The technology favored by the authors is
the hybrid sulfur cycle which is shown in Figure 3, below. This cycle uses both
electricity and a large amount of high temperature process heat, both of which can
be generated by a HTGR.
DC Electricity
O2
H2O
Oxygen
Recovery
SO2
H2 O
O2
Oxygen
Generation
H2SO4  SO2 + H2O +½O2
H2
SO2
H2O
Hydrogen
Generation
(Electrolyzer)
SO2 + 2H2O  H2SO4 + H2
H2SO4
Heat
Sulfuric Acid
Vaporization
Heat
Figure 3 The Hybrid Sulfur Cycle
To make full use of the high temperature process heat in making liquid fuels from
coal, nuclear-powered steam reforming of methane is also needed. This
technology has been investigated by groups in Japan and in Germany over several
decades, but an applicable technology has not yet been commercialized.
The final major building block is making liquid hydrocarbon fuels from hydrogen
and carbon-rich feedstocks, especially coal. Several processes that add hydrogen
directly to coal to make liquid fuels were investigated over twenty years ago.
Major pilot plants were built and operated in the United States, but none of these
technologies was commercialized. The only coal conversion technology that is
currently commercial makes liquid fuels indirectly by gasifying coal to syngas, a
mixture of hydrogen and carbon monoxide. It then uses Fischer-Tropsch
3
chemistry to synthesize liquid hydrocarbons from which finished fuels are made.
A similar process is in use to convert natural gas to liquid fuels.
3. Coal- and Gas-to-Liquids
The ratio of hydrogen to carbon atoms for coal is less than 1:1. For liquid fuels,
this ratio is nearly 2:1. This means that to make liquid fuels from coal, the amount
of hydrogen in the coal must be approximately doubled. For ever atom of carbon
in the coal, an atom of hydrogen must be found. In a process in which coal is the
only feedstock, this hydrogen is generated by means of coal gasification and the
so called “water gas shift.” The reactions are as follows:
CH (coal) + H2O  1½ H2 + CO (steam gasification)
CO + H2O  H2 + CO2 (water gas shift)
These reactions generate five atoms of hydrogen and a molecule of carbon
dioxide for every atom of carbon gasified and “shifted.” In terms of mass, that
means almost 9 pounds of carbon dioxide for every pound of hydrogen needed.
This means that about 20% extra coal must be gasified and “shifted” to provide
the required hydrogen. The water gas shift is carried out only sufficiently to
provide enough hydrogen for the Fischer-Tropsch reaction. The steam gasification
reaction is highly endothermic and additional coal must be oxidized to generate
the necessary heat. Hence, a large amount of oxygen is required for this partial
oxidation. The net result is that about twice as much carbon in the form of coal
enters the gasifier than exits the process as liquid hydrocarbon fuel. The
difference is rejected as carbon dioxide. The flow scheme for this process is
shown in Figure 4, below.
Air
Separation
Air
Nitrogen
CO2
Oxygen
Coal
Gasification
Steam
Water Gas
Shift
Fischer Tropsch
Reaction
Acid Gas
Removal
Product
Recovery
Products
Secondary
Acid Gas
Removal
CO2
Syngas
Natural
Gas
Reforming
Figure 4 Flow Scheme for Coal- and Gas-to-Liquids
4
The Fischer-Tropsch reaction produces a large quantity of methane as well as
hydrocarbon liquids. To recover the hydrogen from unwanted gas, this methane
is “reformed” in a reaction similar to the steam gasification reaction.
CH4 + H2O  CO + 3H2 (steam reforming)
Heat for this reaction is usually provided by partially oxidizing the methane, but
the heat may be added externally. Partial oxidation generates additional carbon
dioxide. Reforming of methane produces twice as much hydrogen as the steam
gasification of coal. Thus, by adding methane to the reforming step (shown by
the dashed line in Figure 4), shifting gas from the gasifier can be avoided and the
gasification load reduced.
Why anyone would want to go through this trouble and inefficiency to transform
coal and gas into liquid fuels becomes obvious when one examines Table 1. The
fuel prices listed are recent prices taken from the DOE Energy Information
Administration (EIA) website [1, 2, 3]. Note that on the basis of energy content,
gasoline is more than three times more costly than natural gas and almost 15 times
more costly than coal. This is a large incentive to process coal and natural gas
into transportation fuels. The carbon dioxide emissions listed are emissions
produced upon complete combustion which must be distinguished from
processing emissions. The price of hydrogen listed is consistent with producing it
from natural gas by means of conventional steam methane reforming.
Table 1 Comparison of Fuels
Current Price
Fuel
Coal, Bituminous
Gasoline
NG
Hydrogen
($/gal
/ton or
/MCF)
Heating
Value
Cost of
Energy
BTU/lb
(LHV)
$/ million
BTU
12,603
18,696
20,267
51,623
$ 1.98
$ 28.77
$ 9.11
$ 19.13
$/lb
$ 50.00 $ 0.025
$ 3.33 $ 0.538
$ 8.45 $ 0.185
$ 5.24 $ 0.987
Carbon Dioxide
Emissions
lbs CO2/
lb fuel
2.95
3.16
2.74
-
lbs CO2/
MM BTU
234
169
135
-
Price of hydrogen is consistent with price of NG
4. Nuclear Coal- and Gas-to-Liquids
Nuclear energy can benefit the coal- and gas-to-liquids processes in ways that are
shown in Figure 5. High-temperature process heat and electricity from a HTGR
can be used to split water to provide both hydrogen and oxygen to the process and
to provide heat for steam reforming of recycle hydrocarbons and externally
supplied natural gas.
5
Coal
Fischer Tropsch
Reaction
Gasification
Steam
Product
Recovery
Products
Syngas
Oxygen
Hydrogen
Water
Water
Splitting
Steam
Reforming
Recycle
Hydrocarbons
Energy
from PBMR
Natural
Gas
Figure 5 Nuclear Coal- and Gas-to-Liquids
The benefits of such an external source of high-temperature process heat and
hydrogen are manifold. First and foremost, by providing adequate hydrogen to the
process, the gasification and shifting of coal providing needed hydrogen can be
avoided. Moreover, carbon oxidized to provide heat in the gasifier can be
hydrogenated so that there are no carbon dioxide emissions produced in
processing. By supplying external heat to the reforming step, partial oxidation of
natural gas is unnecessary and the carbon dioxide emissions so produced can also
be eliminated.
Air
Separation
Air
Nitrogen
CO2
Oxygen
Coal
Gasification
Steam
Water Gas
Shift
Fischer Tropsch
Reaction
Acid Gas
Removal
Product
Recovery
Products
Syngas
Natural
Gas
Reforming
Secondary
Acid Gas
Removal
CO2
Figure 6 Benefits of Nuclear Energy to Coal- and Gas-to-Liquids
In the limiting case, enough natural gas can be introduced so as to eliminate the
need for an external source of hydrogen. The air separation unit, however, would
still be required in this case to provide heat for the gasifiers. The result is that the
blocks and streams shown shaded darkly in
6
Figure 6 are eliminated and the blocks more lightly shaded are reduced in size by
almost half.
5. Economic Incentive for Nuclear Process Heat
The main premise of this analysis is that the pump price of gasoline will be set by
the price of crude oil and the costs associated with processing and distribution.
Information on the fractional contributions of these costs for the years 2004 and
2005 is available from the EIA [4]. Using this information, an estimate was made
of the price of gasoline at $100 per barrel in 2008. No adjustment was made for
inflation. Additional data from the EIA and from other sources were used to
estimate the hydrogen consumption [5,6] and carbon dioxide emissions [7]
associated with producing a gallon of gasoline. The results of these estimates are
shown in Table 2.
Table 2 Contribution of Various Costs to the Pump Price of Gasoline
Distribution & marketing
Refining costs & profits
Taxes
Crude
Gasoline price / gal
12%
18%
23%
47%
Crude price
CO2 emitted in making petroleum products
Hydrogen requirement for making products
2004
$
$
$
$
$
0.222
0.333
0.426
0.870
1.85
$
36.98
9%
19%
19%
53%
2005
$
$
$
$
$
0.204
0.431
0.431
1.203
2.27
$
50.23
2.303 lb/gal
0.0303 lb/gal
6%
20%
11%
63%
2008
$
$
$
$
$
0.222
0.776
0.426
2.373
3.80
$100
calculated from eia data
calculated from eia data
7
$5.00
Gasoline Price
$4.50
$4.00
$3.50
$3.00
$2.50
$2.00
$50
$60
$70
$80
$90
$100
$110
$120
Crude price
Figure 7 Estimated Relationship between the Price of Crude Oil and the
Pump Price of Gasoline
From these estimates, relationships were developed for the price of petroleumbased gasoline based on the effect of the price of crude oil and on a possible
carbon dioxide emissions penalty. This penalty could be in the form of a tax, cost
of capture and sequestration, or other mechanism. These relationships are shown
in Figure 7 and Figure 8, respectively.
8
Pump Price of Gasoline
$4.00
$3.90
$3.80
$3.70
$30
$40
$50
$60
$70
$80
$90
$100
Carbon Dioxide Tax, ($/ton)
Figure 8 Effect of a Carbon Dioxide Tax on the Pump Price of Gasoline
It is clear that the price of gasoline is a very strong function of the price of crude
oil, but is affected only slightly by a carbon penalty on processing emissions. We
assumed that taxes, distribution and marketing costs were inflexible. Refining
costs are somewhat dependent upon crude price because the cost of refining
energy is partially tied to crude cost.
The market price of gasoline was thus set by crude price and carbon dioxide
penalty. By using the relationships described in Section 3 , above, but using more
precise hydrogen-to-carbon ratios based on the actual composition of the various
fuels, and current fuel prices as shown in Table 1, pump price contributions for
the feedstocks, processing fuel costs, distribution and taxes, and carbon dioxide
penalty were calculated for four cases. Distribution and taxes were kept equal to
those estimated for petroleum-based gasoline in all cases. The four cases are coalto-liquids (CTL) using only coal feedstock for both carbon and hydrogen; coaland gas-to-liquids (NG+CTL) using an ideal balance of coal and natural gas (NG)
feed; CTL using nuclear energy to split water to provide enough hydrogen and
process heat to eliminate all carbon dioxide emissions (NCTL); and NG+CTL
using an ideal balance of NG and coal and enough nuclear process heat for steam
methane reforming (SMR) to eliminate all carbon dioxide emissions
(NNG+CTL). In the cases using NG, coal is accounted for as carbon feedstock
and NG is accounted for as hydrogen feedstock. The cost of hydrogen made by
splitting water with nuclear energy was not estimated. This was used as a
parameter in the analysis.
The calculated pump price contributions for each case were subtracted from the
market price for gasoline. The remainder is the processing costs and profit for
9
each case. This represents the amount of money available per gallon of gasoline to
pay for the capital and operating charges of the CTL or NG+CTL plant as well as
profit. Actual capital and operating costs were not estimated. All nuclear
hydrogen capital and operating costs are included in the cost of this hydrogen.
Thus the cost of the nuclear reactors in the NCTL case is included in the cost of
hydrogen, whereas in the NNG+CTL case, it is in the processing costs and profits.
These cases are compared in Figure 9 for a crude price of $100 per barrel, a
carbon dioxide penalty of $50 per ton, and a nuclear hydrogen price of $5 per
kilogram. The market price for gasoline, including carbon dioxide penalty, for
this scenario is $3.85 per gallon. This price is only indicative and is based upon
an extrapolation from approximate data. “Processing Costs and Profits” are also
designated as “Processing Margin” in subsequent figures.
100%
Pump price contribution
80%
60%
40%
20%
0%
Gasoline from
Petroleum
Gasoline from Coal Gasoline from Coal Gasoline from Coal Gasoline from Coal
and NG
and Nuclear Water and NG w. Nuclear
Splitting
SMR
Processing Method
Carbon Feedstock
Hydrogen feedstock
Fuel for gasification & reforming
Distribution & Taxes
Carbon Dioxide Penalty
Processing costs & profit
Figure 9 Comparison of Gasoline Pump Price Contributions
The important point to be gained from this analysis is that the margin for
processing for any CTL or NG+CTL process is very large due to the high price of
crude oil. It is likely, for example, that a CTL plant is feasible at crude prices
well below $100 per barrel. This is illustrated in Figure 10 which is also based on
a carbon dioxide penalty of $50 per ton.
10
$3.50
CTL Processing Margin
$3.00
$2.50
$2.00
$1.50
$1.00
$70
$80
$90
$100
$110
$120
Crude Price
Figure 10 Processing Margin for CTL as a function of Crude Oil Price
As can be seen in Figure 11, the processing margin for these technologies is not
affected nearly as much by carbon dioxide penalties as it is by crude price. It is
primarily a high crude oil price that opens an opportunity for all of these
technologies. The carbon dioxide penalty affects only the relative
competitiveness of nuclear and non-nuclear options.
NG + CTL Processing Margin
$2.10
$2.05
$2.00
$1.95
$1.90
$1.85
$1.80
$20
$30
$40
$50
$60
$70
$80
$90
$100
$110
Carbon Dioxide Penalty ($/ton)
Figure 11 Processing Margin for NG+CTL as a function of Carbon Dioxide
Penalty
11
A comparison of the CTL and NCTL cases in Figure 9 can appear discouraging to
the nuclear water splitting option until one remembers the message of Figure 6.
Roughly 40 % of the CTL processing costs can be eliminated by importing
hydrogen and oxygen from a water splitting plant. From Figure 12 it is clear that
at a carbon penalty of $50 per ton, nuclear hydrogen for a CTL plant can become
competitive at a cost of around $4.50 a kilogram.
Processing Margin
$2.50
$2.00
$1.50
$1.00
$0.50
$1.00
$2.00
$3.00
$4.00
$5.00
$6.00
$7.00
Cost of Nuclear Hydrogen
NCTL
CTL
60% of CTL
Figure 12 Processing Margins for Nuclear CTL as a function of Hydrogen
Cost
The case for the NNG+CTL case is also strong. Although the air separation unit
is not eliminated, its size and the size of the gasifying unit are greatly reduced in
comparison with the non-nuclear option. Higher carbon dioxide penalties will
only increase the competitiveness of the nuclear options.
The high price of crude oil is making all CTL and NG+CTL processes
competitive. If current prices remain in place, there appears to be sufficient
economic margin to consider using nuclear process heat in these applications.
The competitiveness of these nuclear options with respect to commercial CTL and
NG+CTL processes will, of course, depend upon the cost of building and
operating nuclear process heat and water splitting facilities as well as the level of
carbon dioxide penalties. The estimated hurdles that HTGR and water splitting
technologies have to clear to be competitive appear to be well within the range of
current targets and estimates.
12
6. References
1.
“Table 34. Average Price of Coal Delivered to End User Sector by Census Division and
State,” EIA – Coal Data, Reports, Analysis and Surveys, Energy Information
Administration, Official Energy Statistics from the U.S. Government, 2006, March 21,
2008, http://www.eia.doe.gov/cneaf/coal/page/acr/table34.html.
2.
“EIA Natural Gas Weekly Update, Overview (Wednesday, March 12 to Wednesday,
March 19), EIA – Natural Gas Price Data and Analysis, Energy Information
Administration, Official Energy Statistics from the U.S. Government, March 20, 2008,
March 21, 2008 http://tonto.eia.doe.gov/oog/info/ngw/ngupdate.asp..
3.
“Weekly Retail Gasoline and Diesel Prices,” EIA – Petroleum Data, Reports, Analysis,
Surveys, Energy Information Administration, Official Energy Statistics from the U.S.
Government, March 17, 2008, March 21, 2008,
http://tonto.eia.doe.gov/dnav/pet/pet_pri_gnd_dcus_nus_w.htm.
4.
“A Primer on Gasoline Prices,” EIA brochure Reports, Analyses, DOE/EIA-X040, May
2006, Energy Information Administration, Official Energy Statistics from the U.S.
Government, 4pp., March 20, 2008,
http://www.eia.doe.gov/bookshelf/brochures/gasolinepricesprimer/eia1_2005primerM.ht
ml.
5.
Hydrogen Production in Refineries, PERP Report 06/07S1, Nexant: Chem Systems,
White Plains, NY, October 2007, p 59.
6.
“Refinery & Blender Net Production,” EIA – Petroleum Refining and Processing Data
and Analyses, Energy Information Administration, Official Energy Statistics from the
U.S. Government, February 29, 2008, March 20, 2008,
http://tonto.eia.doe.gov/dnav/pet/pet_pnp_refp_dc_nus_mbbl_m.htm
7.
Schipper, Mark, “Energy-Related Carbon Dioxide Emissions in U.S. Manufacturing,”
EIA – Energy Emissions Data & Environmental Analysis , Special Topic Report #:
DOE/EIA-0573(2005), November 2006, 16 pp., Table 2, p.5, March 20,2008
http://www.eia.doe.gov/oiaf/1605/ggrpt/pdf/industry_mecs.pdf.
13