Beruflich Dokumente
Kultur Dokumente
www.elsevier.com/locate/ijhydene
Abstract
The use of hydrogen as an alternative fuel is gaining more and more acceptance as the environmental impact of hydrocarbons
becomes more evident. A life cycle assessment study has been carried out to investigate the environmental aspects of hydrogen
production. Production by natural gas steam reforming and production upon renewable energy sources are examined. Hydrogen
is selected as a future alternative fuel because of the absence of CO2 emissions from its use, its high-energy content and
its combustion kinetics. A very large number of environmental burdens result from the operation of the di7erent hydrogen
production routes. A complete and accurate identi8cation and quanti8cation of the environmental emissions has been attempted.
The use of wind, hydropower and solar thermal energy for the production of hydrogen are the most environmental benign
methods. The bene8ts and the drawbacks of the competing hydrogen production systems are presented.
? 2004 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.
0360-3199/$ 30.00 ? 2004 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhydene.2004.01.016
1444 C. Koroneos et al. / International Journal of Hydrogen Energy 29 (2004) 1443 1450
Storage
Hydrogen Transport Hydrogen Life Cycle Assessment framework
Primary
Renewable Hydrogen Hydrogen Useful Goal Definition and
Oxygen
Energy production
Oxygen
utilization Energy Scoping
Source
Water Water
Environment
Fig. 1. The life cycle of hydrogen from RES. Inventory Analysis Interpretation
H 2 product slipstream
steam
Natural gas
feed stock Catalytic High Low Pressure
ZnO
Hydrogenation steam temperature temperature swing H2
Bed
reforming shift shift adsorption
Natural offgas
gas fuel
can also be determined. Once data are assembled, the inven- Table 1
tory items are added up to provide a total pro8le for each Average air emissions from H2 production by natural gas steam
option. In some LCAs, the inventory is the 8nal product. reforming [8]
However, even though it is di<cult to do an impact analysis Air emission System total (g=kg H2 )
(the 8nal step in the LCA methodology), the inventory can
provide useful information to aid decision makers. Benzene (C6 H6 ) 1.4
Carbon dioxide (CO2 ) 10662.1
Carbon monoxide (CO) 5.9
3. Hydrogen production by natural gas steam reforming Methane (CH4 ) 146.3
Nitrogen oxides (NOx as NO2 ) 12.6
Nitrous oxide (N2 O) 0.04
Steam reforming is at present (and very likely will be in
Non-methane hydrocarbons (NMHCs) 26.3
the future) one of the most important and most economic
Particulates 2.0
way of hydrogen production. In this context, it is of crucial Sulphur oxides (SOx as SO2 ) 9.7
importance that steam reforming induces the least CO2 emis-
sion of all industrial scale processes available at present.
During steam reforming hydrocarbons are catalytically
split in the presence of steam at temperatures of 800900 C
(Fig. 3) [8]. Normally, the split is proceeded with nickel 4. Hydrogen production based upon renewable energy
catalyst in gas-8red ovens. Mostly natural gas is used as
feed but heavier hydrocarbons up to naphtha can also be As stated earlier, about 97% of the worldwide hydrogen
processed. During the catalytic split the so-called syngas production is accomplished by steam reforming of natural
is produced that mainly consists of hydrogen and carbon gas and other fossil primary energy. However, a number of
monoxide. The basic equation is innovative production paths exist for hydrogen production
based upon renewable energy and some of them have been
Cn Hm + nH2 O nCO + (n + m=2)H2 :
assessed in this study by carrying out an LCA of the techno-
Apart from this basic reaction other reactions take place logical systems. The investigated process chains start with
where CO2 and soot are already produced. In the following the extraction of the primary energy carrier, the transporta-
step (the so-called shift-reaction) carbon monoxide from the tion to the hydrogen production plant, the conversion into
syngas is transferred according to the equation hydrogen and the liquefaction before the 8nal use (Fig. 4).
CO + H2 O CO2 + H2 The following renewable energy sources were examined:
into carbon dioxide and hydrogen. The reaction is catalyzed 1. Solar energy using photovoltaics for direct conversion.
using iron oxide. During the terminating puri8cation, the 2. Solar thermal energy.
hydrogen is separated from the product gas. Today, pres- 3. Wind power.
sure swing adsorption (PSA) is the prevailing process. 4. Hydro power.
The remaining product gas is piped back and used as fuel 5. Biomass.
to 8re the steam-reforming reactor. After the fuel gas has
passed several heat exchangers, it is 8nally released into the The comparative assessment of the di7erent hydrogen pro-
atmosphere. duction scenarios was made with the use of the Global
Table 1 is a list of the major air emissions that result from Emission Model for Integrated Systems (GEMIS) database.
the production of H2 by natural gas steam reforming that GEMIS was developed by the Oto-Istitute (Istitute of Ap-
were used for the purpose of this study [8]. plied Ecology) in Germany [9].
1446 C. Koroneos et al. / International Journal of Hydrogen Energy 29 (2004) 1443 1450
Hydrogen Production
by Electrolysis
Energy production upon
Renewable Energy
Energy Liquid Hydrogen Use of
Sources
Transport Transportation Liquid Hydrogen
(Solar, Wind, Hydropower,
Solar thermal, Biomass)
Hydrogen
Liquefaction
Electrolysis is often considered as the preferred method Fig. 5. Typical electrolysis cell.
of hydrogen production as it is the only process that need
not rely on fossil fuels. It also has high product purity, and hydrogen would build up on the electrode and block current
is feasible on small and large scales. Kow.
At the heart of electrolysis is an electrolyzer, consisting A gas separator, or diaphragm, is used to prevent inter-
of a series of cells each with a positive and negative elec- mixing of the hydrogen and oxygen molecules although it
trode (Fig. 5). The electrodes are immersed in water that allows free passage of ions.
has been made electrically conductive, by adding hydrogen The reaction at the cathode are:
or hydroxyl ions, usually in the form of alkaline potassium
hydroxide (KOH). (1) K + + e K A positively charged potassium
The anode (positive electrode) is typically made of nickel ion is reduced.
and copper and is coated with oxides of metals such as man- (2) K + H2 O K + The ion reacts with water to form
ganese, tungsten, and ruthenium. The anode metals allow + H + OH a hydrogen atom and a hydroxyl
quick pairing of atomic oxygen into oxygen pairs at the elec- ion.
trode surface. (3) H + H H2 The highly reactive hydrogen
The cathode (negative electrode) is typically made of atom then bonds to the metal of
nickel, coated with small quantities of platinum as a cata- the cathode and combines with
lyst. The catalyst allows quick pairing of atomic hydrogen another bound hydrogen atom to
into pairs at the electrode surface and thereby increases the form a hydrogen molecule that
rate of hydrogen production. Without the catalyst, atomic leaves the cathode as a gas.
C. Koroneos et al. / International Journal of Hydrogen Energy 29 (2004) 1443 1450 1447
The reactions at the anode are: ment of 0:347 MJ=MJ (0:00244MJ=kg H2 ) is given in the
GEMIS database (refer to 30 bar inKow pressure).
(1) OH OH + e A negatively charged hydroxyl
ion is oxidized.
(2) OH 12 H2 O+ 12 O The ion reacts to form water and 5. Comparative assessment of hydrogen fuel production
an oxygen atom.
(3) O + O O2 The highly reactive oxygen atom During the previous part of the study, the inventory of
then bonds to the metal of the an- di7erent fuel production processes was presented. Hydro-
ode and combines with another gen production from conventional and renewable sources
bound oxygen atom to form an was thoroughly analyzed. The next step of the study is the
oxygen molecule that leaves the impact assessment, to see how the speci8c substances af-
anode as a gas. fect the environment. The impact assessment evaluates the
magnitude and signi8cance of the potential environmental
The rate of hydrogen generation is related to the current den- impacts of the di7erent life cycles under study. It consists
sity (the current Kow divided by the electrode area, measured of three steps: classi8cation, characterization and valuation
in ampere per meter square). In general, the higher the cur- [12]. The categories that have been examined in our study
rent density, the higher the source voltage required and the are four: global warming potential (GWP), acidi8cation ef-
power cost per unit of hydrogen. However, higher voltages fect, eutrophication e7ect and winter smog e7ect. The rea-
decrease the overall size of the electrolyzer and therefore re- son for this is based on the nature of the data collected and
sult in a lower capital cost. State-of-the-art electrolyzers are the importance of these impact categories.
reliable, have energy e<ciencies of 6580%, and operate at
current densities of about 186 A=ft2 (2000 A=m2 ). 5.1. Greenhouse gases emissions
For electrolysis, the amount of electrical energy required
can be somewhat o7set by adding heat energy to the reac- Although CO2 is the most important greenhouse gas and
tion. The minimum amount of voltage required to decom- is the largest emission from this system, quantifying the
pose water is 1:23 V at 77 F (25 C). At this voltage, the total amount of greenhouse gases produced is the key to
reaction requires heat energy from the outside to proceed. examining the GWP of the di7erent systems (Fig. 6). The
At 1:47 V (25 C) no input heat is required. At higher volt- GWP is a combination of CO2 ; CH4 , and N2 O emissions.
ages (and same temperature), heat is released into the sur- The GWP can be normalized to CO2 equivalent emissions to
roundings during water decomposition. describe the overall contribution to global climate change.
Operating the electrolyzer at lower voltages with added As shown from the 8gure, the variation of CO2 eq. emissions
heat is advantageous, as heat energy is usually less costly of di7erent processes is quite large. H2 from natural gas has
than electricity, and can be recirculated within the process. by far the larger emissions.
Furthermore, the e<ciency of the electrolysis increases
with increased operating temperature. For the electrolytic 5.2. Acidi6cation emissions
hydrogen production, the thermodynamic losses are mainly
due to irreversibilities associated with heat production from Acidi8cation is measured as the amount of protons re-
high-quality energy resources (fossil fuels), electricity leased into the atmosphere. The weighting factors are pre-
generation and water splitting [11]. sented either as mol H+ or as kg of SOx equivalent. The
two types of compound mainly involved in acidi8cation are
4.2. Liquefaction process
0.06
pression followed by adiabatic expansion, whereby the gas
0.05
cools down due to the JouleThomson e7ect. A quantity of
0.04
0:97 kWh=kg heat, a condensation enthalpy of 0:13 kW=kg 0.03
and an energy release of 0:2 kW=kg due to the Ortho 0.02
Para-conversion has to be withdrawn for liquefaction of 0.01
hydrogen. The theoretical minimum energy requirement is 0
H2 from
H2 from
H2 from NG
Hydropower
H2 from PV
H2 from
Biomass
Wind
0.00035 0.00025
0.0003
0.0002
SO4 eq. [kg/MJ]
0.0002 0.00015
0.00015
0.0001
0.0001
0.00005 0.00005
0
H2 from 0
H2 from PV
H2 from
Hydropower
H2 from NG
SolarTh
H2 from
H2 from
Wind
Biomass
H2 from
H2 from PV
H2 from
Hydropower
H2 from NG
SolarTh
H2 from
H2 from
Wind
Biomass
Fig. 7. SO4 equivalent emissions during hydrogen production.
Fig. 9. SPM equivalent emissions of hydrogen production.
3.00E-05 Table 2
Eco-indicator 95 normalization and evaluation factors [12]
2.50E-05
2.00E-05
H2 from NG
H2 from PV
H2 from
Hydropower
SolarTh
H2 from
H2 from
Wind
Biomass
sulphur and nitrogen compounds. Chemicals like ammonia, 5.4. Winter smog e;ect emissions
HF, HCl, and NOx contribute to this impact category. SO2
and SOx emissions are considered to have the same e7ect in For evaluating winter smog, the winter smog potentials
this impact category (Fig. 7). In this category H2 from PV (WSP) are used for converting the di7erent chemical emis-
has the highest SO4 eq. emissions. sions (dust, SO2 ) to an equivalent basis. In this case, solid
particulate matter (SPM) is used as the equivalent chemi-
5.3. Eutrophication air emissions cal compound. Fig. 9 displays the equivalent emissions of
SPM during the production of hydrogen. The production of
Nitrogen and phosphorus are essential nutrients for the H2 from photovoltaics is shown to have the highest SPM
regulation of ecosystems. Enrichment (or eutrophication) of eq. emissions and this is due to primarily to the production
water and soil with these nutrients may cause an undesirable stage of PVs.
shift in the composition of species within the ecosystems.
Eutrophication of terrestrial ecosystems is mainly due to
(long distance transport of) atmospheric emissions of NOx 6. Normalization and evaluation
(nature areas) and emissions to soil of nitrogen and phos-
phorus (agricultural areas). Normalization is de8ned as an optional element relating
Nutriphication potentials are available for all important all impact scores of a functional unit to the impact scores of
eutrophying compounds. It is important to note that there a reference situation. The aim of normalization is to relate
are available nutriphication potentials for compounds to air the environmental burden of a product to the burden in its
and to water. For the purposes of this project only the emis- surroundings.
sions which are released to air are studied (Fig. 8). H2 from In this study, the Eco-indicator 95 weighting method for
biomass has the highest value of PO4 eq. emissions due environmental e7ects that damage ecosystems or human
to the fact that biomass combustion results in high NOx health on a European scale is used. The calculation of
emissions. normalization values have been carried out using the data
C. Koroneos et al. / International Journal of Hydrogen Energy 29 (2004) 1443 1450 1449
100%
14% 10% 13%
90% 19%
23%
28% 8% 5%
8%
80%
10%
6%
70% 6%
35%
39%
60% w-smog (air)
42%
50% Eutrophication (air)
54% 44% Acidification (air)
40% 59%
Greenhouse Emissions (air)
30%
48%
20% 43%
36%
10% 22%
17%
12%
0%
H2 from P V S
H2 from H2 from Wind H2 from r
H2 from H2 from NG
So arTh Hydropower Biomass
H2 from NG
H2 from Biomass
H2 from Hydropower
H2 from Wind
H2 from SolarTh
H2 from PV
on resource extraction and emissions, that were collected Table 2 presents the normalization and evaluation weighting
previously in a normalization study carried out for the factors used for the purpose of this study.
Dutch ministry of transport and public works and the Dutch Finally, the evaluation scores are added up to give a total
ministry of Housing, Spatial planning and the Environment impact for each material and process in the assembly. The
(Blonk et al., 1997). These normalization values were indicator graph is showing the total impact scores of all
mostly based on environmental interventions resulting from the hydrogen production paths (Figs. 10 and 11).
European production in 19901994 [12].
Normalization only reveals which e7ects are large and
which e7ects are small, in relative terms. It says nothing of 7. Conclusions
the relative importance of these e7ects. Evaluation factors
are used for this purpose. Weighting factors have been ap- Although hydrogen is generally considered to be a clean
plied in order to scale the seriousness of the results, mea- fuel, it is important to recognize that its method of produc-
sured in indicator points. The standard eco-indicators can be tion plays a very signi8cant role in the level of environ-
regarded as dimensionless 8gures. The absolute value of the mental impacts. Examining the inputs and outputs from the
points is not very relevant as the main purpose is to com- life cycle of di7erent production paths gives a complete pic-
pare relative di7erences between hydrogen production pro- ture of the environmental burdens associated with hydrogen
cesses. The scale is chosen in such a way that the value of production.
one point is representative for one thousandth of the yearly The LCA of the hydrogen systems indicates that the route
environmental load of one average European inhabitant. of production with the use of photovoltaic energy has the
1450 C. Koroneos et al. / International Journal of Hydrogen Energy 29 (2004) 1443 1450