Renewable Hydrogen Production a Technical Evaluation Based on Prodess Simulation

lable at ScienceDirect

Journal of Cleaner Production 18 (2010) S51eS62

Contents lists avai

Journal of Cleaner Production

journal homepage: www.elsevier .com/locate/ jc lepro

Renewable hydrogen production: a technical evaluation based onprocess simulation

Angela Miltner*, Walter Wukovits, Tobias Pröll, Anton FriedlVienna University of Technology, Institute of Chemical Engineering, Getreidemarkt 9/166-2, 1060 Vienna, Austria

a r t i c l e i n f o

Article history:Received 4 December 2009Received in revised form27 May 2010Accepted 28 May 2010Available online 10 June 2010

Keywords:HydrogenBiomassRenewable energyProcess simulation

* Corresponding author. Tel.: þ43 1 58801 15923; fE-mail address: [email protected] (A. M

0959-6526/$ e see front matter � 2010 Elsevier Ltd.doi:10.1016/j.jclepro.2010.05.024

a b s t r a c t

An evaluation of different hydrogen production technologies based on renewable raw materials and/orrenewable energy is presented. The evaluation comprises alkaline electrolysis, steam reforming of bothbiogas and gasification gas, the coupled dark and photo fermentation as well as the coupled dark andbiogas fermentation. Each technology is investigated with different plant layouts and/or different rawmaterials. All examined technologies are designed to produce hydrogen in a quality suitable for the use inmobile fuel cells. The presented evaluation is based on the hydrogen production efficiency and theenergy efficiency of the processes.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

The debate on global warming and the foreseeable depletion offossil reserves have led to worldwide initiatives for the develop-ment of renewable energy carriers. One possible future energycarrier is hydrogen, which could be used to supply households andindustries to produce heat and electricity. But it could also be usedas the future alternative fuel for automobiles. Such a switch fromfossil energy carriers to a hydrogen based energy systemwill causechanges in the energy infrastructure as well as in the technologiesused for the energy production and utilization (Könnölä et al.,2007). But the efforts of changing to a hydrogen society will beworthwhile only if the hydrogen is produced in a sustainable way.This means that the production has to be based on renewable rawmaterials and/or renewable energies as well as on efficientconversion technologies in the proper scale.

At present hydrogen production is mainly based on reforming offossil fuels (Corbo and Migliardini, 2007, 2009; Dalle Nogare et al.,2007; de Jong et al., 2009; Kaila and Krause, 2006; Rabe et al.,2007). The steam reforming of natural gas is the most commonstate of the art technology for hydrogen production (Corbo andMigliardini, 2009; de Jong et al., 2009). But technologies likepartial oxidation (Corbo andMigliardini, 2007; Chen et al., 2007) orautothermal reforming (Kaila and Krause, 2006; Rabe et al., 2007;

ax: þ43 1 58801 15999.iltner).

All rights reserved.

Hoang et al., 2006; Murata et al., 2007) of fossil fuels are alsowell established technologies. Besides the reforming of fossil fuels,there is also electrolysis of water as state of the art technology forthe production of hydrogen (Zeng and Zhang, 2010; Badwal et al.,2006; Chaffin et al., 2006; Grigoriev et al., 2006; Kreuter andHofmann, 1998; Marshall et al., 2007; Roy et al., 2006; de Souzaet al., 2007; Stojic et al., 2003), with alkaline electrolysis as itsmost common type. Hydrogen production by electrolysis can, incontrast to the reforming of fossil fuels, be considered as renewabletechnology, if electricity from renewable resources is used for theoperation of the electrolyzer.

Biomass based hydrogen production technologies are still underdevelopment, hence only few plant and operation data are avail-able up to now. It is, therefore, difficult for policy makers to identifypromising technologies as well as the advantages and disadvan-tages of each technology. One instrument to overcome thisdilemma is the use of process simulation. This tool offers thepossibility to build up theoretical plant layouts based on realprocess data that could, for instance, be taken from lab scaleexperiments. The process simulation results offer the possibility torate different technologies and to identify hurdles and potentials.

In the present work process simulation is used to evaluatedifferent hydrogen production technologies. These technologiescover the state of the art technologies steamreformingof natural gasand alkaline water electrolysis as well as biomass based technolo-gies. The latter comprise the reforming of biogas as well as thereforming of biomass gasification gas, the direct production ofhydrogen from biomass by coupled dark and photo fermentation

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62S52

and the coupled dark and biogas fermentation. All biomass basedtechnologies have been investigated with different kinds of rawmaterials and/or different plant layouts. Each of the processesmeetsthe strict regulations for hydrogen used in mobile fuel cells,requiring a minimum of 99.97 vol% of hydrogen (CaliforniaDepartment of Food and Agriculture, 2008). It is assumed thathydrogen will only be sold in this quality at fuel stations, even ifinternal hydrogen combustion engines would be able to run onlower quality hydrogen. The size range of the biomass conversionplants is chosen in a decentralized scale in order to offer the possi-bility of a sustainable hydrogen production. The steam reforming ofthe produced biogas and gasification gas on the other hand isanalyzed both in a decentralized and a centralized plant size. Theprocess simulation results are used to evaluate all technologiesaccording to their hydrogen production and energy efficiency.

As stated above, the present paper focuses on decentralizedhydrogen production and the application of the produced hydrogenas mobile fuel. It is, therefore, necessary to build up overall supplychains including the hydrogen production, the storage and trans-portation of the hydrogen, the fuelling stations and finally theutilization in the hydrogen vehicles. Each of the steps within thischain, and in particular the hydrogen storage technologies, wouldbe worth a detailed discussion and analysis. But this would bebeyond the scope of the present paper, hence these steps andtechnologies are deliberately not considered.

2. Investigated processes and process simulation

The aim of this paper is to evaluate different processes for theproduction of renewable hydrogen. However, hydrogen is notproduced in a renewable way at present but mainly by reforming ofnatural gas. Hence, the steam reforming of natural gas is included inthis study as a state of the art reference technology.

Alkaline water electrolysis is another proven technology forhydrogen production which is used in a much smaller scalecompared to natural gas steam reforming. At present electrolyzersare typically operated with the common electricity mix thatconsists mainly of non renewable electricity. However, hydrogencan be produced in a sustainable way via electrolysis if electricityfrom renewable sources is used for its operation. Within the pre-sented analysis it is assumed that the alkaline electrolysis is oper-ated with electricity from renewable sources. It is, therefore,treated as a renewable hydrogen production technology.

Hydrogen can also be produced in a renewable way if biomass isused as rawmaterial. However, biomass based technologies are stillunderdevelopment andnoneof theproposed technologieshasbeentested in large scale yet. Among the various developed biomassbased technologies four promising technologies have been chosenfor the present analysis. These are the steam reforming of gas

Fig. 1. Overview of all eva

produced by biogas fermentation, the steam reforming of gasproducedbybiomass gasification, the direct biological production ofhydrogenvia the coupleddark andphoto fermentation aswell as theproduction of hydrogen by dark fermentation coupled with biogasfermentation. The analysis of the biomass based technologies iscarried out for several plant layouts as shown in Fig. 1. The steamreforming of biogas and gas from biomass gasification is eithercarried out on-site, with or without prior CO2 separation, or ina central steam reforming plant that is connected to severaldecentralized gas production plants via the natural gas grid. Thetechnology of coupled dark and biogas fermentation is also includedwith threedifferent configurations. Thebiogas that is produced fromthe residue of the dark fermenter is either used in a gas boiler toproduce process heat onlyor in a gas engine to produce both processheat and electricity. Furthermore, the biogas can be converted tohydrogen by on-site steam reforming of the biogas.

The evaluation of the investigated hydrogen production tech-nologies is based on process simulation. The application of processsimulation offers the possibility to easily analyze different plantlayouts and plant sizes with various rawmaterials. Moreover, it canbe used to gain results for technologies that are not fully developedyet. Within this work the process simulation is carried out with thecommercial simulation program IPSEpro. It is an equation orientedflowsheet simulation package which was originally designed tocarry out thermal power cycle calculations (Perz, 1991). The soft-ware package consists of several modules like a Model Develop-ment Kit or a Process Simulation Environment (Perz and Bergmann,2007). The Model Development Kit (MDK) offers the possibility todevelop and implement new unit models for processes that are notincluded in the standard simulation libraries. These user built unitmodels contain the equations describing the behavior of the units(e.g. mass balances of the components or energy balances) as wellas the graphical symbols of the units. The Process SimulationEnvironment (PSE) on the other hand is used to build up theflowsheets of the processes graphically and to solve the processmodels based on the unit model equations.

In the course of the presented work new user models have beendeveloped for most of the units necessary to build up the investi-gated hydrogen production plants. Only few of the used modelswere available within the standard power plant library of theprogram (Perz and Bergmann, 2003) or were taken from precedingworks (Pröll et al., 2005a, 2007).

A description of all investigated hydrogen production technol-ogies is given in the following sections.

2.1. Steam reforming of natural gas

The plant for natural gas steam reforming (see Fig. 2) containsan adsorber (1) for the removal of hydrogen sulfide and hydrogen

luated process chains.

air

fuel

natural

gas

hydrogen

stack

process

water

1

2

3

4

5

6

Fig. 2. Flowsheet of the natural gas steam reforming plant; 1: adsorber for HCl and sulfur compounds, 2: steam reformer, 3: CO shift reactor, 4: condenser, 5: PSA, 6: steam turbine.

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62 S53

chloride from the natural gas. This unit also includes a precedinghydration step that converts organic sulfur compounds to hydrogensulfide by using a small amount of the produced hydrogen for theconversion. The cleaned natural gas enters an externally heatedcatalytic steam reformer (2) that converts CH4 and steam tohydrogen and CO at a temperature of 880 �C and a pressure of30 bar. At the same time some of the CO reacts with steam tohydrogen and CO2 via the CO shift reaction. After a cooling step thegas is fed into a CO shift reactor (3) where the main part of the CO isconverted to hydrogen via the CO shift reaction at 330 �C to furtherincrease the hydrogen yield of the plant. Subsequently the gasstream is cooled down and dried (4) and enters a pressure swingadsorption (PSA) (5) at 40 �C to purify the hydrogen to the desiredquality. The off-gas of the PSA is recycled to the steam reformer (2)where it is mixed with additional natural gas and used as burningfuel to supply the heat for the reforming reaction.

The steam reforming plant is equipped with a multitude ofheat exchangers for heat recovery and a steam cycle for processsteam generation. The excess process steam is used to generateelectricity in a steam turbine. The plant size is designed to produce289 Mio. mSTP

3 (m3 at standard temperature and pressure) ofhydrogen per year, which has been chosen according to the steamreforming plant at the oil refinery of the company OMV inSchwechat, Austria (Pulletz and Böhme, 2003). The appliedcomposition of the natural gas is given in Table 1.

2.2. Alkaline water electrolysis

A typical alkaline electrolysis plant consists of several units likewater preparation, alkaline cycle and electrolyzer stack. All theseparts have been included within one unit operation model (seeFig. 3). This model contains a detailed modeling of the alkaline

Table 1Applied composition of natural gas.

Volume fraction of component, % Natural gas

CH4 96.210C2H6 1.110C3H8 0.551N2 1.131CO2 0.998H2S 3.258 � 10�7

water electrolyzer itself. All other parts of the electrolysis plant areonly considered with their energy demand.

Two size ranges are analyzed for the alkaline electrolysis:a decentralized plant with a hydrogen production capacity of485 mSTP

3 /h and a small plant producing only 60 mSTP3 /h. The latter is

supposed to be located directly at the hydrogen fuelling stationwhere it covers the total demand of the station. The data for thealkaline water electrolysis plants are taken from Ivy (2004). Thedecentralized plant is based on the data for the Norsk Hydro 5040plant which is operated at 33 bar and 80 �C. The small plant is basedon the Stuart IMET 1000 plant which is operated at 25 bar and80 �C. Both plants require 4.8 kWh of electricity per mSTP

3 ofproduced hydrogen.

Within the present work it is assumed that the alkaline elec-trolysis is operated with green electricity in order to be able to treatthis technology as renewable hydrogen production technology.

2.3. Reforming of biogas

The biogas fermentation plant is operated in continuous modeand comprises a mashing tank (1), an anaerobic fermenter (2),a post digestion tank (3) and a biological desulphurization reactor(4) as shown in Fig. 4.

The mashing tank is used to mix the raw material with water inorder to get a pumpable substrate. The anaerobic fermenter isoperated at 35 �C and produces roughly 80% of the total biogas. Thepost digestion tank serves as storage tank for the biogas and thefermentation residue. Moreover, up to 20% of the total producedbiogas can originate from the post digestion tank. In a typical biogasfermentation plant, the biological desulfurization reactor is oper-ated with air. As the nitrogen content of the air would cause severeproblems during the purification of the subsequently producedhydrogen, it is decided to run the reactor with oxygen instead of air.

unused water

hydrogen rich gas

oxygen rich gas

water

Fig. 3. Flowsheet of the alkaline water electrolysis.

Fig. 4. Flowsheet of the biogas fermentation plant; 1: mashing tank, 2: biogas fermenter, 3: post digestion tank, 4: biological desulfurization.

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62S54

The layout and size of the plant are based on the plant in Bruck ander Leitha, Austria. It is in a size range of about 30,000 tons of rawmaterial per year which corresponds to 800e900 mSTP

3 /h of biogas,depending on the chosen raw material. The biogas yields of thedifferent raw materials as well as the composition of the producedbiogas are based on experimental data (Institut für Energetik undUmwelt, 2006). The data used for the raw materials investigatedin the present study are given in Table 2. The bio waste representsthe separately collected biogenic residue from households inAustria, containing mainly kitchen slops and small fractions of yardwaste.

The produced biogas is used as input in a steam reforming plant.The connection between the biogas plant and the steam reformingplant is examined with three different layouts. The simplest layoututilizes the produced biogas directly in an on-site steam reformingplant. This decentralized steam reforming plant has the samelayout as the natural gas steam reforming plant described above,only showing a smaller size range. Beside the plant size also theplant efficiency is lower for the decentralized reforming plant(0.723 kWH2/kWCH4 compared to 0.838 kWH2/kWCH4 for theplant in the centralized size range, based on lower heating values).The produced biogas is both used as a feed that is converted tohydrogen in the reforming plant and as a combustion fuel, toprovide the necessary heat for the steam reforming reactor.

The biogas contains a huge amount of CO2 that is only deadfreight and requires additional amounts of biogas for its heating inthe steam reformer. The hydrogen yield of the combined plant can,therefore, be increased by CO2 separation prior to the steamreforming plant. This configuration is considered in the secondplant layout, in which the CO2 separation is realized withmembrane modules. The layout of the separation step is shown inFig. 5. It comprises a gas compressor (1) to provide the required

Table 2Biogas yields and biogas composition of the investigated raw materials.

Volume fraction of component, % Raw material

Maize silage Potato peels Bio waste

CH4 53.0 51.4 61.5CO2 40.5 42.9 32.8H2O 5.6 5.6 5.6H2S 0.9 0.1 0.1Biogas yield, mSTP

3 /t raw material 190.0 67.7 100.0

pressure difference across the membrane, a condenser (2) and themembrane modules (3) that separate mainly CO2 from CH4 by gaspermeation. This layout is a simplified version of the process pre-sented by Harasek et al. (2007) and Miltner et al. (2008a).

It is assumed that 4.5% of the entering CH4 are lost due topermeation through the membrane and that 2% of the entering CO2are not separated from the CH4 stream. Hence, the CO2-richpermeate stream of the gas permeation plant still contains smallamounts of CH4. Due to the high global warming potential of CH4, itis advisable to reuse this off-gas stream. In the presented layout thisstream is added to the fuel stream of the on-site steam reformerthat is already a mixture of biogas and off-gas of the pressure swingadsorption of the steam reforming plant.

In the third plant layout the produced biogas is upgraded tonatural gas quality and supplied to the natural gas grid. Theupgraded biogas from several biogas plants is transported toa centralized steam reforming plant via the natural gas grid. Theupgrading to natural gas quality is again done by gas permeationsimilar to the process shown in Fig. 5. The plant is only extendedwith a sulfur adsorption prior to the membrane modules, in orderto meet the specifications of the natural gas grid. Furthermore,a compressor is needed that feeds the gas into the gas grid ata pressure of about 30 bar. The central steam reforming plant isequivalent to the plant described in chapter 2.1. The advantage ofthis pathway is the increased efficiency of the centralized reform-ing plant compared to the decentralized one. One disadvantage ofthe presented configuration is the fact, that the CO2-rich off-gas ofthe gas permeation plant cannot be used internally. This drawbackcould e.g. be eliminated, if a part of the produced biogas would beused on-site in a gas boiler or gas engine to produce the requiredprocess heat and electricity. In this case the off-gas could be mixedwith the biogas assuring that no CH4 would be released to theatmosphere.

2.4. Reforming of gasification gas

Among the multitude of developed biomass gasification tech-nologies, the authors have opted for the Fast Internally CirculatingFluidized Bed (FICFB) gasification technology. The plant layout andsize were chosen according to the plant in Güssing, Austria, whichis designed with a supplied fuel power of 8 MWth (based on lowerheating value of the fuel), as described by Pröll et al. (2007). Themain parts of the gasification plant (see Fig. 6) are the FICFB gasifier,

biogas from

fermentation plant

CO2 rich off-gas

condensate

1

CH4 rich gas to

steam reforming plant

12

3

Fig. 5. Flowsheet of the gas permeation plant for CO2 separation; 1: two stage compressor with intercooling, 2: condenser, 3: membrane modules.

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62 S55

which comprises a gasification chamber (1) and a combustionchamber (2), a dust filter (3) and a tar scrubber (4).

Similar to the biogas based plants, the produced gasification gasis converted to hydrogen in a steam reforming plant. In order toprevent coking within the steam reformer, a second tar scrubber isimplemented additionally to the plant in Güssing, as shown inFig. 6. Moreover, the second tar scrubber further reduces the NH3and H2S content of the gas, as presented by Pröll et al., (2005b).Within the present study the plant is designed to be operated withwood chips at a fuel power of 7.8 MWth (based on lower heatingvalue of the fuel). The applied composition of the produced gasi-fication gas is given in Table 3 and is based on the values given byPröll et al. (2005b) and Rauch (2005). The net power of theproduced gasification gas is calculated with 5.6 MW (based onlower heating value).

Similar to the biogas based technology presented in theprevious chapter, the produced gasification gas is also used as inputin a steam reforming plant. The combination of the gasificationplant and the steam reforming plant is again considered with threedifferent layouts. The first layout uses the tar free gas directly asfeed and burning fuel in an on-site steam reforming plant equiva-lent to the one described in chapter 2.3 concerning the reforming ofbiogas.

raw

material

lime

olivine

1

3

4

2

Fig. 6. Flowsheet of the biomass gasification plant; 1: gasification c

The second plant layout includes again a CO2-separation stepprior to the steam reforming plant. The gasification gas, in contrastto biogas, has a high content of hydrogen (approx. 40 vol%, drybase). Hence, it is not possible to use a gas permeation plant for theseparation of CO2 because this technology would also separate thehydrogen from the methane rich stream, which would reduce theoverall hydrogen yield of the plant. Therefore, a monoethanolamine(MEA) scrubber is used to only remove the CO2 from the gasifica-tion gas. The scrubber is operated with a MEA-water-solution with30 mass% of MEA. It is assumed that the scrubber has a CO2-removal efficiency of 90% and that 3.7 MJ of steam are needed todesorb 1 kg of CO2 in an associated desorption column. The purifiedhydrogen and methane rich gas is then used as feed and burningfuel for an on-site steam reforming plant.

In the third plant layout the gasification gas is upgraded tonatural gas quality by methanation. The upgraded gas from severalgasification plants is transported to a centralized steam reformingplant via the natural gas grid, where it is converted to hydrogen.The layout of the upgrading plant is shown in Fig. 7. The gasificationgas passes through an adsorber (1) to remove sulfur compoundsand HCl before it enters a methanation reactor (2) where hydrogen,CO and CO2 are converted to methane. A subsequent membraneseparation step is used to separate the unconverted CO2 and

cooling or district heating

air

ash

RME

cooling

stack gas

gasification

gas

4

cooling

hamber, 2: combustion chamber, 3: dust filter, 4: tar scrubber.

Table 3Composition of the main components of the raw gasification gas.

Volume fraction of dry component, % Gasification gas

CH4 10.5C2H4 2C3H8 0.6CO 25CO2 20.8H2 39.1N2 1.6H2O 35.1NH3 (clean gas after first tar scrubber) 0.04H2S (clean gas after first tar scrubber) 0.002

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62S56

hydrogen content from the obtained methane rich gas, in order tomeet the gas specifications of the natural gas grid. The gaspermeation plant for the CO2 separation is considered equivalent tothe one used for the CO2 removal from biogas with subsequent on-site reforming (see chapter 2.3).

The simulation of the methanation reactor was based on thepublications of Seemann et al. (2005) and Hofbauer et al. (2006).Within the present work the efficiencies of themethanation reactorare set to 98.6% for the CO conversion and to 82.1% for the chemicalefficiency (energy content of produced CH4 related to energycontent of the input stream to the methanation reactor, both basedon lower heating values). The exiting methane rich gas is assumedto have a methane concentration of 41 vol%, a CO2 concentration of48 vol% and a CO concentration of 0.49 vol% (all compositions ondry basis).

2.5. Coupled dark and photo fermentation

Both dark and photo fermentation are technologies thatproduce a hydrogen rich gas by biological conversion of sugars andorganic acids, respectively. The combination of both processes isvery beneficial as organic acids are produced as a side productduring the dark (thermophilic) fermentation step. These acids canbe utilized as feed for the photo fermentation step whereby thewhole plant efficiency is increased. Both technologies as well astheir combination are still in the laboratory phase. Hence, the plantlayout for the combined process as well as the plant size are basedon a theoretical work presented by Markowski et al. (2008).However the layout has been adapted and simplified in somerespects as shown in Fig. 8. Experimental results for the verificationof the simulation models are taken from publications by Claassenet al. (2000) and Claassen et al. (2005).

Fig. 7. Flowsheet of the methanation plant; 1: adsorber for HCl and sulfur compounds,membrane modules, 6: compressor for grid pressure level.

The first step within the coupled dark and photo fermentationplant is a hydrolysis step (1) that is needed to convert componentslike starch into fermentable sugars. This process comprises severalsteps like liquefaction and saccharification. The hydrolysis processhas been modeled simplified by considering only the overallconversion from the rawmaterial to the hydrolysate and the overallenergy demand, not considering in detail the individual sub-steps.In the next step unfermentable solids are separated from thesubstrate (2). The substrate is then mixed with water in order toobtain a sugar concentration that is nontoxic for the microorgan-isms used in the dark fermenter. After a preheating step thesubstrate is fed into the dark fermenter (3). In the dark fermentersugar and water are converted to gaseous H2 and CO2 as well asliquid organic acids. 15% of the influent sugars are utilized by themicroorganisms for their growth. The fermenter is operated at70 �C and a vacuum of 50 kPa. The vacuum is needed to evaporatea part of the water content of the substrate in order to reduce thepartial pressure of hydrogen in the fermenter. This reduction of thepartial pressure is required since high hydrogen concentrationscause an inhibition of the growth of the microorganisms and thusa decrease in their hydrogen productivity (Willquist et al., 2009).The main part of the evaporated water is condensed in a subse-quent condenser (5) and fed back into the dark fermenter. Aftera cooling step the liquid effluent of the dark fermenter is fed intothe photo fermenter (4) that operates at 35 �C and 1 bar underexposure to light. In the photo fermenter the organic acids, whichhave been produced in the dark fermenter, and water are convertedto gaseous H2 and CO2. The gas streams from both dark and photofermenter are mixed and purified in a vacuum swing adsorption(VSA) plant (6).

Within this paper the calculations for the coupled dark andphoto fermentation are carried out with either 100% conversion ofthe influent sugar within the dark fermenter (implying that nosugar is utilized by the microorganisms for their growth) and 100%conversion of the influent acids within the photo fermenter or with80% conversion in each of the fermenters. The former calculationleads to the theoretical maximum hydrogen amount for a givenraw material, whereas the latter calculation represents a realisticreachable production scenario. It is assumed that the liquid productof the dark fermenter is acetic acid only. Furthermore, thehydrogen recovery of the VSA plant has been varied between 80%and 95% as this part of the plant shows some optimization poten-tial. Moreover, there is a possibility to replace the VSA byamembrane contactor with very low hydrogen loss (Modigell et al.,2008). The composition of the utilized raw materials is given inTable 4.

2: methanation reactor, 3: two stage compressor with intercooling, 4: condenser, 5:

Fig. 8. Flowsheet of the coupled dark and photo fermentation plant; 1: hydrolysis, 2: solids separation, 3: dark fermenter, 4: photo fermenter, 5: condenser, 6: VSA.

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62 S57

2.6. Coupled dark and biogas fermentation

The utilization of the organic acid rich residue from the darkfermenter within a photo fermenter is not the only option. It canalso be utilized in a biogas fermenter to produce a methane richbiogas. The produced biogas can then be used in different ways,whereof three variants are investigated within the present work.The first plant layout utilizes the biogas in a gas engine (7) toproduce electricity and heat that is used to heat up the substrate ofthe dark fermenter (3) and the fermenter itself (see Fig. 9). Thehydrogen rich gas from the dark fermenter is purified in a vacuumswing adsorption (VSA) plant (6) and the VSA off-gas is mixed withthe biogas from the biogas fermenter (4) and fed into the gas engineto use the energy content of the lost hydrogen.

The biogas can also be used in a gas boiler to produce processheat only (see Fig. 10). The process heat is again used to heat thesubstrate of the dark fermenter (3) and the fermenter itself. Asbefore, the off-gas of the VSA (6) for the hydrogen purification ismixed with the biogas from the biogas fermenter (4) and used asa feed for the gas boiler (7).

The third investigated option is to separate the CO2 from theproduced biogas and to convert the obtained gas to hydrogen in anon-site steam reforming plant. In this case a biological desulfur-ization reactor is needed prior to the CO2 separation to reduce theH2S content of the biogas. The CO2 removal is implemented by a gaspermeation plant. The purified biogas is subsequently utilized asfeed and burning fuel for an on-site steam reforming plant. Both thegas permeation and the reforming plant are implemented equiva-lent to the ones described in chapter 2.3. The CO2-rich permeate ofthe gas permeation plant is mixed with the fuel for the steamreformer to prevent CH4 loss to the atmosphere.

The dark fermentation step was modeled equivalent to the darkfermenter presented in the previous chapter. The built upmodel for

Table 4Composition of the utilized raw materials for the coupled dark and photo fermen-tation and the coupled dark and biogas fermentation.

Mass fraction of component, % Raw material

Maize corn Potato peels

C 23.801 4.851H 3.995 0.627O 31.705 4.169N 0 0.308S 0 0.033Water 15.0 89.0Ash and insolubles 25.50 1.023Starch and glucose, dry ash free component, % 100 51.9

the biogas fermenter operated with the organic acid rich effluent ofthe dark fermenter was verified with data taken from literature(Chu et al., 2008; Kyazze et al., 2007; Zhu et al., 2008).

Due to the early stage of development of the coupled dark andbiogas fermentation it was decided, similar to the coupled dark andphoto fermentation, to base the performance calculations ona range of conversion factors rather than on single data taken fromliterature. Hence, the calculations have been carried out with 80%conversion of the influent sugar in the dark fermenter and 70%conversion of the influent acids and sugar in the biogas fermenteras well as with 100% conversion in both fermenters (again implyingthat no sugar is utilized for the growth of the microorganismswithin the dark fermenter). The lower boundaries seem to berealistic conversion ratios and the upper limits of 100% conversionrepresent the theoretical maximum. The lower conversion limit forthe dark fermenter also corresponds to the value chosen for thecoupled dark and photo fermentation. Similar to the simulation ofthe dark and photo fermentation it was assumed that only aceticacid is produced in the dark fermenter. Moreover, the hydrogenrecovery of the VSA for the purification of the hydrogen rich gasfrom the dark fermenter has been varied between 80% and 95% dueto the same reasons as for the coupled dark and photo fermenta-tion. The calculations for the coupled dark and biogas fermentationhave been carried out only with potato peels as raw material. Theinvestigated composition of the potato peels is equivalent to theone used for the coupled dark and photo fermentation (see Table 4).

3. Results and discussion

3.1. Production efficiency analysis

All biomass based technologies are considered in a decentral-ized size range that offers the possibility of sustainable supply withthe needed amounts of rawmaterials. The large electrolysis plant isalso investigated in a decentralized size whereas the small elec-trolysis plant is in a size range that can be used to produce thehydrogen directly at a hydrogen fuelling station. The size range ofthe different hydrogen production technologies is presented inFig. 11 showing selected raw materials and plant layouts.

The results of the specific hydrogen production for the biomassbased processes are shown in Fig. 12. It can obviously be seen thatnot only the choice of the technology but also the choice of the rawmaterial has a strong influence on the production performance.

A comparison between the results of the biogas fermentationplants shows that the hydrogen yield per kg of dry raw material ishigher for maize than for potato peels and bio waste. The consid-erably lower yields of the plants operatedwith biowaste aremainly

Fig. 9. Flowsheet of the coupled dark and biogas fermentation with gas engine; 1: hydrolysis, 2: solids separation, 3: dark fermenter, 4: biogas fermenter, 5: condenser, 6: VSA, 7:gas engine.

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62S58

caused by the low biogas yield per ton of dry bio waste. Fig. 12 alsoshows that the biogas fermentation plant with upgrade to naturalgas quality and a central steam reforming plant has the highestproduction efficiency among the biogas fermentation layouts.

Among the biomass gasification based plants the layout with gasupgrading to natural gas quality and central steam reforming has thelowest hydrogen yield per kg of dry rawmaterial. The reason for thisresult is that the gasification gas has a high hydrogen content ofapproximately 40 vol% (dry basis). In the central steam reforminglayout this hydrogen content is converted to methane in themethanation reactor. The methane is then converted back tohydrogen in the central steam reforming plant with a certainmethane loss due to incomplete conversion and side reactions.Moreover, the conversion from hydrogen to methane is incompleteas well and some of the produced methane and almost the totalunconverted amount of hydrogen are lost in the CO2-separation step.Beyond that, a part of the producedmethane has to be used as fuel toprovide the required heat for the steam reformer. The sum of theselosses leads to a hydrogen yield that is even considerably lower thantheyieldof thegasificationplantwithdirect on-site steamreforming.

A comparison between the results of the biogas fermentationplants and the biomass gasification plants shows that the biomass

Fig. 10. Flowsheet of the coupled dark and biogas fermentation with gas boiler; 1: hydrolysgas boiler.

gasification plants are leading to higher specific hydrogen yields,except for the layout with central reforming. The gasification optionwith direct on-site steam reforming even shows higher yields thanany of the biogas fermentation layouts with CO2 separation and on-site reforming. Moreover, it has a higher production efficiency thanthe biogas fermentation plants with central reforming operatedwith potato peels and bio waste.

The theoretical yield of the coupled dark and photo fermenta-tion of maize corn (100% conversion in each fermenter) is thehighest one of all compared plants, both for 80% and 95% hydrogenrecovery in the VSA plant. A comparison between the dark/photofermentation and the biogas fermentation plants operated with thesame raw material shows that the dark/photo fermentation with80% conversion in each fermenter and 80% recovery in the VSAleads to lower hydrogen yields than any of the biogas fermentationplants. The result for dark/photo fermentation of maize corn with80% conversion and 95% hydrogen recovery in the VSA is slightlylower than that for biogas fermentation of maize silagewith centralreforming but higher than the yield of any of the biogas plantsoperated with potato peels. Comparing the plants operated withpotato peels, only the dark/photo fermentation with 100% conver-sion in the fermenters show higher yields than the biogas

is, 2: solids separation, 3: dark fermenter, 4: biogas fermenter, 5: condenser, 6: VSA, 7:

32,988

60

485

546

1,161

500

257

0.8238

0.0015

0.0121

0.0136

0.0290

0.0125

0.0064

Hydrogen production in mSTP

³/h or kg/s

m³/hkg/s

Natural gas steam reforming

Alkaline electrolysis

small

Alkaline electrolysis

large

Biomass gasification

direct on-site reforming,wood chips

Biogas fermentation

direct on-site reforming, maize silage

Dark-/photo fermentation

potato peels

Dark-/biogas fermentation

potato peels

Fig. 11. Size range of the investigated hydrogen production technologies; the volume flows are given in m3/h at standard temperature and pressure.

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62 S59

fermentation plants. This can be explained by the fact that the darkand photo fermentation utilizes only the starch content of thepotato peels whereas the biogas fermentation is able to use a largerpart of the substrate. Hence, raw materials which offer contentsthat can be converted to biogas but not to hydrogen lead to a largerdifference between the specific yields of these two technologies.

The hydrogen production yields of the coupled dark and biogasfermentation with gas engine as well as gas boiler are, as expected,low as only the dark fermenter produces hydrogen. But theperformance of the layout with on-site reforming is comparable tothe dark and photo fermentation of potato peels. Comparing therealistic lower values of conversion in the fermenters and 80%

0.00.10.20.30.40.50.60.70.80.91.01.11.2

Hyd

ro

gen

pro

du

ctio

n

in

mS

TP³/kg

raw

d

ry m

aterial

Bio

ga

s ferm

en

ta

tio

n, m

aize

s

ila

ge

cen

tral r

efor

min

g

CO

2-se

p., o

n-si

te re

form

ing

on-

site

refo

rmin

g

Bio

ga

s ferm

en

ta

tio

n, p

ota

to

p

ee

ls

cen

tral r

efor

min

g

CO

2-se

p., o

n-si

te re

form

ing

on-

site

refo

rmin

g

Bio

ga

s ferm

en

ta

tio

n, b

io

w

as

te

cen

tral r

efor

min

g

CO

2-se

p., o

n-si

te re

form

ing

on-

site

refo

rmin

g

Ga

sific

atio

n, w

oo

d c

hip

s

on-

site

refo

rmin

g

hydrogen production, dark fermentation based

technologies: 80 % H2 recovery in VSA

hydrogen production, dark fermentation based

technologies: 95 % H2 recovery in VSA

Fig. 12. Comparison of the specific hydrogen production of all biomass based technologietemperature and pressure.

hydrogen recovery in the VSA plants, the yield of the dark/biogasfermentation with on-site reforming is only slightly lower than thecorresponding yield for the dark/photo fermentation of potatopeels.

3.2. Energy efficiency analysis

The energy efficiencies of all examined technologies and plantsare shown in Fig. 13. The energy efficiency relates the energycontent of the produced hydrogen, based on the lower heatingvalue, to the energy input to the process. The figure shows tworesults: one result relating the hydrogen energy content to the raw

cen

tral r

efor

min

g

CO

2-se

p., o

n-si

te re

form

ing

Da

rk

/p

ho

to

fe

rm

en

t., m

aize

c

orn

con

vers

. 100

% e

ach

ferm

ente

r

con

vers

. 80%

eac

h fe

rmen

ter

Dark/p

ho

to

ferm

en

t., p

ota

to

p

ee

ls

con

vers

. 100

% e

ach

ferm

ente

r

con

vers

. 80%

eac

h fe

rmen

ter

Dark/b

io

ga

s fe

rm

en

t., p

ota

to

p

ee

ls

con

vers

. 80/

70%

, boi

ler

con

vers

. 80/

70%

, eng

ine

con

vers

. 100

/100

%, b

oile

r

con

vers

. 100

/100

%, e

ngin

e

con

vers

. 80/

70%

, ref

orm

ing

con

vers

. 100

/100

%, r

efor

min

g

s; the specific hydrogen production is given in m3/kg dry raw material at standard

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62S60

material energy input only (based on the lower heating value(LHV), dry) and one result relating it to the sum of all inputtedenergies (raw material, electricity and heat) minus producedelectricity.

Expectedly the energy efficiency of the state of the art tech-nology natural gas steam reforming is the highest one of allcompared technologies. Furthermore, the steam reforming plantproduces more electricity than needed for its operation. This leadsto a higher energy efficiency related to the total energy input thanrelated to the raw material energy input only. The alkaline elec-trolysis has the highest energy efficiency related to the total energyinput among all regenerative hydrogen production technologies.

The biogas fermentation plant with maize silage as raw mate-rial and central steam reforming plant exhibits the highest energyefficiencies among all biomass based technologies. The onlyexceptions are the energy efficiencies related to the raw materialinput of the dark/photo fermentation plants with 100% conversionand 95% hydrogen recovery in the VSA plant, which are onlytheoretical values. In contrast to the production efficiency, theenergy efficiencies of the biomass gasification plants are lowerthan the efficiencies of the biogas fermentation plants operatedwith maize silage and potato peels. This is caused by the fact, thatthe lower dry heating value of the wood chips is 1.21e1.27 timeshigher than the lower dry heating values of maize silage andpotato peels, respectively. On the other hand the hydrogen yieldrelated to the dry raw material input of the gasification plants,disregarding the plant layout with central reforming, is only0.97e1.25 times higher than the yield of any of the biogasfermentation layouts operated with maize silage or potato peels.Therefore, the energy efficiencies of the gasification plants areslightly lower than the efficiencies of the named biogas fermen-tation plants, although the production efficiencies show theinverse characteristics. It is also remarkable that the energy effi-ciency related to the total energy input of the gasification layout

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

En

erg

y e

ffic

ien

cy

in

M

WH

2/M

Win

Alk

alin

e e

le

ctro

ly

sis

larg

e

smal

l

Na

tu

ra

l g

as

s

te

am

re

fo

rm

in

g

Bio

ga

s ferm

en

ta

tio

n, m

aize

s

ila

ge

cen

tral r

efor

min

g C

O2-

sep.

, on-

site

refo

rmin

g o

n-si

te re

form

ing

Bio

ga

s ferm

en

ta

tio

n, p

ota

to

p

ee

ls

cen

tral r

efor

min

g

CO

2-se

p., o

n-si

te re

form

ing

on-

site

refo

rmin

g

Bio

ga

s ferm

en

ta

tio

n, b

io

w

as

te

CO

2-se

p., o

n-si

te re

form

ing

on-

site

refo

rmin

g

related to raw material energy input (LHV, dry), d

fermentation based technologies: 80 % H2 recove

related to raw material energy input (LHV, dry), d

fermentation based technologies: 95 % H2 recove

Fig. 13. Comparison of the energy efficien

with direct on-site reforming is slightly higher than that of thelayout with preceding CO2 separation. This is caused by the highsteam energy demand of the MEA-desorption applied in the CO2-separation step in the latter plant.

Comparing the reforming of biogas and the dark and photofermentation, only the dark/photo fermentation plants with theo-retical 100% conversion in each fermenter show efficienciescomparable to the biogas fermentation plants. But it has to bementioned that the coupled dark and photo fermentation does notinclude any use of the produced gas or residues for the self-supplyof the heat demand of the plant. The biogas reforming plants on theother hand include the self-supply of the heat demand for bothbiogas fermentation and steam reforming. The only exception is thebiogas layout with central steam reforming plant where no heatintegration is possible between the biogas and the steam reformingplant. The use of the dark/photo fermentation residues and sepa-rated particles like lignin for process heat production couldimprove the energy efficiency of the dark/photo fermentationtechnology. Combined with an improvement in the hydrogenrecovery of the gas upgrading system to the proposed value of 95%,this technology could definitely be competitive to the reforming ofbiogas.

The coupled dark and biogas fermentation plants with gasengine as well as with gas boiler show again very low energy effi-ciencies compared to the other technologies and layouts. Theenergy efficiency related to the total energy input of the layout withgas engine is higher than the efficiency related to the raw materialenergy input, since the gas engine produces more electricity thanneeded for the operation of the plant. The energy efficiencies of thelayout with on-site steam reforming are comparable to those of thedark and photo fermentation. Only the coupled dark and photofermentation with 100% conversion in each fermenter and 95%hydrogen recovery in the VSA plant show considerably higherenergy efficiencies.

cen

tral r

efor

min

gG

as

ific

atio

n, w

oo

d c

hip

s

cen

tral r

efor

min

g C

O2-

sep.

, on-

site

refo

rmin

g o

n-si

te re

form

ing

Dark/P

ho

to

ferm

en

t., m

aize

c

orn

100

% c

onve

rs. e

ach

ferm

ente

r 8

0% c

onve

rs. e

ach

ferm

e te

r

Dark/p

ho

to

ferm

en

t., p

ota

to

p

ee

ls

100

% c

onve

rs. e

ach

ferm

ente

r

80%

con

vers

. eac

h fe

rmen

ter

Dark/b

io

ga

s fe

rm

en

t., p

ota

to

p

ee

ls

80/

70%

con

vers

., bo

iler

80/

70%

con

vers

., en

gine

100

/100

% c

onve

rs.,

boile

r

100

/100

% c

onve

rs.,

engi

ne

80/

70%

con

vers

., re

form

ing

100

/100

% c

onve

rs.,

refo

rmin

g

ark

ry in VSA

ark

ry in VSA

related to total energy input, dark fermentation

based technologies: 80 % H2 recovery in VSA

related to total energy input, dark fermentation

based technologies: 95 % H2 recovery in VSA

cies of all investigated technologies.

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62 S61

4. Conclusions and outlook

The present paper evaluates different hydrogen productionpathways comprising alkaline electrolysis, reforming of biogas,reforming of gasification gas, coupled dark and photo fermentationas well as coupled dark and biogas fermentation. The biomassbased technologies are investigated with different plant layoutsand/or raw materials. The steam reforming of natural gas isincluded as reference technology for hydrogen production.

Alkaline electrolysis shows the highest energy efficiency amongthe renewable technologies and it offers the advantage of inde-pendence of any biomass. This technology offers a good possibilityof sustainable hydrogen production in regions without biomassresources or with stiff competition between the food and theenergy sector. But it has to be kept in mind that a sustainableproduction of hydrogen by alkaline electrolysis can only be guar-anteed if electricity from renewable sources is utilized.

The reforming of biogas obtains good results regarding bothhydrogen production and energy efficiency provided that theproper raw material is chosen. The reforming of gasification gasshows good production efficiencies but in contrast the energyefficiencies are low compared to the reforming of biogas. Theproduction efficiency results of the coupled dark and photofermentation are comparable to those of the reforming of biogasbut their energy efficiencies are lower. However, since this tech-nology is in an early stage of development it still has potential fordevelopment andmight be a real alternative to reforming of biogas.Finally, the best results for the coupled dark and biogas fermenta-tion regarding both hydrogen production and energy efficiency areobtained for the layout with on-site steam reforming of theproduced biogas, showing efficiencies comparable to the dark andphoto fermentation.

The choice of the proper technology will have to be based on theavailability of raw materials, since the kind of raw material hasa strong influence on the performance of the technology, on thecompetition between food and energy production and on thedevelopment of raw material prices. The regional circumstancesshould also be regarded. For example it would not be advisable tobuild a biomass gasification plant in an agricultural region withscarce wood resources even though the production efficiency ofthis technology is high. It should also be kept in mind that thelayouts with centralized steam reforming are ecologically worth-while only if the associated hydrogen fuelling stations can beplaced in a proper range from this central plant. Hence, thisproduction path should only be chosen for regions with very goodinfrastructure and a dense fuelling station network to avoid longhydrogen transportation by trucks.

It would be necessary to extend the presented analysis by aneconomic and ecologic analysis, e.g. a life cycle analysis, to identifythe costs and the impact on the environment of each technologyand thus evaluate the sustainability of the processes. Someeconomic results based on slightly different plant layouts and anolder data base have already been presented by the authors(Miltner et al., 2008b). However, these calculations should beupdated and extended by the coupled dark and biogas fermenta-tion. Moreover, it would be desirable to extend the analysis tofurther raw materials. Especially other refuse materials like foodproduction waste or sewage sludge as well as intertillage would beadvantageous preventing the competition between food andenergy production.

The analysis could also be extended with other hydrogenproduction technologies. For example the presented gasificationprocess could be replaced by the adsorption enhanced reformingprocess (Pfeifer et al., 2007) which is a gasification technology thatproduces a hydrogen rich syngas.

Finally, it will be necessary to determine the amount ofhydrogen that can be produced in a sustainable way withina specific region or even worldwide. The decision whether thefuture world should be based on a hydrogen society or not has to bebased on the evaluation of the general potential to produce theneeded amounts of hydrogen in an economic and sustainable way.

Acknowledgements

We gratefully acknowledge the support by “Austrian AdvancedAutomotive Technology”, a subprogram of the Federal Ministry ofTransport, Innovation and Technology (BMVIT) in cooperationwiththe “Austrian Industrial Research Promotion Fund” (FFG).

We also like to thank H. Hofbauer from Vienna University ofTechnology and the companies Profactor and OMV for theircontribution and support.

References

Badwal, S., Giddey, S., Ciacchi, F., 2006. Hydrogen and oxygen generation withpolymer electrolyte membrane (PEM)-based electrolytic technology. Ionics 12(1), 7e14.

California Department of Food and Agriculture, 2008. Specifications for hydrogenused in internal combustion engines and fuel cells. California Code of Regula-tions Title 4, Division 9, (Chapter 6), Article 8. U.S.A.

Chaffin, J., Bobbio, S., Inyang, H., Kaanagbara, L., 2006. Hydrogen production byplasma electrolysis. Journal of Energy Engineering 132 (3), 104e108.

Chen, L., Hong, Q., Lin, J., Dautzenberg, F.M., 2007. Hydrogen production by coupledcatalytic partial oxidation and steam methane reforming at elevated pressureand temperature. Journal of Power Sources 164 (2), 803e808.

Chu, C.-F., Li, Y.-Y., Xu, K.-Q., Ebie, Y., Inamori, Y., Kong, H.-N., 2008. A pH- andtemperature-phased two-stage process for hydrogen and methane productionfrom food waste. International Journal of Hydrogen Energy 33 (18), 4739e4746.

Claassen, P., Groenestijn, J., Janssen, A., van Niel, E., Wijffels R. Feasibility of bio-logical hydrogen production from biomass for utilization in fuel cells. In: 1stworld Conference and Exhibition on biomass for energy, Industry and Climatechange Protection, Sevilla, 2000, pp. 1665e1667.

Claassen, P., Budde, M., van Niel, E., de Vrije, T., 2005. Utilization of Biomass forHydrogen Fermentation. Biofuels for Fuel Cells: Renewable Energy fromBiomass Fermentation. IWA Publishing, London, pp. 221e30.

Corbo, P., Migliardini, F., 2007. Hydrogen production by catalytic partial oxidation ofmethane and propane on Ni and Pt catalysts. International Journal of HydrogenEnergy 32 (1), 55e66.

Corbo, P., Migliardini, F., 2009. Natural gas and biofuel as feedstock for hydrogenproduction on Ni catalysts. Journal of Natural Gas Chemistry 18 (1), 9e14.

Dalle Nogare, D., Baggio, P., Tomasi, C., Mutri, L., Canu, P., 2007. A thermodynamicanalysis of natural gas reforming processes for fuel cell application. ChemicalEngineering Science 62 (18e20), 5418e5424.

de Jong, M., Reinders, A.H.M.E., Kok, J.B.W., Westendorp, G., 2009. Optimizinga steam-methane reformer for hydrogen production. International Journal ofHydrogen Energy 34 (1), 285e292.

de Souza, R.F., Padilha, J.C., Gonçalves, R.S., de Souza, M.O., Rault-Berthelot, J., 2007.Electrochemical hydrogen production from water electrolysis using ionic liquidas electrolytes: towards the best device. Journal of Power Sources 164 (2),792e798.

Grigoriev, S.A., Porembsky, V.I., Fateev, V.N., 2006. Pure hydrogen production byPEM electrolysis for hydrogen energy. International Journal of Hydrogen Energy31 (2), 171e175.

Harasek, M., Makaruk, A., Miltner, M., Schlager, R., 2007. Process Dynamics ofa Biogas Upgrading System e Simulation and Experimental Practice. AIChEAnnual Meeting, Salt Lake City.

Hoang, D.L., Chan, S.H., Ding, O.L., 2006. Hydrogen production for fuel cells byautothermal reforming of methane over sulfide nickel catalyst on a gammaalumina support. Journal of Power Sources 159 (2), 1248e1257.

Hofbauer, H., Rauch, R., Fürnsinn, S., Aichernig, C., 2006. Power center Güssing.Berichte aus Energie- und Umweltforschung, 79/2006. Bundesministerium fürVerkehr, Innovation und Technologie (in German).

Institut für Energetik und Umwelt, 2006. Bundesforschungsanstalt für Land-wirtschaft, Kuratorium für Technik und Bauwesen in der Landwirtschaft.(Assistance: Production and Utilization of Biogas). Fachagentur für Nachwach-sende Rohstoffe e.V (in German).

Ivy, J., 2004. Summary of Electrolytic Hydrogen Production eMilestone CompletionReport. National renewable energy laboratory. U.S. Department of Energy.

Kaila, R.K., Krause, A.O.I., 2006. Autothermal reforming of simulated gasoline anddiesel fuels. International Journal of Hydrogen Energy 31 (13), 1934e1941.

Könnölä, T., Unruh, G.C., Carrillo-Hermosilla, J., 2007. Toward prospective voluntaryagreements: reflections from a hydrogen foresight project. Journal of CleanerProduction 15 (3), 259e265.

A. Miltner et al. / Journal of Cleaner Production 18 (2010) S51eS62S62

Kreuter,W.,Hofmann,H.,1998.Electrolysis: the importantenergy transformerinaworldof sustainable energy. International Journal of Hydrogen Energy 23 (8), 661e666.

Kyazze, G., Dinsdale, R., Guwy, A.J., Hawkes, F.R., Premier, G.C., Hawkes, D.L., 2007.Performance characteristics of a two-stage dark fermentative system producinghydrogen and methane continuously. Biotechnology and Bioengineering 97 (4),759e770.

Markowski, M., Urbaniec, K., Trafczynski, M., Budek, A., Wukovits, W., Friedl, A.,Ljunggren, M., Zacchi, G. Heat integration of a plant to process biomass tohydrogen. PRES 2008, included in CD-ROM of Full Texts of CHISA 2008, Praha,2008. Lecture K8.1.

Marshall, A., Børresen,B., Hagen,G., Tsypkin,M., Tunold,R., 2007.Hydrogenproductionby advanced proton exchange membrane (PEM) water electrolysers-Reducedenergy consumption by improved electrocatalysis. Energy 32 (4), 431e436.

Miltner, M., Makaruk, A., Harasek, M. Application of gas permeation for biogasupgrade e operational experiences of feeding biomethane into the Austrian gasgrid. In: 16th European Conference on Biomass for Energy, Industry and ClimateProtection, Valencia, 2008a.

Miltner, A., Friedl, A., Wukovits, W., Schnitzhofer, W., Ahrer, W., Pollak, K. Hydrogenfor the automotive sector: evaluation of performance and economy of possibleproduction pathways for Austria. PRES 2008, included in CD-ROM of Full Textsof CHISA 2008, Praha, 2008b. Lecture K7.11.

Modigell, M., Schumacher, M., Beggel, F., Guichard, J., Sequeira, S.-E., Teplyakov,V.M., Zenkevitch, V.B. Evaluation of different gas upgrading technologies inbiohydrogen production processes. PRES 2008, included in CD-ROM of FullTexts of CHISA 2008, Praha, 2008. Lecture K2.7.

Murata, K., Saito, M., Inaba, M., Takahara, I., 2007. Hydrogen production by auto-thermal reforming of sulfur-containing hydrocarbons over re-modified Ni/Sr/ZrO2 catalysts. Applied Catalysis B: Environmental 70 (1e4), 509e514.

Perz, E.A., 1991. Computer method for thermal power cycle calculation. Journal ofEngineering for Gas Turbines and Power 113 (2), 184e189.

Perz, E.W., Bergmann, S., 2003. IPSEpro V4.0 user Documentation. Simtech, Graz.Perz, E.W., Bergmann, S., 2007. A simulation environment for the techno-economic

performance prediction of water and power cogeneration systems usingrenewable and fossil energy sources. Desalination 203 (1e3), 337e345.

Pfeifer, C., Puchner, B., Hofbauer, H., 2007. In-situ CO2-absorption in a dual fluidizedbed biomass steam gasifier to produce a hydrogen rich syngas. InternationalJournal of Chemical Reactor Engineering 5 (A9).

Pröll, T., Rauch, R., Aichernig, C., Hofbauer, H., 2005a. Fluidized bed steam gasifi-cation of solid biomass e analysis and optimization of plant operation usingprocess simulation. In: 18th International Conference on Fluidized BedCombustion. ASME, Toronto Paper FBC2005e 78129.

Pröll, T., Siefert, I.G., Friedl, A., Hofbauer, H., 2005b. Removal of NH3 from biomassgasification producer gas by water condensing in an organic solvent scrubber.Industrial & Engineering Chemistry Research 44 (5), 1576e1584.

Pröll, T., Rauch, R., Aichernig, C., Hofbauer, H., 2007. Fluidized bed steam gasifi-cation of solid biomass e performance characteristics of an 8 MWth combinedheat and power plant. International Journal of Chemical Reactor Engineering5 (A54).

Pulletz, G., Böhme, W., 30.09.2003. Hydrogen in the Mobile Sector. Presented at“Energiegespräche”. Austrian Energy Agency (in German).

Rabe, S., Truong, T.-B., Vogel, F., 2007. Catalytic autothermal reforming of methane:performance of a kW scale reformer using pure oxygen as oxidant. AppliedCatalysis A: General 318, 54e62.

Rauch, R., 2005. Indirectly Heated Gasifiers e the Case of the Guessing Reactor. 1stEuropean Summer School on renewable Motor fuels, Birkenfeld, Deutschland.

Roy, A., Watson, S., Infield, D., 2006. Comparison of electrical energy efficiency ofatmospheric and high-pressure electrolysers. International Journal of HydrogenEnergy 31 (14), 1964e1979.

Seemann, M., Biollaz S., Stucki S., Schaub M., Aichernig C., Rauch R., Hofbauer H.,Koch R. Methanation of BioSyngas and Simultaneous low-temperaturereforming: fist results of long Duration Tests at the FICFB gasifier in Güssing. In:14th European biomass Conference, Paris, 2005.

Stojic, D.L., Marceta, M.P., Sovilj, S.P., Miljanic, S.S., 2003. Hydrogen generation fromwater electrolysisepossibilities of energy saving. Journal of Power Sources 118(1e2), 315e319.

Willquist, K., Claassen, P.A.M., van Niel, E.W.J., 2009. Evaluation of the influence ofCO2 on hydrogen production by Caldicellulosiruptor saccharolyticus. Interna-tional Journal of Hydrogen Energy 34 (11), 4718e4726.

Zeng, K., Zhang, D., 2010. Recent progress in alkaline water electrolysis for hydrogenproduction and applications. Progress in Energy and Combustion Science 36 (3),307e326.

Zhu, H., Stadnyk, A., Béland, M., Seto, P., 2008. Co-production of hydrogen andmethane from potato waste using a two-stage anaerobic digestion process.Bioresource Technology 99 (11), 5078e5084.