Scholarly article on topic 'Achievements of European projects on membrane reactor for hydrogen production'

Achievements of European projects on membrane reactor for hydrogen production Academic research paper on "Chemical engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Journal of Cleaner Production
OECD Field of science
Keywords
{BIONICO / FERRET / FluidCELL / "Fluidized membrane reactor" / "Hydrogen separation" / "Biofuels and natural gas"}

Abstract of research paper on Chemical engineering, author of scientific article — Gioele Di Marcoberardino, Marco Binotti, Giampaolo Manzolini, José Luis Viviente, Alba Arratibel, et al.

Abstract Membrane reactors for hydrogen production can increase both the hydrogen production efficiency at small scale and the electric efficiency in micro-cogeneration systems when coupled with Polymeric Electrolyte Membrane fuel cells. This paper discusses the achievements of three European projects (FERRET, FluidCELL, BIONICO) which investigate the application of the membrane reactor concept to hydrogen production and micro-cogeneration systems using both natural gas and biofuels (biogas and bio-ethanol) as feedstock. The membranes, used to selectively separate hydrogen from the other reaction products (CH4, CO2, H2O, etc.), are of asymmetric type with a thin layer of Pd alloy (<5 μm), and supported on a ceramic porous material to increase their mechanical stability. In FERRET, the flexibility of the membrane reactor under diverse natural gas quality is validated. The reactor is integrated in a micro-CHP system and achieves a net electric efficiency of about 42% (8% points higher than the reference case). In FluidCELL, the use of bio-ethanol as feedstock for micro-cogeneration Polymeric Electrolyte Membrane based system is investigated in off-grid applications and a net electric efficiency around 40% is obtained (6% higher than the reference case). Finally, BIONICO investigates the hydrogen production from biogas. While BIONICO has just started, FERRET and FluidCELL are in their third year and the two prototypes are close to be tested confirming the potentiality of membrane reactor technology at small scale.

Academic research paper on topic "Achievements of European projects on membrane reactor for hydrogen production"

Accepted Manuscript

Achievements of European projects on membrane reactor for hydrogen production

Gioele Di Marcoberardino, Marco Binotti, Giampaolo Manzolini, José Luis Viviente, Alba Arratibel, Leonardo Roses, Fausto Gallucci

PII: DOI:

Reference: To appear in:

S0959-6526(17)31062-4 10.1016/j.jclepro.2017.05.122 JCLP 9661

Journal of Cleaner Production

Received Date: Revised Date: Accepted Date:

06 January 2017 10 May 2017 22 May 2017

Please cite this article as: Gioele Di Marcoberardino, Marco Binotti, Giampaolo Manzolini, José Luis Viviente, Alba Arratibel, Leonardo Roses, Fausto Gallucci, Achievements of European projects on membrane reactor for hydrogen production, Journal of Cleaner Production (2017), doi: 10.1016/j. jclepro.2017.05.122

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Word count: 4803 Character count: 27265

Achievements of European projects on membrane reactor for

hydrogen production

Gioele Di Marcoberardino1(*), Marco Binotti1, Giampaolo Manzolini1, José Luis Viviente2, Alba Arratibel2,3, Leonardo Roses4, Fausto Gallucci3

1Politecnico di Milano, Dipartimento di Energia, via Lambruschini 4, 20156, Milano, Italy 2TECNALIA Research&Innovation, Mikeletegi Pasealekua 2, 20009, San SebastianDonostia, Spain

3Chemical Process Intensification, Department of Chemical Engineering and Chemistry, Eindhoven University of Technology, P.O. Box 513, 5612 AZ Eindhoven, The Netherlands 4HyGear B.V., Westervoortsedijk 73, P.O. Box 5280, 6802 EG Arnhem, The Netherlands

(*) corresponding author. Tel. +39 02 2399 3935, e-mail: gioele.dimarcoberardino@polimi.it ABSTRACT

Membrane reactors for hydrogen production can increase both the hydrogen production efficiency at small scale and the electric efficiency in micro-cogeneration systems when coupled with Polymeric Electrolyte Membrane fuel cells. This paper discusses the achievements of three European projects (FERRET, FluidCELL, BIONICO) which investigate the application of the membrane reactor concept to hydrogen production and micro-cogeneration systems using both natural gas and biofuels (biogas and bio-ethanol) as feedstock. The membranes, used to selectively separate hydrogen from the other reaction products (CH4, CO2, H2O, etc.), are of asymmetric type with a thin layer of Pd alloy (<5 |im), and supported on a ceramic porous material to increase their mechanical stability. In FERRET, the flexibility of the membrane reactor under diverse natural gas quality is validated. The reactor is integrated in a micro-CHP system and achieves a net electric efficiency of about 42% (8% points higher than the reference case). In FluidCELL, the use of bio-ethanol as feedstock for micro-cogeneration Polymeric Electrolyte Membrane based system is investigated in off-grid applications and a net electric efficiency around 40% is obtained (6% higher than the reference case). Finally, BIONICO investigates the hydrogen production from biogas. While BIONICO has just started, FERRET and FluidCELL are in their third year and the two prototypes are close to be tested confirming the potentiality of membrane reactor technology at small scale.

KEYWORDS

BIONICO; FERRET; FluidCELL; Fluidized membrane reactor; hydrogen separation; biofuels and natural gas.

1 INTRODUCTION

The constant increase of electricity and heat demand has led to an intensification of fossil

fuels utilization that today, despite the impressive development of renewables, still account

for more than 80% of the overall primary energy consumptions worldwide (International Energy Agency, 2015). Carbon dioxide emissions associated with the utilization of fossil fuels are considered one of the main responsible for CO2 concentration rise in the atmosphere

and of the consequent greenhouse effect. Several options for anthropogenic CO2 emissions reduction are being investigated: Van Vuuren et al. (2007) although accepting the absence of a silver bullet, indicate carbon capture and storage (CCS) as the most attractive technology. The same vision is also share by Wennersten et al. (2015) that underline how effective communication is crucial to favour public acceptance of this technology. Yong et al. (2016) explore clean electricity production systems either based on renewables or implementing CCS, Karschin and Geldermann (2015) propose and optimize local bioenergy production and distribution systems as function of biomass availability, number of heat customers and heat loss in the system. Dovi et al. (2009) focus on systems based on renewables and on the production of biofuels and hydrogen. As suggested by Dovi et al. (2009) and supported by Maack and Skulason (2006), hydrogen can replace fossil fuels in power generation and transportation in the long term. Hydrogen production should be based on electrolysis exploiting renewable electricity (wind, PV) or on biofuels as biogas. When moving to the short period, one of the most interesting options to reduce CO2 emissions in the residential sector is the combined heat and power generation on micro scale (micro-CHP) that increases fuel exploitation compared to conventional separated heat and electricity production. Alanne and Saari (2004) outlined the potentiality of micro-CHP systems comparing in terms of performance and costs four different technologies (reciprocating engines, Stirling-engines, Fuel Cells, Micro Turbines), while Campanari et al. (2009) evaluated that micro-CHP systems based on fuel cells applied to single-user residential applications could achieve 25% primary energy saving. The advantages in terms of efficiency and CO2 savings are even larger when the electricity is produced by polymer electrolyte membrane (PEM) fuel cells based systems: PEM offer an efficient source of electricity with high primary energy saving potential at micro-cogeneration scale (Campanari et al., 2009a). However, they require non-contaminated (ultra-pure) hydrogen as fuel (CO tolerance below 10 ppm).

Both in short and long term perspectives, the development of efficient and cost effective hydrogen fuel processors is therefore crucial (Taanman et al., 2008). Membrane reactors (MR) for hydrogen production are foreseen as a promising technology for process intensification. Gallucci et al. (2013) reviewed the most suitable membrane materials for hydrogen separation together with commercially available concepts and Paglieri and Way (2002) focused on Palladium membranes for hydrogen separation. The main advantage of membrane reactors is the capability of producing and separating hydrogen in a single reactor with thermodynamic and economic benefits. Recent studies demonstrate that membrane reactors overcome the performance of conventional fuel processors (Di Marcoberardino et al., 2016c; Foresti and Manzolini, 2016). Campanari et al. (2009) calculated a 25% electric efficiency increase when replacing conventional fuel processors based on steam reforming, water gas shift and preferential oxidizer with a membrane reactor. The same advantage was confirmed by Di Marcoberardino et al. (2016), while the adoption of membrane reactor when using bio-ethanol as feedstock has an efficiency increase of about 15% than conventional configurations (Foresti and Manzolini, 2016).

Between the fixed bed and fluidized bed configurations, the latter can significantly reduce heat and mass transfer issues related to reforming reaction and H2 permeation. Fluidization impact on catalyst and membrane and has been explored (Roses et al., 2011). Three EU projects FERRET (Gallucci et al., 2014), FluidCELL (Viviente et al., 2014), BIONICO (Binotti et al., 2015) are investigating the application of the membrane reactor concept to hydrogen production and micro-CHP systems using different fuels. BIONICO project focuses on hydrogen production from biogas produced by landfill or anaerobic digestors. FERRET and FluidCELL deal with the integration of the fluidized membrane reactor to PEM based micro-CHP systems: in FERRET, the flexibility of the membrane

reactor with respect to the diverse NG quality is validated, while in FluidCELL the performance of an off-grid micro-CHP system using bio-ethanol as fuel is evaluated. This work outlines the advantages of membrane reactor in each of the considered applications through the main achievements within the three projects. The paper is organized as follows: a dedicated section will describe the membrane reactor concept and its characteristics and then, three separate sections will summarize each project goals and results.

2 MEMBRANE REACTOR CONCEPT

Conventional fuel processors for hydrogen production or combined with PEM micro-cogeneration systems are usually based on the concept shown in Figure 1. Hydrogen production and purification from hydrocarbons requires several steps carried forward in different reactors: (a) a reformer operating at high temperatures1 either autothermal (ATR) or conventional steam reforming (SR), followed by (b) two water gas shift reactors (WGSR), one at high and one at low temperatures to enhance carbon monoxide conversion to hydrogen, and (c) a final purification step that depends on the hydrogen application. In case of H2 utilization in PEM based micro-CHP systems, a PReferential Oxidizer (PROX) is required to oxidize CO and obtain a final composition at the outlet with 40^60 %mol of H2 diluted with CO2, steam and N2. The net electric and thermal efficiencies of commercial systems based on the abovementioned technology are in the range of 32% and 60% respectively [15]. In case of hydrogen production, the purification is carried out in a Pressure Swing Adsorption system (PSA) splitting a pure hydrogen stream from the other gases. The reference hydrogen production efficiency is of around 60% for a system size of 100 kgH2/day (Di Marcoberardino et al., 2016b).

Figure 1. Conventional fuel processors for hydrogen production

In this work, the adoption of hydrogen selective membrane in an autothermal reforming reactor is considered (see Figure 2). The fluidized bed configuration is preferred thanks to the more uniform temperature achieved. Additionally, bed-to-wall mass transfer limitations, often very detrimental for packed bed membrane reactors, are largely reduced. The fluidized membrane reactor advantages with respect to a conventional fuel processor are the following:

• the entire production and purification process is carried out in one reactor, the membrane reactor;

1 The reforming temperature depends on fuel type. NG and biogas requires temperatures above 800°C while, in the ethanol case, temperature are set around 600°C

• hydrogen separation with membranes drives the reaction conversion towards product side; so the same fuel conversion of conventional batch processes can be achieved at lower operating temperatures;

• the extent of hydrocarbon conversion can be defined in the design phase by means of adequate selection of stream pressures and membrane area.

In the micro-CHP cases, the membrane separates pure hydrogen which can directly feed the fuel cell. Feeding pure hydrogen compared to the diluted one increases the electric conversion efficiency: Minutillo et al. (2008) measured a decrease in the output voltages (about 8-10%) when reformate was used instead of pure hydrogen. Through a detailed PEM modelling (Minutillo and Perna, 2008), it was shown that the presence of carbon dioxide does not only dilute hydrogen, but leads also to carbon monoxide formation of by reverse WGS reaction.

+ Sweep gas

Sweep gas

Retentate

Pd-Ag membranes

Catalyst

Û .o Pre-Reforming zone

« rt t « P -Oxidation zone

Q « G e> «a

Distributor plate

Figure 2. Schematic of a membrane reactor

The higher CH4 conversion and H2 separation factor (SF) of membrane reactors with respect to conventional configurations (w/o membranes) are outlined in Figure 3. Experimental tests are carried out feeding a mixture of pure methane, steam and air for ATR at 550°C and S/C equal to 3 varying the operating pressure between 2 and 4 bar. The SF and the O/C ratio are defined as follows:

X1 ' Fc^

i 2i + 2

Hjperm

r H 2>r et r H 2>p er m

Figure 3 shows as the adoption of the membrane reactor (w/ Mem. case) increases the methane conversion (therefore the hydrogen production) more than 15% with respect to the conventional fuel processor (w/o Mem. case).

On the other hand, to enhance the hydrogen permeation, two different reactor configurations can be designed: with vacuum or with sweep gas at the permeate side, both aiming at decreasing hydrogen partial pressure in order to increase the hydrogen flux and consequently reduce the membrane required area. Further information about the two configurations using NG as feedstock can be found in (Di Marcoberardino et al., 2016c), while the bio-ethanol case is discussed in (Foresti and Manzolini, 2016).

Steam Reforming

o w/ Mem. (Exp.) □ w/o Mem. (Exp.)

-w/ Mem. (Th.) w/o Mem. (Th.)

Pressure (bar)

AutoThermal Reforming

o w/ Mem. (Exp.) -w/ Mem. (Th.)

□ w/o Mem. (Exp.)----w/o Mem. (Th.)

Pressure (bar)

Figure 3. Methane conversion (top) and Separation Factor (bottom) for SR and ATR with (w/) and without (w/o) membranes at different pressures. Experiments carried out at 550°C, S/C=3

and, for ATR, 0/C=0.25.

169 The main characteristics required by the membranes are (i) mechanical and thermal stability,

170 (ii) high hydrogen fluxes and (iii) high perm-selectivity (defined as the ratio between the

171 permeance of hydrogen and the permeance of other species). The technology identified by

172 Tecnalia which addresses all these three requirements is based on asymmetric supported

173 membranes made by Pd alloy. The support is either made of ZrO2 with 100 nm pore size as in

174 FERRET project or alumina supported as in FluidCELL with a 1 to 4 p,m thick Pd-Ag layer:

175 the lower the thickness the higher the flux, but with penalties from selectivity point of view.

176 The adoption of thin Pd-Ag layers requires supports with low roughness having small pores

177 (<200 nm) with uniform pore size distributions, therefore dedicated efforts in the projects are

178 devoted to the support manufacturing process. The developed membranes present H2/N2

179 perm-selectivities larger than 7000 and nitrogen permeance < 8 x 10-11 mol m-2 s-1 Pa-1 at 300

180 °C. The performance of the membranes adopted in FluidCELL and FERRET are aligned with

181 the ones of other manufacturers as reported in two review works (Fernandez et al., 2015a,

182 2015b).

183 In FERRET, 30 membranes of 25 cm length were manufactured while in FluidCELL the

184 number of membranes is 37 and they are 50 cm long. The two membrane types are shown in

185 Figure 4. In BIONICO, the presence of H2S in the biogas feedstock requires the adoption of

186 different Pd alloy currently under investigation. A preliminary calculation assessed that

187 around 100 membranes 50 cm long should be manufactured to guarantee a hydrogen

188 production of 100 kg/day.

190 Figure 4. FERRET (top) and FluidCELL (bottom) membranes to be inserted in the prototype

191 reactor

193 3 FERRET: Flexible micro-CHP system

194 The FERRET project, started in April 2014, aims at developing a micro-CHP system flexible

195 towards the variability of NG compositions in the European Union. The NG compositions

196 variation affects the design of the system, its performance and/or its lifetime. Four different

197 compositions, representative of the different European biogases (see Table 1), were selected

198 for assessing the system flexibility. The UK composition features an average NG, the Italian

199 case is almost pure methane, while the NL and the ES cases have the minimum and maximum

200 Wobbe index respectively. In addition, the considered compositions vary in terms of inert

201 concentration: inert gases reduce the H2 fraction and thus the permeation driving force across

202 the membrane. The definitions of H2 potential , PtH2, and Wobbe index, WI,are as follows.

210 211 212

P*h. =

mol H2

2 mol NG HHV

Table 1. Natural Gas compositions

Species units NG type

NL UK IT ES

CH4 %mol 81.230 92.070 99.581 81.570

C2H6 %mol 2.850 3.405 0.056 13.380

C3H8 %mol 0.370 0.761 0.021 3.670

n-C4H10 %mol 0.080 0.177 0.002 0.400

i-C4H10 %mol 0.060 0.140 0.006 0.290

n-C5H12 %mol 0.020 0.048 0 0

i-C5H12 %mol 0.020 0.061 0.002 0

C6+ %mol 0.080 0.090 0.007 0

CO2 %mol 0.890 0.865 0.029 0

N2 %mol 14.400 2.375 0.296 0.690

LHV MJ/kg 38.0 46.7 49.7 48.6

PtH2 moWmolNG 3.52 4.07 3.99 4.66

WI MJ/Nm3 43.6 52.0 53.1 56.6

The considered layout of the micro-CHP system is shown in Figure 5. It shows two streams of water pumped by P-1 and P-2 for feed and sweep, respectively. Liquid feed water is mixed with compressed air, then evaporated through HX-0 and HX-1 and finally superheated in HX-2. At the reactor inlet , compressed NG and a preheated mix of air and steam are fed from the bottom section. Sweep gas is evaporated through HX-3 and HX-4 and fed separately to the reactor as shown in Figure 5. At the outlet, the retentate and hydrogen exit from the top section of the reactor. After cooling, the remaining fuel in the retentate is combusted in the burner to generate steam in HX-1 and HX-3 closing the cycle. A closed loop for heat recovery which includes HX-7, the fuel cell (HX-9) and HX-6 is designed. Process water is recycled

220 221 222

via condensation in three separators downstream permeate, retentate and exhaust gases cooling.

The system performance was assessed using both UK and NL natural gas compositions outlining their impact on the net electric efficiency. In general, the net electric efficiency is 10% points higher than commercially available micro-CHP systems based on PEM fuel cell of the same size. Focusing on NG composition impact, the results indicate 1% higher electric efficiency for UK case, as consequence of the different NG compressor consumption (lower volumetric flow thanks to the higher WI), and a 7% membrane area reduction due to lower inert gases concentration. Starting from these two system designs, the performances at different NG compositions are evaluated. The NL case, that has the high inert gases concentration, was selected as reference for the system design and the overall performance for different NG composition was assessed (Figure 6). The rationale behind the selection of the worst case (NL composition) as reference is to have a slight increase of the net electric efficiency with other NG compositions, and, in addition, an easier reactor control (Di Marcoberardino and Manzolini, 2017). Limited efficiency variation are obtained, hence demonstrating the flexibility of the membrane reformer.

Flue gas

ATR-MR (600 °C)

Cooling circuit (Rgv)

Exhaust

Figure 5. FERRET Layout of micro-CHP system using sweep gas (Di Marcoberardino and

Manzolini, 2017)

Table 2. ATR-MR system performance using different NG (Di Marcoberardino and Manzolini, 2017)_

Results units UK NL

S/C and Temperature reactor - / °C 2.5/600 2.5/600

Pressure reaction/permeate side bar 8/1.3 8/1.3

NG feed Nm3/h 1.19 1.38

NG power input [LHV base] kW 12.06 12.13

NG power input [HHV base] kW 13.35 13.44

Net AC power output kW 5.00 5.00

Fuel Cell AC power output kW 6.09 6.12

Balance of plant kW 1.09 1.12

Thermal recovery kW 6.78 6.81

Net electric efficiency [LHV base] %LHV 41.48 41.21

Net electric efficiency [HHV base] %HHV 37.45 37.19

Net thermal efficiency %LHV 56.23 56.12

Total efficiency [LHV base] %LHV 97.71 97.33

Total efficiency [HHV base] %HHV 88.22 87.83

Total membrane area m2 0.264 0.283

H2 production/permeation Nm3/h 3.42 3.44

HRF % 92.0 92.1

□ H2 perm ONet El. eff.

— 3.44 A .c

| 3.43 -¥

a! 3.42 A a.

1 3.41 -3.40

UK 8 bar

ES 8 bar

- 41.8%

- 41.6% £

- 41.4% ^

- 41.2% 41.0%

□ Net AC Pow. □ FC gross Pow. DC O NG inlet 6200 -i-r 1

6000 -5800 -5600 -5400 -5200 -5000 -4800

UK 8 bar

□ IT

ES 8 bar

- 1 - 1 - 1 - 1 - 1 - 0. 0.

Figure 6. NL base case with sweep: micro-CHP flexibility under different NG qualities (Di

Marcoberardino et al., 2016a)

Once defined the reference NG composition, the system lay-out and operating parameters, the design was finalized and the construction of the ATR membrane reformer prototype was started. The flexible fuel processor was designed to work with different fuels with WI ranging from 43.6 to 56.6 MJ/Nm3 and with hydrogen potential from 3.52 to 4.66 moln2/mol\(l. A picture of the membrane reactor is shown in Figure 7.

Figure 7. FERRET ATR membrane reactor

252 4 FluidCELL: Advanced Bio-Ethanol micro-CHP

253 FluidCELL project, started in April 2014, aims at developing a high efficient m-CHP system

254 based on PEM fuel cell and integrating a low temperature fluidized membrane reactor. The

255 system is fuelled with bio-ethanol and intended for off-grid applications. A pre-commercial

256 system developed by Helbio, based on conventional fuel processor, is rated 5 kWel with an

257 electric efficiency of 22.5% (Rossetti et al., 2012): detailed simulations of the system showed

258 an increase of the electric efficiency up to 31% when improving heat integration (Rossetti et

259 al., 2015).

260 The layout developed in FluidCELLis depicted in Figure 8. The W/EtOH mixture feed is

261 pumped through a series of heat exchangers: it evaporates cooling the exhausts (HX1) and is

262 then superheated by the retentate flow (HX2). No additional fuel for reactants heating is

263 considered, thus the retentate combustion energy must cover the whole feed heat duty. Air is

264 compressed and directly fed to the reactor. The permeate gas is cooled evaporating the sweep

265 water (HX3), to prevent hydrogen contamination in case of leakages. Then, the sweep stream

266 is heated up to the reactor temperature by the exhaust gases (HX4). Finally, the retentate flow

267 is cooled down to 120°C, throttled and combusted.

Figure 8. FluidCELL Layout with sweep gas (Foresti and Manzolini, 2016)

Main results of the simulations for the FluidCELL layout, together with the reference case based on conventional fuel processor (SR) are summarized in Table 3. The adoption of membrane reactor increases the net electric efficiency by 7% points with respect to the reference case calculated using the same assumptions. On the contrary, the thermal efficiency is lower for the FluidCELL case because the higher electric efficiency reduces the heat that can be recovered.

Table 3. Simulation results on performance of the ATR-MR system (Foresti and Manzolini, _2016)_

Results units Reference case FluidCELL Case Sweep gas

W/EtOH, Temperature reactor - 6/600°C 3.6/500°C 4.2/500°C 3.6/500°C

Pressure reaction/permeate side bar 2 12/1.3 12/1.3 16/1.3

EtOH power input [LHV base] kW 12.29 12.73 12.44 12.44

EtOH power input [HHV base] kW 13.62 14.10 13.78 13.78

Net AC power output kW 5.00 5.00 5.00 5.00

Gross DC power output kW 5.44 5.65 5.65 5.68

Balance of plant kW 0.44 0.65 0.65 0.068

Thermal recovery kW 9.08 6.54 6.89 6.61

Net electric efficiency [LHV base] %LHV 33.1 40.6 39.3 40.2

Net electric efficiency [HHV base] %HHV 29.9 36.6 35.5 36.3

Net thermal efficiency [LHV base] %LHV 63.0 53.1 54.2 53.3

Net thermal efficiency [HHV base] %HHV 57.1 47.9 48.9 48.1

Total membrane area m2 - 0.37 0.30 0.22

H2 production/permeation Nm3/h 4.12 3.18 3.18 3.20

HRF % - 65.9 63.8 65.7

Compared to the FERRET project which uses NG case, the membrane area of FluidCELL case is larger even with a higher feed pressure. This is due to the lower reactor temperature (500°C vs 600°C) which reduces the hydrogen partial pressure from 1.7 bar to 1.1 bar

286 together with the membrane permeance, being the former the most relevant aspect. Once

287 defined the system operating conditions and membrane area requirements, the membrane

288 reactor manufacturing started and it is now complete. A sketch of the fuel processor is shown

289 in Figure 9.

291 ^m

292 Figure 9. FluidCELL ATR membrane reactor

294 5 BIONICO: Biogas MR for decentralized H2 production

295 The BIONICO project, started in September 2015, will develop, build and demonstrate a

296 novel reactor concept integrating H2 production and separation in a single vessel in a biogas

297 production plant. The BIONICO pilot plant will be built in an ENC landfill plant in Portugal

298 and is expected to start-up in July 2018. The hydrogen production capacity will be of 100

299 kg/day. The adoption of biogas as fuel input makes the hydrogen produced green and is

300 justified by the remarkable growth of biogas production expected in the next decades.

301 Roughly 10000 biogas plants in agriculture, industry and waste water treatment are in

302 operation in Europe, but the European potential for biogas is still enormous as the production

303 of biogas could be multiplied by a factor of four to five (European Biogas Association, 2015).

304 Typical applications of biogas are the power generation through internal reciprocating engines

305 or upgrading to biomethane by CO2 separation. When moving to hydrogen production from

306 biogas, the conversion process is more complex since biogas can have variable gas

307 compositions depending on primary matter sources. In addition, traditional reforming based

308 conversion technologies are energy and capital intensive since, as already seen, several

309 process steps are involved. The adoption of a membrane fluidized reactor can increase the

310 hydrogen production efficiency up to 70%, reducing at the same time the system complexity

311 (see Figure 10). As term of comparison, conventional processes for hydrogen production from

312 biogas have an efficiency around 64%. The hydrogen purity target is set at 99.99% equal to

313 the one required by automotive applications which is one of the hydrogen production target.

314 The 100 kg/day production capacity is also aligned with the features of automotive refuelling

315 stations. The presence of sulphur in the biogas requires the development of dedicated

membranes capable of dealing with sulphur content up to few ppm. Tecnalia is focusing on thin film Pd-Ag-Au membranes on top of ceramic supports.

Retentate

HYDROGEN

Figure 10. BIONICO system layout

l2^feed

6 CONCLUSIONS

This paper summarized the activities and achievements carried out in three European projects: FERRET, FluidCELL and BIONICO. The three projects have developed a fluidized membrane reactor to enhance the hydrogen production and micro-CHP system efficiencies. The simulations and laboratory experiments confirmed the potentiality of the technology for the considered applications: the hydrogen production can be improved from 60% to more than 70% and the micro-CHP system net electric efficiency fed with natural gas and ethanol can be as high as 42% and 40% respectively, which is around 10% higher than competitive systems based on the same concept. In addition, the hydrogen production in one single reactor reduces the system complexity with further advantages of this technology over the competitive ones.

7 ACKNOWLEDGEMENTS

The research leading to these results has received funding from the Fuel Cells and Hydrogen Joint Technology Initiative under grant agreements No 621181 (FERRET), No 621196 (FluidCELL). BIONICO has received funding from the Fuel Cells and Hydrogen 2 Joint Undertaking under grant agreement No 671459. This Joint Undertaking receives support from the European Union's Horizon 2020 research and innovation programme, Hydrogen Europe and N.ERGHY. The present publication reflects only the author's views and the FCH JU and the Union are not liable for any use that may be made of the information contained therein.

341 8 NOMENCLATURE

Acronyms

ART-MR

FCH JU

LHV Mem MR

Pressure, bar Temperature, °C

Alternate Current Autothermal reformer Autothermal membrane reformer Combined heat and power Direct Current Spain Ethanol

European Union Fuel Cell

Fuel Cell and Hydrogen Joint Undertaking High Heating value [MJ/kg] Hydrogen recovery factor Heat exchanger Italy

Low heating value [MJ/kg] Membrane Membrane reactor NG Natural Gas

NL The Netherlands

O/C Oxygen to carbon molar ratio

P Pump

PEM Polymer electrolyte membrane type

PSA Pressure Swing Adsorption

PROX Preferential oxidizer

S/C Steam to carbon molar ratio

SF Separation Factor

SR Steam reformer

UK United Kingdom

W/EtOH Water to ethanol molar ratio

WGSR Water gas shift reactor

Greek letters

Pi Density of species or mixtures, kg/m3

Subscripts

perm Permeate ret Retentate

343 9 REFERENCES

344 Alanne, K., Saari, A., 2004. Sustainable small-scale CHP technologies for buildings: the basis for

345 multi-perspective decision-making. Renew. Sustain. Energy Rev. 8, 401-431.

346 doi:10.1016/j.rser.2003.12.005

347 Binotti et al., 2015. Biogas membrane reformer for decentralized H2 production.

Campanari, S., Macchi, E., Manzolini, G., 2009a. Membrane reformer PEM cogeneration systems for residential applications - Part B: techno-economic. Asia-Pacific J. Chem. Eng. doi:10.1002/apj .247

Campanari, S., Macchi, E., Manzolini, G., 2009b. Membrane reformer PEM cogeneration systems for residential applications-Part A: full load and partial load simulation. Asia-Pacific J. Chem. Eng. 4, 301-310. doi:10.1002/apj.246 Di Marcoberardino, G., Gallucci, F., Manzolini, G., van Sint Annaland, M., 2016a. Definition of validated membrane reactor model for 5 kW power output CHP system for different natural gas compositions. Int. J. Hydrogen Energy 41, 19141-19153. doi:10.1016/j.ijhydene.2016.07.102 Di Marcoberardino, G., Manzolini, G., 2017. Investigation of a 5 kW micro-CHP PEM fuel cell based system integrated with membrane reactor under EU natural gas quality. Int. J. Hydrogen Energy. Di Marcoberardino, G., Manzolini, G., Binotti, M., 2016b. D2.2 - Definition of the reference case. Di Marcoberardino, G., Roses, L., Manzolini, G., 2016c. Technical assessment of a micro-cogeneration system based on polymer electrolyte membrane fuel cell and fluidized bed autothermal reformer. Appl. Energy 162, 231-244. doi:10.1016/j.apenergy.2015.10.068 Dovi, V.G., Huisingh, D., Klemes, J.J., 2009. Cleaner energy for sustainable future. J. Clean. Prod. 17,

889-895. doi:10.1016/j.jclepro.2009.02.001 European Biogas Association, 2015. EBA Annual Report 2015.

Fernandez, E., Coenen, K., Helmi, A., Melendez, J., Zuniga, J., Pacheco Tanaka, D.A., van Sint Annaland, M., Gallucci, F., 2015a. Preparation and characterization of thin-film Pd-Ag supported membranes for high-temperature applications. Int. J. Hydrogen Energy 40, 1346313478. doi:10.1016/j.ijhydene.2015.08.050 Fernandez, E., Helmi, A., Coenen, K., Melendez, J., Viviente, J.L., Pacheco Tanaka, D.A., van Sint Annaland, M., Gallucci, F., 2015b. Development of thin Pd-Ag supported membranes for fluidized bed membrane reactors including WGS related gases. Int. J. Hydrogen Energy 40, 3506-3519. doi:10.1016/j.ijhydene.2014.08.074 Foresti, S., Manzolini, G., 2016. Performances of a micro-CHP system fed with bio-ethanol based on fluidized bed membrane reactor and PEM fuel cells. Int. J. Hydrogen Energy 41, 9004-9021. doi:10.1016/j.ijhydene.2016.03.210 Gallucci, F., Fernandez, E., Corengia, P., van Sint Annaland, M., 2013. Recent advances on membranes and membrane reactors for hydrogen production. Chem. Eng. Sci. 92, 40-66. doi:10.1016/j.ces.2013.01.008 Gallucci et al., 2014. A flexible natural gas membrane reformer for m-CHP applications. International Energy Agency, 2015. Key world energy statistics.

Karschin, I., Geldermann, J., 2015. Efficient cogeneration and district heating systems in bioenergy villages: an optimization approach. J. Clean. Prod. 104, 305-314. doi:10.1016/j.jclepro.2015.03.086 Maack, M.H., Skulason, J.B., 2006. Implementing the hydrogen economy. J. Clean. Prod. 14, 52-64.

doi:10.1016/j.jclepro.2005.05.027 Minutillo, M., Jannelli, E., Tunzio, F., 2008. Experimental Results of a PEM System Operated on

Hydrogen and Reformate. J. Fuel Cell Sci. Technol. 5, 14502. doi:10.1115/1.2784324 Minutillo, M., Perna, A., 2008. Behaviour modelling of a PEMFC operating on diluted hydrogen feed.

Int. J. Energy Res. 32, 1297-1308. doi:10.1002/er.1424 Paglieri, S.N., Way, J.D., 2002. Innovations in Palladium Membrane Research. Sep. Purif. Rev. 31, 1169. doi:10.1081/SPM-120006115 Roses, L., Gallucci, F., Manzolini, G., Campanari, S., van Sint Annaland, M., 2011. Comparison between fixed bed and fluidized bed membrane reactor configurations for PEM based micro-cogeneration systems. Chem. Eng. J. 171, 1415-1427. doi:10.1016/j.cej.2011.05.061 Rossetti, I., Biffi, C., Tantardini, G.F., Raimondi, M., Vitto, E., Alberti, D., 2012. 5kWe+5kWt reformer-PEMFC energy generator from bioethanol first data on the fuel processor from a demonstrative project. Int. J. Hydrogen Energy 37, 8499-8504. doi:10.1016/j.ijhydene.2012.02.095 Rossetti, I., Compagnoni, M., Torli, M., 2015. Process simulation and optimization of H2 production from ethanol steam reforming and its use in fuel cells. 2. Process analysis and optimization. Chem. Eng. J. 281, 1036-1044. doi:10.1016/j.cej.2015.08.045

403 Taanman, M., de Groot, A., Kemp, R., Verspagen, B., 2008. Diffusion paths for micro cogeneration

404 using hydrogen in the Netherlands. J. Clean. Prod. 16. doi:10.1016/j.jclepro.2007.10.010

405 Van Vuuren, D.P., Den Elzen, M.G.J., Lucas, P.L., Eickhout, B., Strengers, B.J., Van Ruijven, B.,

406 Wonink, S., Van Houdt, R., 2007. Stabilizing greenhouse gas concentrations at low levels: An

407 assessment of reduction strategies and costs. Clim. Change 81, 119-159. doi:10.1007/s10584-

408 006-9172-9

409 Viviente et al., 2014. FluidCELL, Advanced m-CHP fuel CELL system based on a novel bio-ethanol

410 Fluidized bed membrane reforemer.

411 Wennersten, R., Sun, Q., Li, H., 2015. The future potential for Carbon Capture and Storage in climate

412 change mitigation - an overview from perspectives of technology, economy and risk. J. Clean.

413 Prod. 103, 724-736. doi:10.1016/j.jclepro.2014.09.023

414 Yong, J.Y., Klemes, J.J., Varbanov, P.S., Huisingh, D., 2016. Cleaner energy for cleaner production:

415 modelling, simulation, optimisation and waste management. J. Clean. Prod. 111, 1-16.

416 doi:10.1016/j.jclepro.2015.10.062