High Temperature PEM Fuel Cell - OSTI.GOV

PSO project: 4760

High Temperature PEM Fuel Cell

Final report

- Public part -

Project identification: FU-nr. 4760, Eltra nr. notat elt2001-231a, Energistyrelsens journalnummer 1723/03-0020 Project duration: 01.01.2004 – 30.11.2006 Coordination: Energy and Materials Science Group, Department of Chemistry, Technical University of Denmark Partners: IRD Fuel Cells A/S Danish Power Systems Aps DONG Energy Authors: Jens Oluf Jensen (DTU) Steen Yde Andersen (IRD) Thibault De Rycke (IRD) Morten Nilsson (DPS) Torkild Christensen (DONG)

2

Abstract The main outcome of the project is the development of stacking technology for high temperature PEMFC stacks based on phosphoric acid doped polybenzimidazole membranes (PBI-membranes) and a study of the potential of a possible accommodation of HT-PEMFC in the national energy system. Stacks of different lengths (up to 40 cells) have been built using two different approaches in terms of plate materials and sealing. The stacks still need maturing and further testing to prove satisfactory reliability, and a steady reduction of production cost is also desired (as in general for fuel cells). However, during the project the process has come a long way. The survey of HT-PEM fuel cells and their regulatory power in the utility system concludes that fuel cells will most likely not be the dominating technique for regulation, but as no other technique has that potential alone, fuel cells are well suited to play a role in the system provided that the establishment of a communication system is not too complicated. In order to maintain an efficient power system with high reliability in a distributed generation scenario, it is important that communication between TSO (Transmission System Operator) and fuel cells is included in the fuel cell system design at an early stage. Resumé De vigtigste resultater af projetet er udviklinge af stakketeknologier for højtemperatur PEMFC stakke baseret på phosphorsyredopet polybenzimidazol-membraner (PBI-membraner) og et studie af potentialet for indpasning af HT-PEMFC I det danske energisystem. Stakke af forskellig længde (op til 40 celler) er bygget v.h.a. to forskellige koncepter m.h.t. pladematerialer og forseglinger. Stakkene kræver stadig nogen videreudvikling og test for pålidelighed. Løbende omkostningsreduktioner er tillige ønskelige som ved alle brændselsceller. Projektet har dog bragt teknologien et stort skridt videre på disse områder. Udredningen om HT-PEMFC og deres regulerkraft i energisystemet konkluderer, at brændselsceller sandsynligvis ikke vil blive den dominerende regulerteknik, men da ingen anden teknologi har dette potential alene, er brændselsceller velegnede til at spille en rolle i systemet forudsat, at etableringen af et kommunikationssystem ikke bliver for kompliceret. Hvis et effektivt system skal opretholdes med høj pålidelighed i et eldistributions-scenarium er det vigtigt, at kommunikationen mellem TSO (Transmission System Operator) og brændselscellerne inkluderes i brændselscellesystemerne på et tidligt tidspunkt.

3

Table of content Abstract……………………………………………………………….… ……………3 Resumé…………………………………………………………………...……………3 Content………………………………………………………………………………...4 1 Introduction............................................................................................................5

1.1 Original proposal ...........................................................................................5 1.2 Background for HT-PEMFC..........................................................................5

2 Test station.............................................................................................................7 2.1 The need for test stations ...............................................................................7 2.2 Design and construction of test station ..........................................................7

2.2.1 Flow system ...........................................................................................7 2.2.2 Electrical system ..................................................................................10

3 Stacks build in the project....................................................................................12 3.1 Introduction to stacking ...............................................................................12 3.2 Stacking based on the IRD concept .............................................................12

3.2.1 Stack Construction ...............................................................................13 3.2.2 Leak tests .............................................................................................14 3.2.3 Stack tests.............................................................................................16 3.2.4 Conclusion ...........................................................................................18

3.3 Stacking based on the DTU/DPS concept ...................................................19 3.4 Conclusion ...................................................................................................21

4 Adaptation in the energy system..........................................................................22 5 Application of results, business strategy, publication and dissemination............23 Annexes: Annex 1: Stacking of HT-PEMFC by the IRD concept (confidential annex by IRD). Annex 2 Single cells and stacking of HT-PEMFC by the DTU/DPS concept (confidential annex by DTU/DPS). Annex 3 ”Anvendelse af HT PEM-brændselsceller til systemydelser (final report from DONG Energy – in Danish).

4

1 Introduction

1.1 Original proposal The objectives of the present project were to develop a high temperature PEM fuel cell stack (HT-PEMFC stack) for a fuel cell based electricity/heat co-generator system for domestic use. The rated electrical power will be 2 kW and the stack shall operate at temperatures up to 200ºC with a high tolerance for carbon monoxide (CO) in hydrogen. The high temperature makes the cell suited for heat recovery and the CO tolerance allows for the use of reformat hydrogen from natural gas without CO purification. The temperature and the membrane system make the use of conventional PEM stack materials impossible. An important part of the project is therefore to find suitable materials and to develop a stack design for the high temperature system. The rest of the electricity/heat system will not be build during the project. Instead the stack will be tested with a synthetic gas mixture of H2, CO2 and CO as for a reformer/shift gas without the complicated CO purification. The possibilities for mass production and the cost for it will be estimated. Moreover, the prospects of integration of HT-PEM into the Danish system of decentralized heat and power production are looked into.

1.2 Background for HT-PEMFC Today there are great expectations to fuel cells in a variety of fields. Of the many types of fuel cells the polymer fuel cells (PEMFC) is the one closest to commercialization, among other things because the moderate working temperature allows for a broad selection of construction materials. Possible applications range from portable electronics over power back-up and vehicles to CHP (combined heat and power) units and smaller decentralized power plants. It is expected that CHP units for domestic use will be among the first applications hitting the market. Moreover, it is anticipated that such units will provide an inexpensive decentralized production form. So far PEMFC has been based on a polymer membrane of the material Nafion or similar perfluorinated sulphonic acid polymers. Nafion is a good material, but its conductivity depends strongly on high water content. This limits the operating temperature in practical systems to ca. 80ºC because the electrolyte will otherwise dry out and loose conductivity. The low temperature makes the anode catalyst very sensitive to certain impurities like carbon monoxide (CO) in the fuel gas. If high purity hydrogen is used as fuel, CO is not a problem, but in case reformate hydrogen derived from carbon containing fuels like natural gas, it certainly is. The reason is that when organics are steam reformed, traces of CO are always present in the product gas. Even after a low temperature shift reactor about 0.5 % CO is present. A Nafion cell only tolerates CO in the order of 20 ppm (0.002 %). Removal of CO is a complicated and energy consuming process, which adds to complexity and cost of the system. The Nafion cell it therefore most suited for pure hydrogen.

5

Over recent years Department of Chemistry at DTU and Danish Power Systems have been involved in the development of a high temperature electrolyte based on the polymer polybenzimidazole (PBI) doped with phosphoric acid. With this system cells can be operated at temperature up to 200ºC and no water management is needed. Moreover, the tolerance towards CO is outstanding, and cells are hardly affected by 3 % (30,000 ppm) CO in the fuel gas at that temperature. This makes the system ideal for CHP application base on reformate natural gas. No CO clean-up is needed, and the heat produced is more easily utilized due to the high temperature.

6

2 Test station

2.1 The need for test stations Fuel cell test stations are typically dimensioned for a limited range of fuel cell sizes. The important size dependent factors are:

• Mass flow controllers • Electronic load • Heating/cooling system • Water management system • Individual cell voltage monitoring system (for stacks)

The flow controllers (for fuel and air) are calibrated and reliable within a certain range between a minimum and a maximum rate. For electronic mass flow controllers this range is typically in the order of 1 to 20 (e.g. 100 to 2000 ml pr. minute). Accurate flow rates are not crucial when new fuel cell materials are tested in laboratory test cells with high gas stoichiometry. In such tests all performance parameters that are mostly independent on the flow channel design can be studied. The gasses will be present at all times outside the gas diffusion electrode. However, when it comes to stack testing the response to the fuel and air stoichiometry are important parameters because they must be kept as low as possible without sacrificing too much of the performance due to local starvation. This means that for the testing of, e.g. new membranes the flow rates just need to be at a large enough magnitude. DTU has a number of test stations with different size capabilities, but none of them were suitable for stacks in kW sizes. The need was a test station with the capability for testing fuel cells with 256 cm2 electrodes ranging from single cells to stacks in the order of 50 cells with measurement of individual cell voltages. The flow rates should be accurate in the whole dynamic rage of the cell or stack tested. An oil circuit for initial heating and subsequent cooling should be integrated.

2.2 Design and construction of test station

2.2.1 Flow system The flow diagram of the test station is shown in Figure 1. Fuel and air are supplied to the stack via two arrays of mass flow controllers with different ratings. The connection labelled “Fuel 2 in” is for the addition of other gasses, like CO. “Fuel 3 in” is an auxiliary inlet connection without flow controllers. A reformer system can be connected here. The off-gasses can optionally be directed through back pressure controllers in order to run cells pressurized. The oil loop is driven by the pump, P1. The cylindrical oil sump contains a heating element, which is used for preheating the stack via the coolant.

7

A secondary coolant loop with the heat exchanger HX2 and the pump P2 is not fitted yet. All 6 connections to the stack are flexible hoses and a pressure transducer and a thermocouple are fitted at each one for logging pressures and temperatures of all flows in and out of the stack.

Figure 1. Flow diagram for test rig. V1-5: valves, P1-2: pumps, MFC’s: mass flow controllers, P: pressure transducers, T: thermocouples, BPC1-2: back pressure controllers, HX1-2: heat exchangers. Table 1. Technical specifications of the components Make Range Flow controllers, air (or O2)

Brooks 0.1-2 l/min ±0.02 1.3-26 l/min ±0.2 30-600 l/min ±6

Flow controllers, fuel (H2)

Brooks 0.05-1.0 l/min ±0.01 1.0-20 l/min ±0.2 15-300 l/min ±3

Pressure transducers

Vegabar 14 0-4 bar relative

Back pressure controllers

Brooks 0-8 bar overpressure Max 24 l/h

Tubing and fittings Swagelok 12 mm stainless

8

steel or cupper Coolant Shell Thermia B

Figure 2. The test rig with the 40 cell stack mounted.

9

2.2.2 Electrical system The electric system is based on control and data logging by PC. The PC communicates with the physical setup via a serial port and 15 communication modules of the make ICP-DAS (for RS232/RS485). The total wiring diagram is very complicated, but Figure 3 shows the overall principle. The communication modules are connected directly to the physical sensors and controllers. A few parts are not at present connected to the PC control system. These are the oil pump and its controller, the oil heater and two single cell heaters. The oil pump speed can be controlled continuously, but direct PC control requires a physical modification, which is planned. The oil heating element is controlled by a separate controller. There is no point in integrating it with the PC as temperature stepping is a slow process and setting may just a s well be carried out manually. The two channel independent temperature controllers are meant for thermal management (heating) of single cells or small stacks which do not need active cooling. PD controllers for both single cell heaters and the oil heater are programmed manually. The electronic load was build at the Department of Chemistry during a previous project. It is rated 2kW or 200A, whatever higher. It can easily be replaced with other units if needed.

Figure 3. Principle of electrical control system. See text for details. The data logging capabilities are as follows:

10

• Stack voltage: ± 5V or any higher levels by voltage dividers. • Cell voltages: 40 cells (can be extended to 80). • Stack/cell current: up to 200 A (for higher currents the load can be

exchanged). • Local stack temperatures: 24 channels (Thermocouples type K anywhere in

stack). • Pressures: 6 channels (in and out of fuel, air and oil). • System temperatures: 7 (in and out of fuel, air and oil, and in oil sump).

Figure 4. Connections for cell voltages and internal temperatures in the stack.

Figure 5. Manual controls for gas flow and current. The shift between manual and PC control is an option.

Figure 6. Communication modules.

Figure 7. Electronic load.

11

3 Stacks build in the project

3.1 Introduction to stacking IRD Fuel Cells has developed PEMFC stacks over the years and their stacks have reached a rather mature state. Today stacks for both hydrogen and direct methanol are available on commercial terms. The change from low-temperature to high-temperature PEMFC does not in principle involve a fundamental redesign of the stacking components. Nevertheless, the increased temperature as well as the phosphoric acid in the membrane makes it impossible to use the same materials for sealing and plates as for conventional stacks. After identifying the new suitable materials it became evident that design modifications were necessary in order to adapt to the new materials and to the different membrane electrode assemblies properties. All together a significant effort has been made on the stack development at IRD. DTU and DPS have also previously worked with stacking. The Approach here has been quite different although the general idea with flat plates and internal manifolding is the same. The difference is on the bipolar plate materials and the sealing. The technology is not at all as mature and several challenges remain, especially in terms of leak tightness. However, the potential for low cost mass production is strong. A special challenge with high-temperature PEMFC stacks is the cooling circuit. If air cooling is applied there is no big difference, but in order to utilize the produced heat efficiently, a liquid heat transfer medium must be used. With low-temperature stacks the liquid coolant is always water, and minor leaking into the electrodes is not a problem, because water is desired there anyway. It is even supplied to the cells on purpose to keep up the conductivity of the membrane. At temperatures up to 200ºC water can in principle be applied, but the vapour pressure is over 15 bar, which is not convenient. Instead, less volatile liquids like heat transfer oil must be used, and even the smallest leaking into the electrode pores will be lethal to the stack. When the project was initially formulated it was intended to pursue both ideas and during the project decide whether one should prevail, or if they could be merged into one improved concept. Toward the end of the project it was decided among the partners to keep both concepts open to get the most out of the project, and therefore the stacking results are reported as two lines although close contact has been held and the difficulties of both systems have been discussed openly.

3.2 Stacking based on the IRD concept IRD’s stack design is based on graphite composite bipolar plates with internal manifolds and sealed with fluorinated silicon based rubber gaskets. IRD has selected an internal oil-based cooling circuit integrated into each cell in the stack. The main advantages of this concept are the simple construction and the liquid based cooling allowing an easy integration into CHP systems. A homogenous stack

12

temperature can be achieved since temperature gradients through the active cells and temperature gradients along the active cell areas can easily be regulated by the cooling flow. The closed cooling circuit gives also a more efficient way to warm up the stack before operation. However it requires special sealing materials to sustain the harsh conditions and to prevent oil from leaking out of the stack or through the internal manifold into the fuel cells. The selected fluorinated silicon based rubber material is a good compromise between life time stability, performance and relatively low hardness (< 60 IRHD) as long as it is not in direct contact with the HT-PEM electrolyte. The cooling flow fields are situated between each cell, at the backside of the cathode flow field. The anode and cathode flow fields are based on a serpentine channel design, while the cooling flow field is designed with straight parallel channels. The pressure drops across the flow fields are tolerable at high currents and doesn’t get below a minimum level at low currents to enable a homogenous fluid distribution. The stack design is modular and stacks ranging from 1 cell to 40 cells can be made without modifications of manifolds or flow fields.

3.2.1 Stack Construction

Figure 8. Exploded view of IRD HT-PEM stack. The stack constructed contains 7 cells (Figure 8). The graphite bipolar plates were machined with flow fields, manifolds and sealing grooves. The polymer composite end plates were machined likewise and equipped with Swagelok connectors for air,

13

fuel and cooling liquid. The HT-PEM MEAs were reinforced in the sealing area to prevent the MEAs to be damaged by the seal compression. The stack was assembled with tie rods and disk springs and compressed to 6 bar. The assembled stack was investigated for defects before actual operation of the stack under load.

3.2.2 Leak tests The MEAs provided by DTU were reinforced with polymer masks around the membrane sealing edges. The membrane edges and the polymer masks together gave an uneven surface with creases which could not be made smooth. The cells in the stack could therefore not be sealed sufficient enough. A new polymer mask was developed which much better could accommodate the creases at the membrane edges and provide a flat surface for the sealing. The 7-cell stack made incorporated the different attempts to improve the tightness. The leak testing of the 7-cell stack showed acceptable leak rates for the anode/cathode circuit (Figure 9) and for the cooling circuit (Figure 10) to the outside, but cross-over leakage rate between the anode and cathode rate was to high (Figure 11) .

PSO 7-cell stackleak test anode and cathode

0

0,1

0,2

0,3

0,4

0,5

0,6

0 100 200 300 400 500 600 700 800 900

Time [s]

Pres

sure

[bar

(g)]

cathodeanode

Figure 9. Leak test of anode and cathode of the 7-cell stack

14

PSO 7-cell stackleak test cooling

0

0,5

1

1,5

2

2,5

0 100 200 300 400 500 600 700 800 900

Time [s]

Pres

sure

[bar

(g)]

Figure 10. Leak test of cooling of the 7-cell stack

PSO 7-cell stackcross-over test

0

0,1

0,2

0,3

0,4

0,5

0,6

0 50 100 150 200 250 300

Time [s]

Pres

sure

[bar

(g)]

cathodeanode

Figure 11. Cross-over test of the 7-cell stack During this cross-over test, the anode circuit is pressurized keeping a constant gas flow, while the cathode circuit is closed to the outside. The pressure rise in the cathode circuit is observed versus time. It was decided to use the stack never the less since the cross-over leakage rate was not too high. Only short time performance tests were going to take place where a careful control of anode and cathode pressures would counteract the cross-over.

15

3.2.3 Stack tests The stack has been operated for two days, and a total time of 8 hours of operation. The tests were carried out at ambient pressure, with air and hydrogen.

PSO 7-cell stack27/03/2007

0

1

2

3

4

5

6

7

9:00 9:30 10:00 10:30 11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00

time (hh:mm)

Stac

k Vo

ltage

[V]

0

30

60

90

120

150

180

210

Tem

pera

ture

[ºC

] - P

ower

[W] -

Cur

rent

[A]

Stack voltage

Load Current

Oil Inlet Temperature

Oil Outlet Temperature

Stack power

Figure 12. 7-cell stack, day 1 test profile. The stack was tested at a temperature of 150ºC. The stack reached at this temperature a performance of 150W (3V at 50A) corresponding to a power density of 0.1 W/cm2 at a current density of 0.23 A/cm2.

16

PSO 7-cell stack

0

1

2

3

4

5

6

7

10:30 11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00 15:30

Time (hh:mm)

Stac

k an

d ce

ll vo

ltage

[V]

0

30

60

90

120

150

180

210

Tem

pera

ture

[ºC

] - C

urre

nt [A

] - P

ower

[W]

Stack voltage

Min cell voltage

Max cell voltage

Oil Inlet Temperature

Oil Outlet Temperature

Power

Load Current

Figure 13. 7-cell stack, day 2 test profile. A performance test at 150ºC, with air and hydrogen at ambient pressure was performed on the second day of testing (air stoichiometry: 6, hydrogen stoichiometry: 1.5)

PSO 7-cell stack@ Ambient pressure, 150ºC, λAir=6, λH2=1.5

0

1

2

3

4

5

6

7

8

0,00 0,05 0,10 0,15 0,20 0,25 0,30

Current density [A/cm2]

Stac

k vo

ltage

[V]

0

20

40

60

80

100

120

140

160

Stac

k Po

wer

[W]Vstack

P stack

Figure 14. 7-cell stack performance curve.

17

PSO 7-cell stack@ Ambient pressure, 150ºC, λAir=6, λH2=1.5

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

0,00 0,05 0,10 0,15 0,20 0,25 0,30

Current density [A/cm2]

Stac

k vo

ltage

[V]

0,000

0,015

0,030

0,045

0,060

0,075

0,090

0,105

0,120

0,135

0,150

Pow

er d

ensi

ty [W

/cm

2 ]

UcellminUcellmaxMax cell (DTU)PcellminPcellmax

Figure 15. Individual cells performance curves The stack cells showed different levels of performance but overall performance of the individual cells are satisfactory as compared to the data given by DPS for the individual MEA performances. The best cell reached a performance of 180 W at 60 A, but could not be tested at higher current because of the low and limiting performance of other cells in the stack. On the air side, a maximal pressure drop under these conditions (42 l/min) was 13 mbar; the pressure drop on the anode side (4.5 l/min) was 8 mbar.

3.2.4 Conclusion A stack design based on the use of graphite bipolar plates has been developed. Other stack construction materials suitable for high temperature operation up to at least 180oC were identified and the stack components developed. A 7-cell stack was constructed and tested verifying the stack design. The leak tightness of the stack was acceptable apart from the crossover leakage between the anode and cathode circuits. The crossover leakage was however controlled by adjustment of the anode and cathode circuit pressure and stack was tested. The test showed that cells in the stack performed at the same level as measured by the individual MEA performances. The stack performance was however limited by low performing cells. The pressure drops across the anode and cathode circuits were low even at high stoichiometry allowing ambient pressure operation. The stack construction and components must be further developed especially the sealing concept in order to have a reliable and producible stack

18

3.3 Stacking based on the DTU/DPS concept The alternative stacking idea at DTU and DPS came up during a nationally funded project at DTU, “Udvikling af lille mobil brændselscellebaseret kraft/varme-enhed” 2001-2002 (STVF 5013-00-0027). The project was funded evenly by Danish Research Council and Danish Power Systems Aps. During that project cells up to 100 cm2 were housed using alternative stacking materials, and a 3 cell stack with reasonable performance was build and later a 20 cell stack with sever leak problems. The project ended with a promising but very immature stacking technology. The technique was attempted applied in an EU project (AMFC) as a consequence of a partners (Proton Motor, a stack builder) withdrawal from the project by January 2003 because they found that upgrading their stacking technology to high temperature would involve too many resources. Nevertheless, A 2 kW stack had to be built in a very short time. Alternatively the project had to be closed. DTU took up the challenge and made a stack with the new 16x16 cm cells that where developed in the same period in synergy with the present PSO project. Unfortunately, the time was very short, and the stack suffered from severe leaking. The present project had a small overlap with the AMFC project, and one of the first things to do was to perform a post mortem inspection of the failing stack. Based on this MEA’s and plate geometry were improved in a sequence of steps. Several single cells and short stacks with 4 to 10 cells were manufactured at DTU, as well as a 40 cell stack. The progress during the project period was significant. The overall challenge was to make the cells tight, and tightness was detected as either hydrogen in the cathode off gas or as an OCV (open circuit voltage) depression. Some single cells and short stacks could be pressurized without signs of crossover. In the 40 cell stack, however, a number of cells were leaking and consequently, no proper testing could be performed. Nevertheless, 1700 W was obtained for a short period. One of the 4 cell stacks is shown in Figure 16, and test runs a 160 and 200ºC can be seen in Figure 17. The 40 cell stack is shown in Figure 18. In order to maintain the option for protecting the intellectual propertied of the stacking technique, the details of the stack design is kept confidential and reported separately in an annex to the present report.

19

Figure 16. A stack containing one 4 cell unit and two cooling plates.

0

0.5

1

1.5

2

2.5

3

3.5

0 20 40 60 80 100 120

Current (A)

Volta

ge (V

)

0

10

20

30

40

50

60

70

80Po

wer

(W)

200oC

160oC

Figure 17. Polarization of the 4 cell stack at 160 and 200°C with air and hydrogen. The flow rates corresponded to 0.5 A/cm2, and the gasses were not preheated. The large gap in the measurement is due to experimental problems with electronic feed back resonance from the load.

20

Figure 18. The 40 cell stack built.

3.4 Conclusion Stacking by the DTU/DPS concept is a promising alternative to other stacking techniques, and small stacks have been successful. However, a higher degree of reproducibility and reliability when it comes to avoiding internal leaks is desired when larger stacks are to be build. It is expected that these issues can be solved in the near future. The next step will then be to optimize the flow distribution and the gas stoichiometry and to perform durability experiments.

21

4 Adaptation in the energy system The study of the adaptation in the energy system was conducted by DONG Energy (former Elsam). The work of DONG Energy is reported in: Dong Energy Generation, 19 December 2006: “Anvendelse af HT PEM-brændselsceller til systemydelser” Doc. no. 261597/TORCH (In Danish) It is added as an annex to this report, but the Final project report is not released yet.

22

5 Application of results, business strategy, publication and dissemination

The developed stacking techniques are subjected to further refining in the EU Project “FURIM” and the succeeding PSO 2007 project “Development of HT-PEMFC components and stack for CHP unit”. In FURIM a 2 kW stack will be build in 2007 for integration with a diesel reformer in an auxiliary power unit for Volvo. In the next PSO project a stack will be build for CHP purposes. The partners need to treat key elements for the stacking technology as confidential, and patents applications are under consideration. Consequently, no publications on the stacking technologies are submitted to international journals. However, the following publications and presentations are made with support form this project.

• J. O. Jensen, Q. Li, C. Pan and N. J. Bjerrum. A. P. Vestbø, K. Mortensen, H. N. Petersen, C. L. Sørensen, T. N. Clausen, J. Schramm and N. J. Bjerrum. Recent Progress in High Temperature PEMFC and the Possible Utilization of the Excess Heat for Fuel Processing. In press, Int. J. Hydrogen Energy (2007).

• Q. Li, R. He, C. Pan, J. O. Jensen, L. N. Cleemann, M. Nilsson and N. J.

Bjerrum. Gas Diffusion Electrodes for PBI Cells. Abstract to the 210th ECS Meeting, Cancun, Mexico, October 29-November 3, 2006 (poster).

• Q. Li, P. Noyé, J. O. Jensen, C. Pan, and N. J. Bjerrum. Recent Progress in

Preparation and Characterization of PBI membranes for PEMFC. 210th Electrochemical Society Meeting, Cancun, Mexico, Oct.29-Nov.03, 2006 (oral).

• J. O. Jensen, Q. Li, C. Pan, P. Noyé and N. J. Bjerrum. High Temperature

PEMFC. BrintDanmark. Danish Hydrogen Associations Conference, Herning, 30. March, 2005 (oral).

• J. O. Jensen. High temperature PEM fuel cells. Running PowerPoint

præsentaion at Hydrogen Demonstratorium in Herning. Open from March 2006 (-).

• Q. Li, J. O. Jensen, P Precht Noyé, C. Pan, and N. J. Bjerrum. Proton

exchange membranes for fuel cells: - challenges and recent developments. The 15th International Symposium on Fine Chemistry and Functional Polymers. Shanghai, China, 2005 In: Journal of Fudan University (Natural Science) ; 44 (5), pp. 666-667 - Shanghai, 2005 (proceeding).

• Q. Li, J. O. Jensen, P. P. Noyé, C. Pan and N. J. Bjerrum. Proton Conductivity

and Operational Features of PBI-based Membranes. Solid State Electrochemistry: Proceedings of the 26th Risoe International Symposium on Materials Science. Eds.: S. Linderoth, A. Smith, N. Bonanos, A. Hagen, L. Mikkelsen, K. Krammer, D. Lybye, P. V. Hendriksen, F. W. Poulsen, M.

23

Mogensen and W. G. Wang. Risø National Laboratory, Roskilde, Denmark, 2005 267-272 (proceeding).

One additional outcome of the project is a high degree of confidentiality between the partners. This has lead to more collaboration in later R&D projects (EU FP6, IP project Furim as well as nationally funded projects). The openness and unrestricted discussions between DTU/DPS and IRD on stacking and other fuel cells issues, can to a large extent be attributed the present project.

24