FCH JU Grant Agreement Number: 325262

Test Protocols for MEAs

______

FCH JU Grant Agreement number: 325262

Project acronym: CISTEM

Project title:Construction of Improved HT-PEM MEAs and Stacks for Long Term Stable Modular CHP Units

Work package:3 -Degradation with SoA Materials

Deliverable:3.1MEA test protocol

Period covered: 01.06. – 30.11.2013

Name, title and organization of the scientific representative of the project's coordinator:

Peter Wagner

NEXT ENERGY · EWE Forschungszentrum für Energietechnologie e.V.

Carl-von-Ossietzky-Str. 15

26129 Oldenburg / Germany

Tel: + 49 441 999 06 316

Fax: + 49 441 999 06 109

E-mail:

Project website address:

MEA – Test protocols

This document defines the different testing protocols and experimental conditions for all participants within CISTEM with respect to MEA investigation. The agreement to use the same protocols and test procedures allows an easy comparison of data between different labs, institutions and participants. Furthermore, literature data can be interpreted according to the existing test protocols.

The necessity to agree to standardized test protocols have been recognized much earlier, for example by the International Electrotechnical Commission (IEC). This CISTEM document takes into account the IEC//S 62282-7-1 “Fuel Cell Technologies – Part 7-1: Single cell test methods for polymer electrolyte fuel cell (PEFC)” document. Although some parts from the EIC document can easily be included in the test procedures it is necessary to adapt the CISTEM test protocols to the specific needs for high temperature PEM fuel cells.

Therefore, this document is only related to general conditions and testing of single cells and MEAs. The testing of short stacks, stacks and the CHP unit will be described in a second document to be published later within deliverable 2.4.

The description of the protocols is strictly related to the different tasks within the work packages and follows the goals of the DoW.

The content of the document will be updated regularly by the work package leaders according to the project progress.

Version / Date / Updated by / Remarks
01 / 30.11.2013 / NEXT ENERGY

1.General conditions and MEA testing

Within the work packages 2 and 3 of the CISTEM project different testing methods for MEAs, including long term durability and degradation issues, are involved, covering various aspects from improvement of efficiency by utilization of oxygen enriched cathode air up to the increase of backpressure and fuel switching betweenreformate and pure hydrogen.

One overall intention within CISTEM is to achieve an improvement in long term durability to more than 20.000 hours of operation and electrical efficiency increase.

In addition, together with intensive membrane and catalyst research and long term durability improvement,the participants will decide after 24 months (deliverable D3.2)which is the “best MEA” that will be manufactured for the final stacks for the CISTEM CHP setup.

All these different approaches and test methods require the usage of commonly agreed test protocols and experimental conditionsfor all project participants. Because the number of test benches available for the CISTEM participants is limited the consortiumwill strictly focus on the main projects tasks only.

The question of “benchmarking” has been decided by the participants. Benchmarking will be done with state-of-the-art MEAs – status 30.11.2013 - from Danish Power Systems. Improvement of MEA design from DPS and UCLM will always be investigated in comparison to these MEAs marked SoA. In some specific cases BASF P1100 MEAs may also be used within the followingtest procedures. But these MEAs will onlybepartially used as bench marking and are intended in most cases to allow comparison to literature data and/or former research results.

Some general definitions within the project are required:

In order to perform MEA testing, especially on a long term basis, it is necessary to agree to terms like begin-of-life (BoL) andend-of-life (EoL).

Definition: Begin-of-lifeof MEAs starts when the first initial IV curve directly after break-in will be taken.BoL reference value will be taken at 0.3 A/cm². This means that the time for the break-in procedure is not included within presentations as a function of time.

Definition: End-of-lifeof MEAs has been reached, when the performance (voltage) at 0.3 A/cm² has decreased more than 10% from the initial starting value which is determined with the initial IV curve directly after break-in.

All degradation calculations will be done by taking into account the time between BoL and EoL.

2.Definition of test parameters

a.Definition of MEA sizes (DPS + UCLM)

MEA active area for lab investigation at NEXT: 25 cm²

MEA active area for lab investigation at UCLM: 25 cm²

Active area for MEAs for short stacks and stacks:200 cm²

(seeAnnex I for details on MEA design)

b.Definition of operating temperature

Although temperature is a highly variable parameter the consortium agrees todefine 160°C for all investigations.

If the test results for improvement of long term durability require reduction of the operating temperature in order to increase the operational time, the participants will have to decide on the new operating temperature.

c.Definition of break-in procedures

Investigators working with CISTEM MEAs (regardless of the provider) should follow the standard break-in procedures given by DPS (or at least until stable conditions have been reached).

Start-up procedures

Temperature: Heat to 120°C before addition of fuels;
Current draw: min. 100 mA/cm2 when fuel is present;
λ (H2)>1.5 to avoid H2 starvation;
Do not exceed a voltage of 800 mV;
Duration under hydrogen as fuel: 48 hours;
Duration under synth. reformate: 48 hours.

d.Definition of stoichiometries

Standard stoichiometry in lab investigations with H2 as fuel is:

BASF MEAs

λ = 1.2or 1.5/2.0 H2/Air

1.5/2.0Synth. reformate/air

1.2 or 1.5/9.5 H2/pure O2

1.5/9.5Synth. reformate/O2.

DPS MEAs

λ = >1.5/2.0 H2/Air

1.5/2.0Synth. reformate/air

1.5/9.5 H2/pure O2

1.5/9.5Synth. reformate/O2.

UCLM MEAs

λ = 1.5/2.0H2/Air

1.5/2.0Synth. reformate/air

1.5/9.5 H2/pure O2

1.5/9.5Synth. reformate/O2.

During testing please observe that the consortium has agreed to follow only one break-in procedure for all MEAs!

A stoichiometry of 9.5 with pure oxygen is necessary to have the same volume flow rates like under air.

Stoichiometry under oxygen enriched cathode air:

BASF MEAs

λ = 1.5/2.0 H2/Air

1.5/2.85H2/30%O2

1.5/4.28H2/45% O2

1.5/7.13H2/75% O2

1.5/9.5 H2/pure O2.

DPS MEAS

λ = 1.5/2.0 H2/Air

1.5/2.85H2/30% O2

1.5/4.28H2/45% O2

1.5/7.13H2/75% O2

1.5/9.5 H2/pure O2.

UCLM MEAS

λ = 1.5/2.0 H2/Air

1.5/2.85H2/30% O2

1.5/4.28H2/45% O2

1.5/7.13H2/75% O2

1.5/9.5 H2/pure O2.

The stoichiometry values have been calculated and are currently evaluated. Results will be shown within the first reports.

The oxygen content of the cathode air is regularly tested and determined by calibrated GC measurements with an error of ± 0.5%.

e.Definition of contact pressure variation

If necessary, 0.2, 0.5, 1.0 and 1.5 MPa compression forces will be applied. Sometimes NEXT ENERGY will operate with even higher compression forces to better understand the limits of the MEAs. But this is no specific project related task.

For all general long term testing, ASTs and start/stop cycling the contact pressure is 0.75 MPa.

f.Definition of electrochemical testing protocols

Detailed protocols for IV, EIS, CV and LSV have been given in papers published by NEXT ENERGY. The conditions are repeated here:

Current voltage curves

Istart : 0.2 A/cm²
Umin : 0.4 V
Imax : 20 A
Istep : 0.5 A
Waittime : 30 sec
Conditioning time : 10 min

Electrochemical impedance spectroscopy

AC perturbation amplitude 10 mV;
Frequency range from 100 kHz to 100 mHz;
Current densities: 0.03, 0.1, 0.2, 0.3 and 0.4 A/cm2.

Cyclic voltammetry

Nitrogen for the cathode (working electrode);
Hydrogenfor the anode (counter and pseudo reference electrode);
Gas flows: 0.1/0.1 l/min N2/H2.
8 CV scans – scan number 6 will be evaluated;
0.05V to 1.0V with a scan rate of 100 mV/s;

Linear sweep voltammetry

Nitrogen for the cathode (working electrode);
Hydrogen for the anode (counter and pseudo reference electrode);
Gas flows: 0.3/0.3 l/min N2/H2.
LSV potential sweep between initial rest potential and 0.5 V with a scan rate of 2 mV/s;

g.Definition of MEA test procedures

Break-in

The consortium agreed on only one break-in procedure, so we follow the procedures given by DPS with all other MEAs.

Start-up procedure

Temperature: Heat up to 120°C before addition of fuels;
Current draw: min. 100 mA/cm2 when fuel is present;
λ (H2)>1.5 to avoid H2 starvation;
Do not exceed a voltage of 800 mV;
Duration under hydrogen as fuel: 48 hours;
Duration under synth. reformate: 48 hours.

h.Definition of general test routines of MEAs

After Break-in always follow the same routines (where applicable):

IV and EIS with H2/air;
Switch gases to H2/O2;
IV and EIS again;
Switch gases to H2/N2 and perform CV and LSV;
Reset to H2/air
Adjust to a constant load of 0.2 A/cm²;
Set new parameter (for example new compressive force or new gas composition on the cathode side);
Equilibrationof the MEA overnight;
Start again at 1.

i.Definition of test routine for long term testing at constant load

Break-in according to 2.c;
Compression at 0.75 MPa (at NEXT);
Initial electrochemical characterization with IV, EIS, CV and LSV;
Overall duration of the test is 1000 hours at 0.3 A/cm² constant load conditions;
Electrochemical investigation: IV, EIS, CV and LSV, once a week until 1000 hours or EoL have been reached.

j.Definition of test routine for long term testing with load cycling

Break-in according to2.c;
Compression at 0.75 MPa (at NEXT);
Initial electrochemical characterizationwith IV, EIS, CV and LSV;
Overall duration of the test is 1000 hours with
4 minutes at OCP = 0 A/cm²;
16 minutes at 0.3 A/cm²;
Electrochemical investigation: IV, EIS, CV and LSV, once a week until 1000 hours or EoL have been reached.

The consortium suggests at least one final long term testing at constant load conditions for at least 3600 hours or until EoL has been reached. This should be done in the final project year with the by that time defined “best” MEA.

k.Definition of test protocols to determine the optimum for O2 enriched cathode air under hydrogen

Duration: one week.

Start with standard stoichiometry – use the factors given before.

After break-in over the weekend;
Initial electrochemical characterization IV, EIS, CV and LSV with air/O2;
Switch to 30% O2 concentration;
equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with enriched air/O2;
Switch to 45%;
Equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with enriched air/O2;
Switch to 75%;
Equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with enriched air/O2.

After determination of the optimized O2 enrichment we perform two long term investigations:

FF serpentine

a)1000 hours with constant load at 0.3 A/cm²,

b)1000 hours with load cycling 4 min at OCP and 16 min at 0.3 A/cm².

m.Definition of test protocols to determine the optimum for O2 enriched cathode air under synthetic humidified reformate

The gas composition will be taken from the ICI composition values further down from 2.o.

Duration: one week.

Start with standard stoichiometry with synthetic humidified reformates and continue with the calculated stoichiometries given in 2.d.

After break-in over the weekend;
Initial electrochemical characterization IV, EIS, CV and LSV with air/O2;
Switch to 30% O2 concentration;
equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with O2 enriched air/O2;
Switch to 45%;
Equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with oxygen enriched air/O2;
Switch to 75%;
Equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with oxygen enriched air/O2.

After determination of the optimized O2 enrichment we perform two long term investigations:

a)1000 hours with constant load, FF serpentine

b)1000 hours with load cycling, FF serpentine

n.Definition of test protocols to determine the optimum backpressure with air/synthetic reformate

Duration: one week.

No change of stoichiometry.

After break-in over the weekend;
Initial electrochemical characterization I/V, EIS, CV and LSV with air/O2 at open cathodeand anode;
Adjust to constant load 0.3 A/cm²;
Switch to 2bara of backpressure in cathodeand anode;
Equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with air/O2;
Adjust to constant load 0.3 A/cm²;
Switch to 2.75 bara;
Equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with air/O2;
Adjust to constant load 0.3 A/cm²;
Switch to 3.5 bara;
Equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with air/O2;
Adjust to constant load 0.3 A/cm²;
Switch to 4.25 bara;
Equilibrate overnight;
Perform all electrochemical testing IV, EIS, CV and LSV with air/O2.

After determination of the optimized cathode/anode backpressure we perform two long term investigations:

a)1000 hours with constant load, FF serpentine;

b)1000 hours with load cycling, FF serpentine.

o.Definition of real/synthetic reformate compositions

The following reformate gas compositions from the reformer have been given by ICI:

Reformate from reformer:

CH4 / H2O / H2 / CO / CO2 / Remarks
% / 1,37 / 30,97 / 53,92 / 1,04 / 12,70 / Humidified
% / 1,98 / - / 78,11 / 1,50 / 18,40 / Dry

These values will be considered as being the optimum.

The min/max variation in gas composition needstbd.by ICI.

Synthetic reformate:

Because the fuel cell test stations at NEXT ENERGY do not have a possibility to mix methane into the anode side, we use a synthetic reformate by adding the CH4 percentage to the CO2 amount.

In addition, for the first test periods we will also neglect carbon monoxide by increasing the CO2 content on the CO values.

So compositionsare:

CH4 / H2O / H2 / CO / CO2 / Test conditions
% / 0 / - / 78,11 / 0 / 21.89* / Dry
% / 0 / - / 78,11 / 1,50 / 20.39* / Dry
% / 0 / 30.97 / 53.92 / 0 / 15.11* / Humidified
% / 0 / 30.97 / 53.92 / 1.04 / 14.07* / Humidified

*recalculated

NEXT will use either dry synthetic reformate or humidified synthetic reformate with 31% H2O.

If required, NEXT will do one “live”-test with real reformate and hydrogen with randomly switching between the fuel supplies on a “best” MEA. This needs to be defined later.

p.Definition of test protocols to investigate start/stop cycling under reformate

The first technical meeting of OWI and NEXT raised another key issue. It is important to investigate how a MEA reacts to start/stop cycling under reformate. The investigation will be done at NEXT.

Detailed experimental conditions will be defined by inhouse/OWI:

60 completed daily start/stop cycles
Standard break-in procedure with wet reformate (tbd. by inhouse);
Only once: initial electrochemical characterization IV, EIS, CV and LSV with air/O2;
Keep it over night with constant load 0.3 A/cm²;
Next Day => Start of Experiment: first polarization curve (I/V)
Start and hold constant load 0.3 A/cm² for 3 hours
Second polarization curve (I/V)
Follow the shutdown procedure including flushingstrategy (tbd.by inhouse) and cool the system down to RT or proposed idling temperature.
Start heating up next morning and equilibrate for two hours
Starting procedure with wet reformate
Repeat 5 to 8 and continue until 59 days (cycles) have been completed or EoL has been reached;
Last experimental day: Skip 8 and keep it over night with constant load 0.3 A/cm²;
Final day: Electrochemical characterization before final shutdown IV, EIS, CV and LSV with air/O2;
Proposed idling temperatures of 60 and 100 °C after shutdown will be tested by NEXT and compared to room temperature (Master Thesis).

q.Definition end-user scenario – fuel switching

Our project is related to the CHP operation with varying fuel supplies. One of the major goals is to demonstrate the capabilities of the CHP to switch and operate between the two fuels, pure hydrogen and reformate.

For this it is obligatory that we do sometests with MEAs and determine the ability and the required operating conditions to achieve the best performance. The proposed operating strategy will be investigated in more detail.

Catalyst loading

Usually reformate usage and pure hydrogen require different catalyst loadings to counterbalance the efficiency loss with reformate against platinum. In case of HT-PEM this has not really been thoroughly investigated yet, according to our knowledge.

The consortium agreed on same platinum loading for both operations: hydrogen and reformer based.

Thestrategy to counterbalance through operational parameters will be investigated.

Nevertheless, the catalyst loading needs to be defined. According to the DoW it is no more than 0.3 mg/cm² in total (anode and cathode). It is suggested to UCLM to start with equal distribution first before varying between anode and cathode.

In the progress of the project it is suggested that both MEA suppliers try to find and provide an optimized catalyst loading solution.

First tests with BASF and current SoA MEAs from DPS and UCLM will give information about the response and requirements to fuel switching.

r.Definition of an emergence shutdown procedure

The emergency shutdown procedure including flushing strategy of a stack/CHP will be defined by OWI together with inhouse.

3.Annex

MEA-Drawing A

MEA Drawing B

______

Version 01: date 30.11.20131