Fluid Management component improvement for Back up fuel cell systems

PROJECT FINAL REPORT

Publishable

FCH JU Grant Agreement number: 301782

Project acronym: FLUMABACK

Project title: Fluid Management component improvement for Back up fuel cell systems

Funding Scheme: Collaborative Project

Period covered: from 1ST July 2014 to 30th June 2015

Name, title and organisation of the representative of the project's coordinator1:

Ilaria Rosso, Chief Innovation Officer, Electro Power Systems S.p.A.

Tel: +39 011 9899853

E-mail: ilaria.rosso@electropowersystems.com

Project website2 address: http://www.flumaback.eu

1 Usually the contact person of the coordinator as specified in Art. 8.1. of the grant agreement 2 The home page of the website should contain the generic European flag and the FCH JU logo which are available in

electronic format at the Europa website (logo of the European flag: http://europa.eu/abc/symbols/emblem/index_en.htm;

logo of the FCH JU, available at: http://ec.europa.eu/research/fch/index_en.cfm). The area of activity of the project

should also be mentioned.

2

1.1 Executive summary (1 page max)

The FluMaBack (Fluid Management component improvement for Back up fuel cell systems) project aims at improving the performance, lifetime and cost of balance of plant (BOP) components of back up fuel cell systems specifically developed for countries where long blackouts occur (1000 hours/year). Due to this long time operation requirement the improvement of system components addressed in this project will benefit both back-up and CHP applications. The project focuses on new design and improvement of BOP with the largest potential for performance improvement and cost reductions: air blower, hydrogen recirculation pump, humidifier and heat exchanger. Specific targets in terms of efficiency and cost have been pointed out for each BOP component to be developed in the project. The objective for system and BOP component lifetime is ten years, i.e. 10.000 hrs.

The consortium consists of large and small entities which are R&D centers: Environment Park, JRC, Foundation for the Development of New Hydrogen Technologies in Aragon, Jozef Stefan Institute, University of Ljubljana- Faculty for Mechanical Engineering; BoP components industrial developers and manufacturers: Domel for air and hydrogen blower, Tubiflex for humidifier and Onda for heat exchangers; fuel cells stack and fuel cell system developers and manufacturers Nedstack and Electro Power Systems (coordinator).

Three successive releases of air blower, hydrogen blower and humidifier and one heat exchanger has been developed and delivered. Main results are:

The air blower developed in the project shows significant improvements respect to SoA in terms of cost, efficiency, lifetime and it is ready to be used in commercial fuel cell system products.

The hydrogen blower developed in the project has proper flow capacity and presents lower consumption than SoA, but further development is necessary to improve lifetime.

The humidifier developed in the project has been proved to be very promising in terms of material, performance, design and manufacturing costs. Further development activities are required in the manufacturing process to improve lifetime.

The H2/air heat exchanger development has been interrupted as no benefits were observed, while several disadvantages exist (increase in costs and system space and reduction of efficiency).

Two releases of 3kW and 6 kW fuel cell systems including the successive releases of developed BoP components have been developed and tested. Test characterization confirmed the proper performance of developed BOP components for operation in the fuel cell systems and advantages in terms of BOP power consumption have been measured (8.6% power consumption achieved). A computer model has been developed in Simulink/Matlab environment for fuel cell stack, air blower, humidifier, H2 blower, according to theoretical equations and laws, and improved, when possible, with real results from the tests carried out during the project. The validation process resulted in a good matching between the experimental values and those obtained in the simulation. A dynamic approach has been followed to perform an efficiency assessment in a complete set of scenarios, as a function of external ambient conditions (related to specific markets) and operating variables. Specific activities related to market preparation and environmental sustainability assessment have been performed: RCS report analysis with the full range of regulations, codes and standards that apply stationary fuel cell systems; LCA report analysis with evaluation of each component of the fuel cell system regarding material composition, production processes, supply of fuel including evaluation of tie-up time of material resources and system overall energetic efficiency; End-of-Life (EoL) assessment for main components of 3kW Flumaback fuel cell system, taking into account reverse logistics process and legislation; a detailed market analysis for each BOP components and fuel cell systems including potential business cases for Flumaback fuel cell systems in North Africa and North Europe in the telecommunication sector. For dissemination purpose, Flumaback dedicated website was created (www.flumaback.eu).

Contact details: ilaria.rosso@electropowersystems.com, mitja.mori@fs.uni-lj.si

3

1.2 Summary description of project context and objectives (4 pages max)

Stationary fuel cells can efficiently convert pure hydrogen, biogas, natural gas or other gaseous hydrocarbons

into electricity and heat – often in cogeneration, i.e. combined heat and power generation.

Due to their flexibility, as pointed out in the FCH JU – Fuel Cell Distributed Generation Commercialisation

Study “Advancing Europe's energy systems: Stationary fuel cells in distributed generation”3, stationary fuel

cells are highly efficient technology to transform today’s fossil fuels and tomorrow’s clean fuels into power and

heat, with the potential to be one of the enablers of Europe’s transition into a new energy age.

In the last decade, stationary power applications using hydrogen have advanced and reached market

penetration status in different geographical areas. Indeed, hydrogen fuel cells represent an optimal solution for

frequent blackouts or off-grid applications being able to store excess energy and instantaneously release it

when a power dip or outage occur providing reliable power when needed. These advantages have been

recognized by industries.

Indeed, according to The Fuel Cell Industry Review 2014, 70.200 units of fuel cells were shipped all in each

region of the world in 2014 from which 45.600 were stationary fuel cells4 (table 1).

Table 1: Fuel cells shipments

From market perspective point of view, growth in the fuel cell market continues to accelerate after 2013 and

2014 saw rising demand in portable, transportation, and stationary applications in particular. Stationary

applications, which vary widely by country and region, include utility-scale, industrial/commercial building, and

residential power fuel cells. According to a recent report from Navigant Research, fuel cell systems for all

applications are expected to generate nearly $57.8 billion in annual revenue by 20235.

Despite the significant increase in number of shipped units all around the world, and the attractive market

perspective, fuel cells still have challenges to overcome and some issues to be addressed:

• Cost: Except in premium applications such as back-up power generation for major financial institutions,

systems costs need to be reduced in order to become attractive for backup power and base load power

generation;

3 Roland Berger Strategy Consultants, Advancing Europe's energy systems: Stationary fuel cells in distributed generation, March 2015 4 The Fuel Cell Industry Review 2014, E4tech, Nov. 2014 5 Fuel Cells Annual Report 2014 Stationary, Portable, and Transportation Fuel Cell Sectors: 2013 and 2014 Global Market Developments, Navigant Research, 2014

4

• Lifetime: Long time operation of fuel cell systems has been demonstrated, but cannot be taken for granted

yet;

• Reliability: The reliability of fuel cell system under a broad range of ambient conditions needs to be

guaranteed for full adoption by end users;

• Novelty: In most conservative markets, any new technology requires significant support and public

understanding in order to compete;

• Infrastructure: Refuelling, large-scale manufacturing processes and support infrastructures, such as trained

personnel, are not yet available for fuel cell systems.

Up to today, the applied research has focused mostly on the core components of fuel cell systems, such as

stacks and cells, not considering potential benefits behind the improvement on the sub-components of fuel cell

systems such as the BOP components. On top of this, there is a clear need for the targeted industry-oriented

development for improved BOP components. Current fuel cell back-up systems include BoP components not

specifically developed for fuel cell systems (BOP components that derive from research applications that could

fit well fuel cell system working conditions that are very expensive, or/and BoP components that derive from

other industrial sectors with a lower cost and reduced performances).

Both these approaches affect the reliability and the cost-competitiveness of the entire fuel cell system,

compromising their massive adoption in the business continuity market.

The objectives behind the Flumaback project can clearly address the above-mentioned challenges and needs.

Indeed, the FluMaBack (Fluid Management component improvement for Back up fuel cell systems) project

aims to improve the performance, life time and cost of BOP components of back up fuel cell systems

specifically developed to face black-out periods of around 1,000h/year for specific markets: USA, Africa and

North Europe where hard operative conditions are present (high and low temperatures). Due to this long time

operation requirement the improvement of system components addressed in this project will benefit both back-

up and CHP applications.

The project focuses on new design and improvement of BOP components for utilization in PEMFC based

stationary power applications, aimed at:

- improving BOP components performance, in terms of reliability;

- improving the lifetime of BOP component both at component and at a system level;

- reducing cost in a mass production perspective;

- simplifying the manufacturing/assembly process of the entire fuel cell system.

While in recent years, the performance and durability of the PEMFC have increased and the cost has

decreased at the same time, performance, durability and costs of BOP components have basically stayed the

same. So, for improvements on performance, durability and cost of the fuel cell system, R&D dedicated on

BOP components have become essential. The project is focused on the most critical BOP components with

the largest potential for performance improvement and cost reductions:

- Air and fluid flow equipments, including subcomponents and more specifically blower and recirculation

pumps

- Humidifier

- Heat exchanger

The specific project targets are summarized in the table below:

5

*cost is referred to single kW of fuel cell system output power, that means more than 1200 € total savings for each 6 kW fuel cell system, plus savings coming from

further downsizing of the fuel cell stack (about 10 %) thanks to efficiency improvement.

Table 2: Flumaback Project Specific Targets

The 3 years project duration guarantees the achievement of all project targets.

The consortium consists of large and small entities that are R&D centres Environment Park, JRC, Foundation

for the Development of New Hydrogen Technologies in Aragon, Jozef Stefan Institute , University of Ljubljana-

Faculty for Mechanical Engineering and BoP components industrial developers and manufacturers: Domel,

Tubiflex and Onda, fuel cells stack and fuel cell system developers and manufacturers, Nedstack and Electro

Power Systems.

The partners are located throughout the EU: Italy, Spain, The Netherland and Slovenia.

This consortium represents a real opportunity for developing a strategic alliance of industrial actors that in

future can collaborate fostering the technological evolution of the components towards more efficient and

flexible BOP components that will allow fuel cell industry to exploit the huge opportunities both for EU market

and for emerging countries.

6

1.3 Description of the main S&T results/foregrounds (25 pages max) 1. Experimental and Economic Assessment of Functional Requirements of BOP components

The technical and economical specifications of each BoP components to be developed have been defined with

the aim to arrive at “best value for money” BOP components, in terms of performance (reliability and

efficiency), lifetime, cost and assembly simplification.

Starting points have been:

- the process scheme of the fuel cell system to be developed by Electro Power Systems

- operating conditions and characteristics of Nedstack fuel cell stack

1.1: Technical and economical specification of blower

Starting from the project targets reported in the DOW:

Topics Current state of the art Advances

Blower

Efficiency 15% >30%

Noise (dB) 65 < 60

Lifetime (h) 6.000 10.000

Cost (Euro/kW) 60 50

The performance specifications for Blower have been defined according to functional requirement of both fuel

cell stack and fuel cell system:

Table 3: Performance requirements for the blower to be developed for 3kW and 6kW fuel cell systems

The fuel cells application requires several features hard to reach at the same time. The system circuit pressure

7

drop is relatively high and the needs of relatively high pressure, proper flow and efficiency together with long

lifetime are considerably challenging. Furthermore, the target of reducing the cost is very important.

Several blower manufacturers are able to offer solutions satisfying only one or two requirements at the same

time. However offering all the above reported features is considerably more challenging and a product

satisfying more than one or two features at the same time at low cost is not yet present in the market.

1.2: Technical and economical specification of recirculation pump

Starting from the project targets reported in the DOW:

Topics Current state of the art Advances

Recirculation pump

Lifetime (h) 6.000 20.000

MTBF 6.000 20.000

Cost (Euro/kW) 40 34

The performance specifications for recirculation pump have been defined according to functional requirement

of both fuel cell stack and fuel cell system:

Table 4: Performance requirements for the recirculation pump to be developed for 3kW and 6kW fuel cell systems

At the current state of the art, the reference and most used technology for the hydrogen recirculation is the

membrane pump. Several companies offer membrane pump-based solutions in the market. However this

technology does not guarantee long lifetime due to degradation of elastomeric membrane mechanical

properties that leads to the membrane break after a certain number of cycles. From this point of view, the

blower technology does not include delicate moving parts and can theoretically guarantee longer lifetime.

However, to the consortium knowledge no hydrogen blower with the reported requirements is present in the

8

market. In fact, the use of blower with hydrogen is challenging due to some difficulties in making the bearings

or the grease compatible with humid hydrogen, Furthermore there are considerably more moving part to be

perfectly sealed from the hydrogen respect to a membrane pump. Thus, a deep development activity is

required in order to satisfy their challenging technical requirements.

1.3: Technical and economical specification of humidifier

Starting from the project targets reported in the DOW:

Topics Current state of the art Advances

Humidifier

Cost (Euro/kW) 300 <100 Euro

Size (dm3) 3 3

flexibility 2 sizes with different factor forms

4 sizes with same factor forms

The performance specifications for humidifier have been defined according to functional requirement of both

fuel cell stack and fuel cell system:

Table 5: Performance requirements for the Humidifier to be developed for 3kW and 6kW fuel cell systems

At present time, only a few companies in the world provide reliable and performing humidifier for fuel cell

application. The main issue in this case is the relative high cost of component due mostly to the high cost of

the perfuorosulfonic membrane and its hard manufacturability.

In this contest developing a humidifier with a lower cost and a higher flexibility (more power size with the same

component) is demanding in order to contribute to the fuel cell system price reduction.

1.4: Technical and economical specification of heat exchanger

According to the original DOW, the development of a heat exchanger directly located at the head of the fuel

cell stack was foreseen because at the time of proposal presentation and negotiation the fuel cell system

developed by the final user presented two cooling circuits: the first one was thermally connected with fuel cell

stack coolant through a liquid/liquid heat exchanger, the second included a split (air/liquid) connected with

ambient.

According to the current process scheme of the fuel cell system, one cooling circuit is now present, therefore

during the kick-off meeting it was agreed to develop the remaining external heat exchanger and a new one to

9

pre-heat hydrogen before inlet into the stack. The evaluation of the advantage/disadvantage of such

introduction represents one of the scope of the project.

The performance specifications for Air/H2 heat exchanger humidifier have been defined according to

functional requirement of both fuel cell stack and fuel cell system:

Table 6: Performance requirements for the H2/Air heat exchanger to be developed for 3kW and 6kW fuel cell systems

According to functional requirement of fuel cell system the performance specifications for external heat

exchanger (Air/Glysantin) are reported in Table 7.

Table 7: Performance requirements for the external heat exchanger to be developed for 3kW and 6kW fuel cell systems

As such heat exchanger is usually already available as commercial product, additional specifications regarding

non corroding with glysantine and presence of thermostatic by-pass valve have been requested.

2. BoP components development

2.1 Development of the successive releases of air blower (leader: Domel)

Three iterations of air blower has been developed and delivered in order to achieve the expected technical and

economical specifications.

10

The air blower prototype is a three stage blower designed to achieve required airflow and pressure

performance in system operating point (Table 1). The prototype design is an optimal compromise between

efficiency, lifetime and economical requirements. Domel developed three stage blower with three impellers to

provide sufficient pressure difference. CFD analyses were done by UL to verify flow circumstances in channels

inside of blower and regularity of all components construction.

To achieve also the economic target the blower was designed so it can be used for 3kW and 6kW system with

only a small change of regulation voltage. At airflow exit from the blower the threaded hole was made to

enable fixation of pressure sensor. If required to avoid a negative impact of pressure sensor on the air flow we

suggest having additional adapter added to this hole to prevent sensor being placed into the air stream. For

the blower performance measurement this hole was closed.

The most critical blower components for lifetime specification are bearings. With current design the rotational

speed has been reduced as much as possible, without compromising the efficiency. This new design should

achieve much better lifetime than currently existing blowers and should achieve more than 20000h. The

lifetime is strongly dependent also on environmental conditions (temperature, dust particles, humidity…) and

should be proven by testing.

Air blower

3 kW fuel cell 6 kW fuel cell

Type Centrifugal, 3 stage

Max pressure at working point 17 kPa 20 kPa

Max flow at working point 5 l/s 8,3 l/s

RPM at working point 18 500 20 000

Voltage 48 V DC

Input power (at WP) 530 W 800 W

Input power (max) 700 W 950 W

ΔT on outlet 15 °C (estimated)

Noise (sound power level) 90 dB (A)

Geometry Φ165 x 176,5 [mm]

Weight 2,5 kg

Lifetime 10 000+ h

Cost 225 eur (100 p.a.), 192,5 eur (800 p.a.) Table 8: Technical and economical specification of the air blower

The second iteration of the air blower is optimised for NEW working points, which were provided by Nedstack

(required pressure is lower due to removal of the heat exchanger, see task 3.5).

OLD working point NEW working point

3 KW FC 6 kW FC 3 KW FC 6 kW FC

Q [l/s] 5 8,3 5 8,3

p [kPa] 17 20 15 16 Table 9: Old and new working points for air blower

To improve efficiency new impeller geometry was designed, sealing at the impeller inlet (between compressor

stages) was improved, and new motor (stator winding) was used. The rotational speed is reduced for about

1000 rpm at WP 6kW and for about 500 rpm at WP 3kW, which also means the lifetime will be higher.

11

The third iteration of the air blower is optimized for reduced lifetime, which is now 10 000 hours. Due to the

reduced lifetime the rotational speed could be increased up to 30 000 rpm (second iteration has 17 000 rpm).

To reach higher efficiency impeller and diffuser geometry were developed with CFD simulations, new type of

impeller inlet sealing was designed and new EC motor and electronic were used. Due to increased rotational

speed impellers with smaller diameter could be designed, which means the complete blower has smaller

dimensions and reduced weight (up to 340 g compared to the second iteration).

The successive iterations of air blowers that have been developed and delivered are shown in the Figure 1.

Figure 1: Successive iteration of air blower prototypes

Preliminary tests performed at Domel demonstrated the improved performance of successive iterations: lower

energy consumption and a better efficiency, as presented in the figures below.

Figure 2: Air blower - Efficiency – 6 kW FC

mod. 497.3.267-821 – initial blower η = 18,5 % mod. 497.3.266-441 – blower it1 η = 23,5 % mod. 497.3.266-441 – blower it2 η = 25 % mod. 497.3.220-15 - blower it3 η = 33,5 % Δη = +8,5% (it2 – it3) Δη = +15% (initial – it3)

12

Figure 3: Air blower –- Efficiency – 3 kW FC

Figure 4: Input Power – 6 kW FC

Figure 5: Input Power – 3 kW FC

In addition to this, Domel provided cost estimation for the three iterations of air blower developed as reported

in the table here below:

mod. 497.3.267-821 – initial blower η = 14 % mod. 497.3.266-441 – blower it1

η η = 19 % mod. 497.3.266-441 – blower it2 η = 19 % mod. 497.3.220-15 - blower it3 η = 31 % Δη = +12% (it2 – it3) Δη = +17% (initial – it3)

mod. 497.3.267-821 – initial blower P = 640 W mod. 497.3.266-441 – blower it1 P = 480 W mod. 497.3.266-441 – blower it2 P = 400 W mod. 497.3.220-15 - blower it3 P = 250 W

ΔP = -390 W (initial – it3)

mod. 497.3.267-821 – initial blower P = 920 W mod. 497.3.266-441 – blower it1 P = 750 W mod. 497.3.266-441 – blower it2 P = 540 W mod. 497.3.220-15 - blower it3 P = 420 W ΔP = -500 W (initial – it3)

13

Pos Domel Code Decsription of Goods Samples Up to

10pcs

100 pcs €/

blower

800 pcs €/blower

2.500 pcs €/blower

1 497.3.266-342 700W, 48V DC, 2 fan stage, regulation 0-10 V DC, tangential discharge, speed

output. Peformance and dimensions as per attached datasheets.

465,00

€/each

225,00

€/each

192,50

€/each

173,50

€/each

2 497.3.266-441 700W, 48V DC, 3 fan stage, regulation

0-10 V DC, tangential discharge, speed output. Peformance and dimensions as

per attached datasheets.

465,00 €/each

229,00 €/each

195,80 €/each

176,80 €/each

3 793.3.285-xxx Low voltage, 3 fan stages. Much Better efficiency in the working point (790

motor, 793 new controller, aerodynamic part - few new toolings needed)

465,00

€/each

280,00

€/each

240,80

€/each

216,72

€/each

Table 10 Cost estimation of three iterations of air blower developed by Domel

The target cost of 50€/kW present in DOW has been fully achieved for all blower iterations to be employed in

the 6kW fuel cell system in production samples of 100 pieces. It is not fully achieved if the air blower is

employed in 3kW fuel cell system even if production samples are of 2500 pieces.

Such cost estimation have been used in WP6 for the market analysis described in deliverable D 6.4 – Market

study report and Implementation plan.

In order to perform end-quality control line for blowers JSI developed two test points to track every produced

item according to different parameters: vibration quality, bearings, impeller, sound emission, anechoic

chamber, electrical properties, voltage, current and power. End-quality control line for blowers installed at

Domel facility is shown in figure 6.

. Figure 6: end-quality control line for blower

14

2.2 Development of successive releases of hydrogen recirculation blower (leader: Domel)

The hydrogen recirculation blower is a completely novel design. To overcome the problems of existing

recirculation pump designs (low life time, low efficiency and high noise), a centrifugal blower solution (instead

of membrane or side channel) has been developed with three stage innovative impeller channel design. To

prevent hydrogen leakage and material degradation appropriate materials and components were used and

materials sensitive to hydrogen were protected.

Development activities have been performed in strict co-operation among Domel, EP, Nedstack and JSI. The

Ansys software tool was used for the CFD analysis by UL to optimize aerodynamics parts and to provide

structural and modal analysis of components. Three successive iterations have been developed and delivered

to achieve the project targets.

Table 11: Technical and economical specification of the hydrogen blower

Testing activities on the first iteration detected several problems: bearings are not appropriate for use in

hydrogen environment due to grease which became more viscous; the blower could not reach the full speed;

High temperature of casing; high energy consumption; inlet and outlet connectors are too small (1/8‘‘ thread);

resistance of hydrogen flow.

Second iteration of the hydrogen blower has improved cooling, changed inlet and outlet connectors and added

fixation points on housing that the installation with rubber dampers is possible to reduce vibrations and noise.

Second iteration also has integrated electronic with sensors to more stable control and added electric contact

to reduce leakage of hydrogen.

Second iteration blowers successfully performed tests with dry and humidified hydrogen, but then stopped

operating after ½ h. The issue were damaged bearings, due to the water, which came through bearing inside

of the motor and removed grease out of it. The problem appeared although the labyrinth sealing was added

and therefore it has been decided to develop new iteration of blower.

The third iteration is redesign of second iteration, which means that the main parts, such as motor, impellers,

diffusers and return channels are the same, different is only a position of electric motor. The motor is now

placed above of the aerodynamic part to prevent damage of bearings due to the water. Due to the relocation of

the motor the hydrogen inlet connector is not anymore on top, but it is now on the side, between motor and

aerodynamic part.

15

Figure 7: Successive iterations of hydrogen blower prototypes

Test results will be presented in paragraph 3.

Estimation of manufacturing cost has been performed as well.

Table 8: Cost estimation of hydrogen recirculation blower developed by Domel

Prices for manufacturing of 2,500 and 20,000 pieces are budgeted prices that should add costs for several

tooling and production equipment plus additional R&D activities.

Anyway, the target cost of 34€/kW present in DOW should be fully achieved for 6kW fuel cell system (and

almost achieved for the 3 kW fuel cell system) in production samples of 20,000 pieces.

2.3 Development of successive releases of humidifier (leader: Tubiflex)

Humidifier is completely new development for the industrial partner involved. Tubiflex started their activity

performing patent investigation about existing humidifier patents (tubular and planar) made by INTERPATENT.

The investigation showed that there are no patents active for applications of interest on either planar nor

tubular humidifier.

Analysis of Tubiflex continued on selection of proper material alternative to Nafion. The investigation

conducted to use a hollow membrane whose geometry is not affected from pressure changes and that has

good water permeability and high selectivity between water and nitrogen. For this purpose, polymer with an

asymmetric structure and polisulfone based with the ideal composition of a polysulfone and polyvinyl-

pyrrolidone mix in a 70/30 ratio seemed to be the best option.

Three releases have been developed and delivered by Flumaback. Development activities have been

supported by CFD calculations carried out by UL. Concept design has been considered with aim to comply:

performance requirements (humidification, pressure drop) of 3 kW and 6 kW fuel cell system; Overall length of

the humidifier equal or less than stack length; Inlet and outlet connections design to optimize general piping

layout.

Successive improvements and changes implemented in the three iterations are summarized in the table

below:

16

Table 12: Main characteristics of successive iteration of humidifier prototypes.

The three iteration of humidifiers developed and delivered are shown in the pictures of figure 8.

Figure 9: Successive iterations of humidifier prototypes

A detailed manufacturing cycle has been defined by Tubiflex together with a preliminary economical evaluation

taking into account the current manufacturing cycle and raw materials procurement. The final cost of the

humidifier prototype employing Membrana fibers is 347.5 €, resulting in a cost for humidifier for the 3kW fuel

cell system of 116 €/kW and for the 6 kW fuel cell system of 58 €/kW. The target cost (Euro/kW) <100 €, in

DOW, is thus achieved for both 3 kW and 6 kW fuel cell system even with a not optimized manufacturing

process and for limited number of pieces (up to 10).

This result is therefore very promising for next mass manufacturing step.

17

2.4 Development of successive releases of heat exchanger (leader: Onda)

First release of Air/H2 heat exchanger prototypes for 3 kW and 6 kW fuel cell systems have developed and

delivered according to the technical specifications defined in 1.4 paragraph.

First tests performed by Nedstack and EP pointed out that the high pressure drop over the heat exchanger

inhibits successful application. A second release has been delivered but it still appeared difficult to meet both

principal specifications for the internal heat exchanger at the same time: low pressure drop on the air side and

high temperature increase on the hydrogen side. No benefits were observed when applying this component,

while several disadvantages exist (increase in costs and system space and reduction of efficiency). Therefore,

as per decision during the Mid Term Review, this component and further development was eliminated.

Figure 10: Successive iterations of heat exchanger prototypes

3. BOP component testing activities 3.1 Test protocol definition (leader: EP)

The test protocol has been developed. It reports specific test procedures for each BoP component and fuel cell

system in full accordance with manufacturers and the end-user in order to validate the performance criteria of

the components concerned in the development activities. Special attention has been put in order to avoid

overlapping among tests performed by individual manufacturer, Nedstack (in charge of validation of basic

suitability of BOP components for stack operation) and research centres. The definition of the testing

procedure includes also a strategy for performing accelerated tests on the components to validate anticipated

lifetime. Specifically, an ageing model has been defined and agreed with manufacturers and end user.

The protocol includes 3 types of test for each these BOP components:

A) fine characterization test (FCT) particularly used for the first release of each BOP component

B) life/ageing test (LAT)

C) periodical performance reference test (for validation purpose) (PRT)

A revised test protocol has been proposed in a second phase of the project taking into account results of the

market analysis. It came out that the new addressed markets for the developed fuel cell systems are North-

Europe remote locations and North Africa, instead of Asia. This means that real time of back-up is around

1,000 h/year, so that 10 years lifetime is about 10,000 h instead of 20,000 h. The modified test protocol on

18

BoP components (air blower and H2 pump) considered these changes and the accelerated aging tests

followed a new protocol, in order to reach 10,000 h in the timeframe of the project. Full details are reported in

the public deliverables D5.1 Test protocol definition and D5.6 Revised test protocol.

3.2 Test of air blower (successive releases) (leader: EP)

A dedicated test-bench has been built-up at EP facilities to perform the fine characterization test of successive

iteration of blowers developed and delivered.

Summary of test results from fine characterization of the three iterations of air blower are reported in figure 11

and 12. The Figure 11 reports the engine consumptions measured during the fine characterization at EP of the

three air blower releases in comparison with the reference one. The figure 12 reports the static efficiency

calculated in the working fuel cell system points.

Figure 11: EP consumption data: comparison between the reference, first, second and third release air blower

Figure 12: EP static efficiency data: comparison between the reference, first, second and third release air blower

19

From the results data analysis it figures out significant improvements has been achieved in the successive

iterations with regards to the reference, commercial, air blower.

The third air blower has highest capacity and presents increased efficiency very close to the target project

specification.

EP test results are compared with Domel and Nedstack results in figure 13 and 14. The Figures report the

engine consumptions at 3kW and at 6 kW measured during the fine characterization at EP compared with the

data collected by Nedstack during the test with the fuel cell and the data from Domel.

The data compared between EP and Domel are similar, only for the first release are quite different.

The consumption reported by Nedstack is lower in Figure 13, but this is related because it has measured

pressure drop lower than the target point and also it has actual flows that are 10% below reported values due

to offset of the Bosch airflow meter. The consumptions reported by Nedstack in Figure 14 are higher, because

they are measured at new working point of 545 l/min and 180 mbar. Due to small differences in experimental

setup and conditions among partners, also small differences in energy consumption are reported. However,

the general trend of a significant reduction in energy consumption from reference to the final, third iteration is

confirmed for all partners and hence are valid under all conditions.

Figure 13: Engine consumption at 3kW working point: data collected and compared by EP, Nedstack and Domel

Figure 14 Engine consumption at 6kW working point: data collected and compared by EP, Nedstack and Domel; X data collected at new target point measured by Nedstack

20

The ageing test protocol has been performed on the third release of air blower.

A middle and end-time characterization was performed on the same bench used for the fine characterization

and no deviations were detected: same pressure drop, consumptions and static efficiency. Thus lifetime

expectancy of 10,000 hours has been confirmed.

Summarizing, the air blower developed in the project shows significant improvements from the reference

performance of commercial product and achieved the expected target in terms of both power consumption and

expected lifetime.

3.3 Test of hydrogen recirculation blower (successive releases) (leader: EP)

The successive releases of the recirculation blower have been characterized on a test bench system set up by

EP capable of measuring the performance of the device in terms of pressure drop, flow rate, temperature and

energy consumption. Problems already described in paragraph 2.2 have been detected by both EP and

Nedstack.

The table 13 summarizes the performances and the energy consumption of first, second and third release of

hydrogen blower:

Table 13: H2 blower specification comparison at 3 kW and 6 kW power system.*Data collected with H2; **Data collected

with wet H2 (only on first and third release)

Because of the limited time available, only EP has tested the third release.

Mainly it is clear that development results are as follows:

- reduced energy consumption from first to third releases, better with dry hydrogen;

- reduced problems occurred during the previous blowers, due to the increased temperatures;

- no interruptions due to corrosion;

- start up and tests repeated 5 times: data repeatability.

From the results data analysis it figures out that the third blower is well designed for the purpose: it has proper

flow capacity and presents lower consumption, close to the target project specification; however because of

21

the additional time requested for the development activities and related late delivery of the third release, the

ageing tests on the third release haven’t been performed, so lifetime target has not be evaluated.

3.4 Test of humidifier (successive releases) (leader: EP)

The humidifier was characterized on a dedicated bench test system, where air humidity can be controlled, by a

separate thermodynamic equilibrium humidifier. The purpose of the test is a fine evaluation of the water

transfer capabilities of the device from the wet to the dry stream. Humidification performance has been

checked by testing humidifier with the air blower (Configuration of the project).

Figure 14 reports the summary of sum of the pressure drop (tube and shell side) measured during the fine

characterization at EP of the three humidifier releases in comparison with the reference one (Permapure).

The figure 15 reports the tube side pressure drops, and the figure 16 the shell side pressure drops.

Figure 14: EP pressure drop measured data: comparison between the references, first, second and third release humidifier

Figure 15: EP tube side pressure drop measured data: comparison between the reference, first, second and third release

humidifier

22

Figure 16: EP shell side pressure drop measured data: comparison between the reference, first, second and third release

humidifier

With regard to the pressure drops the three humidifiers and also the reference one, present, at equal flow,

lower pressure drop on tube side; 4mbar at 3kWFCS working point and between 5 and 6 mbar at 6kWFCS

working point.

Regarding the analysis on the shell side pressure drops:

The third new release presents pressure drop comparable with the target points; the other releases present

pressure drops higher than the target, due to the higher number of the fibers (second release) and to the low

area for secondary flow inlet and outlet.

The sum of the losses is lower than the target points; this allows the blower to work at a lower voltage, with a

reduction in consumption.

In addition, the first, third new and the reference humidifiers have shown a good performance according to the

project targets: a good transport efficiency of 85-90% is reached matching the initial specification of

90%RH@55°C. By contrast, the second release, due to hollow fibers increase, is negatively affect by the

humidity exchange.

Regarding the ageing tests of humidifier, the ageing of the component is mainly due to the fibers exposed to

extreme conditions, where the shutdown procedure has not expelled all the water and this goes to crystallize

inside the fibers in dramatic weather conditions.

The revised protocol provided ageing activities on the first humidifier release, because the first and third

releases present the same hollow fibers number and the same hollow fibers surface.

After the tests on the 6kW fuel cell system (1st release), see chapter 4, the humidifier was removed from the

system and mounted on a dedicated test bench to be tested in order to follow the protocol.

During preliminary characterization part of the airflow passes from dry to wet side inside the humidifier.

When the humidifier was opened, many tubes were broken, probably due to an ageing of the material,

consequently to wet and dry cycles during the tests on the system.

Ageing tests was performed only on the fibers in order to perform degradation study for the material. To the

purpose wet and dry cycles has been performed.

23

After 10 cycles of wet/dry cycles, no significant degradation at material level has been observed and it can be

concluded that the degradation occurred close to the resin and the fibers.

As a conclusion, the humidifier developed in FluMaBack Project is very promising in terms of identified

material (alternative to Nafion), design and manufacturing costs. Further development activities are required in

the manufacturing process to improve lifetime.

4. Fuel cell system development and testing

4.1 Development of the first release of the 3kW and 6kW fuel cell systems including first release of BoP

components (leader: EPS)

The starting point of the development of the first release of 3 kW and 6 kW fuel cell systems has been to build

the system in a cabinet very similar to the one used by EPS for its standard products. However after

preliminary evaluation, it was noted that this approach presented some weak points, e.g. the small space for

maintenance or substitution of components; the impossibility to mount the AIR – H2 heat exchanger on the

lowest part of the system (for condensing water removal) or the impossibility of components displacement or to

change their position.

For the above reasons it was decided to manufacture the first release of fuel cell systems using a more flexible

structure made of extruded aluminum profiles that will allow to change the component position, to replace

them easily and to place sensors and gauges where necessary. Moreover, it was decided to make it bigger

than the standard cabinet on each dimension to make an easy access to the components and to include also

the air-coolant heat exchanger. The easy access to the components facilitates the measurements during the

tests and the inclusion of measurements instruments. Furthermore it makes also easier the implementation of

any

modifications able to improve the system performance and/or reliability.

The two prototypes of 3kW and 6 kW fuel cell systems have been assembled as shown in figure 17 including

the first release of BOP components. They has been preliminarily tested to verify the correct set up and then

delivered to EP for testing (see paragraph 4.3).

Figure 17: First release of 3 kW and 6 kW fuel cell systems

24

4.2 Development of the second release for the 3kW and 6kW fuel cell systems including second

release of BoP components (leader: EPS)

The development of the second release of 3 kW and 6 kW fuel cell systems including second release of BoP

components required a close cooperation between EPS, the fuel cell system manufacturer, and EP that tested

the 1st release of 3 kW and 6 kW and suggested improvements applied on the 2nd release of 3 kW and 6 kW.

On 1st release fuel cell system several problems related to the BoP software configuration have been pointed

out

The 2nd release of prototypes of 3kW and 6 kW fuel cell systems have been assembled in a standard cabinets,

close box, instead of open box configuration as it was for 1st release. The cabinets used are larger than

standard cabinets used for EPS products in order to facilitate the assembly operations of the BoP

components.

Contrary to the 1st release that used an open box configuration, the cabinet used for the 2nd release is provided

with air recirculation fans and relative controls integrated into the software to ensure no explosive area inside

the cabinet. In addition to this, the cabinet has been equipped with a heating system device to prevent freezing

for outdoor operations that will be simulated in climatic chamber tests.

The 2nd release of prototypes of 3kW and 6 kW fuel cell systems includes third iteration of both air blower and

humidifier. By contrast, as 2nd iteration of hydrogen blower still has overheating, noising and starting problem,

it was agreed to assembly in the system a commercial hydrogen recirculation pump available on the market.

The two prototypes of 3kW and 6 kW fuel cell systems have been assembled as shown in figure 18 and

preliminarily tested to verify the correct set up and then delivered to EP for testing (see paragraph 4.3).

Figure 18: Second release of 3 kW and 6 kW fuel cell systems

25

4.3 Test of prototype FCS (first and second release) (leader: EP)

Test procedures used are derived from the FCH-JU funded FITUP project.

The main target of these tests is the evaluation of FC system behaviour in case of grid failures. These tests

should confirm the functionality, reliability and performance of the system with new developed BoP

components.

The first releases are only “open prototypes”, in order to measure additional external measurements:

- Air blower performances;

- Electric performances;

- Evaluation of system efficiency

A proper test set-up has been established for performing tests.

During the test of the first release of 3 and 6 kW fuel cell systems (Open box configuration) at EP several

problems occurred.

The main problem concerned management of stack drying and flooding, which caused poor stack

performance, problem confirmed also by Nedstack. Initially, excessive airflows due to limited controllability of

the blower resulted in stack drying and instable and poor performance. Later, the problem was detected as

water droplets condensing in the cathode inlet pipes, such that liquid obstructed bipolar plates channels.

Changes in the blower management, as operated by the controller, was required, since the original EPS

software was programmed to increase lower set point voltage, thus increasing airflow rate when the cell

voltage showed symptoms of flooding.

This control logic is based on the fact that liquid water is formed on MEA’s catalyst sites, as by‐product of

hydrogen and oxygen reaction, and can sometimes obstruct reactants (mainly oxygen/air) flow through the

cell, leading to a cell voltage reduction (starvation phenomena taking place). The problem was tackled with an

ad-hoc solution, since the root cause is the particular ‘exploded’ layout of the release 1: exposed steel piping

allowed the humidified airflow to condense and dew formation. In addition, the control logic had to be tuned to

avoid excessive airflows and instable operation.

Further support from EPS to EP was given in the overall hardware configuration, to allow installation of

current, voltage drop and RH measurement equipment.

Finally, the 75‐cell direct monitoring was necessary at EP for external data acquisition of fuel cell stack of 6 kW

FCS. This is because original EPS software was designed to allow the external data acquisition up to 64 cell.

In addition, air blower excess flow, probably caused prolonged MEA drying, leading to poor system

performance.

At the end it has been demonstrated that the very poor 6 kW fuel cell system and related stack performances

are due to damage of humidifier: almost 45% of the air goes to the shell side without flowing through the stack

so that a low stoichiometry is available; in addition almost 50% of airflow is leaking from dry to wet side inside

the humidifier, leading to poor humidification.

Summarizing, conclusions on test performed on the first release of 3 and 6 kW FCS are reported in the table

here below:

26

Table 14: test results of the first release of 3 and 6 kW FCS

The second releases (3 and 6kW) are only “closed systems”; the tests have followed the test protocol, and the

“lessons learned” during the first releases characterization allowed to complete the analysis.

In the second release the BOP consumption has been carefully evaluated.

For the 3kW FCS it has been measured that:

Fuel cell stack data: current=135A, V = 27.4, Power=3700W.

Pload =2950W.

Efficiency=79% (Pload/Pstack).

Auxiliary consumption and conversion losses: 790W.

Considering the power electronics efficiency, supplied by EPS, the single contribution of auxiliaries and

conversion losses can be evaluated. The DC/C converter has, in fact, a 94.5% efficiency when operating at

3kW, so the power supplied to the auxiliaries is ~550 W.

For the 6 kW fuel cell system it has been measured that:

STACK DATA: Vstack=47.6V, Stack current 150A, P=7163W,

Pload=6000W

Efficiency=84% (Pload/Pstack).

Ancillaries and Conversion losses: 1163 W.

Analizing the DC/DC converter efficiency curve, at 6 kW ouput it is 91%. This allows the ancillaries

consumption evaluation, resulting in ~520 W, representing the 8.6% of the net system output (7.2% compared

to the stack gross output).

27

The target of BOP power consumption equal to 8.3% of DOW related to 6 kW fuel cell system output power is

therefore substantially achieved.

Second release of 3 kW fuel cell system has been tested in Climatic chamber at JRC as well.

Tests have been performed at hot and dry condition (45°C and 10%RH) and hot and humid condition (45°C

and 80%RH). Freezing startup (- 40°C) was not conducted due to the problem of the coolant circuit leakage

occurred during the previous test.

Hydrogen consumption evaluation has been performed at both EP and JRC with similar results of about 245

g/h.

5. Computer model development

A prototype model in Matlab Simulink has been developed by FHA through the module called Thermolib,

which allows treatment of all the devices with liquid or gaseous fluids quickly and easily.

Each of the modules or blocks represents every device in the system, including the features of each of them

so they can be modified quickly in case of equipment modifications.

First, the complete BOP interconnected systems was proposed using the available Thermolib block models

(Matlab/Simulink environment) for every component of the BOP.

To comply further with the equipment description, the main components have been specifically modeled with

updated Simulink blocks: Updated models for air blower, H2 recirculation pump, gas heat exchangers, fuel cell

stack and humidifier are proposed.

These models are developed according to theoretical equations and laws, and improved when possible with

real results from the tests carried out during the project.

Specifically, models have been developed for:

Fuel cell stack, modifying the basic behavior of the library’s block (beyond the standard

characterization according to voltage-current curves) to adapt it to

o Relative humidity influence

o Pressure (oxygen pressure, atmospheric pressure)

o Air ratio influence (lambda)

o End of life and beginning of life real behavior

o Pressure loss

o Temperature corrections (different temperatures profiles for coolant, stack and gases) to

represent the real behavior of the stack

Air blower, modifying the standard library’s block to adapt it to the different Flumaback prototypes’

results, specifically, taking into account:

o Inlet air conditions (pressure, humidity)

o Efficiency and temperature increase in the air

o Relationship between control voltage (blowers’ electronics), characteristic curve and real

power consumption

Air humidifier: several blocks and functions have been developed from scratch in the project, because

the technology used in Flumaback is not represented in the Simulink library. The model has

considered:

o Mass and enthalpy balances for the air (humid/dry) streams

o Pressure drop

28

o Approach temperature and approach dew point temperature, based on the manufacturers

data sheet.

Hydrogen recirculation pump and heat exchangers have been modeled along the project duration but

were finally discarded for the final simulations. In this aspect:

o Hydrogen recirculation pump used was the objective or theoretical one, as some empirical

problems made difficult its fully characterization for modeling purposes

o Flumaback developments for an External HX (for coolant) and the initially studied gas/gas HX

were modeled, but not furtherly used during the second half of the project.

Control blocks were developed for modelling purposes, in order to exploit the dynamic capabilities of

the model and in order to adapt each separate block (stack, blowers, etc) to work together as a real

BOP, reaching the set points selected by the user.

The validation process included the comparison between the results obtained in the simulation of the

developed fuel cell model with the information provided by the partners of the real tests performed. In

the results provided, a good matching between the experimental values and those obtained in the

simulation can be observed.

An efficiency assessment was done regarding a complete set of scenarios, as a function of external

ambient conditions (correlated to addressed markets i.e. North Europe and North Africa) and

operating variables (air temperature, air pressure and humidity, ageing of the stack).

The dynamic approach of the model could be furtherly used to improve the process of developing a

control loop in the real system, but also could be exploited as a predicting tool, in order to detect if

control and alarms are well designed for every working condition.

29

1.4 The potential impact (including the socio-economic impact and the wider societal implications of the project so far) and the main dissemination activities and exploitation of results (10 pages max)

As above stated, one of the main objectives of the Flumaback project is to improve the performance, life time

and cost of BOP components of back up fuel cell systems specifically developed to face black-out periods of

around 1,000h/year for specific markets: USA, Africa and North Europe where hard operative conditions are

present (high and low temperatures).

Accordingly, several activities have been performed within WP6 and WP7 related to Regulation Codes and

Standards (RCS) that apply to fuel cells and hydrogen and specifically stationary fuel cell systems; life cycle

analysis (LCA); End –of-Life Assessment (EoL) and dissemination respectively. All related deliverables are

public.

1. Regulation, Codes and Standards (RCS), key legal aspects for the technology deployment (leader: FHa)

The main goal of this task was to present proper report on RCS based on a detailed study of the following

point:

- Analysis of the full range of regulations, codes and standards that apply fuel cells and hydrogen;

- Definition of a stationary fuel cell system;

- Determination the most important RCS that apply this application;

- Analysis of the most important RCS that apply this application.

Considering fuel cell and hydrogen RCS that apply at this moment, are just several European Directives that could apply, but not closely related to hydrogen or fuel cells but to the application. Besides, there are a small number of European standards about stationary fuel cell systems:

- Directive 94/9/EC of the European Parliament and the Council of 23 March 1994 on the approximation of the laws of the Member States concerning equipment and protective systems intended for use in potentially explosive atmospheres. (Commonly named ATEX 95, ATEX Equipment Directive).

- Directive 1999/92/EC of the European Parliament and of the Council of 16 December 1999 on minimum requirements for improving the safety and health protection of workers potentially at risk from explosive atmospheres (15th individual Directive within the meaning of Article 16(1) of Directive 89/391/EEC). (Commonly named ATEX 137, ATEX Workplace Directive).

- Directive 97/23/EC of the European Parliament and of the Council of 29 May 1997 on the approximation of the laws of the Member States concerning pressure equipment.

- Directive 2004/108/EC of the European Parliament and of the Council of 15 December 2004 on the approximation of the laws of the Member States relating to electromagnetic compatibility and repealing Directive 89/336/EEC Text with EEA relevance.

- Directive 2006/42/EC of the European Parliament and of the Council of 17 May 2006 on machinery, and amending Directive 95/16/EC (recast) (Text with EEA relevance).

In addition to the Directives, the following European standards have been identified:

- IEC/TS 62282-1 Ed. 2.0. Fuel cell technologies. Terminology.

- IEC/TC 62282-2 Ed. 2.0. Fuel cell technologies. Fuel cell modules.

30

- IEC 62282-3-100 Ed. 1.0Stationary fuel cell power systems. Safety

- IEC 62282-3-200 Ed. 1.0 Stationary fuel cell power systems. Performance test methods.

- IEC 62282-3-300 Ed. 1.0 Stationary fuel cell power systems. Installation.

- IEC 62282-3-201 Ed.1.0 Stationary fuel cell power systems – Performance test methods for small fuel cell power systems.

- IEC 62040 – 1 Uninterruptible power systems (UPS) – Part 1: General and safety requirements for UPS.

- IEC 62040 – 2 Uninterruptible power systems (UPS) – Part 2: Electromagnetic compatibility (EMC) requirements.

Despite the extensive legislation at European level applicable to any process or equipment involved in the manufacture and development of electric fuel cell power generation, there is still no explicit legislation regarding such systems.

European standards collect this type of system, as seen herein; terminology, installation, security and types of tests to be performed in determining the performance are defined.

There remains the transposition of those standards to specific regulations on this issue at European level that provide enough features for the specific devices in this application, indicating ranges of variables for each level of power established.

Regarding the analysis taken, it could be conclude that the prototype will be developed with the aim to be compliant with all the regulations and standards related to this application. All related deliverable are public. 2. Life Cycle Assessment (LCA) (Leader: FHa, UL) The objective of the LCA assessment is to evaluate each component of the fuel cell system regarding material composition, production processes, supply of fuel, waste management and recycling of the unit components, including evaluation of tie-up time of materials resources in the society and system overall energetic efficiency. LCA has been carried out using available data from component production and Flumaback hydrogen technologies UPS (HT-UPS) assembly process. This assessment will help to understand critical points in the terms of environmental impact and potential reductions in emissions within the production process and operation.

The goal of the study was to determine environmental impacts of the HT-UPS. The system’s physical boundaries include the UPS’s subsystems: fuel cell stack, BoP components (air humidifier, blower, H2 recirculation blower and external heat exchanger), monitoring and assembly process with auxiliary equipment, etc. In terms of phases the study was done for manufacturing process of HT-UPS and operation phase of 10,000 h with hydrogen production by means of electrolysis in defined location of operation sites: Norway (Oslo) & Morocco (Marrakesh).

The LCA Assessment is based on an LCA model created using Gabi 6 software and is presented in figure 19 below:

31

Figure 19: Flumaback LCA model

The updated LCA model calculates the environmental impacts for both end site location Norway and Morocco. Main results are presented in table 15.

Table 15: Contribution of manufacturing FluMaBack UPS system, transport to Norway or Morocco and operation of 10.000h in total impacts

As presented in tables above, In the case of Norway the most influential phase is manufacturing phase. Just in the case of GW where 82% of impact comes from operational phase due to electricity production, whereas in the case of Morocco the contribution of operational phase is more influential because electricity in Morocco is 91 % produced from fossil fuels.

For the manufacturing phase Global Worming (GW) caused by manufacturing process had been considered. The main conclusion are presented in the graph below:

Graph 1: Global warming in kg CO2-eq. for manufacturing process (total of 1762,2 kg of CO2-eq.)

32

As can been seen from the graph, in manufacturing process: • main impacts comes from fuel cell, battery and cabinet production; • impact can be directly linked to the mass of the component; • fuel cell and battery production, the high energy consumption, is a very important parameter; and • transport of all components to assembly site in Torino represents just 1 % of total GW.

The LCA assessment analysed the CO2 emission that had been made comparing CO2 emission per 1 kWhe from several technologies (global averages).

10 7 14 11 39 41 66

456

8001.000

664

335

4.234

em

issi

on

sC

O2

per

1 k

Wh

ele

ctri

city

in g

/kW

h

Figure 20: CO2 emissions for 1kWh produced by different technologies and 3 kW UPS system in both end-sites

As shown in figure 20 above, it is evident that 1kWh electricity produced with FluMaBack 3 kW UPS system that is installed in Norway has significantly lower CO2 emission that electricity from fossil fuels. In contrary if system is installed in Morocco impact is almost 13 times bigger.

Summarizing, main conclusions are that the manufacturing phase is more influential in all environmental indicators in the case of Norway because of hydro energy mix in electricity production. By contrast, in Morocco the influence of used electricity (91% from fossil fuels) for H2 production is dominant. Whereas, transport has a negligible influence in all environmental impacts.

All the environmental LCA data resulting from this study will be available to the ILCD Data Network.

All related deliverable are public 3. End-of-life assessment (EoL) (leader FHa,UL) As for common daily hardware, the necessary actions for disposal of the fuel cell system at its end of useful lifetime have to be foreseen. This study implied a revision of the current environmental legislation and future trends, a thorough classification of the components regarding their materials and its level of environmental hazard, the potential of valorisation of the materials, the reverse logistics for hazardous and not hazardous items, lay out of specifications for the works of disassembly and disposal of parts and components, as well as recommendations regarding packaging, transport, and information display for the final users. Environmental legislation in the European Union (EU) is a set of decisions, regulations, directives, etc. which have to protect, regulate rescue and safeguard the Union's citizens from environment-related pressures. The EU has put in place a broad range of environmental legislations due to an intense effort. Approximately 200 or so environmental laws cover most eventualities nowadays. A full analysis of different of regulation codes and standards that apply to fuel cell systems for EoL analysis has been performed both in terms of current environmental legislation and future trends.

33

Regarding current environmental legislations, the EU has developed a framework which helps to group together different directives, regulations, etc, the Environment Action Programme, and currently it is in effect the 7th Environment Action Programme (7th EAP) that will be working until 2020, guiding EU environment policy. The 7th EAP is exposed at Decision nº 1386/2013/EU of the European Parliament and of the Council of 20 November 2013 on a General Union Environment Action Programme to 2020 ‘Living well, within the limits of our planet’. Regarding future trends the 7th EAP indicates a fully implementation of EU’s waste legislation. The Flumaback project identified necessary actions for disposal of the fuel cell system with respect to EU laws. On this regards, the waste generated by the fuel cell at the end of its life has to fit actual EU legislations: i.e. Directive 2008/98/EC on waste; Directive 1999/31/EC on the landfill of waste; Directive 2006/66/EC on batteries and accumulators and waste batteries and accumulators; Directive 2012/19/EU on waste electrical and electronic equipment. Regarding future trends on EU laws, the so-called “Circular economy”, a new economic model related with products defined by European Commission had been identified. Formally called “Towards a circular economy: A zero waste programme for Europe”, it is at level of proposal for directive, and wants to join all directives part of 7th AEP. The EoL assessment continued with the identification of materials masses and types used in the system. Three-stage hazardous scale was set up to identify all hazardous materials and their presence in the components. The EoL assessment was done also using the LCA model from Task 6.2, where manufacturing phase is modelled in detail. Main results are summarized in following graph and figures.

Graph 2: Material used in 3kW UPS fuel cell system

Three stage scale of hazardous materials were defined to delineate hazardousness of material and/or component or process. Three stage scale are:

34

After that, hazardous stage scale for materials used in UPS system and polymers used in the FluMaBack UPS system have been identified. As can be noticed from figure 21 below, most of the materials used in UPS system have the level of 1 or 2 in three stage hazardous scale. Sulfuric acid, electronic equipment and platinum are considered as very hazardous materials that require special attention; additionally platinum is bonded in fuel cell.

Figure 21: hazardous materials

The analysis was performed also for each BoP Component. As shown in table 22 below.

Figure 22: hazardous materials – BoP components

The outcome of this analysis is that the most present material is steel and the most hazardous one are Platinum and electronic equipment both at UPS fuel cell system level that at components level. After evaluation of reverse logistics process and legislation, an EoL assessment scenario for main components of 3kW Flumaback fuel cell system has been prepared, as summarized in the following tables.

Table 16: External heat exchanger EOLA scenarios

35

Table 17: Humidifier EOLA scenarios

Table 18: Battery EOLA scenarios

Table 19: FC stack EOLA scenarios

Table 20: Hydrogen blower EOLA scenarios

Table 21: Air blower EOLA scenarios

Table 22: UPS Assembly EOLA scenarios

36

In conclusion, after the disassembly process of the 3 kW UPS System, each material of the BoP component is checked and classified according to the possibility to reuse theme, re-manufacture them or landfill. Most of them are metal made, consequently, once re-use is rejected, materials have to be transformed into basic metals. These components are: cabinet, compressors, pumps, fans, DC/DC converter, control panel, wiring, cooling fan, pressure regulator, valves-fittings-piping, external heat exchanger. The final destination of materials must be determined to maximize, in a broad sense, the performance that can be obtained, or to minimize the social and environmental impact. All related deliverable are public. 4. Market Preparation

In order to determine the business cases, the market and commercialisation options for the fuel cell systems

developed in the Flumaback project, a market analysis and marketing strategy have been outlined.

The historical numbers of shipments related to fuel cells, mostly stationary fuel cells, have been recovered.

Stationary fuel cells have been divided into three different groups: backup fuel cells, auxiliary power units and

combined heat and power. It has been shown that most of the shipments from 2009 until 2013 are related to

the last group of stationary fuel cells, mainly motivated by the ENE-FARM project develop in Japan,

representing 67% of the total market of this period.

In a 2020 horizon, different scenarios have been developed to obtain the estimations of shipments for each

type of fuel cell. In this context, an optimistic scenario throws a value of 25 000,00 shipment for back-up units

for telecom market all around the world which greatly clashes with the 21 000 units of CHP in Europe and the

150 000 of CHP in Japan or the 23 000 units that will be sell as auxiliary power units in the world for the same

time horizon.

Three potential business cases have been detected for Flumaback fuel cell systems. North Africa offers

opportunities to provide back-up solutions as many outages take place in that area. Blackouts in the North

Europe are an important problem for telecom operators, where the statistics show that single events of up to

200 hours per year are occurring. The third possible case has to deal with the possibility of using the system

during catastrophic events.

An important analysis has been done related to the different stakeholders interested in fuel cells market. In this

sense, managers and shareholders are identified as the most important influencers, however; investors,

buyers, suppliers and competitors keep close.

Related to the FluMaBack characteristics and its price, an important evaluation has been done taking into

account all the competitors nowadays present in the market.

It is shown that for the complete fuel cell system, the 6 kW fuel cell is more price competitive than the 3 kW

one. Different scenarios have been considered having as sensitivity variables the percentage of profit and the

number of units produced as it will reduce the production cost. When developing a market analysis, one of the

most important parts is about the substitute products that coexist. Related to the potential business cases, the

costs associated in each case for different scenarios have been calculated. In the case of telecom back-up

systems the fuel cell has been compared with batteries and diesel generators in five hourly scenarios; 8, 24,

72 and 200 hours of yearly operation. Extrapolating the results to a ten years amortisation scenario shows that

the most economical technology is the diesel one, although no incentives for new technologies or additional

taxes for the pollutant one have been considered. However, this does not represent big differences in cost; as

for example in the 24 hours scenario, the total cost for the 3 kW system in ten years will be comparable to the

diesel one.

37

In the outage case in the North Africa, two scenarios have been calculated for each of the rated power of the

fuel cell: 1 000 hours and 3 000 hours of operation. The comparison here has been done considering only a

diesel generator, as the number of hours of operation is higher, it does not make sense to consider batteries.

As the number of hours increase, the differences in cost in these cases are extremely high, being more

competitive the diesel generator. Anyhow, coupling the FluMaBack system with a hydrogen generator, in order

to completely eliminate logistics costs and hence decreasing the OPEX burden will improve FluMaBack

system's TCO performance, highlighting the flexibility of fuel cells, which can be easily adapted to different

operational contexts. In the catastrophic events case, renewable electricity generation (PV), electrolyser and

fuel cell have defined an emergency box as the one compound. For a complete climatological scenarios have

been included: Sahara Desert, Nicaragua, Indonesia and Alaska. Hard environmental conditions in Alaska are

one of the worst possible scenario, focused on winter months where the number of sunny hours less than 2.

On the other hand, very high isolation on Sahara and Nicaragua ensure a stable hydrogen production along

the year.

At least, after analysing all the cases it can be concluded that important opportunities will appear in the next

years for fuel cells in the telecommunication base stations market. At the same time, in order to be competitive

with its substitutes (mainly diesel), some incentives need to be included in new installations in the first years of

commercialization.

All related deliverables are not public.

5. General promotion activities (Leader: UL) Flumaback dissemination activities followed Flumaback Guidelines for Dissemination Strategy and has the objective to reach a broad audience and so making project results widely visible. In this context, a monthly newsletter has been sent to interested subjects (e.g. researchers, universities, stakeholders). First newsletter has been sent on November 2013 whereas last newsletter has been sent on July 2015.

Coordinator and project partners took part to some public events in which Flumaback project, objectives and activities have been presented such as:

Gwangju, Metropolitan City, Korea, June 2014 (event attended by FHa)

o 20th World Hydrogen Energy Conference 2014, Project presentation on RCS study

Brussels, 10th and 11th November 2014 (event attended by the Coordinator)

o Project Review Days, Poster session

Brussels, 11th and 12th November 2013 (event attended by the Coordinator)

o Project Review Days, Poster session

Moreover, the Coordinator has been invited by FHC2JU to give an oral presentation about Flumaback project

and its results at the next Project Review Days in Brussels next 17th and 18th November 2015. In addition to

this, Flumaback project results will be presented in the same occasion at the Poster session.

All related deliverables are public.

38

6. Development and maintenance of internal and public Project portals (Leader: UL)

Within this task a dedicated website communication tools have been developed. Flumaback dedicated website

is reachable through the web page www.flumaback.eu. The web site results updated containing up-to-date

information about project progress and results, participation at relevant events, relevant information on

involved partners and members.

In addition to this, an intranet portal has been developed too representing the communication and material-

exchanging tool used by all project partners.

Indeed, on the intranet portal is possible to find all relevant information of the project such as:

Presentation regarding all Progress Meetings

Administrative documents (i.e. Grant Agreement, Annex I, Annex D)

Financial rules

Deliverables

Progress report

All relevant material useful/related to the project.

All related deliverables are public 7. Dissemination of project results in scientific and professional journals and conferences (Leader: UL)

The consortium planned to publish the results of Flumaback project in various scientific and professional

journals. As per DoW, 10 publications were mandatory. Realized publications are the sum of all activities

regarding scientific and expert papers, master thesis, conferences, fares, workshops, etc.

Some of the scientific papers, expert publications, master thesis and press releases published are:

Mitja Mori, Boštjan Drobnič, Boštjan Jurjevčič, Mihael Sekavčnik, Influence of End Site Location on

Environmental Impacts of Fuel Cell Backup Power Supply System (scientific paper)

A. Debenjak, P. Boškoski, B. Musizza, J. Petrovčič and Đ. Juričić: Fast measurement of PEM fuel cell

impedance based on PRBS perturbation signals and Continuous Wavelet Transform, Journal of Power

Sources, 254 (2014) 112 – 118 (scientific paper)

Improving the operation of Balance of Plant components, EU Research, Dec. 2014 (publication on EU

Research) http://issuu.com/euresearcher/docs/eu_research_06_digital_mag?e=14613198/13425690)

Presentation of the Flumaback project and its results on the Slovenian Journal of Fluid Power,

Automation and Mechatronics, June 21th, 2015/03 (Technical journal publication)

FluMaBack activities and LCA results presented at the Der Langen Nacht der Forschung, Österreichs,

Klagenfurt, Austria, April 2014 (poster session)

Flumaback project description and results on the “Result in Brief" written by CORDIS science editors

http://cordis.europa.eu/project/rcn/104283_en.html (other dissemination activities)

Flumaback project description and results on the University of Ljubljana, Department of Energy

Engineering, YearBook 2014 (other dissemination activities)

All related deliverables are public.