Economic aspects of Fuel Cell Based Stationary Energy Systemskth.diva-portal.org/smash/get/diva2:881401/FULLTEXT01.pdfbroad market, longer stack lifetime, the possibility of selling

Economic Aspects of Fuel Cell-Based Stationary Energy Systems

SUAT SEVENCAN

Doctoral Thesis, 2016 KTH Royal Institute of Technology School of Chemical Science and Engineering Department of Chemical Engineering and Technology Applied Electrochemistry SE-100 44 Stockholm Sweden

TRITA-CHE Report 2016:1 ISSN 1654-1081 ISBN 978-91-7595-754-8 Akademisk avhandling som med tillstånd av KTH i Stockholm framläggs till offentlig granskning för avläggande av teknisk doktorsexamen fredagen den 15 januari 2016, kl. 10.00 i Kollegiesalen, Brinellvägen 8, Stockholm

”ليس شيء مطلقا ، كل شيء ممكن.“

حسن صباح

“Nothing is absolute, everything is possible.”

Hassan Sabbah

to my mother, for her constant prayers for my success and happiness… to my father, for his teachings of resourcefulness and self-reliance… to Linda, for her great love and support throughout my journey…

Abstract

It is evident that human activity has an important impact on climate.

Constantly increasing energy demand is one of the biggest causes of

climate change. The fifth assessment report of the Inter-governmental

panel on climate change states that decarbonisation of electricity

generation is a key component of climate change mitigation. Increased

awareness of this fact and escalating concerns around energy security has

brought public attention to the energy industry, especially sustainable

power generation systems.

Future energy systems may need to include hydrogen as an energy carrier

in order to achieve necessary levels of CO2 emission reductions, and

overcome the challenges renewable energy systems present. Fuel cells

could be a corner stone of future hydrogen inclusive energy solutions.

New solutions like fuel cells have to compete with existing technologies and

overcome the shortcomings of emerging technology. Though these

shortcomings are well-recognised, fuel cells also have many advantages

which makes continued research and development in the field highly

worthwhile and viable. Key to their adoption is the identification of a niche

market to utilise their advantages while overcoming their shortcomings

with continuous research and development.

This thesis aims to evaluate some of the stationary fuel cell applications

and determine whether one could become the niche market as an entry

point for fuel cells. This is achieved by economic evaluations of real and

hypothetical applications.

Results of the studies here imply that to decrease the total life cycle impacts

of fuel cells to more acceptable levels, resource use in the manufacturing

phase and recycling in decommissioning should be shown more attention.

Results also present a picture showing that none of the applications

investigated are economically feasible, given the current state of

technology and energy prices. However, fuel cell-based combined cooling,

heating and power systems for data centres show the potential to become

the niche market that fuel cells need to grow. A further conclusion is that a

broad market, longer stack lifetime, the possibility of selling electricity

back to the grid and governmental subsidies are essential components of

an environment in which fuel cells can permeate through the niche market

to the mainstream markets.

Keywords: Fuel cells, niche market, stationary applications, feasibility,

multi-generation

Sammanfattning

Det är uppenbart att mänsklig aktivitet har en viktig påverkan på klimatet. Den

femte bedömningsrapporten från FN:s klimatpanel slår fast att utfasning av

fossila bränslen i elproduktionen är en nyckelkomponent för att mildra

klimatförändringarna. Samtidigt är konstant ökad efterfrågan på energi en av

de största orsakerna till klimatförändringarna. Ökad medvetenhet om detta

faktum och eskalerande oro gällande energisäkerhet har lett till att

allmänheten fått upp ögonen för energiindustrin, i synnerhet hållbara

kraftproduktionssystem.

Framtida energisystem kommer troligen att behöva inkludera vätgas som

energibärare för att nå nödvändiga sänkta nivåer av koldioxidutsläpp, och

övervinna de utmaningar som förnybara energisystem uppvisar. Bränsleceller

skulle kunna vara en hörnsten i framtida väteinkluderande energilösningar.

Nya lösningar som bränsleceller måste tävla med befintliga tekniker och

övervinna de brister som framväxande teknik för med sig. Även om deras

brister är väl kända, så har bränsleceller många fördelar vilket gör fortsatt

forskning och utveckling inom det här området högst meningsfull och

bärkraftig. En nyckel till en ökad spridning av bränsleceller är identifieringen

av en nischmarknad där man kan dra nytta av bränslecellernas fördelar medan

bristerna övervinns genom fortsatt forskning och utveckling.

Syftet med den här doktorsavhandlingen är att utvärdera några stationära

bränslecellsapplikationer och avgöra om någon av dessa skulle kunna vara en

möjlig nischmarknad som också skulle kunna vara en inkörsport för

bränsleceller. Utvärderingen genomförs medelst ekonomisk analys av verkliga

och hypotetiska applikationer.

Resultaten av studierna här visar att för att minska de totala livscykeleffekterna

av bränsleceller till mer acceptabla nivåer, borde mer uppmärksamhet

fokuseras på resursanvändningen i tillverkningsfasen och på återvinning vid

avveckling. Resultaten visar också att ingen av de undersökta applikationerna

är ekonomiskt genomförbar med nuvarande teknik och energipriser. Däremot

har bränslecellsbaserade system som kombinerar kraft, kyla och värme för

datacenter, potential att bli den nischmarknad som bränsleceller behöver för

att växa. En annan slutsats är att en bredare marknad, längre livslängd för

bränslecellsstackarna, möjlighet att sälja elektricitet tillbaka till elnätet och

statliga subventioner är grundläggande komponenter i en miljö där

bränsleceller kan sprida sig från nischmarknaden till en bredare marknad.

Özet

Beşeri aktivitenin iklim üzerinde önemli bir etkisi olduğu aşikâr. Sürekli

artan enerji talebi iklim değişikliğinin en büyük sebeplerinden biri.

Hükümetler arası iklim değişikliği panelinin beşinci değerlendirme

raporunda elektrik üretiminin karbonsuzlaştırılmasının iklim

değişikliğinin azaltılmasında anahtar öğe olduğu ifade ediliyor. Bu

gerçeğin farkındalığının artması ve yükselen enerji güvenliği kaygısı

kamunun dikkatini enerji endüstrisine, özellikle yenilenebilir güç üretim

sistemlerine çekiyor.

Gerekli CO2 salımı azaltımına erişebilmek ve yenilenebilir enerji

sistemlerinin zorluklarıyla baş edebilmek için geleceğin enerji

sistemlerinin enerji taşıyıcı olarak hidrojeni içermesi gerekmektedir. Yakıt

pilleri hidrojeni de kapsayan enerji çözümlerinde bir köşe taşı olabilir.

Yakıt pilleri gibi yeni çözümler mevcut teknolojilerle yarışmak ve yeni

teknolojinin noksanlıklarının üstesinden gelmek zorundalar. Bu bilinen

noksanlıklara rağmen, yakıt pillerinin sahip oldukları birçok avantaj

alanda devam eden araştırma ve geliştirmeyi bir hayli değerli kılıyor. Yakıt

pillerinin benimsenmesinin anahtar ögesi araştırma ve geliştirmeyle

noksanlıkların üstesinden gelinirken bu avantajların değerlendirileceği bir

niş pazarın tanımlanmasıdır

Bu tez, bazı sabit yakıt pili uygulamalarını değerlendirerek hangisinin yakıt

pillerinin pazar giriş noktası olacak niş pazar olabileceğini belirlemeyi

hedeflemektedir. Bu, gerçek ve farazi sistemlerin ekonomik

değerlendirilmesi ile gerçekleştirilmiştir.

Buradaki çalışmaların sonuçları yakıt pillerinin toplam çevresel etkisinin

azaltılabilmesi için üretim evresindeki kaynak kullanımına ve demontaj

evresinde geri dönüşüme daha çok ilgi gösterilmesi gerektiğini ortaya

koymaktadır. Ayrıca sonuçların ortaya koyduğu resim mevcut teknoloji ve

enerji fiyatlarıyla incelenen uygulamaların hiçbirinin ekonomik olarak

makul olmadığını göstermektedir. Velakin veri merkezlerinin güç ve

soğutma ihtiyacını karşılayan yakıt pili tabanlı birleştirilmiş soğutma

ısıtma ve güç sistemi, yakıt pillerinin gelişebilmek için ihtiyaç duyduğu niş

market olma potansiyeli göstermiştir. Çalışmalardan çıkarılabilecek diğer

bir sonuç da; yaygın pazar, daha uzun yığın ömrü, şebekeye elektrik satma

imkânı ve devlet sübvansiyonlarının yakıt pillerinin niş marketten ana

markete doğru gelişebilmeleri için bir zorunluluk olduğudur.

List of Appended Papers

Paper I A Preliminary Environmental Assessment of Power Generation Systems

for a Stand-Alone Mobile House with Cradle to Gate Approach

Sevencan S, Altun-Çiftçioğlu G, Kadırgan N.

Gazi University Journal of Science. 2011; 24: 487-94.

Paper II Life Cycle Assessment of Power Generation Alternatives for a Stand-Slone

Mobile House

Sevencan S, Altun-Çiftçioğlu G.

International Journal of Hydrogen Energy. 2013; 38: 14369-79.

Paper III A Preliminary Feasibility Study of a Fuel Cell Based CCHP System.

Sevencan S, Altun-Çiftçioğlu GA, Kadırgan MAN.

Gazi University Journal of Polytechnic. 2011; 14: 199-202.

Paper IV Fuel cell based cogeneration: Comparison of electricity production cost

for Swedish conditions.

Sevencan S, Guan T, Lindbergh G, Lagergren C, Alvfors P, Ridell B.

International Journal of Hydrogen Energy. 2013; 38: 3858-64.

Paper V An Economical Comparison of Power-to-gas Alternatives in Bozcaada –

Turkey

Sevencan S, Lindbergh G, Lagergren C, Alvfors P.

Manuscript submitted to Energy.

Paper VI Economic Feasibility Study of a Fuel Cell-Based Combined Cooling,

Heating and Power System for a Data Centre

Sevencan S, Lindbergh G, Lagergren C, Alvfors P.

Manuscript accepted in Energy and Buildings.

http://dx.doi.org/10.1016/j.enbuild.2015.11.012

__________________________________________________

I did the data collection, calculations and interpretation for all the papers

except for Paper IV where Tingting Guan participated in the data

collection.

Abbreviations

AFC Alkaline Fuel Cell

CCHP Combined Cooling Heating and Power

CHP Combined Heat and Power

DALY Disability-Adjusted Life Year

DG Distributed Generation

EACF Equivalent Annual Cash Flow

FC Fuel Cell

FC-CCHP Fuel Cell based Combined Cooling Heating and Power

FC-CHP Fuel Cell based Combined Heating and Power

LCOE Levelized Cost of Energy

MCFC Molten Carbonate Fuel Cell

MCFC-CHP Molten Carbonate Fuel Cell based Combined Heating

and Power

MEA Membrane Electrode Assembly

MJ SE Mega Joule Surplus Energy

NG Natural Gas

NPV Net Present Value

O&M Operating and Maintenance Cost

PAFC Phosphoric Acid Fuel Cell

PEMFC Proton Exchange Membrane Fuel Cell

PEMFC-CHP Proton Exchange Membrane Fuel Cell based Combined

Heating and Power

PDF Potentially Disappeared Fraction

PTG Power To Gas

POO Probability of Occurrence

PV Photovoltaic

R-MCFC Reverse Molten Carbonate Fuel Cell

Contents

Introduction ________________________________________ 1

Compelling global demand for energy alternatives ..................... 1 The rise of hydrogen .................................................................. 2 The environmental aspects of fuel cells ..................................... 4 Status of fuel cell economics ...................................................... 6 Fuel cells in the growing IT sector .............................................. 8 Power-to-gas and Island applications ........................................ 9 The search for a niche market for fuel cells .............................. 11 Aim of the thesis ...................................................................... 13

Fuel cells __________________________________________ 15

Proton exchange membrane fuel cells ..................................... 18 Molten carbonate fuel cells ...................................................... 19

Applications of fuel cells _____________________________ 21

Transport applications .............................................................. 21 Portable applications ................................................................ 22 Stationary applications ............................................................. 22

Distributed generation ....................................................................... 23

Multi-generation ____________________________________ 25

Power-to-gas ........................................................................... 26 Combined heat and power ....................................................... 26 Combined cooling, heating and power ..................................... 28

Thermally driven cooling technologies .............................................. 28

Methodology _______________________________________ 33

Environmental assessments .................................................... 33 Life cycle assessment .......................................................................... 33

Energy related calculations ...................................................... 39 Cooling and heating load calculations ............................................... 39 Adsorption cooling components sizing .............................................. 40 Electricity from photovoltaics and wind ............................................ 43

Economic calculations ............................................................. 43 Payback period .................................................................................... 43 Cost of electricity comparison ............................................................ 44 Equivalent annual cash flow............................................................... 44

Results and Discussion ______________________________ 47

Environmental impacts of fuel cells .......................................... 47 Fuel cell-based combined cooling, heating and power ..................... 56

Fuel cell-based combined heating and power ......................... 59 Power-to-gas ........................................................................... 61

Conclusions and recommendations ____________________ 65

Acknowledgements _________________________________ 69

References ________________________________________ 71

Introduction | 1

Introduction

Compelling global demand for energy alternatives

The evidence of human influence on the climate is stronger than ever. The

anthropogenic emissions of greenhouse gases in recent decades are at the

highest levels of history. The impacts of climate change on human and

natural systems are immensely widespread. To avoid these profound and

increasingly destructive changes, it is imperative to act swiftly and

decisively.

The Earth’s surface temperature has, for the last three decades, been

getting higher than any preceding decade since recording began in 1850.

Figure 1 shows observed globally averaged combined land and ocean

surface temperatures. The 30-year period between 1983 and 2012 was

likely the warmest of the last 1400 years [1].

The world’s energy demand is constantly increasing due to the steady

growth of population and increased quality of living. Today more than ever,

with the increasing global awareness of climate change over the last

decade, development of sustainable power generation systems is a great

concern. Another issue that attracts the public’s attention to sustainable

power generation is energy security. As the conventional fossil fuel

resources are concentrated in a limited number of geopolitical regions and

nuclear power plants have fallen from the public’s grace, many countries

are looking for alternatives such as renewable energy.

2 | Economic Aspects of Fuel Cell-Based Stationary Energy Systems

Figure 1 - Observed globally averaged combined land and ocean surface

temperature anomaly 1850–2012 [1]

The rise of hydrogen

The fifth assessment report Intergovernmental panel on climate change

states decarbonisation of electricity generation is a key component of

climate change mitigation [2]. Figure 2 shows the CO2 emission reduction

levels needed between three scenarios in the International Energy Agency’s

2012 Energy Technologies Perspective report. In this report it is stated

that, to limit the global temperature rise from 6 oC to 2 oC or even 4 oC, a

Introduction | 3

tremendous amount of CO2 emission savings is necessary and the lion

share in these savings should be in electricity decarbonisation [3]. We have

the technological means, such as alternative energy generation methods,

to limit climate change and give a better chance to future generations.

Figure 2 - CO2 emissions and reductions in buildings [3]

Abandoning conventional methods of power generation is an imperative

step in taking control of climate change. Although renewable energy

systems are promising solutions for all these issues, they are sporadic in

nature. In order to overcome the challenges renewable energy systems

present and to achieve considerable CO2 emission reductions, it is likely to

be imperative that future energy systems include hydrogen due to the

benefits it offers as a clean energy carrier [4]. Self-proclaimed ‘hydrogen

romantic’ Veziroğlu prophesized an optimistic projection of hydrogen

replacing all fossil fuels by 2074, shown in Figure 3 [5]. Of course, it is

impossible to say whether or not his prophecy will be fulfilled, but it is

obvious that fuel cells, electrochemical devices that convert chemical

energy of a fuel, usually hydrogen, into electrical energy without

combustion, will play a key role in such endeavours [4, 6, 7].


Figure 3 – Veziroğlu’s projection for the use of hydrogen in the world [5]

The environmental aspects of fuel cells

Emphasising environmental benefits of fuel cells as added value may be

the key to market entry [8-11]. There are Life Cycle Assessment (LCA)

studies in the literature that evaluated the environmental advantages of

fuel cells. Pehnt [12] investigated the polymer electrolyte fuel cell stack

production process to identify environmental impacts of stack components

and materials in comparison with the utilisation of the stacks in vehicles.

Pehnt concluded that, although the advantages of fuel cell vehicles,

especially for urban areas with severe environmental conditions, should

not be underestimated, the environmental impacts of stack production

cannot be neglected in relation to the impacts from utilisation of said stack

in vehicles and the platinum group metals as catalysts are the main cause

of these impacts. Although the impact per kW from stacks produced for

stationary applications are higher than the ones produced for mobile

applications, because of their longer lifetime in per kWh impacts

comparison they are better. Also the impact of the production process of

the stack has much less importance in stationary applications than in

mobile systems. Pehnt also concluded that by recycling platinum group

metals environmental impacts can be decreased substantially. Lunghi and

Bove [13] did an LCA study of a Molten Carbonate Fuel Cell (MCFC) stack

Introduction | 5

to identify process steps that have high environmental impact and find

improvement possibilities. They concluded that the highest environmental

impacts are due to the anode production and, to a lesser extent, cathode

production and that there is a need for more energy efficient processes with

higher recycling of the materials. Khan et al. [14] studied a fuel cell – wind

turbine hybrid system with process chain analysis, focusing on greenhouse

gas emissions. They concluded that although the system they investigated

is a zero emission system in operation, considering the production phase

of the components it has significant GHG emission in total. However, it

still has far lower global warming potential than a conventional diesel

system. Khan et al. calculated that the system has 40.6 gCO2eq/kWh GHG

emission and 0.372 MJ/ kWh energy use. The fuel cell contributed most to

the global warming potential (49 %), followed by the electrolyser (26 %),

and the wind turbine (25 %). Staffell and Ingram [15] studied

environmental impacts of an alkaline fuel cell-based residential

cogeneration system using the Impact2002+ indicator considering only

manufacture and disposal stages. They concluded that in relation to other

components of the cogeneration system the alkaline fuel cell stack

production has considerable environmental impact, of which sulphur

dioxide and other respiratory inorganics present the greatest

environmental burden. For the entire system stainless steel and other

structural material dominates the impacts. They also mention the large

influence of catalyst recycling on environmental impact and the possibility

of decreasing the catalyst load, a system optimization to reduce the use of

steel and design change to enable easy disassembly and recycling. Staffel

et al. [16] made a comparative LCA between a solid oxide fuel cell-based

cogeneration system with boiler and the UK national grid combination to

determine energy and carbon payback times. They concluded that although

a complete 1 kW solid oxide fuel cell cogeneration system, including stack

replacements over the lifetime, emits 1.8-2.9 tonnes of CO2 and requires

29-55 GJ of primary energy, carbon and energy payback times averaged

0.5 and 1.5 years for displacing coal fired generation, 1.7 and 2 years for

displacing the average grid mix, and up to 3.5 years for displacing

combined cycle gas turbine units. Mori et al. [17] did a cradle-to-end-of-

utilisation environmental assessment of an Uninterruptable Power Supply

(UPS) using a Polymer Electrolyte Membrane Fuel Cell (PEMFC) and

compared it to an internal combustion engine-based UPS using the

CML2001 impact assessment method. Their results pointed out that a


hydrogen UPS is an environmentally sound system compared to an

internal combustion engine UPS. They also concluded that the impacts of

hydrogen UPS system are mainly from the manufacturing phase while

internal combustion engine systems impacts are mainly at the operational

phase. Cox and Treyer [18] compared cracked ammonia-fuelled alkaline

fuel cell with a diesel generator for off-grid use. Their results show that the

LCA results are dominated by the production of ammonia, and that

renewable ammonia production pathways improve environmental

performance considerably. They also pointed out the environmental

benefits of recycling of platinum group metals and electrode substrates as

well as the economic ones. Strazza et al. [19] evaluated environmental

impacts of a 230 kW Solid Oxide Fuel Cell (SOFC) and compared it to a

micro gas turbine. They concluded that the SOFC system has

environmental and economic benefits over micro gas turbines. The

contribution analysis they did for the natural gas fuelled SOFC system,

revealed that the fuel supply is a large part of the environmental impact

and a biogas-fed system shows environmental benefits, depending on the

power and energy mix used during the digestion process.

Status of fuel cell economics

For fuel cells to take their rightful place in the global energy system,

environmental advantages will not be enough, they have to be

economically viable. Combined Heat and Power (CHP) and Combined

Cooling Heating and Power (CCHP) are promising applications for fuel cell

systems. Several studies have been made to evaluate economical aspects of

fuel cell CHP and CCHP systems. Silveira and Gomes [20] made a techno-

economic analysis of a CCHP system with an MCFC. They concluded that

an investment cost of more than 1500 USD/kW renders the system

unfeasible with more than 6 years payback time. Lipman et al. [21]

investigated electricity generation from a PEMFC as a distributed

generation system as well as a fuel cell vehicle generating electricity when

parked in California. Their analysis showed that a residential PEMFC

system can break-even at the investment cost of 1200 USD/kW and a 6

USD/GJ natural gas cost as the commercial system’s break-even point is at

700 USD/kW investment cost and 4 USD/GJ natural gas price. For a fuel

cell vehicle scheme to be viable the system should work at above 30 %

(LHV) overall efficiency and natural gas prices must be in the order of 3-4

Introduction | 7

USD/GJ. Alanne et al. [22] made a comparative financial assessment of a

SOFC system versus gas, oil and electricity heating systems for a single-

family dwelling considering different capital costs, interest rates, gas, oil

and electricity prices. They concluded that a 1-2 kWe SOFC system is viable

when compared to electrical and oil furnaces. Zabalza et al. [23] presented

the results of a feasibility analysis of fuel cell CHP systems for different

loads, single family house, apartment blocks, hotels and department

stores, comparing them with a combination of central boiler for heat and

utility for electricity. They calculated the total cost in EUR/year for the

SOFC system to be greater than the conventional alternative for every size.

Their sensitivity analysis showed a large influence of capital cost, the need

to change fuel cell stacks every 7-8 years and surplus electricity sale price

on the feasibility of the system. They also concluded that determining the

working hours as well as the thermal and electrical capacity of the fuel cells

is also important for achieving optimum energy use. Staffel et al. [24]

investigated the economic performance of four fuel cell technologies at 1

kWe as part of a residential CHP system in the UK to find cost targets using

demand profiles of 102 house-years. Estimated annual savings for these

technologies were in the range of 225-250 EUR giving a viable target cost

range of 350-625 EUR/kW for fuel cells with a minimum of 2.5 years of

lifetime. They concluded that increasing lifetime may bring the cost to a

significantly more favourable level. It was also emphasised that in case of

using one house-year demand, the results of savings and target costs can

deviate by up to 28 % of the mean values that were presented in the study.

Ren and Gao [25] analysed gas engine and fuel cell-based 1 kW residential

micro-CHP application in Japan. They looked into minimum cost and

minimum CO2 emission operation modes for both systems. They

concluded that the fuel cell system is the better option for the examined

residential building from both economic and environmental points of view.

In minimum cost mode the annual cost was calculated to be 24 % and 26

% lower using gas engine and fuel cell, respectively. The Fuel cell system in

minimum emission mode lowered annual CO2 emission by 9 % where gas

engine lowered CO2 emission by 2 % in the same mode. They also

concluded that the economic benefits of the systems they investigated were

mainly due to the governmental incentives. Pruitt et al. [26] did an

operational savings versus total installed cost analysis to determine

conditions for fuel cell CHP to be economically viable considering eight

scenarios with different sizes, locations, building types, and net metering


models. Their analysis indicated that to reach positive operational savings

a market with higher electricity price relative to the natural gas price and

net metering is essential. Furthermore, greater electrical and thermal

efficiencies are necessary for better energy savings. Staffell and Green [27]

explored the cost of fuel cell CHP systems to determine the gap between

the numbers in the studies and targets set by agencies and the reality by

gathering past estimates, current prices and targets. Their investigation

showed that the US Department of Energy’s targets of 1500 USD/kW by

2015 and 1000 USD/kW by 2020 are untenable. Unless the market

quadruples by then, 1000 USD/kW will not be reached. They concluded

that these targets do not include the components and the complexity of the

balance of plant; when these are included in the estimations the target

comes to a more reasonable level of 3500/kW. Chen and Ni [28] made an

economic analysis of a fuel cell cogeneration (power and cooling) and

trigeneration (power, cooling and hot water) system for a hotel in Hong

Kong. They concluded that the cogeneration system has a payback period

of 10 years and that total system efficiency may be improved to 94 % from

84 % by transforming the system from cogeneration to trigeneration. The

trigeneration system’s payback time is 5.7 years. These payback periods are

subject to subsidies, electricity price, natural gas price and the balance of

plant component prices. The trigeneration system’s payback period may

vary between 2.7 and 13.9 years. The results of a techno-economic study of

PEMFC and SOFC-based micro-CHP systems, performed by Napoli et al.

[29] show that these systems decrease the primary energy use of domestic

users considerably. Load-following ability increases the positive effects of

these systems especially with PEMFC, due to its low operation temperature

and higher flexibility. When choosing between PEMFC and SOFC systems,

from a primary energy point of view, the SOFC system is the better option.

Although the SOFC system has a higher investment cost it also has higher

yearly savings.

Fuel cells in the growing IT sector

The possibility of employing fuel cells in the IT sector has been attracting

attention for a while [30-36]. Many of the studies in the literature are

focussing on meeting just the power load of data centres. In their studies

of a 10 kW in-rack fuel cell application Zhao et al. [37, 38] concluded that

it is possible to eliminate grid dependence and the internal power

Introduction | 9

distribution system of a server rack with an in-rack fuel cell. They also

concluded that in-rack fuel cells can achieve 53 % energy efficiency for a

single server rack. Riekstin et al. [39] looked into the possibility of in-rack

fuel cells for a whole data centre. They concluded that such a system may

reach above 20 % cost reduction even with conservative projections. There

are also studies focussing on the possibility of using multi-generation for

data centres. Looking into powering and cooling a data centre via

cogeneration Schmidt et al. [40] concluded that the feasibility of such a

system is dependent on natural gas and electricity prices. Xu and Qu [41]

did a comprehensive assessment of a gas turbine-run CCHP system

meeting power and cooling demands of a data centre. Guizzi and Manno

[42] modelled a fuel cell CCHP system for a data centre with constant

power and cooling load. The system was estimated to make a 47 % cost

reduction, above 20 % primary fuel use reduction and more than 20 % CO2

emission reduction. They concluded that the stable nature of the demand

and the low power-to-cooling-demand ratio make these centres a good

match for CCHP use; however, to make CCHP feasible for data centre use,

the electricity price per MWh should be higher than the price of natural

gas. They further concluded that CCHP, for the case they were

investigating, may bring an operational cost saving of up to 54 %.

Power-to-gas and Island applications

Another possible application for fuel cell technologies is using them to

complement power fluctuations of a wind and/or photovoltaic energy

system. This application is especially attractive for places where the grid is

not available, like remote islands. In their paper introducing the

Renewislands project, Chen et al. [43] emphasize how hydrogen and fuel

cells may be the solution for the intermittent nature of renewable energy

systems. In their review article of power-to-gas Götz et al. [44] investigated

polymer electrolyte membrane, solid oxide and alkaline electrolysis

systems as well as methanation technologies. Their cost analysis concluded

that the literature indicates lower specific investment for an alkaline

system than for both polymer electrolyte membrane and solid oxide

electrolysis systems. Kalinci [45] investigated energy generation

alternatives for an island in Turkey, two of which were fuel cell systems.

She concluded that the wind/FC system investigated needed a larger

hydrogen tank and electrolyser to cope with intermittency during the


summer months and added photovoltaic panels to complement the

summer intermittency and increase the reliability. The wind/PV/FC

system has an electricity cost of 0.836 USD/kWh. Loisel et al. [46]

inspected a hybrid off-shore wind power system with different uses of

hydrogen, power generation via fuel cell, injection into gas-line and as fuel

for boats and cars. Their results showed loss for all the scenarios

investigated. The production cost of hydrogen was calculated as 4.2

EUR/kg H2 in the most viable case. Another conclusion they drew was that

combining many usages that need different infrastructure may cause

losses. Enevoldsen and Sovacool [47] studied a hybrid hydrogen-wind

power system for the island of Mykines. Their calculations showed that

comparing to the conventional diesel generators on the island, the

wind/FC system has a higher cost of electricity, 0.23-0.39 USD/kWh and

0.53 USD/kWh, respectively. They concluded that for the hybrid system to

compete with the conventional system an increase of efficiency of both

electrolyser and fuel cell was needed as well as a decrease of the capital

costs. They also suggested that with the environmental benefits and the

self-sufficiency it brings, the hybrid system may be worth the electricity

price difference.

There are also several studies focussing on the cost of hydrogen production

in power-to-gas applications. Glatzmaier et al. [48] made a report on

hydrogen production from concentrated solar energy for the National

Renewable Energy Laboratory. They estimated a cost of hydrogen,

depending on the cost of the photovoltaic system, between 5.78 USD/kg H2

and 23.27 USD/kg H2. Greiner et al. [49] presented a method for the

evaluation of wind-hydrogen systems. They claimed it enables optimised

component sizing as well as calculation of hydrogen production cost. Their

results revealed that the hydrogen costs were 2.8 and 6.2 EUR/kg H2 for

grid-connected and isolated systems respectively. The techno-economic

analysis of a Wind-FC system by Bernal-Agustin and Dufo-Lopez [50]

showed that an electricity price of 1.71 EUR/kWh is required to make the

system feasible. In their study of hydrogen from off-shore wind systems

used as transportation fuel, Mathur et al. [51] showed that farm capacity is

inversely proportional to hydrogen price: 1.53 INR (Indian Rupee)/MJ H2

for a 300 MW system and 2.33 INR/MJ H2 for a 5 MW system. They

concluded that these costs cannot compete with petrol prices. The potential

for hydrogen and electricity production from wind in Kırklareli, Turkey

Introduction | 11

was investigated by Gökçek [52]. An annual hydrogen production of 102.37

kg H2 and an annual electrical energy production of approximately 15000

kWh was found possible. The grid-integrated system with a hub height of

36 m they studied had hydrogen production costs of 0.35 – 4.49 USD/kg

H2. Hydrogen production from wind in Western Canada was studied by

Olateju and Kumar [53] They showed the optimum size for electrolysers to

be 240 kW and 360 kW with a hydrogen production costs of 10.15 USD/kg

H2 and 7.55 USD/kg H2, respectively. Hydrogen production from solar

electricity using alkaline electrolysers was investigated by Negrou et al.

[54]. The case study of Algeria showed hydrogen costs between 3.25

EUR/kg H2 and 11.45 EUR/kg H2 depending on the production capacity.

They also showed that hydrogen from solar electricity becomes less

expensive than hydrogen from conventional sources at minimum of 8000

tH2/year production capacity. G. Genç et al. [55] made a cost analysis of a

wind-electrolyser-FC system for Kayseri, Turkey. Their study showed that

the price of hydrogen was in the range of 8.8 – 30.8 USD/kg H2, depending

on size of system and height of hub. M.S. Genç et al. [56] did a review of

wind-hydrogen production in central Anatolia for five cities with three

different hub heights and three different numbers of electrolysers with a

total rated power of 40 kW and four wind turbines of different rated power.

Their results showed a trend for lower hydrogen prices for higher hubs,

higher wind turbine power and a group of electrolysers with smaller rated

power instead of one big electrolyser.

The search for a niche market for fuel cells

New technological options such as fuel cells generally represent solutions

to new requirements or new constraints and they have to compete with the

existing technologies [57, 58]. The technologies fuel cells need to compete

with, currently have unsatisfactory environmental performances but work

to improve these technologies is ongoing and they may someday reach an

acceptable level of environmental performance. In the current framework

shaped to best utilise the incumbent technologies, fuel cell systems are

simply too expensive. It is necessary to find a niche market for fuel cells

that values their non-standard attributes. Subsequently, research and

development efforts will improve fuel cell attributes to satisfy the

mainstream market. However, for the research and development to

continue, a steady stream of investments is essential. A suitable niche


market which is consistent and large enough should ensure this steady

stream and, if close enough to the mainstream markets could allow fuel

cells to penetrate these markets. There are several studies presenting

possible niche markets for fuel cells. Tso and Chang [59] conclude that fuel

cell scooters might be a solution for Taiwan’s search for a zero emission

scooter as well as a viable niche market for fuel cells. Cigolotti et al. [60]

compare three cogeneration system with an internal combustion engine,

gas turbine and a molten carbonate fuel cell, and they concluded that the

MCFC system with biogas from anaerobic digestion at a cost of 3000

EUR/kW shows promise to be a niche market. In their market survey,

Ramírez-Salgado and Domínguez-Aguilar [61] concluded that the market

for small handheld devices in Mexico has an encouraging size of 107

Million USD for direct methanol fuel cells. Klebanoff et al. [62] presented

the success of a fuel cell powered mobile light tower project that will be

commercialized soon. In their investigation on fuel cell-based residential

combined heat and power, Brown et al. [63] assessed four countries’-

Germany, the UK, the USA and Japan - relative strengths and concluded

that, with its industry-government collaborations, Japan is the leading

country for the prospect of commercialisation of such systems. Cottrell et

al. [64] recommended continuing demonstration projects to increase

consumer awareness and suggested residential combined heat and power

and remote stand-alone applications, such as mobile telecommunication

towers, as possible niche markets.

Economic viability is probably the most important aspect in the search for

a niche market that will allow fuel cells to grow. There are many ongoing

studies trying to find an application for fuel cells that is economically

viable. However, many of these studies use estimations instead of real life

costs: this renders their results less accurate. The work presented here

consists of studies that mainly used real life data from actualised

applications and/or already commercialised products; this allows us to

draw more accurate conclusions about the future of fuel cells.

Introduction | 13

Aim of the thesis

This thesis aims to evaluate some of the stationary fuel cell applications

and determine which one could become the niche for the fuel cell market

entry point. This is achieved by economic evaluation of real and

hypothetical applications. This work also aims to shed light on

environmental impacts and the current economic state of fuel cell systems

and what would be required for them to compete with alternatives in

stationary applications.


Fuel cells | 15

Fuel cells

Fuel cells were invented in 1830’s but the first commercial use did not

occur until the 1960s when NASA used fuel cells in space missions to

generate power for probes, satellites and space capsules [65]. Since then,

fuel cells have been used in many applications, portable and stationary.

Every type of fuel cell is characterized by its materials, dimensions, and

geometry; yet, the core of the device remains the same: it consists of two

electrodes and an electrolyte. The electrolyte should have such properties

that it will insulate the two half-cell reactions electrically while

simultaneously allowing ionic passage between the anode and cathode

sides. The electrolyte should also be impassable for gases to prevent

unwanted reactions. Lastly the electrolyte should not react with any of the

reactants or products of the process.

As the electrolyte is not electrically conductive, the electrons that are

released from the anode reaction need to follow a longer route to the

cathode; through the load that is connected between two electrodes. This

makes it possible to harvest their energy directly in the form of electrical

current. Since there is no combustion, there is no need for extra energy

conversions (heat to mechanical, mechanical to electrical), increasing

efficiency to much higher levels than combustion-based electricity

generation methods.

Electrode materials vary depending on the type of fuel cell. The common

aspect of the electrodes is their porous nature so that they have a high

surface area and the reaction zone is maximized.

Fuel cells have many advantages over conventional electricity generation

systems. As explained above, fuel cells have higher efficiency than

conventional systems because of direct energy harvesting; they are not

producing electricity through a thermal energy conversion process. This

makes it possible for small scale systems to be as efficient as large ones.

This feature makes fuel cell an important contender for small-scale

applications. Due to the lack of combustion, another important advantage

of fuel cells is no or comparatively very low emissions at the point of use. If

the hydrogen is generated via electrolysis using electricity from renewable


energy sources, such as solar or wind power, emissions at the use stage are

eliminated. The lack of moving parts is another important advantage of fuel

cells due to the simplicity, increased reliability and decreased need for

regular maintenance. It also makes fuel cells very quiet systems; this is an

important feature for Distributed Generation (DG) applications where the

power generation system is very close to, or sometimes inside, the building

it is powering. An average fuel cell system including a fuel reformation unit

is around 60 decibel at one meter distance.

Although fuel cell types are named and usually distinguished by the

electrolyte used in them, there are other important differences. For

different purposes and applications, different types of fuel cells have

emerged from research.

Basic information about the five principal types of fuel cells is given in

Table 1.

Proton Exchange Membrane Fuel Cell (PEMFC),

Molten Carbonate Fuel Cell (MCFC),

Solid Oxide Fuel Cell (SOFC),

Phosphoric Acid Fuel Cell (PAFC),

Alkaline Fuel Cell (AFC).

Table 1 - Types of fuel cells and selected features (adapted from [65, 66])

FC

Type Electrolyte

Mobile

Ion

Operating

Temperature

PEMFC Solid Polymer

Membrane H+ 30 – 110 oC

MCFC (Na,K)2CO3 CO32- 500 – 700 oC

SOFC (Zr,Y)O2-δ O2- 600 – 1000 oC

PAFC H3PO4 H+ 150 – 250 oC

AFC KOHaq OH- 50 – 200 oC

Fuel cells | 17

Figure 4 – Fuel cell types for different applications (adapted from [65])

Different types of fuel cell are suitable for different applications according

their capacity, materials and operating conditions. Figure 4 illustrates the

compatibility of different fuel cell types with different applications.

Since the studies reported here are focussed on two types of fuel cells, only

these principal fuel cell types will be described in detail in this thesis:

PEMFC (papers I-V) and MCFC (papers IV-VI).

Types of

Applications

Power (W) 1 10 100 1k 10k 100k 1M 10M

Main Advantages

FCs Bring to

Application

Cars, boats, and

domestic CHP

Distributed power

generation, CHP, CCHP

Different types of

FCs that can be

employed for the

application

Higher energy density than

batteries, Faster recharging

Potential for zero

emissions, Higher

efficiency

Higher efficiency, low

noise, low pollution

Portable electronics

Equipment

AFC MCFC

SOFC

PEMFC

PAFC


Proton exchange membrane fuel cells

PEMFCs have a wide range of capacity and many transport, stationary and

portable applications [67]. Their distinctive features are low working

temperature and polymer membrane electrolyte. The basic principle of the

PEMFCs is shown in Figure 5.

Figure 5 - Basic principle of PEMFCs

Sulfonated polymers are used as the electrolyte in PEMFCs, most

commonly Nafion. On both sides of the membrane there are electrodes,

commonly carbon cloth which supports the platinum catalyst particles.

Outside the electrodes there is an electrically conductive gas diffusion layer

which ensures effective diffusion of reactants to the catalyst layer. This

sandwich as a whole is called the Membrane Electrode Assembly (MEA).

In the MEA, hydrogen supplied to the anode goes through the gas diffusion

layer and is split into protons and electrons at the electrode. Protons are

then transported to the cathode side through the membrane in form of

CathodeAnode

H3O+

4e-

O2

Fuel Air

Excess Fuel

Water VapourExcess air

Heat

H2

H2

H2O

H2O4e-

H3O+

H3O+

H3O+

Electrolyte

Fuel cells | 19

H3O+ while the electrons go through the external circuit producing

electricity. In the meantime oxygen from air goes through the gas diffusion

layer and reacts with protons and electrons forming water at the cathode.

Hydrogen oxidation at anode: 2𝐻2 → 4𝐻+ + 4𝑒−

Oxygen reduction at cathode: 𝑂2 + 4𝐻+ + 4𝑒− → 2𝐻2𝑂

Overall reaction: 2𝐻2 + 𝑂2 → 2𝐻2𝑂

Molten carbonate fuel cells

MCFC uses an electrolyte consisting of liquid molten carbonate salt

immobilised in a chemically inert porous ceramic made of LiAlO2. MCFCs

operate at 500-700 oC and have CO32- as mobile ions. The high temperature

makes them a good candidate for cogeneration applications. The basic

principle of the MCFC is shown in Figure 6.

Figure 6 - Basic principle of MCFCs

CathodeAnode

4e-

FuelAir,CO2

Excess Fuel,

Water Vapour,

Heat,CO2

Excess Air4e-

Electrolyte

O2

CO2

CO2

CO2

H2O

CO2

H2O

CO32-

CO32-

H2

H2


Ionic salt releases carbonate ions when heated. These flow to the anode and

combine with hydrogen to form CO2, H2O and electrons. Oxygen and CO2

recycled from the anode react with electrons coming from the anode

through the external circuit, giving back the carbonate ion to the

electrolyte.

Anode reaction: 2𝐻2 + 2𝐶𝑂32− → 2𝐻2𝑂 + 2𝐶𝑂2 + 4𝑒−

Cathode reaction: 𝑂2 + 2𝐶𝑂2 + 4𝑒− → 2𝐶𝑂32−

Overall reaction: 2𝐻2 + 𝑂2 → 2𝐻2𝑂

As may be seen in the reactions, CO2 is involved both anode (product) and

cathode (reactant) reactions. CO2 formed at the anode is recycled to be

used in the cathode by burning anode exhaust. The resulting mixture of

steam and CO2 is fed to the cathode after mixing with fresh air. The heat

from the burning of the anode exhaust is used to pre-heat the reactant air,

improving the efficiency and maintaining the operating temperature of the

system.

Applications of fuel cells | 21

Applications of fuel cells

Applications of fuel cells can be categorised into three main areas:

Transport applications

As well as their environmental advantages, fuel cells have several

properties that make them suitable for many transport applications. There

are several demonstration applications, from fuel cell scooters to passenger

carts, from boats to busses, from three-wheelers to forklifts. Fuel cell cars

are coming out of demonstration stage to become commercial. Currently

there are two companies in the market that have started initial commercial

sales of their fuel cell cars. Figure 7 shows a number of fuel cell vehicles.

Fuel cell car – Toyota Mirai Fuel cell scooter – Suzuki Burgman

Fuel cell passenger cart – Istanbul, Turkey Fuel cell forklift – Istanbul, Turkey

Figure 7 - Some fuel cell vehicles


Portable applications

Fuel cells are also suitable for portable applications, where they are built

into or charge up products that are designed to be moved. The size of these

applications may range from 500 kW to less than 5 W, e.g. auxiliary power

units for trucks, personal devices such as laptops, radios, etc., small

personal electronics, cameras, mp3 players, portable chargers, military

applications to power wearable electronics, toys, educational kits. There

are already commercial auxiliary power units with fuel cell technology as

well as toys, educational kits and portable chargers. Figure 8 shows some

portable applications of fuel cells.

A laptop powered by a direct methanol fuel

cell

An educational fuel cell toy car

A fuel cell portable charger

Figure 8 - Portable fuel cell applications

Stationary applications

Units that provide energy to a location-bound load are called stationary

units. Fuel cells may be used as prime movers, systems that convert energy

from fuel to other forms of energy, in several stationary applications: main

power units, backup power systems, uninterruptible power systems. There

are many demonstration projects as well as several commercial ones.

Figure 9Figure 9 - Several stationary fuel cell power systems shows some

of them.

Applications of fuel cells | 23

Bloomenergy – 200 kW SOFC –

California, USA [68]

FuelCell Energy – 300 kW MCFC –

Connecticut, USA [69]

Ballard – 1.1 MW PEMFC –

California, USA [70]

UTC – 400 kW PAFC –Connecticut,

USA [15]

ACCAGEN – 21 kW PEMFC –

Bozcaada, Turkey [71]

IDATECH – 5 kW PEMFC –

Istanbul, Turkey [72]

Figure 9 - Several stationary fuel cell power systems

Distributed generation

Central power plants, which are located close to energy sources and far

from populated areas, comprise an integral part of the current electric grid.

They supply the distribution network which distributes the power to load

centres and to consumers. These systems were preferred due to the fact

that it was less expensive to take advantage of economies of scale and

distribute power than to transport fuel and integrate power generation

technologies with populated areas. Technological innovations increasing


efficiency of power generation systems in small scales made it economical

to generate power at or close to the consumer, which in turn increased the

focus on distributed generation (DG). Generating power at or close to the

load site brings a substantial decrease in transmission and distribution

losses of electricity and savings on infrastructure [73-76].

DG can be at different scales according to level of proximity to the

consumer; from district level to building level, to even household level; this

brings many advantages. DG can put off large investments with long

payback such as central power plants, substations and infrastructure. DG

is an important opportunity for promoting renewable energy systems. It

may also create possibilities for utilising local fuel alternatives instead of

fuel transportation. Increased DG in a local grid will increase energy

sources on the grid, decreasing energy dependency and vulnerability to

grid failures, service interruptions, vandalism and external attacks. It will

also bring flexibility against price fluctuations [73, 74, 77, 78]. Systems

investigated in this thesis are all DG systems at different levels: district,

building, household and single consumer. Co-location of energy systems

with load leads to the possibility of multi-generation.

Multi-generation | 25

Multi-generation

Today’s energy problems cannot be solved without a multidimensional

solution that addresses efficiency, pollution, and energy security. Multi-

generation can address these problems as well as create economic appeal

for alternative energy systems. Multi-generation systems are systems that

combine prime movers with key processes enabling them to use various

sources and generate a multitude of marketable products as well as power,

such as heating, cooling, synthetic fuels, desalination and chemicals [79].

A representation of possible outputs of a multi-generation system is shown

in Figure 10.

Figure 10 - Possible outputs of a multi-generation system

Compared to separate generation, multi-generation brings many benefits

such as higher total efficiency, ability to utilise different primary energy

sources, minimised emissions, etc. The multi-generation systems that are

Multi-generation


investigated in this thesis are Combined Heat and Power (CHP), Combined

Cooling, Heating and Power (CCHP), and Power-to-gas (PTG).

Power-to-gas

Power-to-gas (PTG) is converting electricity to another energy carrier in

gaseous form, primarily hydrogen. PTG is a method for storing energy as

hydrogen or other gases such as methane. It is a fitting solution for

sporadic power generation from renewable energy sources such as solar

and wind energy. Combined with such systems, PTG is used to store energy

as hydrogen when generated energy exceeds the demand and used for

electricity generation in peak hours via a fuel cell and/or hydrogen internal

combustion engine coupled to electricity generators [80, 81]. Hydrogen

may be fed to existing gas lines, used as transportation fuel, stored in tanks

or used to synthesise hydrocarbons [81-83]. Figure 11 shows components

of a PTG system and uses of the hydrogen produced.

Figure 11 - Power-to-gas system (Paper V)

Combined heat and power

CHP produces and utilises heat and power from a single energy source

which is enabled by co-locating supply and demand. Due to the fact that it

Electrolyser H2 Storage

Methanisation

Industry

H2 Fuelling

Fuel Cell, H2-fuelled Generator

Gas Network

CH4

Renewable electricity


is easier to transfer excess electricity than excess heat, CHP systems are

designed to meet a given heat demand. This limits CHPs to high heat-

demand users. Conventional separate generation of heat and power has

losses at power plant, distribution and heat generation. Since power and

generation is combined and located at/close to point of use, CHP systems

have smaller losses and may reach much higher total efficiency levels.

Figure 12 illustrates an example comparison between CHP and

conventional heat and electricity generation.

Figure 12 - Sankey energy flow diagram of a combined heat and power

system

A CHP system consists of a prime mover, an electricity generator, a heat

recovery system and a control system. The prime mover generates

electricity via driving the generator and also creates usable heat that is

recovered and used for heating purposes. Another benefit of CHP is being

able to utilise almost any primary fuel depending on the prime mover.

Some prime movers can even use multiple fuel types, which provides

flexibility against price volatility and addresses energy security concerns.


Combined cooling, heating and power

CCHP is a derivation of CHP. It has all the advantages of a CHP system and

utilises the recovered heat further providing space cooling or cooling for

processes, increasing the total efficiency. Figure 13 shows a Sankey energy

flow diagram comparing CCHP to conventional separate generation. In

short, a CCHP system is a CHP coupled with a thermally driven cooling

system.

Figure 13 - Sankey energy flow diagram of a combined cooling heating

and power system (Paper VI)

Thermally driven cooling technologies

Thermally driven cooling systems are designed to use heat for cooling

purposes. They have been commercially available for decades. There are

different types of thermally driven cooling that use different

thermodynamic cycles in varying capacities. Although compared to

conventional vapour compression systems, they have lower efficiency and

higher capital costs, their ability to utilise waste energy makes them

attractive especially when there is large amount of waste heat available

[84]. There are several types of thermally driven cooling systems. The types

that are studied in this thesis are open cycle adsorption and absorption

cooling.


Open cycle adsorption cooling

Open cycle adsorption cooling is based on dehumidification of outside air

with an adsorbent such as silica gel or LiCl. After dehumidification with the

help of a sorption wheel, the air is precooled with maximally humidified

exhaust air and cooled down to feeding temperature by evaporative

cooling. Low temperature heat is used to heat exhaust air to be used in

regeneration. The adsorbent is then regenerated with this heated air [15].

Figure 14 illustrates the operating principle.

Figure 14 - Operating principle of open cycle adsorption cooling (adapted

from [85])

Sorption wheel

The sorption wheel is the key technology behind open cycle adsorption

cooling’s continuous nature. Adsorbent is embedded into a wheel which

has two air flows; the air that needs to be dehumidified and the

regeneration air that desorbs the moisture from the desiccants. The wheel

rotates with a constant speed changing the part of embedded adsorbent

that is in contact with the flows continuously, constantly drying the

adsorbent with regeneration while using the regenerated part to

dehumidify the other flow. The speed of the wheel is adjusted with the

temperature of the regeneration air and humidity conditions to prevent

heat inhibition, inhibition of the adsorbents caused by over desorption and

heating [15, 16]. Figure 15 is an illustration of a commercially available

sorption wheel.

Heat

Source

Sorption

Wheel

Heat

Exchanger

Humidifiers

Inlet

Air

Exhaust

Air

Outside Air

Waste air


Figure 15 - Sorption wheel [86]

Open cycle adsorption systems are mainly used where there is low

temperature heat available, for instance waste heat from a PEMFC or heat

from solar thermal collectors. The system may only be used at low outside

air humidity. If the outside air humidity is higher than the system can

handle, it is possible to couple it with an absorption chiller.

Absorption Cooling

Absorption cooling is very similar to conventional vapour compression

cooling, but instead of a mechanical compressor absorption system it uses

an absorber to compress the refrigerant vapour. The systems has four main

elements: generator, condenser, evaporator and absorber. In the generator

refrigerant - ammonia or water - is separated from solvent - water, LiBr -

by evaporation with the heat input. The refrigerant in vapour form then

goes to the condenser where it is condensed back into liquid form. Liquid

refrigerant flows to the evaporator through a nozzle where it takes heat

from a chilled water circuit. Once again vaporised refrigerant goes into the

absorber, being absorbed back into the refrigerant-weak solution which


comes from the generator and is pumped to the generator as refrigerant-

rich solution. Figure 16 illustrates the working principle of the system.

Figure 16 - Working principle of absorption cooling (Paper VI)

Cooling Water

Cooling Water

Chilled Water

Hot Water/Steam

Refrigerant-rich solution

Refrigerant-poor solution

Absorber

Evaporator

High Pressure Chamber

Generator

Condenser

Liquid Refrigerant

Low Pressure Chamber


Methodology | 33

Methodology

Evaluation of environmental impacts of fuel cell systems is done by Life

Cycle Assessment using Eco-indicator 99 method. Economic aspects of the

systems are assessed using capital budgeting techniques. Both of these

assessments required calculations of energy demand and power generation

for sizing of systems and/or defining parameters.

Environmental assessments

Life cycle assessment

The environmental life cycle of a product consists of all the stages from raw

material extraction to its waste management. LCA is an evaluation method

of environmental impacts related to a product that considers all the energy

and material inputs and wastes to the environment throughout the

product’s life cycle; from extraction and processing of raw material to

manufacturing, from distribution to recycling and final disposal of it [87].

An LCA study is performed in four phases as shown in Figure 17.

Figure 17 - LCA procedure (adapted from [87])

The LCA Procedure

Goal and Scope

Definition

Inventory

AnalysisInterpretation

Impact

Assessment


Goal and scope definition

Goal and scope definition is the phase where the product/process/system

that is going to be assessed are stated. It should include details such as

purpose of the study, intended audience and specifications of the flow

model to be constructed in the subsequent inventory analysis phase, such

as, the functional unit, system boundaries, data sources, impact categories.

It is important to be as clear as possible in these definitions, leaving no

space for ambiguities and being consistent with the intended application.

Figure 18 shows the flow chart that is generally part of the output of this

first phase

Figure 18 - Initial flowchart (adapted from [87])

Raw Material

Acquisition

Processes

Transports

Manufacture

Use

Waste

Management

Resources e.g.

Raw materials,

Energy,

Land resources

Emissions to

Air

Water

Ground

System

Boundary

Distribution

Methodology | 35

Inventory analysis

In this phase a system model according to the requirements of first phase

is built. This model is built upon the rudimentary flow chart developed in

the first phase. In the inventory analysis phase this flow chart is expanded

to show all modelled activities and flows between them [88]. Then

input/output data for these activities are collected, such as energy, raw

materials, intermediate products, solid wastes and emissions to water and

air, and used to calculate the amount of resources used and pollutant

emissions of the system in relation to the functional unit.

Impact assessment

The impact assessment phase aims to describe the impacts of the

environmental loads quantified in the previous phase. This allows the

practitioner to convert the results from inventory analysis, from emissions

and resource use to impacts on environment, and combine them into fewer

parameters. In a classification step inventory results are sorted according

to the environmental impact they cause. Then, in a characterisation step

the impact contributions of these groups are calculated using scientific

models of cause-effect chains in the natural systems of varying complexity.

These results may then be combined into a limited number of impact

categories chosen in the first phase. Life cycle impact assessment is thus a

stepwise aggregation of the information from the previous step as shown

in Figure 19.


Figure 19 - Illustration of the stepwise aggregation of information in the

impact assessment phase (adapted from [87])

Interpretation

This phase summarises the results from previous phases to create a basis

for conclusions and recommendations for decision makers according to the

statements made in the goal and scope definition. This may be done by

selecting the most important parameters to present or to present them in

weighted impact assessment results.

Variants of LCA

There are several variants of LCA according to choices made in the goal and

scope definition phase. Figure 20 illustrates the different variants. Two of

these variants were used in this work.

ClassifiedInventory

Results

CharacterisationResults

WeightingResults

H+ equiv.

Acidificationpotential

One-dimensionalindex

Eutrophication Potential

SO2

NOX

HCl

Etc.

NOX

NH3

P

Etc.

CO2

Etc.

CH4

CFC5

Global Warming Potential

N equiv.

GWP

Etc.

Methodology | 37

Cradle-to-Grave: A full scope LCA from the extraction of raw material,

cradle, used in the manufacturing of the product to disposal, grave, of the

product.

Cradle-to-Gate: An assessment that covers only part of the life cycle, that

is from raw material extraction until the product exits the factory gate.

Figure 20 - LCA variants [89]

Eco-indicator 99 impact assessment method

Figure 21 illustrates the steps of the impact assessment method used in this

work: the Eco-Indicator 99 methodology with sub-procedures and

intermediate results from inventory analysis data to single score indicator.

Eco-Indicator 99 describes natural resource use and pollutant emissions

quantitatively and then analyses and classifies them into 11 impacts. In the

next step these impacts are grouped into three damage categories: Human

Health (Disability-adjusted life years), Ecosystem Quality (in Potentially

disappeared fraction per square meter per year) and Resources (in MJ

surplus energy) [90]. These impacts on the three damage categories are

then normalised using a reference system and further aggregated to a

single score by weighting.


Figure 21 - General representation of the Eco-Indicator 99 methodology

(adapted from [90])

Ext

ract

ion

of

min

eral

s an

d f

oss

il fu

els

Lan

d-u

se:

occ

up

ati

on

an

d

tra

nsf

orm

ati

on N

Ox

SOx

NH

3

Pe

stic

ide

s

He

avy

me

tals

CO

2

HC

FC

Nu

clid

es

SPM

VO

Cs

PA

Hs

Co

nce

ntr

atio

n m

ine

rals

Foss

il fu

el a

vaila

bili

ty

Ch

an

ge in

hab

ita

t si

ze

Co

n. u

rban

,ag

ri.,

nat

., so

il

Co

n. o

zon

e d

ep

l. ga

ses

Dif

f. p

H –

nu

trie

nt

ava

il.

Co

n. g

reen

ho

use

ga

ses

Co

n. r

adio

nu

clid

es

Co

n. S

PM

an

d V

OC

s

Co

n. i

n a

ir, w

ate

r, f

oo

d

Loca

l eff

ect

on

va

scu

lar

pla

nt

spec

ies

Aci

dif

/eu

tr. (

occ

urr

ence

tar

get

spec

ies)

Eco

toxi

city

: to

xic

stre

ss (

PA

F)

Clim

ate

chan

ge

(dis

eas

es-d

isp

lace

m.)

Ozo

ne

laye

r d

epl.

(ca

nce

r an

d c

atar

act)

Re

spir

ato

ry e

ffe

ct (

case

s an

d t

ype

)

Car

cin

oge

nes

is (

can

cer

case

s an

d t

ype)

Ion

iz. R

ad

iati

on

(ca

nce

r ca

ses

and

typ

e)

Surp

lus

en

ergy

fo

r fu

ture

ext

ract

ion

Surp

lus

en

ergy

fo

r fu

ture

ext

ract

ion

Re

gio

n e

ffe

ct o

n v

ascu

lar

pla

nt

spe

cies

Dam

age

to m

iner

al

and

foss

il re

sou

rces

[M

J su

rplu

s e

ne

rgy]

Dam

age

to

ecos

yste

m q

ualit

y [P

DF*

m2*y

r]

Dam

age

to h

uman

he

alth

[d

isa

bili

ty

adju

sted

life

yea

rs

(DA

LY)]

Sing

le S

core

In

dic

ato

r

Re

sou

rce

an

aly

sis

Lan

d-u

se a

na

lysi

sFa

te a

nal

ysi

s

Exp

osu

re a

nd

E

ffec

t a

nal

ysis

Da

ma

ge A

na

lysi

sN

orm

alis

atio

n a

nd

W

eigh

tin

g

Methodology | 39

Energy related calculations

Cooling and heating load calculations

The amount of energy that needs to be removed or added from a space by

the air-conditioning system to maintain the level of comfort, are called

cooling and heating loads, respectively [91]. Accurate calculation of the

cooling and heating loads is important for the selection and sizing of

cooling and heating system components. In the cases of CHP or CCHP it

also affects the selection and sizing of prime mover and heat recovery

systems.

There are several components that should be taken into account when

calculating cooling and heating loads.

The heat coming from outside:

o through walls, floor, ceiling, windows and doors by

conduction, convection and radiation.

o infiltration

o ventilation

the heat generated inside:

o lighting

o electrical devices/appliances

o occupants

Since the studies focussed on economical evaluation, for these calculations

relatively simple methods were selected.

Heat gain/loss through surfaces was calculated using following formula

[92]:

𝑄 = 𝑈 ∗ 𝐴 ∗ ∆𝑇 (1)

where Q is heat (W), U is the overall heat transfer coefficient (𝑊𝑚2𝐾⁄ ), A

is the surface area in contact with outside air (𝑚2), ∆𝑇 is the temperature

difference between the two sides of the surface (𝐾).

The overall heat transfer coefficient is calculated via the formula [92];


1

𝑈=

1

ℎ𝑖+ ∑

𝑑𝑥𝑖

𝑘𝑖+

1

ℎ𝑜 (2)

where hi and ho are inner and outer convection heat transfer coefficients

(𝑊𝑚2𝐾⁄ ), dxi is the thickness of the individual material layer (m), ki is

thermal conductivity of the material.

The heat gain/loss via ventilation was calculated via the following formula

[92];

𝑄 = �̇� ∗ 𝑐𝑎𝑖𝑟 ∗ ∆𝑇 (3)

where �̇� is mass air flow (kg/h), cair specific heat of air (0.28 𝑊ℎ𝑘𝑔𝐾⁄ ), ∆𝑇

is the temperature difference between outside and inside air (K).

Heat gain from lighting and electrical devices is assumed to be equal to the

power of the lighting and devices. Heat gain from occupants in an office

environment is assumed to be 115 W/person [93].

The sum of these components gives the heating/cooling load that will be

met by the air-conditioning system. This sum will be positive for cooling

load and negative for heating load.

Adsorption cooling components sizing

Properly sizing a cooling system is important environmentally as well as

economically. To choose the correct components for an adsorption cooling

system, it is imperative to know the air flow the system will need to meet.

Figure 22 - Steps of open cycle adsorption cooling (adapted from [85])

Heat

Source

Sorption

Wheel

Heat

Exchanger

Humidifiers

Inlet

Air

Exhaust

Air

Outside Air

Waste air

1 2 3 4

569 78

Methodology | 41

Figure 22 shows the numbered steps that correspond to points on the

Mollier Diagram in Figure 23. By dividing cooling load by the difference

between the enthalpies of the 4th and 5th steps, which is the amount of heat

taken from the air-conditioned space in Wh/kg, the mass flow of the air in

the system is calculated. These enthalpy values may either be calculated

using the formulae given in [85] or using a Mollier diagram. The air flow is

also important to ensure that the system will ventilate the space sufficiently

to keep the CO2 levels below health limit, 2.7g CO2/m3.

Another important value that may be learned from the Mollier diagram is

the absolute humidity difference between steps 1 and 2. This gives the

amount of water the sorption wheel needs to take from outside air, which

determines the size and amount of desiccant in the sorption wheel.


Figure 23 - Mollier diagram of an adsorption cooling system (Paper III)

Methodology | 43

Electricity from photovoltaics and wind

Determining the amount of electricity generated from photovoltaics (PV)

panels and wind turbines is important to define the system components

and the size of them and to make sure that the system will be able to meet

the power load. The first step in this is making an availability analysis,

determining the available sun and wind power at the site. Solar irradiation,

sunshine duration and wind speed data are collected as hourly average,

daily average or monthly average according to the sensitivity needed for

the analysis. This data is then used in the formulae below to calculate the

electricity expected to be generated by PV panels and wind turbine.

Energy from PV panels:

𝐸𝑃𝑉 = (∫ 𝐼𝑟(𝑡)𝑑𝑡)𝑆ƞ𝑃𝑉ƞ𝑖 (4)

where S is the surface area of the PV cells, PV is the efficiency of the PV

panels, i is the efficiency of the inverter, and rI is the irradiance (W/m2)

[94, 95].

Electricity produced from wind:

𝐸𝑊 = (∫1

2𝜌𝐴𝐶𝑝(𝜆, 𝜃)𝑣(𝑡)3𝑑𝑡) ƞ𝑆 (5)

where S is the system efficiency with battery and inverter, is the air

density, A is the rotor sweep area, PC is the power coefficient - a function

of tip speed ratio( ) and pitch angle( ) - and v is the wind velocity (m/s)

[94, 95].

Economic calculations

Payback period

Payback period is the amount of time required to recover the funds that

were spent on an investment, or to reach the break-even point. In cases of

comparison of a new system with high investment costs to a conventional

system the formula below is used [96].


𝑃𝑎𝑦𝑏𝑎𝑐𝑘 𝑃𝑒𝑟𝑖𝑜𝑑(𝑦𝑒𝑎𝑟𝑠)

=𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝑐𝑜𝑠𝑡 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 (𝑐𝑢𝑟𝑟𝑒𝑛𝑐𝑦)

𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑐𝑜𝑠𝑡 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 (𝑐𝑢𝑟𝑟𝑒𝑛𝑐𝑦

𝑦𝑒𝑎𝑟) (6)

It is an approximate measure of the rate of cash flow. Although it does not

tell anything about the project’s earning rate after the payback, and does

not consider the size and profitability of the project, it is enough to have an

idea on what to expect from the project.

Cost of electricity comparison

Making sure that a project is profitable and its return is higher than that of

the opportunity cost of capital is essential when evaluating energy systems.

The aim is to find the alternative system that meets the energy

requirements of the project with the least cost. When comparing

alternatives with different energy output, the cost of electricity production

per kWh is used as reference [96]. The calculations made can be simplified

to the following formula:

𝐶𝑜𝐸:𝑃𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 (𝑖𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡 + 𝑎𝑛𝑛𝑢𝑎𝑙 𝑐𝑜𝑠𝑡𝑠 ∗ 𝑦𝑒𝑎𝑟𝑠)

𝑇𝑜𝑡𝑎𝑙 𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑖𝑡𝑦 𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑒𝑑 𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑖𝑡𝑦 𝑜𝑣𝑒𝑟 𝑙𝑖𝑓𝑒𝑡𝑖𝑚𝑒 (7)

Equivalent annual cash flow

Another method for evaluating an energy system is to calculate how much

benefit the system will bring yearly. As a derivation of equivalent annual

cost, Equivalent Annual Cash Flow (EACF) is a method that incorporates

one time and irregular costs such as initial investments or large part

changes, into the annual cash flow and calculated using the following

formulae [97]:

𝐸𝐴𝐶𝐹 =𝑁𝑃𝑉

𝐴𝑙,𝑖

(8)

𝑁𝑃𝑉(𝑖, 𝑙) = ∑𝑅𝑡

(1 + 𝑖)𝑡

𝑙

𝑡=0

(9)

𝐴𝑖,𝑙 =1 − 1 (1 + 𝑖)𝑙⁄

𝑖 (10)

Methodology | 45

where NPV is Net Present Value of cash flow throughout the lifetime of the

system including the capital investment, A is the present value of the

annuity factor, l is expected lifetime of the project, i is interest rate, t is

period, and Rt is cash flow (income-expenses) for a given period.


Results and Discussion | 47

Results and Discussion

There are several applications of fuel cells that may be a market entry point.

This body of work focussed on stationary applications such as backup

power, CCHP, CHP and PTG systems.

Environmental impacts of fuel cells

To gain insight into the implications of environmental impacts fuel cells

cause a preliminary environmental comparison between PV panels, wind

turbine, diesel generator and a fuel cell was made using cradle-to-gate

approach with 1 kW power generation capacity as the functional unit.

Impact assessment was made according to the Eco-indicator 99

methodology. Results were assessed in three damage categories: Human

health, Ecosystem quality and Resources.

Figure 24 - Impacts of the systems on human health (Paper I)

Figure 24 demonstrates the impact of the systems investigated on human

health in Disability-Adjusted Life Years (DALY). The impact of PV panels

and PEMFC are much larger than for both diesel generator and wind

0,00E+00

1,00E-04

2,00E-04

3,00E-04

4,00E-04

5,00E-04

6,00E-04

7,00E-04

DieselGenerator

PEMFC PV WindTurbine

Human Health

DALY


turbine. PV panels have the highest impact on human health, this is mostly

due to the emissions of volatile organic compounds and sulphuric peroxide

mixture in the production line. Comparing the backup power options,

PEMFC and diesel generator, the PEMFC has almost three times more

impact on human health than the diesel generator.

Figure 25 - Impacts of the systems on ecosystem quality (Paper I)

Calculations of the impacts on ecosystem quality includes land use,

pesticides, heavy metals and concentration changes of these in water and

soil; it gives results in Potentially disappeared fraction per square meter

per year (PDF/m2year). This is the proportion of species that might be lost

because of the impact. Figure 25 shows that PV panels have the higher

impact in this category and the diesel generator the lowest.

0,00E+00

1,00E+01

2,00E+01

3,00E+01

4,00E+01

5,00E+01

6,00E+01

DieselGenerator


Ecosystem Quality

PDF*m2yr


Figure 26 - Impact of the systems on resources (Paper I)

Although life cycle inventory studies indicate that the primary energy used

for Si feedstock and wafer production for PV panels is a large factor [98,

99], PV panels still have the least impact on resources. In Figure 26 wind

turbine and diesel generator impact on resources are shown as the highest.

0,00E+00

2,00E+01

4,00E+01

6,00E+01

8,00E+01

1,00E+02

1,20E+02

1,40E+02

1,60E+02

1,80E+02

DieselGenerator


Resources

MJ SE


Figure 27 - Combined impacts of the systems in single score (Paper I)

The results in Figure 27 show that the order of the burden from the system

production is wind turbine > diesel generator > PEMFC > PV panels. The

largest impact for all power generation systems are in the resources

damage category. Resources impact category measures the amount of extra

energy needed to extract the resources from nature by the next

generations. As the study’s scope is cradle-to-gate, not including recycling

possibilities at the decommissioning, and the materials used at both diesel

generator and wind turbine production are highly recyclable in comparison

to the ones used in PV and PEMFC production, these results are not

enough to give a fair comparison.

0

100

200

DieselGenerator

PEMFC PVWind

Turbine

Resources 200,73 148,61 140,60 211,86

Ecosystem Quality 0,92 1,98 3,10 1,10

Human Health 13,14 38,22 40,04 16,40

Pt

Single Score Results


In light of this a more comprehensive study was needed to be made to

obtain a better insight into the full impact of these systems on the

environment. A full LCA study including combinations of these power

generation systems was performed. Table 2 shows the list of combinations

that were investigated. The results were evaluated to compare the

contribution of system components toward the total impact.

Table 2 – Combinations of power generation systems (Paper II)

Main Power Source Backup Power

System

Hydrogen

Storage

Photo Voltaic + Wind

Turbine Diesel Generator -----


Turbine Fuel Cell 150 bar


Turbine Fuel Cell Metal hydride tank

Photo Voltaic Diesel Generator -----

Photo Voltaic Fuel Cell 150 bar

Photo Voltaic Fuel Cell Metal hydride tank

Wind Turbine Diesel Generator -----

Wind Turbine Fuel Cell 150 Bar

Wind Turbine Fuel Cell Metal hydride tank


Figure 28 - Distribution of the system impact over damage categories –

Comparison of main power sources (Paper II)

To compare the main source alternatives the diesel generator was kept

constant as backup system. Since it uses the least amount of diesel, PV-W

was expected to have the least impact, as may be seen in the Figure 28.

However it has the highest impact on the environment. Results indicate

that a wind turbine is a more environmentally friendly option as the main

power source.

0,00

500,00

1000,00

1500,00

2000,00

2500,00

PV-W+DG PV+DG W+DG

Resources 2337,09 2248,25 2185,50

Ecosystem Quality 19,18 17,11 11,65

Human Health 204,65 185,55 130,43

Sin

gle

Sco

re In

dic

ato

r



Comparison of backup power systems (Paper II)

Figure 29 illustrates the comparison between backup systems, which is

made with the combination of PV panels and wind turbine as the main

power source for its lower backup power requirement. Among the three

options, the diesel generator has the least impact on the environment.

Compared to a diesel generator, fuel cell systems have higher, by more than

three times, impact in the human health damage category. In both

combinations fuel cells constitute the largest part of the human health

impact.

0,00

500,00

1000,00

1500,00

2000,00

2500,00

3000,00

3500,00

PV-W+DGPV-

W+FC/MeHPV-

W+FC/PH2

Resources 2337,09 2247,09 2079,34

Ecosystem Quality 19,18 43,47 39,27

Human Health 204,65 711,42 640,99

Sin

gle

Sco

re In

dic

ato

r



Comparison of hydrogen storage methods (Paper II)

Figure 30 shows the results of the comparison between hydrogen storage

alternatives. Metal hydride storage tanks have more impact than

pressurised hydrogen tanks in every damage category due to the

components such as lanthanum compounds used in them.

When all systems were compared together, results show that PV panels

with wind turbine coupled with a fuel cell system as backup with metal

hydride hydrogen storage has the highest impact on environment. The

system with a wind turbine as the main power source and a diesel generator

as backup power has the least overall impact. This is probably due to the

highly recyclable materials used in these systems compared to the others.

Figure 31 shows the overall comparison results for all nine alternatives.

0,00

500,00

1000,00

1500,00

2000,00

2500,00

3000,00

3500,00

PV-W+FC/MeH PV-W+FC/PH2

Resources 2247,09 2079,34

Ecosystem Quality 43,47 39,27

Human Health 711,42 640,99

Sin

gle

Sco

re In

dic

ato

r


Figure 31 - Overall LCA results for nine alternatives (Sorted by total

impacts points) (Paper II)

When looking at damage categories separately;

in resources damage category: PV panels with wind turbine as main

power coupled with a diesel generator has the most, wind turbine

coupled with fuel cell with pressurised hydrogen has the least impact.

in ecosystem quality damage category: PV panels with wind turbine

with fuel cell using metal hydride has the most, wind turbine with

diesel generator has the least impact.

and in human health damage category: again PV panels with wind

turbine with fuel cell using metal hydride has the most, wind turbine

with diesel generator has the least impact.

Between the individual components: the battery bank has the highest

impact in the resources damage category. Fuel cells have the highest

impact in both the ecosystem quality and the human health categories

and second highest in the resources category.


Fuel cell-based combined cooling, heating and power

The first application economically evaluated was a hypothetical PEMFC

cogeneration system powering an office building in Turkey. Coupled with

an Open Cycle Adsorption Cooling system, the system meets the building’s

cooling demand as well as its power and heating demands. The preliminary

assessment was based on a traditional grid electricity – boiler heating –

electric air conditioner triad. The results showed that the difference in the

capital costs may be paid in full by the yearly savings in operational costs

in less than 13 years. Expected rises in electricity prices may cause a more

rapid payback in the future. Table 3 and Table 4 show capital and

operational costs for a conventional system and PEMFC-CHP.

Table 3 - Capital costs comparison (Paper III)

Capital Costs (EUR)

Fuel cell-based CHP 257.000 EUR

Desiccant cooling 85.500 EUR

CHCP Total 342.500 EUR

Boiler system for heating 7.000 EUR

AC 16.000 EUR

Conventional System Total 23.000 EUR

Table 4 - Operational costs comparison (Paper III)

Operational (EUR/year)

Fuel cell-based CHP maintenance 5.900 EUR

Desiccant cooling maintenance 1.250 EUR

Fuel Cost 31.100 EUR

Avoidance by selling electricity 17.900 EUR

CCHP Total 20.350 EUR

Fuel cost for heating 22.900 EUR

Electricity (Including A/C) 22.600 EUR

Conventional System Total 45.500 EUR


Although it is technologically possible to use fuel cells as prime mover for

CCHP systems, high capital costs make them unfeasible for the current

state of technology and energy prices. Although Governmental incentives

for environmentally friendly technologies are not taken into account in the

analysis, they may decrease the payback time making the system more

economically feasible. The FC-CCHP has a substantially lower operating

cost.

Excess electricity that is assumed to be sold to the grid is a significant part

of the savings in operational costs of a FC-CCHP. Without electricity

sellback the payback takes 16 years. From this result it may be extrapolated

that, to encourage investors to put money into such DG systems, it is

imperative to realise the electricity sellback schemes.

With the results of this study the question of ‘What are the conditions that

make FC-CCHP economically feasible?’ has been partially answered. A

high and consistent thermal and electrical demand were the essential

requirements for making FC-CCHP feasible. The next step in the search

was to find an application with a steady thermal and electrical demand;

cooling and powering of a data centre could meet the requirement. So the

next application to be investigated was a system that powers and cools a

data centre. An MCFC-based CHP system coupled with an absorption

cooling system was assessed to see if it could meet the energy demand of a

data centre at KTH – Royal Institute of Technology, campus Stockholm,

Sweden.

EACF for the system was calculated using the savings over the existing

methods (district cooling and grid electricity) as income. Calculations

revealed that the system saves around 815 000 EUR on electricity, cooling

and heating expenses as well as maintenance of backup systems compared

to the existing methods. However, considering the capital and O&M costs,

the system has an equivalent annual loss of approximately 317 000 EUR.

This means that, with current energy prices and state of technology, such

investments are not feasible. Assuming no change in the natural gas (NG)

price and a yearly 3 % rise in electricity and 2 % rise in cooling and heating

prices as well as in O&M costs, the system becomes feasible giving an EACF

of approximately 428 000 EUR.


As a data centre has a high thermal energy demand, the high electrical

efficiency of a fuel cell causes an excess of generated electricity when the

system is set to meet the cooling demand. Since Sweden does not have an

electricity sell-back scheme, the excess electricity is assumed to be used in

the office part of the building where the data centre is located and also in

the surrounding buildings on campus. It is also possible to use extra

electricity in electric cooling to decrease the cooling demand or to meet the

power demand and supplement the extra cooling from district cooling.

Both cases are proven to be less economic by approximately 83 000

EUR/year and 430 000 EUR/year, respectively.

The lifetime for fuel cell stacks is increasing with continuous research and

the next generation is expected to have 7 years of lifetime. When this

change in lifetime is taken into account, i.e. considering stacks with 7 years

lifetime, the EACF is increased by approximately 131 000 EUR. There is

also a 10 year warranty from the manufacturer that includes the first fuel

cell stack change for free. If this deal is to be incorporated into the

calculations the loss becomes more tolerable at 94 000 EUR/year. Table 5

shows the EACF results for all six scenarios.

Table 5 - EACFs of different scenarios (Paper VI)

Scenario EACF

Original Scenario -317 000 EUR

Energy Price Projections +428 000 EUR

Extra El. used for cooling -400 000 EUR

Power Load Mode -745 000 EUR

7 Years Stack Lifetime -186 000 EUR

Fuel Cell Energy Deal -94 000 EUR

High capital costs and O&M costs together with low lifetime of fuel cell

stacks causes the FC-CCHP to be unfeasible. Nonetheless, in the near

future these systems may become feasible through technological

improvements and decrease in O&M and capital costs due to mass

production triggered by a broader market. A 1 % yearly change in natural

gas and electricity prices brings a change of 50 % and 56 % in EACF,

respectively, meaning a substantial change in natural gas and electricity

prices may cause the FC-CCHP to become feasible.


Although the high thermal demand of data centres makes it possible to use

less electrically efficient prime movers, advantages such as low emission

and noise makes fuel cells a major competitor for data centre CCHP. Using

FC-CCHP as the main energy system with grid and district cooling as

backup will also increase energy reliability for the data centre.

The projected changes for natural gas and electricity prices, possible

savings due to heat recovery from the server racks and absorption cycle

exhaust and also possible governmental subsidies and sustainability tax

incentives could make FC-CCHP for data centres a feasible investment.

Fuel cell-based combined heating and power

The unit cost of electricity is used as a criterion for comparing Fuel Cell-

based Combined Heat and Power (FC-CHP) and other small-scale

electricity generation technologies. The data used was collected from

literature, personal interviews, technical reports and product brochures.

Results were grouped by technology, MCFC-based systems and PEMFC-

based systems, and time, 2010 and 2020 projections. The average

electricity cost in 2010 from MCFC-CHP and PEMFC-CHP is 0.29

EUR/kWhel and 0.41 EUR/kWhel, respectively. Projections indicate that

this will be halved by 2020. It will be 0.14 EUR/kWhel for MCFC-CHP and

0.24 EUR/kWhel for PEMFC-CHPi. Table 6 shows a comparison of the

results of Paper IV to results for semi-commercial small-scale technologies

from a report [101] published in 2011 and using the same model. Although

with more scope for improvement a larger reduction in electricity cost for

PEMFC-CHP and MCFC-CHP is expected, in the case of PEMFC-CHP the

electricity cost will be higher than that of other technologies in 2020 as well

as 2010. The MCFC-CHP, on the other hand, might be comparable to some

of the currently semi-commercial technologies such as solar cells and

biogas internal combustion engine. These results also indicate that MCFC-

CHPs might reach below the upper limit of levelised cost of energy for fuel

i The calculations were made in Swedish Crowns (SEK) and then the results were converted to euro [100]


cells published in Lazard’s LCOE Analysis – V5 (0.17 EUR/kWh), meaning

that MCFC-CHP systems might make profit albeit small.

Table 6 - Electricity costs of different semi-commercial small scale

technologies from [101] compared with the results of (Paper IV)

(EUR/kWhel) Capacity

(MW)

2010 2020

Biogas – Internal combustion engine 1 0.20 0.16

Natural gas motor 1 0.08 0.08

Natural gas motor 0.1 0.11 0.11

Biogas organic rankine cycle 2 0.14 0.13

Waste heat organic rankine cycle 0.5 0.08 0.08

Solar cells 0.05 0.55 0.23

MCFC-based CHP ii 0.29 0.14

PEMFC-based CHP ii 0.41 0.24

Although in comparison to other small scale semi-commercial systems,

FC-CHPs have higher electricity costs, it is projected that the FC-CHPs will

have larger improvements and thus larger cost reduction in the future. It

is also imperative to consider all the advantages fuel cells if fuel cells are to

make it to the market on commercial grounds. Table 6 shows this

comparison.

It can be seen in Figure 32 that cost contribution of fuel increases as

expected between 2010 and 2020. Contrary to the expectation however,

the share of O&M cost also increases. This is because of the unreliable

nature of O&M costs of demonstration projects, which are simply higher

than for commercial applications. The limited number of trained personnel

for O&M and the low maturity of fuel cell systems are the two main causes

for high O&M costs for a young technology. This may be overcome by a

broader market, increasing the number of skilled personnel and product

maturity. It will of course also bring higher availability and reliability for

ii Capacity of the systems investigated in this study varies.


these systems. Collection of more realistic O&M data and their evolution

over time may help a great deal in solving this uncertainty. This is an

important area for further studies.

Figure 32 - Comparison of average cost contribution

Figure 32 also shows that the share of capital cost declines with time, as

expected. This decline will surely be greater, due to the fact that current

projections for fuel cell stack lifetime are higher than they were when the

study was performed.

Power-to-gas

Power-to-gas is a promising stationary application for fuel cells. PTG with

fuel cells may be the solution to the sporadic nature of wind and solar

power. To determine the economic feasibility of FC-PTG an existing case


in Bozcaada, Turkey was investigated. Three hypothetical scenarios were

taken into consideration as well as the existing system.

Figure 33 shows the block diagrams of the scenarios investigated.

Figure 33 - Block diagrams of the scenarios studied. (A) Scenario 1:

Existing system of PTG in Bozcaada (B) Scenario 2: PEM fuel cell and

electrolyser replaced by a Reverse Molten Carbonate Fuel Cell (R-

MCFC) (C) Scenario 3: No electricity generation from H2, it is sold as

chemical or fuel (D) Scenario 4: Only PV panels and Wind (Paper V)

The system generates a 179 MWh/year electricity and has a 62.4

MWh/year demand. Table 7 shows the per year results of the power

generation calculations for each scenario. The energy calculation results

are same for scenarios 1 and 2 since the only difference between the

20 kW PV Panels

30 kW Wind Turbine

AC

=

~

~

~

Governor s Building

Net Metering

Grid

(D)

20 kW PV Panels

30 kW Wind Turbine

21 kW R-MCFC

H2 @30 Bar

AC

=

~

=

~

~

~

Governor s Building

Net Metering

Grid

(B)

Compressor

(A)

20 kW PV Panels

30 kW Wind Turbine

21 kW FC

H2 @30 Bar

11 Nm3/h Electrolyzer

AC

=

~

=

~

~

~

Governor s Building

Net Metering

Grid

Compressor

(C)

Grid

20 kW PV Panels

30 kW Wind Turbine

H2 @200 Bar

11 Nm3/h Electrolyzer

=

~

~

~

Governor s Building

Net Metering

AC

Compressor


scenarios is the fuel cell and electrolyser technologies. Scenario 4, which

does not have power-to-gas capabilities, has the highest amount of

electricity sold and bought from the grid because there is no energy storage.

Table 7 - Results of energy calculations (per year) (Paper V)

Scenarios 1 - 2 Scenario 3 Scenario 4

Sold to grid 104 MWh 2 MWh 128 MWh

Bought from grid 2 MWh 12 MWh 12 MWh

Used in electrolyser 24 MWh 126 MWh -

Sold as H2 - 75 MWh -

The results showed that all of the scenarios with PTG have negative EACF

and, therefore, will result in loss over their lifetime; hence they are

unfeasible. PTG with R-MCFC has the highest EACF, due to the lower

capital cost compared to PEMFC + electrolyser. Although the PV-Wind

system itself is a relatively good investment, almost feasible with a 1:1 feed-

in tariff, the lifetimes of electrolysers and fuel cells are not long enough and

capital costs are too high. Hence investing in PTG systems for a small-scale

system like this is not feasible. Table 8 lists the EACF values of the

scenarios for different feed-in tariff cases.

Table 8 – EACF values of the scenarios in different feed-in tariff cases

(Paper V)

EACF (USD/year) Scenario 1 Scenario 2 Scenario 3

Scenario 4 Diesel Chemical

No feed-in tariff -85 492 -63 487 -65 019 -69 489 -17 467

1:1 feed-in tariff -75 801 -53 797 -64 791 -69 201 -796

2:1 feed-in tariff -66 111 -44 106 -64 563 -69 033 15 874

Incorporating an increase of 4-6 % in electricity price and 2-3 % on O&M

costs shows that, although the results are sensitive to such changes, this is

not enough to turn the system feasible.

Figure 34 shows the effect of feed-in-tariff on EACF of the scenarios. This

impact is highest in Scenario 4, where the amount of electricity sold to the


grid is highest. For this scenario the no feed-in-tariff case causes a loss of

17 500 USD a year. A 1:1 feed-in tariff increases the savings by 3.5 times

and, if the feed-in tariff is raised to double the electricity price, as it was in

Germany between 2000 and 2009 [102], the savings in electricity bill goes

up 6 times and the system makes a profit of approximately 16 000 USD per

year.

Figure 34 - Effect of feed-in tariff on EACF (Paper V)

-90 000

-80 000

-70 000

-60 000

-50 000

-40 000

-30 000

-20 000

-10 000

0

10 000

20 000

0 : 1 1 : 1 2 : 1

EAC

F

FEED-IN TARIFF RATIO

Scenario 1

Scenario 2

Scenario 3(Diesel)

Scenario 3(Chemical)

Conclusions and recommendations | 65

Conclusions and recommendations

Promising applications were looked into during this work to find the niche

application that will give fuel cells the opportunity of market penetration.

The Results of these studies pointed to the following conclusions.

Results from the environmental studies imply that, although during the

use phase fuel cells have next to no emissions, when the manufacturing and

decommissioning phases are included in the calculations they are not so

environmentally friendly. The largest part of the fuel cells’ environmental

impact lies in the resources used, mainly platinum group metals (PGM).

Decreasing loading per geometric electrode area and waste minimisation

during the catalyst deposition and, more importantly, increasing recycling

of PGM are ways to improve this situation. Turning to non-PGM catalysts

might also be a solution. Research should take into account not just the

economics of using PGM as catalysts but also the environmental impact it

brings.

Market size is the key to the future of fuel cells. In all the studies that

constitute this work, capital cost comprised the largest share of the fuel cell

systems’ total cost. This is due to the nature of the ‘production to order’

style of the current fuel cell market. Without a considerable market size it

is not possible to obtain economies of scale. The advantage of an increase

in market size would not only be a decrease in capital costs but also in O&M

costs. Currently, the limited number of trained personnel that deal with

fuel cell system in operation are highly educated people, engineers with

master’s degrees and even PhDs. This brings the O&M costs to a higher

level due to the high salaries of these personnel. A bigger market size will

increase the number of skilled personnel and also decrease the cost of spare

parts, which in turn will bring down the O&M costs. Another reason for

high O&M costs is the low product maturity. However, as the market size

and throughput of fuel cells coming out of factories increase, the product

maturity will also increase, which of course will bring greater reliability and

lower maintenance needs. These are the reasons that finding a niche

market for fuel cells is of the utmost importance.

Without governmental subsidies fuel cell market penetration would be

crippled. Governmental subsidies for fuel cells specifically, as well as


general schemes such as carbon trade, green certificates, tax rebates for

green technologies, may counteract economic handicaps of fuel cells such

as capital costs and stack lifetime. This will make fuel cells attractive for

investors and therefore speed up the market penetration process.

Feed-in tariff is essential for fuel cells to be feasible, especially for small-

scale co-generation applications. Having high electrical efficiency makes

fuel cells a great candidate for being a co-generation prime mover.

However CHP and CCHP systems with fuel cells end up with excess

electricity when the operation mode is selected to meet thermal load.

Without an electricity sellback scheme it is possible to set the system to

meet the electricity load and meet the excess thermal load with an auxiliary

system such as a boiler, but this method would be inefficient compared to

the thermal load operation mode. In cases where there is no electricity

sellback possibility selling/giving the excess electricity to surrounding

buildings/loads might be a solution. This would limit the sites for fuel cell-

based co-generation. Generating hydrogen with the excess electricity is

another solution to the no sellback issue. Depending on the pricing of

hydrogen and electricity this solution might be more profitable on a yearly

basis, but the capital cost needed for electrolyser and hydrogen

pressurising and storage equipment would render it unfeasible.

Fuel cell stack lifetime has to be improved. It is evident from the results of

the studies that lifetime of the fuel cell stack is an important factor for

feasibility. The stack of a fuel cell system corresponds approximately to 30

% of the total cost. Currently the lifetime of fuel cell stacks prevent fuel cell

systems from ever paying their initial capital cost, due to the cost of fuel

cell stack changes over the lifetime - approximately three stack changes

over the lifetime of the system. The next generation fuel cells are expected

to have almost 1.5 times longer lifetime but it is still not enough when

compared to competing technologies. Lifetime should be the prominent

focus of research going forward.

Decision makers should bear in mind the other benefits and impacts as well

as the economic ones. It might be possible to increase the attractiveness of

fuel cells to investors through environmental subsidies and incentives.

Emphasising the added value of fuel cells for the consumer may be a way

to offset the high costs of fuel cells. Steinberger-Wilckens [9] stated that

Conclusions and recommendations | 67

“Only if the added value and environmental performance are convincingly

displayed will fuel cells be able to claim a substantial part of their potential

market.” It is imperative to show the consumer / investor / decision maker

the added values they will find make it worth the price difference from the

incumbent technologies.

In conclusion, although the current state of technology and energy prices

renders it economically unfeasible, FC-CCHP for data centres, may be a

suitable niche market to allow fuel cells to blossom due to their steady

thermal and power load and constant increase in data centres number.

This niche market may provide the pathway for fuel cells market expansion

and bring mass production and subsequently lower capital and O&M costs.

With these in mind and considering the current projections for stack

lifetime and energy prices, it is expected that an FC-CCHP system meeting

the energy demands of a data centre will be soon economically feasible and

attractive to investors, thus giving fuel cells the opportunity of market

penetration.


Acknowledgements | 69

Acknowledgements

I would like to express my deep and sincere gratitude to my supervisors;

Göran Lindbergh, Carina Lagergren and Per Alvfors, whose guidance and

support from the initial to the final level enabled me to develop a better

understanding of the scientific process and the subject.

I would like to thank to Bengt Ridell for the insights he has given, Gökçen

Altun for the help she has given during my studies, Gert Svensson for his

support on data centre project, James Wetherhilt and Osman Atanur for

their help with Bozcaada project.

Thanks to Neşet Kadırgan for his support and encouragement of a young

student and inspiration in the field of energy, which set me on my course

toward PhD studies in this field

My deepest gratitude goes to my wife Linda for her unflagging love and

support on every aspect of my life; this dissertation was simply impossible

without her.

I am also grateful to my family for their support throughout my journey in

life; my father for being a role model and teaching me the values of integrity

and hard work, my mother for her constant love and care, my brother

Murat for “infecting” me with love of science, my sisters Reyhan and Fatma

for making me feel loved and never alone.

A special thanks to Rebecca Wilcox Gwynne, who suffered my rantings,

kept me calm, and helped me keep myself healthy.

I offer my regards and gratitude to all of those who believed in me and

supported me in any respect during this long journey with ups and downs.

Finally, my dog Merlin for reminding me of the need to go outside once in

a while and walk in the fresh air while writing a dissertation.


References | 71

References

[1] IPCC, 2013. Summary for Policy Makers. Climate Change 2013:

The Physical Science Basis Contribution of Working Group I to the

Fifth Assessment Report of the Intergovernmental Panel on Climate

Change. Cambridge, United Kingdom and New York, NY, USA.:

Cambridge University Press; 2013.

[2] IPCC, 2014. Summary for Policymakers. Climate Change 2014:

Mitigation of Climate Change Contribution of Working Group III to

the Fifth Assessment Report of the Intergovernmental Panel on

Climate Change. Cambridge, United Kingdom and New York, NY,

USA.: Cambridge University Press; 2014.

[3] OECD/IEA. Energy Technology Perspective. Paris2012.

[4] Ball M, Weeda M. The hydrogen economy – Vision or reality?

International Journal of Hydrogen Energy. 2015;40:7903-19.

[5] Nejat Veziroǧlu T. Quarter century of hydrogen movement

1974–2000. International Journal of Hydrogen Energy.

2000;25:1143-50.

[6] Ball M, Martin Wietschel e. The Hydrogen Economy:

Opportunities and Challenges. New York: Cambridge University

Press; 2009.

[7] The Hydrogen Economy: A Non-technical Review: United

Nations Environment Programme; 2006.

[8] Hardman S, Shiu E, Steinberger-Wilckens R. Changing the fate

of Fuel Cell Vehicles: Can lessons be learnt from Tesla Motors?


[9] Steinberger-Wilckens R. Not cost minimisation but added value

maximisation. International Journal of Hydrogen Energy.

2003;28:763-70.

[10] Hardman S, Chandan A, Steinberger-Wilckens R. Fuel cell

added value for early market applications. Journal of Power Sources.

2015;287:297-306.

[11] Hardman S, Steinberger-Wilckens R, van der Horst D.

Disruptive innovations: The case for hydrogen fuel cells and battery

electric vehicles. International Journal of Hydrogen Energy.

2013;38:15438-51.


[12] Pehnt M. Life-cycle assessment of fuel cell stacks. International

Journal of Hydrogen Energy. 2001;26:91-101.

[13] Lunghi P, Bove R. Life Cycle Assessment of a Molten

Carbonate Fuel Cell Stack. Fuel Cells. 2003;3:224-30.

[14] Khan FI, Hawboldt K, Iqbal MT. Life Cycle Analysis of wind–

fuel cell integrated system. Renewable Energy. 2005;30:157-77.

[15] Staffell I, Ingram A. Life cycle assessment of an alkaline fuel

cell CHP system. International Journal of Hydrogen Energy.

2010;35:2491-505.

[16] Staffell I, Ingram A, Kendall K. Energy and carbon payback

times for solid oxide fuel cell based domestic CHP. International


[17] Mori M, Jensterle M, Mržljak T, Drobnič B. Life-cycle

assessment of a hydrogen-based uninterruptible power supply

system using renewable energy. Int J Life Cycle Assess.

2014;19:1810-22.

[18] Cox B, Treyer K. Environmental and economic assessment of a

cracked ammonia fuelled alkaline fuel cell for off-grid power

applications. Journal of Power Sources. 2015;275:322-35.

[19] Strazza C, Del Borghi A, Costamagna P, Gallo M, Brignole E,

Girdinio P. Life Cycle Assessment and Life Cycle Costing of a

SOFC system for distributed power generation. Energ Convers

Manage. 2015;100:64-77.

[20] Silveira JL, Gomes LA. Fuel cell cogeneration system: a case

of technoeconomic analysis. Renewable and Sustainable Energy

Reviews. 1999;3:233-42.

[21] Lipman TE, Edwards JL, Kammen DM. Fuel cell system

economics: Comparing the costs of generating power with stationary

and motor vehicle PEM fuel cell systems. Energy Policy.

2004;32:101-25.

[22] Alanne K, Saari A, Ugursal VI, Good J. The financial viability

of an SOFC cogeneration system in single-family dwellings. Journal

of Power Sources. 2006;158:403-16.

[23] Zabalza I, Aranda A, Dolores de Gracia M. Feasibility analysis

of fuel cells for combined heat and power systems in the tertiary

References | 73

sector. International Journal of Hydrogen Energy. 2007;32:1396-

403.

[24] Staffell I, Green R, Kendall K. Cost targets for domestic fuel

cell CHP. Journal of Power Sources. 2008;181:339-49.

[25] Ren H, Gao W. Economic and environmental evaluation of

micro CHP systems with different operating modes for residential

buildings in Japan. Energ Buildings. 2010;42:853-61.

[26] Pruitt KA, Braun RJ, Newman AM. Establishing conditions for

the economic viability of fuel cell-based, combined heat and power

distributed generation systems. Applied Energy. 2013;111:904-20.

[27] Staffell I, Green R. The cost of domestic fuel cell micro-CHP

systems. International Journal of Hydrogen Energy. 2013;38:1088-

102.

[28] Chen JMP, Ni M. Economic analysis of a solid oxide fuel cell

cogeneration/trigeneration system for hotels in Hong Kong. Energ

Buildings. 2014;75:160-9.

[29] Napoli R, Gandiglio M, Lanzini A, Santarelli M. Techno-

economic analysis of PEMFC and SOFC micro-CHP fuel cell

systems for the residential sector. Energ Buildings. 2015;103:131-

46.

[30] Mick J. Microsoft's New Fuel Cell Partner is Ready to Blow

Away the Bloom Box -

http://www.dailytech.com/Microsofts+New+Fuel+Cell+Partner+is

+Ready+to+Blow+Away+the+Bloom+Box/article36118.htm.

DailyTech; June 26, 2014.

[31] Microsoft looks at fuel cell power in data centres via server rack.

Fuel Cells Bulletin. 2013;2013:10.

[32] Fuji fuel cell operating at Frankfurt data centre. Fuel Cells

Bulletin. 2013;2013:1.

[33] eBay turns on first fuel cell powered data centre, has Bloom

tech. Fuel Cells Bulletin. 2013;2013:3-4.

[34] Apple doubles fuel cell capacity at NC data centre with Bloom.


[35] Verizon adds Bloom fuel cell units at three facilities in

California. Fuel Cells Bulletin. 2013;2013:7.

http://www.dailytech.com/Microsofts+New+Fuel+Cell+Partner+is+Ready+to+Blow+Away+the+Bloom+Box/article36118.htm

http://www.dailytech.com/Microsofts+New+Fuel+Cell+Partner+is+Ready+to+Blow+Away+the+Bloom+Box/article36118.htm


[36] Equinix installs 1 MW Bloom unit at Silicon Valley data centre.


[37] Zhao L, Brouwer J, James S, Peterson E, Siegler J, Kansal A, et

al. Servers Powered By a 10kW In-rack Proton Exchange Membrane

Fuel Cell System. ASME 2014 8th International Conference on

Energy Sustainability & 12th Fuel Cell Science, Engineering and

Technology Conference. Boston: ASME; 2014.

[38] Zhao L, Brouwer J, Liu J, James S, Siegler J, Kansal A, et al.

Fuel Cells for Data Centers: Power Generation Inches From the

Server. Microsoft Research Technical Report MSR-TR-2014-37.;

2014.

[39] Riekstin AC, James S, Kansal A, Liu J, Peterson E. No more

electrical infrastructure: towards fuel cell powered data centers.

SIGOPS Oper Syst Rev. 2014;48:39-43.

[40] Schmidt R, Iyengar M, Steffes J, Lund V, Asme. Co-generation

- Grid Independent Power and Cooling for a Data Center. Ipack

2009: Proceedings of the ASME Interpack Conference 2009, Vol 2.

New York: Amer Soc Mechanical Engineers; 2010. p. 1063-75.

[41] Xu D, Qu M. Energy, environmental, and economic evaluation

of a CCHP system for a data center based on operational data. Energ

Buildings. 2013;67:176-86.

[42] Guizzi GL, Manno M. Fuel cell-based cogeneration system

covering data centers’ energy needs. Energy. 2012;41:56-64.

[43] Chen F, Duic N, Manuel Alves L, da Graça Carvalho M.

Renewislands—Renewable energy solutions for islands. Renewable

and Sustainable Energy Reviews. 2007;11:1888-902.

[44] Götz M, Lefebvre J, Mörs F, McDaniel Koch A, Graf F, Bajohr

S, et al. Renewable Power-to-Gas: A technological and economic

review. Renewable Energy. 2015.

[45] Kalinci Y. Alternative energy scenarios for Bozcaada island,

Turkey. Renewable and Sustainable Energy Reviews. 2015;45:468-

80.

[46] Loisel R, Baranger L, Chemouri N, Spinu S, Pardo S. Economic

evaluation of hybrid off-shore wind power and hydrogen storage

system. International Journal of Hydrogen Energy. 2015;40:6727-

39.

References | 75

[47] Enevoldsen P, Sovacool BK. Integrating power systems for

remote island energy supply: Lessons from Mykines, Faroe Islands.

Renewable Energy. 2016;85:642-8.

[48] Glatzmaier G, Blake D, Showalter S. Assessment of methods

for hydrogen production using concentrated solar energy. Golden,

Colorado: National Renewable Energy Laboratory - U.S.

Department of Energy; January 1998.

[49] Greiner CJ, Korpås M, Holen AT. A Norwegian case study on

the production of hydrogen from wind power. International Journal

of Hydrogen Energy. 2007;32:1500-7.

[50] Bernal-Agustín JL, Dufo-López R. Hourly energy management

for grid-connected wind–hydrogen systems. International Journal of

Hydrogen Energy. 2008;33:6401-13.

[51] Mathur J, Agarwal N, Swaroop R, Shah N. Economics of

producing hydrogen as transportation fuel using offshore wind

energy systems. Energy Policy. 2008;36:1212-22.

[52] Gökçek M. Hydrogen generation from small-scale wind-

powered electrolysis system in different power matching modes.


[53] Olateju B, Kumar A. Hydrogen production from wind energy

in Western Canada for upgrading bitumen from oil sands. Energy.

2011;36:6326-39.

[54] Negrou B, Settou N, Chennouf N, Dokkar B. Valuation and

development of the solar hydrogen production. International Journal

of Hydrogen Energy. 2011;36:4110-6.

[55] Genç G, Çelik M, Serdar Genç M. Cost analysis of wind-

electrolyzer-fuel cell system for energy demand in Pınarbaşı-

Kayseri. International Journal of Hydrogen Energy. 2012;37:12158-

66.

[56] Genç MS, Çelik M, Karasu İ. A review on wind energy and

wind–hydrogen production in Turkey: A case study of hydrogen

production via electrolysis system supplied by wind energy

conversion system in Central Anatolian Turkey. Renewable and

Sustainable Energy Reviews. 2012;16:6631-46.


[57] Christensen CM. The Innovator's Dilemma: When New

Technologies Cause Great Firms to Fail: Harvard Business School

Press; 1997.

[58] Avadikyan A, Cohendet P, Héraud JA, appliquée ULPdSBdéte.

The Economic Dynamics of Fuel Cell Technologies. Strasbourg:

Springer; 2003.

[59] Tso C, Chang S-Y. A viable niche market—fuel cell scooters in

Taiwan. International Journal of Hydrogen Energy. 2003;28:757-62.

[60] Cigolotti V, Massi E, Moreno A, Polettini A, Reale F. Biofuels

as opportunity for MCFC niche market application. International


[61] Ramírez-Salgado J, Domínguez-Aguilar MA. Market survey of

fuel cells in Mexico: Niche for low power portable systems. Journal

of Power Sources. 2009;186:455-63.

[62] Klebanoff LE, Breit JS, Roe GS, Damberger T, Erbel T,

S.Wingert, et al. Fuel cell mobile lighting: A fuel cell market

transformation project. International Journal of Hydrogen Energy.

2014;39:12948-72.

[63] Brown JE, Hendry CN, Harborne P. An emerging market in fuel

cells? Residential combined heat and power in four countries.

Energy Policy. 2007;35:2173-86.

[64] Cottrell CA, Grasman SE, Thomas M, Martin KB, Sheffield

JW. Strategies for stationary and portable fuel cell markets.


[65] Larminie J, Dicks A. Fuel cell systems explained. West Sussex:

J. Wiley; 2003.

[66] Haile SM. Fuel cell materials and components. Acta Materialia.

2003;51:5981-6000.

[67] Dincer I, Zamfirescu C. Chapter 4 - Hydrogen and Fuel Cell

Systems. In: Zamfirescu ID, editor. Advanced Power Generation

Systems. Boston: Elsevier; 2014. p. 143-98.

[68] Singh B, Guest G, Bright RM, Stromman AH. Life Cycle

Assessment of Electric and Fuel Cell Vehicle Transport Based on

Forest Biomass. J Ind Ecol. 2014;18:176-86.

[69] Wilson BP, Lavery NP, Jarvis DJ, Anttila T, Rantanen J, Brown

SGR, et al. Life cycle assessment of gas atomised sponge nickel for

References | 77

use in alkaline hydrogen fuel cell applications. Journal of Power

Sources. 2013;243:242-52.

[70] Zucaro A, Fiorentino G, Zamagni A, Bargigli S, Masoni P,

Moreno A, et al. How can life cycle assessment foster

environmentally sound fuel cell production and use? International


[71] Hussain MM, Dincer I, Li X. A preliminary life cycle

assessment of PEM fuel cell powered automobiles. Appl Therm Eng.

2007;27:2294-9.

[72] Ally J, Pryor T. Life-cycle assessment of diesel, natural gas and

hydrogen fuel cell bus transportation systems. Journal of Power

Sources. 2007;170:401-11.

[73] Chicco G, Mancarefla P. Distributed multi-generation: A

comprehensive view. Renew Sust Energ Rev. 2009;13:535-51.

[74] Mancarella P, Chicco G. Distributed Multi-generation Systems:

Energy Models and Analyses. Hauppauge: Nova Science Publishers;

2009.

[75] OECD/IEA. Distributed Generation in Liberalised Electricity

Markets. Paris, France: International Energy Agency (IEA); 2002.

[76] Pepermans G, Driesen J, Haeseldonckx D, Belmans R,

D'haeseleer W. Distributed generation: definition, benefits and

issues. Energy Policy. 2005;33:787-98.

[77] Ackermann T. Distributed resources and re-regulated electricity

markets. Electric Power Systems Research. 2007;77:1148-59.

[78] Asif M, Muneer T. Energy supply, its demand and security

issues for developed and emerging economies. Renewable and

Sustainable Energy Reviews. 2007;11:1388-413.

[79] Dincer I, Zamfirescu C. Chapter 9 - Multigeneration Systems.

In: Zamfirescu ID, editor. Advanced Power Generation Systems.

Boston: Elsevier; 2014. p. 517-73.

[80] Yilanci A, Dincer I, Ozturk HK. A review on solar-

hydrogen/fuel cell hybrid energy systems for stationary applications.

Prog Energ Combust. 2009;35:231-44.

[81] Gahleitner G. Hydrogen from renewable electricity: An

international review of power-to-gas pilot plants for stationary


applications. International Journal of Hydrogen Energy.

2013;38:2039-61.

[82] Dursun E, Acarkan B, Kilic O. Modeling of hydrogen

production with a stand-alone renewable hybrid power system.


[83] Strategieplattform Power to Gas - http://powertogas.info/.

Berlin, Germany: Deutsche Energie-Agentur GmbH; January 15,

2014.

[84] Best R, Rivera W. A review of thermal cooling systems. Appl

Therm Eng. 2015;75:1162-75.

[85] Eicker U. Solar Technologies for Buildings. West Sussex: J

Wiley; 2006.

[86] Rotary heat Exchanger http://3.imimg.com/data3/NO/WD/MY-

3973675/heat-exchanger-500x500.jpg: Retrieved January 28, 2015;

n.d.

[87] Bauman H, Tillman A-M. The Hitch Hiker’s Guide to LCA. 1:3

ed. Malmö, Sweden: Holmbergs; 2009.

[88] Goedkoop M, Schryver AD, Oele M, Durksz S, Roest Dd.

Introduction Into LCA with SimaPRO 7. Amersfoort, Netherlands:

PRé Consultants; 2010.

[89] Morawicki RO. Handbook of Sustainability for the Food

Sciences. West Sussex: Wiley; 2012.

[90] Goedkoop M, Spriensma R. Eco-Indicator 99 – A damage

oriented method for life cycle impact assessment: Methodology

Report. 3 ed. Amersfoort, Netherlands: PRé consultants; 2001.

[91] Burdick A. Strategy Guideline: Accurate Heating and Cooling

Load Calculations. Pittsburgh, Pennsylvania: IBACOS, Inc.; June

2001.

[92] Özkol N. Uygulamali Sogutma Teknigi ( Applied Cooling

Technology). Ankara, Turkey: TMMOB Makina Mühendisleri

Odası; 1999.

[93] Heat gain from people -

http://www.engineeringtoolbox.com/persons-heat-gain-d_242.html.

15 February 2009.

[94] Eroglu M, Dursun E, Sevencan S, Song J, Yazici S, Kilic O. A

mobile renewable house using PV/wind/fuel cell hybrid power

http://powertogas.info/

http://3.imimg.com/data3/NO/WD/MY-3973675/heat-exchanger-500x500.jpg:

http://3.imimg.com/data3/NO/WD/MY-3973675/heat-exchanger-500x500.jpg:

http://www.engineeringtoolbox.com/persons-heat-gain-d_242.html

References | 79

system. International Journal of Hydrogen Energy. 2011;36:7985-

92.

[95] Dursun E. Analysis of a stand-alone wind-solar and hydrogen

hybrid power system [Doctoral Thesis]. Istanbul, Turkey: Marmara

University 2013.

[96] Khatib H, Engineers IoE. Economic Evaluation of Projects in

the Electricity Supply Industry, Revised Edition: Institution of

Engineering and Technology; 2003.

[97] Brealey RA, Myers SC, Allen F. Principles of Corporate

Finance: McGraw-Hill/Irwin; 2011.

[98] Alsema EA, De Wild-Scholten MJ. Environmental impacts of

crystalline silicon photovoltaic module production. Mater Res Soc

Symp P. 2006;895:73-81.

[99] Jungbluth N, Stucki M, Frischknecht R. Photovoltaics. Data

v2.1. In: Dones R, editor. Sachbilanzen von Energiesystemen:

Grundlagen für den ökologischen Vergleich von Energiesystemen

und den Einbezug von Energiesystemen in Ökobilanzen für die

Schweiz Dübendorf, CH.: Swiss Centre for Life Cycle Inventories;

2009.

[100] Yearly Average Exchange Rates - Oz Forex Foreign Exchange

- http://www.ozforex.com.au/forex-tools/historical-rate-

tools/yearly-average-rates. September 15, 2011.

[101] Nyström O, Nilsson P, Ekström C, Wiberg A, Ridell B,

Vinberg D. El från nya och framtida anläggningar. Stockholm:

Elforsk AB; 2011.

[102] Wirth H. Aktuelle Fakten zur Photovoltaik in Deutschland

[Current figures on photovoltaics in Germany]. Freiburg, Germany:

Fraunhofer-Institut für Solare Energiesysteme; 2015.

http://www.ozforex.com.au/forex-tools/historical-rate-tools/yearly-average-rates

http://www.ozforex.com.au/forex-tools/historical-rate-tools/yearly-average-rates

Documents

Economic aspects of Fuel Cell Based Stationary Energy Systemskth.diva-portal.org/smash/get/diva2:881401/FULLTEXT01.pdfbroad market, longer stack lifetime, the possibility of selling