Sra Photo Voltaic En

8/14/2019 Sra Photo Voltaic En

1/76

A Strategic Research Agenda

or Photovoltaic Solar

Energy Technology


2/76

Introduction

The European Photovoltaic Technology Platorm is supported by the Sixth European Framework

Programme or Research and Technological Development.This publication was nanced under contract number: 513548, PV SEC

Cover image courtesy o Anna-Lena Thamm / Brochure design: ACG Brussels

More inormation on the European Union is available on the internet (http://europa.eu).

Luxembourg: Oce or Ocial Publications o the European Communities, 2007

ISBN 978-92-79-05523-2

European Communities, 2007

Reproduction is authorised provided the source is acknowledged.

Printed in BelgiumPrinted on white chlorine-ree paper

Europe Direct is a service to help you fnd answers

to your questions about the European Union

Freephone number (*):

00 800 6 7 8 9 10 11(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.


3/76

Introduction

A Strategic Research Agenda

or Photovoltaic Solar

Energy Technology

Research and development in support o realising

the Vision for Photovoltaic Technology

Prepared by Working Group 3

Science, Technology and Applications

o the EU PV Technology Platorm


4/76

Preace

This Strategic Research Agenda (SRA) was prepared by the Science, Technology and

Applications Group o the EU PV Technology Platorm, based on thorough consultations with

representatives o research, industry and other stakeholders. The members o the Working

Group are experts in PV technology, working as senior researchers in the public and private

sector. Feedback obtained in the public consultation on the rst drat o this report ollowing

the 2006 General Assembly o the Platorm has been discussed and taken into account as

ar as the Working Group considered this justied.

Although the group has attempted to cover all the most important parts o PV science, technol-

ogy and applications and to address all the most important research topics, the reader may

nd some aspects insuciently treated. Comments are, thereore, welcome through the PV

Technology Platorm secretariat (see www.eupvplatorm.org).

This SRA will be updated as required in order to refect developments in the photovoltaic

solar energy (PV) sector.

On behal o the Working Group,

Pro. Dr. Wim C. Sinke

Chairman, Science, Technology & Applications Working Group

June 2007

Preface

n Christophe Balli (CH)n Andreas Bett (DE)n Bernhard Dimmler (DE)n Doriana Dimova-Malinovska (BG)n Peter Fath (DE)n Nigel Mason (UK)n

Francesca Ferrazza (IT)n Hansjrg Gabler (DE)

n Maria Hall (SE)n Antonio Mart Vega (ES)n Enn Mellikov (EE)n Anton Milner (DE)n Paul Mogensen (FR)n Christoph Panhuber (AT)n

Nicola Pearsall (UK)n Je Poortmans (BE)

n Christos Protogeropoulos (GR)n Guy Sarre (FR)n Dominique Sarti (FR)n Wim Sinke (NL)n Philipp Strauss (DE)n Marko Topic (SI)n

Ronald van Zolingen (NL)n Tadeusz Zdanowicz (PL)

Members o the Science, Technology & Applications Working Group:


5/76

Contents

1 Executive summary 1

2 Introduction 7

3 Governing principles o the SRA 15

4 PV development options, perspectives and R&D needs

(short-, medium- & long-term) 17

4.1 Cell & module technologies 17

4.2 Waer-based crystalline silicon 18 4.3 Existing thin-lm technologies 25

4.3.1 Introduction 25

4.3.2 Common eatures o all existing thin-lm technologies 25

4.3.3 Thin-lm silicon (TFSi) 29

4.3.4 Copper-indium/gallium-diselenide/disulphide and

related I-III-VI compounds (CIGSS) 33

4.3.5 Cadmium telluride (CdTe) 37

4.4 Emerging and novel PV-technologies 41

4.4.1 Emerging PV technologies 41

4.4.2 Novel PV technologies 45

4.5 Concentrator technologies (CPV) 49

4.6 Balance-o-System (BoS) components and PV systems 57

4.7 Standards, quality assurance, saety and environmental aspects 63

4.8 Socio-economic aspects and enabling research 65

5 Research Funding 67

6 Appendices 70

7 Reerences 70

Co

ntents


6/76

1 Executive summary

The direct conversion o sunlight into electricity is a very elegant process to generate environmentally-

riendly, renewable energy. This branch o science is known as photovoltaics or PV. PV technology

is modular, operates silently and is thereore suited to a broad range o applications and can contribute

substantiallyto our uture energy needs.

Although reliable PV systems are commercially available and widely deployed, urther development o PV

technology is crucial to enable PV to become a major source o electricity. The current price o PV systems

is low enough or PV electricity to compete with the price o peak power in grid-connected applications

and with alternatives like diesel generators in stand-alone applications, but cannot yet rival consumer

or wholesale electricity prices. A drastic urther reduction o turn-key system prices is thereore needed

and ortunately possible. This was emphasised in the document A Vision or Photovoltaic Technology,published by the Photovoltaic Technology Research Advisory Council (PV TRAC) in 2005 [PVT 2005]

and reerred to requently in this report. Further development is also required to enable the European PV

industry to maintain and strengthen its position on the global market, which is highly competitive and

characterised by rapid innovation.

Research and Development - R&D - is crucial or the advancement o PV. Perorming joint research ad-

dressingwell-chosen issues can play an important role in achieving the critical mass and effectiveness

required to meet the sectors ambitions or technology implementation and industry competitiveness.

This led the PV Technology Platorm to produce a Strategic Research Agenda (SRA) to realise the Vision

reerred to above. The SRA may be used as input or dening the EUs Seventh Framework Programme

or Research (the main source o unding or collaborative research between European countries), but also

to acilitate a urther coordination o research programmes in and between Member States.

The table below summarises the key targets contained in the SRA.

The gures are rounded and indicative.

1980 Today 2015 2030 Long termpotential

Typical turn-key system price(2006 /Wp, excl. VAT)

>30 5 2.5 1 0.5

Typical electricitygeneration costs southern Europe(2006 /kWh)

>2 0.30 0.15(competitive

with retailelectricity)

0.06(competitive

with wholesaleelectricity)

0.03

Typical commercial fat-platemodule eciencies

up to 8% up to 15% up to 20% up to 25% up to 40%

Typical commercial concentratormodule eciencies

(~10%) up to 25% up to 30% up to 40% up to 60%

Typical system energy pay-backtime southern Europe (years)

>10 2 1 0.5 0.25

Flat plate reers to standard modules or use under natural sunlight, concentrator reers to systemsthat concentrate sunlight (and, by necessity, track the sun across the sky).

Executive

summar

y


7/76

Current turn-key system prices may vary rom ~4 to ~8 /Wp, depending on system type (roo-top

retro-t, building-integrated, ground-based,...), size, country, and other actors. The gure o 5 /Wp,

however, is considered representative. Similarly, prices in 2015 may range between ~2 and ~4 /Wp.

All prices are expressed as constant 2007 values.

The conversion rom turn-key system price to generation costs requires several assumptions. This report

assumes:

n an average perormance ratio o 75%, i.e. a system yield o 750 kWh/kWp/yr at an insolation

level o 1000 kWh/m2/yr. In southern Europe, where insolation is typically 1700 kWh/m2/yr,

a perormance ratio o 75% translates into 1275 kWh/kWp/yr

n 1% o the systems price will be spent each year on operation & maintenance

n the systems economic value depreciates to zero ater 25 years

n a 4% discount rate

The overall aim o short-term research is or the price o PV electricity to be comparable to the retail

price of electricity for small consumers in southern Europe by 2015. Continued price reduction after

2015 implies that this situation will apply to most o Europe by 2020. This state, where prices are

comparable, is known as grid parity. Larger systems and ground-based PV power plants that are not

connected directly to end-consumers will generally need to produce electricity at lower prices beore

they can be said to have reached grid parity.

To reach these targets, the SRA details the R&D issues related to:

n PV cells and modules:

- materials

- conversion principles and devices

- processing and assembly (incl. equipment)

n Balance o System (BoS):

- system components and installation

- materials installation

- operation and maintenance

n concentrator systems

n environmental quality

n applicability

n socio-economic aspects o PV

A range o technologies can be ound in commercial production and in the laboratory. No clear

technological winners or losers can yet be identied, as evinced by the investments being made

worldwide in production capacity based around many dierent technologies, and in the numerous

concepts developed in laboratories that have large commercial potential. Thereore it is important to

support the development o a broad portolio o options and technologies rather than a limited

set. The development o PV is best served by testing the dierent options and selecting on the basis

o the ollowing criteria:

n the extent to which the proposed research is expected to contribute to reaching the overall targets set

n the quality o the research proposal and the strength o the consortium or research group(s) involved

E

xecutive

summary


8/76

Concerning cells and modules, a distinction is made between existing technologies (waer-based

crystalline silicon, thin-lm silicon, thin-lm CIGSS and thin-lm CdTe) and emerging and novel technolo-

gies (advanced versions o existing technologies, organic-based PV, intermediate band semiconductors,

hot-carrier devices, spectrum converters, etc.).

It is noted that in addition to the cost o PV electricity generation the value o the electricity gener-

ated is important. The latter may be enhanced, or instance, by matching PV supply and electricity

demand patterns through storage.

The main R&D topics per technology area that are addressed to realise the Vision are summarised

below. The detailed descriptions can be ound in subsequent chapters.

1.1 Cells and modules

1.1.1 Topics common to all technologies

n Eciency, energy yield, stability and lietime

Since research is primarily aimed at reducing the cost o PV electricity it is important not to ocus

solely on initial capital investments (/Wp), but also on the energy yield (kWh/W

p)

over the economic or technical lietime.

n High productivity manuacturing, including in-process monitoring & control

Throughput and yield are important parameters in low-cost

manuacturing and essential to achieve the cost targets.

n Environmental sustainability

The energy and materials requirements in manuacturing as well as the possibilities

or recycling are important or the overall environmental quality o the product.

n Applicability

Achieving a degree o standardisation and harmonisation in the physical and electrical

characteristics o PV modules is important or bringing down the costs o installing PV.

Ease o installation as well as the aesthetic quality o modules (and systems) are

important i they are to be used on a large scale in the built environment.

1.1.2 Waer-based crystalline silicon technology

n Reduced specic consumption (g/Wp) o silicon and materials in the nal modulen New and improved silicon eedstock and waer (or waer equivalent) manuacturing technologies,

with careul consideration o cost and quality aspects

n Devices (cells and modules) with increased eciency

n New and improved materials or all parts o the value chain, including encapsulation

n High-throughput, high-yield, integrated industrial processing

n Sae, low-environmental-impact processing

n Novel and integrated (cells/modules) device concepts or the longer term

Executive

summar

y


9/76

1.1.3 Existing thin-lm technologies

1.1.3.1 Common aspects

n Reliable, cost-eective production equipment or all technologies

n Low cost packaging solutions both or rigid and fexible modules

n Low cost transparent conductive oxides

n Reliability o products: advanced module testing, and improved module perormance assessment

n Handling o scrap modules, including re-use o materials

n Developing replacements or scarce substances such as indium

1.1.3.2 Thin-lm silicon (TFSi)

n Processes and equipment or low-cost large area plasma deposition o

micro/nanocrystalline silicon solar cells. The interplay between theeects o plasma, devices and upscaling should be ully mastered

n Specic high-quality low cost transparent conductive oxides suitable or large,

high perormance modules (greater than 12% eciency)

n Demonstration o higher eciency TFSi devices (meaning greater than 15% at laboratory-scale),

improved understanding o interace and material properties, o light trapping,

and o the theoretical perormance limits o TFSi based materials and devices

1.1.3.3 Copper-indium/gallium-diselenide/disulphide (CIGSS)

n Improvement o throughput and yield in the whole production

chain and standardisation o equipment

n Modules with eciencies greater than 15%, developed through a deeper

understanding o device physics and the successul demonstration o

devices with eciencies greater than 20% at laboratory-scale

n Alternative or modied material combinations, o process alternatives like roll-to-roll

coating and o combined or non-vacuum deposition methods

n Highly reliable and low cost packaging to reduce material costs

1.1.3.4 Cadmium telluride (CdTe)

n Alternative activation/annealing and back contacts or simpler,quicker and greater yield and throughput

n New device concepts or thinner CdTe layers

n Enhanced undamental knowledge o materials and interaces or advanced

devices with high eciencies (up to 20% at laboratory scale)

1.1.4 Emerging and novel technologies

1.1.4.1 Emerging technologies

n Improvement o cell and module eciencies and stability

to the level needed or rst commercial application

n Encapsulation materials and processes specic to this amily o cell technologies

n Product concepts and rst generation manuacturing technologies

E

xecutive

summary


10/76

1.1.4.2 Novel technologies

n Demonstration o new conversion principles and basic operation o new device concepts

n Processing, characterisation and modelling o (especially) nanostructured materials

and devices; understanding o the morphological and opto-electrical properties

(including development o theoretical and experimental tools)

n Experimental demonstration o the (potential) eectiveness o

add-on eciency boosters (spectrum converters)

1.1.5 Concentrator technologies

1.1.5.1 Materials and components:

1. Optical systems - Find reliable, long-term, stable and low-cost solutions or fat and concave

mirrors, lenses and Fresnel lenses and their combination with secondary concentrators2. Module assembly - Materials and mounting techniques or the assembly o

concentrator cells and optical elements into highly precise modules that are

stable over the long-term using low-cost, ully automated methods

3. Tracking - Find constructions which are optimised with respect to size,

load-capacity, stability, stiness and material consumption

1.1.5.2 Devices and eciency

Develop materials and production technologies or concentrator solar cells with very high eciencies,

i.e. Si cells with eciencies greater than 26 % and multi-junction III-V-compound cells with eciencies

greater than 35 % in industrial production and 45 % in the laboratory. Find the optimum concentration

actor or each technology.

1.1.5.3 Manuacturing and installation

Find optimised design, production and test methods or the integration o all system components; methods

or installation, outdoor testing and cost evaluation o concentrator PV systems.

1.2 Balance-o-System (BoS) components and PV systems

It is important to understand better the eect o BoS components on turn-key system costs and prices. BoS

costs vary according to system type and - at least at present - the country where the system is installed,

which impact on whether the cost targets or BoS may be deemed to have been reached or not. This

report recommends that a study o BoS aspects be undertaken to quantiy in detail the cost reduction

potential o PV technology beyond 2030 (see Table I). BoS and system-level research should aim to:

n Increase inverter lietime and reliability

n Harmonise the dimensions and lietimes o components

n Increase modularity in order to decrease site-specic costs

at installation and replacement costs over the systems lietime

n Assess and optimise the added value o PV systems or dierent system congurations

n Produce workable concepts or maintaining the stability o electrical grids at high PV penetrations

Executive

summar

y


11/76

n Devise system components that enhance multiunctionality and/or minimise losses

n Develop components and system concepts or island PV and PV-hybrid systems

BoS research should include research into new storage technologies or small and large applications

and the management and control systems required or their ecient and reliable operation.

1.3 Standards, quality assurance, saetyand environmental aspects

n Identication o perormance, energy rating and saety standards or PV modules,

PV building elements and PV inverters and AC modules

n Common rules or grid-connection across Europe

n Quality assurance guidelines or the entire value chain

n A cost-eective and workable inrastructure or the reuse and recyclingo PV components, especially thin-lm modules and BoS components

n Analysis o lietime costs especially o thin-lm and concentrator PV and BoS components

over the short term and or emerging cell/module technology over the longer term

1.4 Socio-economic aspects and enabling research

n Identiying and quantiying the non-technical (i.e. societal,

economic and environmental) costs and benets o PV

n Addressing regulatory requirements and barriers to the use o PV on a large scale

n Establishing the skills base that will be required by PV and associated industries

in the period to 2030 and developing a plan or its provision

n Developing schemes or improved awareness in the

general public and targeted commercial sectors

The countries associated to FP7 should use this SRA as a reerence when developing or ne-tuning

their national R&D programme(s). By interpreting the research priorities described here in their own

national contexts (national R&D strengths, presence o industry), they can align their publicly-unded R&D

with the SRAs recommendations, to the benet o PV in Europe.

PV solar energy is a technology that can be used in many dierent products, ranging rom very smallstand-alone systems or rural use, to building-integrated grid-connected systems, and large power plants.

PV will make a very large contribution to the global energy system in the long term and will be a key

component o our uture, green energy supply system. The rapid growth o the PV sector oers eco-

nomic opportunities or Europe that must be seized now unless they are to be ceded to other regions

o the world. The coming years will be decisive or the uture role o the European PV industry.

E

xecutive

summary


12/76

Introduction

2 Introduction

2.1 What is photovoltaic solar energy (PV)?

The direct conversion o sunlight into electricity is a very elegant process to generate environmentally-

riendly, renewable energy. This branch o science is known as photovoltaics or PV. PV technology

is modular, operates silently and is thereore suited to a broad range o applications and can contribute

substantiallyto our uture energy needs. Although the basic principles o PV were discovered in the

19th century, it was not beore the 1950s and 1960s that solar cells ound practical use as electric-

ity generators, a development that came about through early silicon semiconductor technology or

electronic applications. Today, a range o PV technologies are available on the market and under

development in laboratories.

Complete PV systems consist o two elements: modules (also reerred to as panels), which contain

solar cells, and the Balance-o-System (BoS). The BoS mainly comprises electronic components,

cabling, support structures and, i applicable, electricity storage or optics & sun trackers. BoS costs also

include the labour costs o installation.

2.2 Why a Strategic Research Agenda (SRA)?

Although reliable PV systems are commercially available and widely deployed, urther development

o PV technology is crucial to enabling PV to become a major source o electricity. The current price o

PV systems is low enough or PV electricity to compete with the price o peak power in grid-connected

applications and with alternatives like diesel generators in stand-alone applications, but cannot yet rival

consumer or wholesale electricity prices. A major urther reduction o turn-key system prices is thereore

needed and ortunately possible. This was emphasised in the document A Vision for Photovoltaic

Technology, published by the Photovoltaic Technology Research Advisory Council (PV TRAC) in 2005

and reerred to requently in this report [PVT 2005]. Further development is also required to enable

the European PV industry to maintain and strengthen its position on the global market, which is highly

competitive and characterised by rapid innovation.

Research and Development - R&D - is crucial or the advancement o PV. Perorming jointresearchaddressing well-chosen issues can play an important role in achieving the critical mass and e-

fectiveness required to meet the sectors ambitions for technology implementation and industry

competitiveness. Thereore the PV Technology Platorm has decided to produce a Strategic Research

Agenda (SRA) to realise the Vision document reerred to above. The SRA may be used as input or the

denition o the 7th Framework Programme o the EU (the main source o unding or EU joint research),

but also to acilitate a urther coordination o research programmes in and between Member States.


13/76

Introduction

2.3 PV historic development, state-o-the-art and potential

PV modules and other system components have undergone an impressive transormation, becoming

cheaper, greener and better perorming.

This is evinced by the module and BoS price reductions shown in so-called learning curves (see A

Vision or Photovoltaic Technology, mentioned above), by the increase o power conversion ecien-

cies and energy yields, by enhanced system availabilities, by drastically shortened energy pay-back

times (the period needed or a system to amortise the energy required or its manuacture), and by a

variety o other indicators. Nevertheless, PV technology has by no means demonstrated its ull potential.

Table I gives an indication o where PV was 25 years ago, where it stands today and what it could

realistically achieve over the next 25-50 years. The gures in the column long term potential are more

uncertain than the others.

Table 1. Expected development o PV technology over the coming decades - gures are rounded

and indicative, and should be interpreted with reerence to the provisos listed below.

1980 Today 2015/2020 2030 Long termpotential

Typical turn-key system price(2007 /Wp, excl. VAT)

>30 5 2.5/2.0 1 0.5

Typical electricity generationcosts southern Europe(2007 /kWh)

>2 0.30 0.15/0.12(competitive

with retailelectricity)

0.06(competitive

with wholesaleelectricity)

0.03

Typical commercial fat-plate

module eciencies (see below)

up to 8% up to 15% up to 20% up to 25% up to 40%

Typical commercial concentratormodule eciencies (see below)

(~10%) up to 25% up to 30% up to 40% up to 60%

Typical system energy pay-backtime southern Europe (years)

>10 2 1 0.5 0.25

Flat plate reers to standard modules or use under natural sunlight, concentrator reers to systemsthat concentrate sunlight (and, by necessity, track the sun across the sky).

Current turn-key system prices may vary rom ~4 to ~8 /Wp, depending on system type (roo-top

add-on, building-integrated, ground-based, ...), size, country, and other actors. The gure o 5 /Wp,

however, is considered representative. Similarly, prices in 2015 may range between ~2 and ~4 /Wp.All prices are expressed as constant 2007 values.

The conversion rom turn-key system price to generation costs requires several assumptions. This report

assumes:

n an average perormance ratio o 75%, i.e. a system yield o 750 kWh/kWp/yr at an insolation

level o 1000 kWh/m2/yr. In southern Europe, where insolation is typically 1700 kWh/m2/yr,

a perormance ratio o 75% translates into 1275 kWh/kWp/yr

n 1% o the systems price will be spent each year on operation & maintenance

n that the systems economic value depreciates to zero ater 25 years

n a 4% discount rate


14/76

Introduction

2.4 The value o PV or Europe and the world

2.4.1 Energy and climate

The solar energy resource is larger than all other renewable energy resources[UND 2000][WBG

2003]. In one ambitious scenario, PV covers 20% o global electricity consumption by 2040. Already

in 2020 PV may contribute to the reduction o CO2-emissions by the equivalent o 75 average-sized

coal-red power plants or 45 million cars [EPI 2004]. Since PV is deployable within Europe, it

can play an important role in improving the security o Europes energy supply. Moreover, PV is

very well suited to providing access to energy in rural areas, thus enabling improved healthcare

and education and providing economic opportunities. It may bring electricity to hundreds o millions

people in developing countries by 2020 to 2030.

2.4.2 PV and the Lisbon Agenda

In 2000 the European Council adopted the Lisbon strategy, the aim o which is, by 2010, to

make the EU the most dynamic and competitive knowledge-based economy in the world, capable

o sustainable economic growth with more and better jobs and greater social cohesion, and respect

or the environment.

The European photovoltaic industry and research communitys knowledge, diversity, creativity and enthu-

siasm are key actors in the creation o a competitive advantage. At the European Summit in Barcelona

in 2002, European Heads o State and Government set themselves the goal o increasing Europes

overall level o investment in research to 3% o GDP by 2010, two thirds o which were to come rom

the private sector. The highly innovative and competitive PV sector in Europe contributes towards these

goals. It is clear that it can only maintain and strengthen its position on the world market i continuous,

substantial R&D investments are made.

2.4.3 Economy and jobs

The global PV sector has grown by an average o 25% per year over the past two decades and by

almost 50% per year over the past ve years. This has happened because several countries have put in

place successul market development policies. In Europe, these countries are Germany, Spain, Portugal,

France, Italy, Greece and Belgium - perhaps soon to be joined by many more. Outside Europe, Japan,

Korea and the USA provide good examples o well-developed or emerging markets.

A report rom 2004 orecast that the photovoltaics sector stands a realistic chance o expanding rom

5.8 billion in 2004 to 25 billion in 2010 with 5.3 GWp in annual sales[SPS 2004]. Within

the last 8 years, employment in the photovoltaics sector in Germany rose rom approximately 1,500

to 30,000. By 2005, 6,300 jobs had been created in Spain, bringing the European total to ap-

proximately 40,000 (data rom German industry associations and IEA [BSW 2006], [IEA 2005]).

The sector needs a diverse and qualied workorce ranging rom technicians close to the customers

who install and maintain the PV systems to expert semiconductor specialists working in high-tech solar

cell actories. The photovoltaic industry has the potential to create more than 200,000 jobs in the

European Union by 2020 and ten times this number worldwide. Although the labour intensity will

decrease with decreasing system prices, the rapid market growth will guarantee a strong increase

in the number o jobs in Europe.


15/760

Introduction

The European Commission has acknowledged[COM 2005]:

The renewable energy sector is particularly promising in terms o job and local wealth creation. The

sector invests heavily in research and technological innovation and generates employment, which to

a very high degree means skilled, high quality jobs. Moreover, the renewable energy sector has a

decentralised structure, which leads to employment in the less industrialised areas as well. Unlike other

jobs, these jobs cannot be globalised to the same extent. Even i a country were to import 100% o

its renewable energy technology, a signicant number o jobs would be created locally or the sale,

installation and maintenance of the systems. A number of studies on the job creation effects have already

been published and dierent estimates have been provided[EPI 2004],[ERE 2004], [UVE 2005]

The German Solar Industry Association has reported that despite the act that more than 50% o the

solar cells installed in PV systems in Germany are imported, 70% o the added value stays within the

German economy [BSW 2006].

Electricity generated with photovoltaic systems has additional benets or the European economy in

the long term. First, it can help to reduce the European Unions dependence on energy imports. The

European Commissions report urther acknowledges:

Rising oil prices and the concomitant general increase in energy prices reveals the vulnerability and

dependency on energy imports o most economies. The European Commissions DG ECFIN predicts

that a $10/bbl oil price increase rom $50 to $60/bbl would cost the EU about 0.3% growth and

the US 0.35%. For the European Union, the negative GDP eect would be roughly40 billion rom

2005 to 2007. Further price increases would worsen the situation. The European renewable energy

association (EREC) estimates that140 billion in investment would be required to reach the 2010 goal

o 12% renewable energy consumption [ERE 2004]. This would ensure uel cost savings o20 billion

(not even taking into account the substantial price increases since 2003) and reduce external costs by

30 to77 billion. I we add the employment benets, the overall costs or society can be estimated

to be positive compared to a negative result i no RES were introduced. There are several studies that

examine the dicult issue o quantiying the eect o the inclusion o RES in an energy portolio and

the reduction in the portolio energy price. This is in addition to the economic benets o avoided uel

costs and external costs (GHG), money which could be spent within the economy and used or local

wealth creation [AWE 2003].

Secondly, electricity rom photovoltaic systems is generally produced during peaks in daily demand,

when the marginal costs o electricity production are highest. In southern European climates, the seasonin which electricity demand peaks - summer - is the season when the output o photovoltaic capacity is at

its greatest. PV can be relied upon in rare cases o extreme heat and water shortage when thermoelectric

power plants have to reduce their output due to a lack o cooling water.

During the heat wave that aected Europe in July 2006, peak prices paid at the European electricity

exchange (EEX) spot market exceeded the PV eed-in tari paid in Germany. This will happen more

requently in uture as the eed-in tari decreases, meaning that the occasions when PV electricity is a

cost-eective orm o generation even without support mechanisms will become more common.


16/76

Introduction

2.4.3.1 International competition

Europes photovoltaic industry competes with companies rom Asia, the USA and other parts o the

world. Two o these countries have instituted programmes to support their domestic PV industry - Japan

and China. The eectiveness o the programme sponsored by Japans Ministry o Economy, Trade

and Industry, METI, is already apparent. Due to long term planning, support schemes, investment

security, and a substantial domestic market, the Japanese PV industry has around 50% o the world

market share in PV products.

China is the second country with an industrial strategy geared towards building up a highly competitive

PV industry. China wants to cover the whole value chain rom silicon eedstock to complete systems.

The ruits o this relatively new strategy are already visible. Chinese cell and module manuacturers are

rapidly establishing a signicant share o the world market and their production capacity increases

are unrivalled.

I Europe does not react to this challenge, there is the danger that PV production will move to China, in

common with many other manuacturing technologies. So ar, Europe still has a competitive edge due

to the excellent knowledge base o its researchers and engineers. However, without steady and reli-

able R&D unding and support rom the public purse, this advantage could be eroded in a short time.

More support or innovation and clearer long-term strategies are needed or the European PV industry

to continue to invest in Europe and to ensure that European companies increase their market shares

and become world leaders.

2.4.3.2 PV or development and poverty reduction

The Plan o Implementation o the United Nations World Summit or Sustainable Development, which

took place in Johannesburg in August 2002 contains the commitment rom the UNs Member States to

improve access to reliable, aordable, economically viable, socially acceptable and environmentally

sound energy services and resources, taking into account national specicities and circumstances, through

various means, such as enhanced rural electrication and decentralised energy systems, increased use

o renewables, cleaner liquid and gaseous uels and enhanced energy eciency[WSS 2002]. At the

same time, the European Union announced a $700 million partnership initiative on (renewable) energy

and the United States announced that it would invest up to $43 million in 2003.

The International Conerence or Renewable Energies ollowed up the UNs conerence in June 2004.

It yielded the International Action Programme, which includes some 200 specic actions and voluntarycommitments or developing renewable energy, pledged by a large number o governments, interna-

tional organisations and stakeholders rom civil society, the private sector and other stakeholder groups.

Photovoltaic o-grid systems are the preerred option or rural electrication in developing countries,

where they are crucial in providing energy or light, drinking water, rerigeration and communication.

More than 1 billion people in the world do not have access to electricity.

PV is a cost-eective way o meeting the rapidly expanding electricity demand o developing countries,

while minimising the environmental impact o this demand. Delivering aordable modern energy services

or health, education and social and economic development is central to international aid objectives.


17/76

Introduction

As a clean, ree-uel (except or sunlight) energy source, PV has the potential to create economic and

political stability, with clear implications or improved international security. It should also be emphasised

that robust demand rom less developed countries could make an important contribution to reducing

the costs o PV technology through economies o scale rom increased cell and module production.

The goal or European industry is to capture a 40% market share o the annual market or rural use by

2010 and to keep this level thereater.

2.5 Targets and drivers or PV development

2.5.1 What are the PV implementation targets?

In 1997, the European Commission envisaged 3 GWpPV being installed across Europe by 2010[COM

1997]. It is now clear that as a result o successul market incentives in Germany and other countries the

capacity by that date will probably be more than 5 GWp. For the longer term, [PVT 2005] oers anambitious, though realistic target or 200 GW

pinstalled in the EU by 2030 (o an estimated 1000

GWpworldwide). In relation to the new EU renewable energy targets and the increased overall sense

o urgency it has been stated recently, though, that the rapid development o the PV industry may give a

considerable upward potential or the 200 GWpgure i adequate policy measures are implemented

and R&D support is strengthened.

Japan aims at 5 GWp

by 2010, and has developed roadmaps or 50-200 GWp

o PV capacity by

2030 [NED 2004]. The Korean government has set a target o 1.3 GWpby 2012.

Few other countries have specic targets or PV implementation. Usually PV is an unspecied part o

the (renewable) energy portolio.

2.5.2 Conditions or PV to meet the targets

Very large-scale deployment o PV is only easible i PV electricity generation costs are drastically reduced.

However, because o the modular nature o PV, the possibility to generate at the point-o-use, and the

specic generation prole (overlap with peak electricity demand), PV can make use o lead markets

on its way to eventually becoming as cheap as wholesale electricity. In particular PV may compete with

peak power prices and consumer prices in the short and medium term. The corresponding PV system

price targets are thereore very important or the rapid deployment o PV. Ambitious targets are also

crucial or the global competitive position o the European PV industry sector.

The evolution o turn-key PV system prices outlined in A Vision or Photovoltaic Technology (Figure 1)

provides an excellent starting point to dene underlying cost targets to be addressed in this SRA. It is

noted that research and technology transer to industry directly infuences manuacturing and installation

costs (as well as some other parameters), but not directly turn-key prices. The latter are also determined

by market orces. Cost reduction targets are nevertheless essential to enable price reduction.


18/76

Introduction

Figure 1. Possible evolution o turn-key PV system prices (adapted rom A Vision or Photovoltaic Technology)

The SRA reers to time-scales using the ollowing denitions:

n 2008 ~ 2013: short term

n 2013 ~ 2020: medium term

n 2020 ~ 2030 and beyond: long term

The year 2013 has been chosen because it coincides with the end o the European Commissions cur-

rent programme or unding research, FP7. The start o short term does not coincide with the start o

FP7 (2007) because no signicant results are expected rom FP7 in its rst year. The convention used

in this report is to reer research priorities to the time horizons in which they are rst expected to be

used in commercial product, not to the year by which widespread use is expected.

A technology is said to meet a cost target i pilot-scale production at that cost has been achieved. This

implies the technology will one or two years later be ready or commercial production at that cost.

The overall target o the short-term research described in this SRA is or PV electricity to be competitive

with consumer electricity (grid parity) in Southern Europe by 2015. Specically, this means reaching

PV generation costs o 0.15/kWh, or a turn-key system price o 2.5/Wp

(Table I).

This system price arises rom typical manuacturing and installation costs o


19/76

Introduction

Secondly, the lietime o system components must be considered. High technical lietimes not only

help to reach cost targets, it also increase the overall energy produced and eases the integration o

PV in buildings.

Thirdly, it is essential that energy and materials consumption in manuacturing and installation be ad-

dressed. Further shortening o the energy pay-back time o systems will add to the advantages o PV as

an energy source and, in the longer term, its ability to avoid carbon dioxide emissions. Avoiding the use

o scarce or hazardous materials, or i that is not possible, closing material-use cycles, is an important

topic with great R&D challenges.

Finally, the ability to combine PV components and systems and integrate them with building components

may be signicantly improved. This requires standardisation and harmonisation, but also fexibility in

system design, and should be accomplished without additional engineering (costs).

In addition to the technicalissues described above, the document addresses socio-economic aspectsrelated to the large-scale implementation o PV.

In summary, this SRA identies and addresses the ollowing drivers or PV development:

Electricity generation costs and value

n turn-key investment costs (price):- modules- BoS- system engineering

n operation & maintenance costs(& planned replacement i applicable) /technical lietime

n value:- e.g. possibilities or supply-on-demandor at peak prices

n energy yield

n (actors out o scope o this SRA:interest rate, economic lietime,)

Environmental quality

n energy pay-back time:- modules- BoS

n substitution o hazardous materials

n options or recycling

Integration

n method and ease o mounting,cabling, etc. (also or maintenanceand repair)

n fexibility / modularity

n aesthetics and appearance

n lietime

Socio-economic aspects

n public and political awareness

n user acceptance

n training and education

n nancing


20/76

3 Governing principles o the SRA

Short-term research should be ully dedicated to the competitiveness o the EU industry.

The coming decade is expected to be decisive or the uture prospects o the EU PV industry. The global

PV sector will grow to maturity and achieve multi-billion dollar turn-overs. Competition will be erce.

Rapid innovation and high production volumes are crucial to establish leadership.

No exclusivity

PV comes and will come in dierent ormats. The SRA does not exclude technologies but sets overall

targets that each PV ormat must reach and describes the research priorities or each ormat in

order or it to succeed in meeting the targets.

There is a need or public money to und short-, medium- and long-term research into all parts othe value chain(s), as well as research into socio-economic issues. Since drastic cost reductions are

needed or all elements o the PV system, research should address all parts o the value chain, rom

raw materials up to the complete system, and even beyond. In addition, public unding agencies

should make a strategic top-down decision on how to allocate unding between short-,medium-

and long-term research. Industry will push or short-term research to be the main beneciary o unding

and as the PV industry grows, this pressure may become stronger. Governments must, however, look

ahead to the medium and long term and set aside xed budgets or research applicable to these

time-rames. This report recommends that the combined research spending o the public and private

sector should be distributed between topics with commercial relevance in the short, medium and long

term in the typical ratio 6:3:1 in the near term, moving to the ratio 10:5:1 as private sector unding

increases. See Chapter 5 or details.

Based on a detailed analysis o cost reduction potentials, the working group decided that the same

cost targets shall be used or all fat-plate PV module technologies considered: 0.8-1.0 /Wpor

technology ready by 2013 and implemented in large-scale production in 2015, 0.60-0.75 /Wp

in 2020, and 0.3-0.4 /Wpin 2030. The targets are expressed as a range in order to refect the e-

ciencies o dierent types o module. To meet the overall, cross-technology cost targets, lower eciency

modules need to be cheaper than higher eciency modules, due to the area-related component o the

BoS costs. These targets should not be interpreted as predictions. It is possible that some technologies

will even exceed them. The eciency targets quoted later in the SRA or each technology and are to be

considered as perormance targets that should be met in order to meet the cost target. System costs andprices, it should be noted, are dependent on the specic application that the system is put to. Thereore

the costs and prices mentioned in the SRA are only approximate.

Governing

princip

lesofthe

SRA


21/76

The Balance-o-System (BoS) costs are strongly dependent on, among other actors, the type o system (e.g.

roo-top, building-integrated, ground-based), the eciency o the modules used, and the country where it

is sited. This makes it dicult to ormulate general targets. Indicative targets or the BoS costs or roo-top

systems are: 0.9-1.1 /Wpin 2013, 0.75-0.9 /W

pin 2020, and under 0.5 /W

pin 2030. The

ranges in system cost targets mainly correspond to the range in module eciencies mentioned in the previous

paragraph, refecting the act that part o the BoS cost is system area-related and thus aected by module

eciency (a given system power requires a smaller area when using higher eciency modules).

Possible dierences in BoS cost structure and gures between EU Member States are not taken into

account here.

For concentrator systems target costs it is not meaningul to distinguish between modules and

BoS. Thereore indicative targets or the turn-key cost (not price) o ull systems have been identied:

1.2-1.9 /Wp

in 2013, 0.8-1.2 /Wp

in 2020 and 0.5-0.8 /Wp

in 2030. Considerable

uncertainty exists or these numbers because, with little industrial production at present, extrapolationrom what does exist carries high uncertainty. Furthermore, no clear denition so ar exists o the watt

peak (Wp) power rating o concentrator technology, because, unlike other technologies, they unction

only under direct sunlight. Some attempts to dene a Wp rating have, however, been made and are

the basis or the cost targets above.

Governing

principlesofth

e

SRA


22/76

4 PV development options,perspectives and R&D needs(short-, medium- & long-term)

4.1 Cell & module technologies

PV modules are the basic building blocks o fat-plate PV systems. Modules consist o solar cells, abri-

cated on waers or rom thin active layers on an inert, low cost substrate. For the end-user, the nature

o the cell technology used is seldom their main concern. The parameters they nd more important are

the price per watt-peak o module, the energy yield per watt-peak under eld conditions, the modules

eciency, its size and weight, fexibility or rigidity, and appearance. Customers will also be interestedin the retailers provisions or taking back and recycling the modules at the end o their lives. On the

other hand, or the R&D community, sound understanding o dierent cell and module technologies is

crucial in dening the work to achieve cost reduction, perormance enhancement and an improved

environmental prole. Dierent technologies require their own research and development activities. The

categories o technology chosen or the SRA are:

1. Waer-based crystalline silicon;

2. Existing thin-lm technologies;

3. Emerging and novel technologies (including boosters to technologies in the rst and second category)

The R&D issues surrounding concentrator systems need to be addressed in a dierent, more integrated

manner, so they are covered in a separate subchapter, 4.2. Research conducted under the Emerging

and novel technologies category is also relevant or concentrator systems.

In the ollowing paragraphs the R&D needs o the our technology categories are analysed in detail. The

technology categories have a number o R&D issues in common, which are briefy summarised here:

4.1.1 Eciency, energy yield, stability and lietime

Research aims to optimise combinations o these parameters rather that one parameter at the expense o

another. This implies careul analysis o the costs and benets o individual technological improvements.Since research is primarily aimed at reducing the cost o electricity generation, it is important not to

ocus only on initial costs (/Wp), but also on the systems energy yield (kWh/Wp) over its economic

or technical lietime.

4.1.2 High productivity manuacturing, including in-process monitoring & control

Throughput and yield are important parameters in low-cost manuacturing and are essential to achieving the

cost targets. In-process monitoring and control are crucial tools or increasing product quality and yield.

4.1.3 Environmental sustainability

The energy and material requirement o manuacturing as well as recyclability are important parameters

in the overall environmental quality o the product. Shortening still urther the energy pay-back time o

PV

development


23/76

modules (table 1), designing products in a way that makes them readily recyclable and, where practical,

avoiding o the use o hazardous materials are the most important issues to be addressed here.

4.1.4 Integration

As discussed in more detail in the chapter on BoS, standardisation and harmonisation o specications

will help bring down the costs o PV. Eorts to standardise module specications should also be made

and may make installation easier. Finally, the appearance o modules (and systems) will grow in impor-

tance as they become a more common sight in the built environment.

4.2 Waer-based crystalline silicon

4.2.1 Introduction

Waer-based crystalline silicon has dominated the photovoltaic industry since the dawn o the solar PV

era. It is widely available, has a convincing track-record in reliability and its physical characteristsics

are well understood, in part thanks to its use in the hal-century-old microelectronics industry. A learning

curve or the progress in silicon waer-based technology can be drawn that spans three decades. It

shows that the price o the technology has decreased by 20% or each doubling o cumulative installed

capacity. Two driving orces are behind this process: market size and technology improvement. Such

progress was not made by chance but is the combined result o market-stimulation measures and research,

development and demonstration activities with both private and public support.

The total PV market has increased by an order o magnitude in the last decade, growing by almost 50%

per year in the last ve years, with crystalline silicon accounting or more than 90% o the total volume.

Crystalline silicon modules are manuactured in six-steps: (i) silicon -production, (ii) -purication, (iii) -crystal

growth, (iv) waer slicing, (v) cell abrication and (vi) module assembly. Although considerable progress

has already been made in each step, they may all be signicantly urther improved.

For example, waers have decreased in thickness rom 400 m in 1990 to 200 m in 2006 and have

increased in area rom 100 cm2 to 240 cm2; modules have increased in eciency rom about 10%

in 1990 to typically 13 % today, with the best perormers above 17%; and manuacturing acilities

have increased rom the annual outputs o typically 1-5 MWpin 1990 to hundreds o MW

por todays

largest actories. Plans or GWp-scale actories have been announced.

Three main routes to cost saving have been ollowed in recent years and need to be ollowed urther and

aster: reduction in material consumption, increase in device eciency and advanced, high-throughput

manuacturing. Other important measures that should receive attention include reducing embedded energy

content (and hence the energy pay-back-time), the environmental riendliness o PV systems over their lie

cycle, the denition o accepted standards or crystalline silicon products and advanced manuacturing

practices such as process automation and advanced process control.

Crystalline silicon is a technology with the ability to continue to reduce its cost at its historic rate. Direct

production costs or crystalline silicon modules are expected to be around 1 /Wp in 2013, under

0.75 /Wpin 2020 and lower in the long term. This will happen i R&D eort is directed towards the

issues o greatest strategic concern.

PV

developmentoptions,

perspectivesand

R&Dneeds


24/76

4.2.2 Materials and components

Puried silicon (polysilicon) is the basic ingredient o crystalline silicon modules. It is melted and solidied

using a variety o techniques to produce ingots or ribbons with dierent degrees o crystal perection.

The ingots are shaped into bricks and sliced into thin waers by wire-sawing, or by laser i the aim is to

use the waers to make ribbons. Waers and ribbons are processed into solar cells and interconnected

in weather-proo packages designed to last or at least 25 years. The processes in the manuacturing

chain have improved signicantly during recent years but can improve yet urther. For the past ew years

the availability o polysilicon eedstock has been a critical issue or the rapidly growing PV industry. The

tight supply has caused very high polysilicon spot market prices and has limited production expansion

or part o the industry. On the other hand, it has triggered rapid innovation in waer production and

cell manuacturing, as evinced by the lower silicon consumption per Wp

o module power produced.

Silicon usage is currently 10 g/Wp, whereas it was typically 13 g/W

pjust a ew years ago.

The development o new, lower cost and less energy-intensive techniques or silicon eedstock preparationis underway. This eedstock is expected to come at prices in the range 10-20 /kg (now 30-50 /kg)

and will thus be a key enabler or uture PV growth and cost reduction. As well as having to contend with

high eedstock prices, cell manuacturers are aced with the problem that they lose 50% or more o the

polysilicon starting material during the manuacturing process, even ater recycling. To improve casting

and waering, it is necessary to reduce waste during polysilicon crystallisation, recycle saw dust and other

silicon o-cuts, and improve material handling in the production process through automation. It should be

possible to reach polysilicon consumption below 2 g/Wpin the long term. The targets or the short and

medium term are achievable even at moderate to high eedstock prices. For waer equivalent technolo-

gies, the challenge is to develop high-throughput low-cost silicon lm deposition techniques with suitable

low-cost substrates.

Module assembly is also material-intensive. The assembly must protect the cells rom the outdoor environ-

ment or a minimum, typically, o 25 years while allowing the cell to unction as eciently as possible.

The current standard design, using rigid glass-polymer encapsulation in an aluminium rame, ulls these

basic requirements, but represents about 30% o the overall module cost, contains a lot o embedded

energy (increasing the energy pay-back time o the module) and is a challenge to manuacture on

automated lines even at current waer thicknesses.

New cheaper, more fexible, highly durable encapsulation materials with improved optical properties

are expected to be developed. They may also be better suited or high-throughput manuacturing than

the materials currently used.

New materials and techniques or connections between cells need to be developed to improve the

automated assembly o very thin waers. Metal contact cell geometries may depart signicantly rom

the traditionally H shaped ront-rear structure. The use o back-contacted cells may avour automation

and simpliy processes by reducing the complexity o cell interconnection. Simpler schemes or electrical

interconnection, due in part to improved cell design and to newly developed metallisation techniques,

may eliminate discrete soldering steps because the interconnection scheme could be embedded in

encapsulation sheets.

Future, very large-scale manuacturing may require alternatives to be ound or scarce chemical elements

currently used in module manuacture, such as silver, which is consumed at an average o 80-90 mg/Wp,

or some 130 tonnes/year.

PV

development


25/760

Materials 2008 - 2013 2013 - 2020 2020 - 2030

and beyondIndustrymanuacturingaspects

n Polysilicon targetsConsumption 5 g/WpCost 15-25 /kg(dependent on quality)Waer thickness


26/76

4.2.3 Perormance and devices

Cell and module eciency directly impact on the overall /Wpcost (and price) o a PV module and

have historically been a ocus or technological development. Increasing the eciency o the solar

cells and the power density o the modules, together with the reduction o the specic consumption o

silicon, are the main paths to cost reduction. An increase o 1% in eciency alone is able to reduce

the costs per Wp

by 5-7%.

Small cells with eciency values up to 24.7% have been produced in expensive clean room acilities

with vacuum technologies used or the deposition o metal contacts.

Only three o these high eciency cell processes have so ar been demonstrated at production scale,

in non-clean-room manuacturing environments. All three use monocrystalline silicon, whilst the majority

o commercial cells use a low-cost screen-printing process on multicrystalline silicon waers.

Commercial module eciency values (dening eciency on the basis o the total outer dimensions) are in the

range 12-14% or screen-printed cells and 15-17.5% or the best perorming cells. Device designs capable

o achieving module eciencies o over 18% or multicrystalline silicon, and over 20% or monocrystalline

silicon are expected to be achieved at production scale in the short to medium term. Promising candidates

or such developments are heterojunction cells o crystalline silicon waers with doped amorphous silicon

layers and all-back-contacted cells on both mono- and multicrystalline substrates.

In the long term silicon technology is expected to continue to play an important role in the PV sector.

However, there is uncertainty regarding the precise module eciency, silicon consumption, cell and

module architecture and nature o the cell raw materials ater 2020, when the market size is expected

to be around several tens o GWp/year. It is likely that silicon technology will by this time incorporate

technologies covered under the heading Emerging and novel PV technologies that are currently only

at very early stages o development. Also, the separate steps or cell abrication and module assembly

may become one single integrated production step with thinner waers or waer-equivalent approaches.

In the long term, it is expected that module eciency will exceed the current laboratory record. This

may only be possible by incorporating technologies at the periphery o the device such as up- or down-

converters. For this reason, basic and applied research on advanced concepts and materials should

be included in crystalline silicon research programmes.

PV

development


27/76

Table 3. Research priorities or waer silicon cells & modules - time horizons or rst expected

application o research results in (pilot) manuacturing and products

PV

developmentoptions,

perspectivesand

R&Dneeds

Cells & Modules 2008 - 2013 2013 - 2020 2020 - 2030 and beyond

Industrymanuacturingaspects

n Module eciency >17%n mono & multi/ribbon

>15%n In-line high-yield

processingn Standardisationn Sae processing

and products

n Module eciency >20%n mono & multi/ribbon

>18%n High speed processesn Frameless structuren Sae processing

and products

n Module eciency> 25%

n Energy pay-back time< 6 months

n Sae processingand products

Applied/ advancedtechnology aspects

n Back-contact cellstructures

n New technologiesor electrical contacts

n Heterojunctions oremitters and passivation

n Contact/surace

passivationn Roll-to-roll/ automatic

module manuacturingn Low cost raming/

mounting

n Lietimes > 35 yearsn Metal contacts

(processes, schemesand materials)

n mproved devicestructures andinterconnection

schemes or modulesn Metal contacts

(processes, schemesand materials)

n Improved devicestructures integratingcells/modules

Basic researchand undamentals

n Epitaxial Si lms onlow cost waers

n Low recombinationcontacts

n New device structuresn New passivation

techniques

n Epitaxial Si lmson low cost waers

n Recrystallised Sion ceramics

n Low recombinationcontacts

n New device structures

n New device structuresincluding up/downconverters and othernovel concepts

Casting crucible lledwith silicon (400 kg)Source : Photowatt

Silicon eedstockSource : Deutsche Cell


28/76

4.2.4 Manuacturing and installation

Material consumption must be reduced to avoid scarcity, reduce costs and reduce the energy pay-back

time and other environmental impacts associated with PV module production.

Investment costs or manuacturing plant represent an inevitable part o the cost breakdown o crystalline

modules but should reduce with manuacturing equipment standardisation. It is expected that specic

plant investment costs will reduce rom 1 /Wpor more in current actories to less than 0.5 /W

pin

the long term.

The size o actories is also important or reducing costs suciently to meet to the overall targets. It is

expected that the current plant capacity o typically 100 MWp/year plant will grow to 500-1000

MWp/year in the short term and probably an order o magnitude higher in the long term. It is likely

that in each parallel production line in these plants multiple processes will be perormed concurrently.

Batch processes will tend to disappear. Module assembly, or instance, will likely become an automatedsequence in which sheets o encapsulating materials will be applied on reels and spools. Reaching such

production scale will require great eort.

The SRA emphasises cost reduction but attention should also be given to product and process saety

and the environmental impact o PV. In order that the large-scale use and production o PV nds popular

support, saety must be designed into uture products. Materials, manuacturing and installation must

be sae and environmentally riendly. Recycling chemicals and system components may result in cost

savings and play an important role in increasing the publics acceptance o PV.

4.2.5 Summary

In conclusion, research into crystalline silicon photovoltaic technology will primarily have to address

the ollowing subjects:

n Reducing the specic consumption o silicon and materials in the nal module

n New and improved silicon eedstock and waer (or waer equivalent)

manuacturing technologies, that are cost-eective and o high quality

n Increasing the eciency o cells and modules and, in the longer-term,

using new and integrated concepts

n New and improved materials or all parts o the manuacturing chain, including encapsulation

n High-throughput, high-yield, integrated industrial processingn Finding sae processing techniques with lower environmental impact

PV

development


29/76

PV

developmentoptions,

perspectivesand

R&Dneeds

Cell productionSource : Deutsche Cell

Silicon waersSource : Solar World

Full size moduleSource : ECN


30/76

4.3 Existing thin-lm technologies

4.3.1 Introduction

Thin-lm solar cells are deposited directly on large area substrates, such as glass panels (square metre-sized

and bigger) or oils (several hundred meters long). Thin-lm PV has an inherent low-cost potential because

its manuacture requires only a small amount o active (high cost) materials and is suited to ully integrated

processing and high throughputs. There are three major inorganic thin-lm technologies, all o which have

been have been manuactured at pilot scale (1-2 MWp) and are being or have been transerred to high

volume production (10 MWpto over 50 MW

p). The three technologies are amorphous/microcrystalline

silicon (TFSi - 13% eciency), and the polycrystalline semiconductors CdTe (16.5% eciency) and CIGSS

(an abbreviation o Cu(In,Ga)(S,Se)2- 19.5% eciency). They share a number o common eatures. Each

technology requires only small amounts o semiconductor material: the lm thickness is typically 1/1000

mm (1 m). They have all shown long-term stability under outdoor conditions. They require minimal energy

inputs: the energy pay-back time o thin-lm modules is already around 1.5 years in central Europe and1 year in southern Europe and could reach 3 months in the uture.

At present, the market share o thin-lm PV within total PV production is below 10%, but might grow

to 20% by 2010 and beyond 30% in the long term. The availability o large-area deposition equip-

ment and process technology, as well as the experience available rom within the architectural glass

industry and the fat panel display industry, oer signicant opportunities or high-volume and low-cost

manuacturing. The monolithic series interconnection o cells to produce modules simplies assembly in

comparison with waer-based technologies. Flexible lightweight modules can also be produced using

thin polymer or metal substrates and roll-coating techniques.

Thin-lm technology thus has a great potential or cost reduction.

The challenges acing thin lms are to be ound mainly in the realm o up-scaling production capacity.

The global production capacity o thin lms is expected to reach 1 GWp/year in 2010 and 2 GW

p/

year in 2012. It is being installed mainly in Japan, the USA and Europe. Europe already has excellent

thin-lm R&D inrastructure and a number o thin-lm actories.

Taking account o the increase in production acility sizes, improvements in module eciency and di-

erences in the calculation methods used by the PV industry, in 2010, the total manuacturing costs will

most likely be in the range o 1-1.5 /Wp. Further cost reduction to below 0.75 /W

pin 2020 and

0.5 /Wp by 2030 can be reached. Little dierence in cost between the dierent thin-lm technolo-gies is expected in the long term.

In summary, low cost and high-volume production o thin-lm PV modules is achievable and should en-

able costs to reach 0.5 /Wpin the long term i intensive R&D work is carried out.

4.3.2 Common eatures o all existing thin-lm technologies

As thin-lm PV modules have broadly similar structures and the key steps in their production steps re-

semble one another, R&D eort directed at one technology could be applied to another, increasing its

useulness. This section analyses aspects common to the three thin-lm technologies, with requirements

specic to each one given in later sections.

PV

development


31/76

4.3.2.1 Manuacturing and product issues

Production equipment plays a crucial role in cost reduction. Standard equipment needs to be devel-

oped in conjunction with well-dened processes to achieve higher throughputs and yields. Equipment

manuacturers will play a vital role in this development and knowledge gained in relevant industries

outside PV should be exploited. Deposition equipment rom the fat panel display industry is an example.

Another is sputtering equipment originally developed or coating glass but which can also be used or the

deposition o transparent and metallic contacts on thin-lm PV technologies. A nal example is the use

o roll-to-roll coating equipment, developed or the packaging industry, to manuacture fexible modules

on oils. Productivity parameters such as process yield, uptime and throughput have to be improved

by optimising existing processes and developing new processes. Quality assurance procedures and

in-line monitoring techniques need to be developed urther to improve production yield and module

eciency. The integration and automation o production and processing steps into one line should also

help reduce production costs.

Standardising substrates or modules and other common elements or the dierent technologies will help

reduce the capital cost o production plant. Jointly pursuing the standardisation o equipment and o

building construction elements is necessary and will have positive eects on overall system costs.

Low-cost fexible modules on alternative substrates oer urther potential or cost reduction and enable

new module designs. The equipment and processes or the manuacture o such thin-lm products on

polymer and metal lms have to be developed and improved to take ull advantage o roll-to-roll produc-

tion technologies and monolithic interconnection.

Low-cost module encapsulation (also known as packaging) needs to be developed. Packaging

includes the backsheet, bypass diodes, rames and laminating oils. In addition new module concepts

are required that or example allow higher system voltages or that better tolerate shading.

Common R&D needs or all thin lms are summarised as ollows:

n Standardised product sizes to make the handling o

products easier or vacuum deposition equipment

n Standardised deposition techniques employed by the equipment

n Lower cost transparent electrode materials with better optical and electrical properties,

cycle times and yields and nding ways to include them in manuacture

n Patterning processes or monolithic integration that reduce electrical andarea loss and improve cycle time and yield (e.g. laser scribing)

n Polymer or alternative sealing solutions with longer lietimes

that are suitable or higher throughput manuacturing

n New and low-cost packaging o the active layers: barrier coatings, polymer oil to reduce

material cost and to enhance productivity, in-line processes or packaging

n Concepts or easier mounting and interconnection o modules

n BIPV module designs that look more attractive and perorm optimally;

while being adapted to current practice in the construction sector

n Quality control methods and inline quality assurance

n Supply chain logistics (primarily outside the actory) or large-scale production

PV

developmentoptions,

perspectivesand

R&Dneeds


32/76

4.3.2.2 Eciency and material issues

The undamental properties o inorganic thin lms (with the exception o amorphous Si) are only partially

understood. There is less accumulated knowledge o how to process thin lms compared with the situation

or crystalline Si. Fundamental research is needed to improve device quality and module eciency and

to develop a better understanding o the relationship between the deposition processes and parameters,

the electrical and optical properties o the deposited materials, and the device properties that result.

n Better undamental understanding o the electronic properties o

the three amilies o materials and their interaces is needed

n Improvement o the quality and stability o transparent conductive oxide (TCO)

layers is required, while at the same time reducing the TCOs cost

n More advanced methods or optical connement are required so that active layers can be

thinner and cost less. Advanced optical and electronic modelling o heterostructures is required

and technologies or the implementation o these methods needs urther developmentn High-eciency concepts using materials with dierent band gaps or wide spectrum

absorption should be developed. Alternative absorber materials should be explored

n Novel high-eciency concepts are required, including,

with a view to the long-term, spectrum conversion

4.3.2.3 Perormance

Although thin-lm modules have been in use or over 25 years, eld experience o todays technology

is limited. The material modications and continuous process optimisations in thin-lm module produc-

tion are changing the characteristics and perormance o devices. Although no undamental problems

have so ar arisen, there is a need or accelerated ageing tests o new thin-lm modules to assess their

designs, requiring a better basic understanding o their ageing mechanisms. The development o standard

procedures or measuring the perormance and energy yield o thin-lm modules is also important.

4.3.2.4 Recycling and energy pay-back time

As with any new product, dedicated recycling processes need to be developed both or production

waste and modules that have reached the end o their lives. The processes should minimise the genera-

tion o toxic waste and allow high value metals and module elements to be recuperated.

PV

development


33/76

Table 4. Research priorities or thin lm PV - common aspects - time horizons or rst

expected application o research results in (pilot) manuacturing and products

2008 - 2013 2013 - 2020 2020 - 2030 and beyond


n Optimisation oproduction lineswith respect toproductivity andproduct quality.

n Standardisationo equipment.

n Assessment oadvanced materialsand processes.

n Integration o advancedquality control.

n Pilotingroll-to-roll-concepts

n Proo o conceptsor modied/newdeposition methods,and combinationso processes andmaterials.

Applied/advancedtechnology andinstallation(incl. O&M)aspects

n Increase o moduleeciency by 3%(absolute).

n Improvement odeposition andpatterning componentsand concepts.

n Development o quality

control methods.

n Increase o moduleeciency by a urther2% (absolute).

n Trials o other depositionmethods and substrate/sealing concepts.

n Demonstration olow-material-cost

manuacture.

n Increase o moduleeciency by a urther2% (absolute).

Basic research /undamentalscience

n Basic understandingo the physics andchemistry o TFmaterials.

n Optical connement.n Development o other

concepts or deposition,patterning, sealing.

n Diversication awayrom fat glass asa substrate

n Multi-junctions.n Spectrum conversion

concepts.n Technology merging

PV

developmentoptions,

perspectivesand

R&Dneeds

Crystalline silicon modules on roos inHeerhugowaard, The NetherlandsSource : Roos Magazine

WIM


34/76

4.3.3 Thin-lm silicon (TFSi)

TFSi modules are based on amorphous silicon (a-Si) or silicon-germanium (a-SiGe) alloys, microcrystalline

Si (c-Si), and on processes involving the large-scale recrystallisation o Si. Following two decades o

slow growth, the TFSi industry is reviving, thanks to new technologies like c-Si and the development o

large-area production. Major companies in the US and Japan are oering high quality products manu-

actured using equipment and processes that they have developed with substantial government support.

The TFSi sector benets directly rom the advances that have been made in the fat panel display sector

with plasma enhanced chemical vapour deposition (PECVD), which can be applied to the deposition

o a-Si on large areas. Several EU companies have recently announced the start o mass production

using such equipment. The presence in Europe o a competent pool o module producers, equipment

manuacturers and research institutes creates a avourable environment or the advancement o TFSi.

4.3.3.1 Materials and components

The long-term cost o TFSi modules is determined by the cost o active layer material, the modules eciency,

the choice o encapsulation and packaging materials and the investment cost o production equipment. To

reduce these costs, research should ocus on improving the active layer material, especially o technologies

based on c-Si, nding ways to produce such materials at industrial scale, and developing adequate

transparent conductive oxides (TCOs) and substrates. The most important aims are listed here:

n Lower-cost plasma deposition in the manuacture o high quality micro- or nano-crystalline

Si solar cells, or example using higher deposition rates and simplied processes

n Specic high-quality TCO or glass/TCO-stacks suitable or high-perormance cells,

as well as materials suitable or reversed congurations in which the active cell layers

are deposited on a fexible and/or non-transparent substrate (e.g. or roll-to-roll)

n Improved understanding o the properties o materials, or example, the transport o

electrons in c-Si and o interaces in single- and multi-junction devices:

- low recombination loss junctions

- the use o optical refectors between cell stacks

n New lower cost materials/components or packaging

n New layers and materials, or example c-SiGe, SiC, nanocrystalline-diamond,

layers with quantum dots, spectrum converters

n Evaluation o alternative, potentially lower-cost approaches or the

deposition o high quality layers (or example without plasma)

4.3.3.2 Perormance and devices

Several possibilities exist or improving the eciencies o single-junction amorphous silicon modules. For

instance, eatures such as microcrystalline junctions may be added, or the modules may be combined

with SiGe alloys. The introduction o these advanced eatures at low cost and the achievement o higher

module eciencies are key to the long-term success o the technology. A promising concept is the a-Si/c-Si

tandem cell. The best typical stabilised laboratory conversion eciencies are currently in the range o 9.5%

(a-Si), 12% (tandem a-Si:/c-Si) and 13% (triple junction using SiGe alloys). This translates into commercial

module eciencies o 6.5, 8.5 and 7%, respectively but large area module eciencies up to 11% have

been demonstrated at the prototype scale. Achieving a laboratory-scale perormance rom production

modules and mastering the production o multi-junction devices are the major challenges acing TFSi.

PV

development


35/760

Table 5. Research priorities or thin-lm silicon (TFSi) - time horizons or rst expected

use o research results in (pilot) manuacturing and products

PV

developmentoptions,

perspectivesand

R&Dneeds

2008 - 2013 2013 - 2020 2020 - 2030 and beyond


n PECVD systemsto depositmicrocrystalline-Si

n High-rate c-Sideposition

n Produce high-qualityTCO

n Low cost packagingsolutions/reliability

n Produce technology(interconnection/cleaning,...)

n Roll-to-roll processing

Target :

line demonstration< 0.95 /Wpor 100 MW

p,

>10% (glass substrate)

< 0.75 /Wpor 50 MW

p,

> 9% (fexible substrate)

n Demonstrate nextgeneration equipmentwith lower material use,higher throughput andhigher eciency

n Simplied productionprocesses

n Ultra-low costpackaging

Target :

concept demonstration< 0.65 /Wpor 200 MW

p,

> 12% (rigid)

< 0.5 /Wpor 100 MW

p,

> 11% (fexible)

Target :

concept< 0.4 /Wpat 500 MW

p,

>15% (rigid)

< 0.3 /Wpat 500 MW

p,

> 13% (fexible)

Applied/advancedtechnology aspects

n Large area plasmaprocesses oramorphous andmicrocrystalline SiPlasma process control

n Improved substrates andTCO (light trapping)

n

Advanced embeddingmaterialsn Alternative techniques

or absorber deposition

Target :Demonstrate moduleswith > 12%

n New depositionreactor concepts

n Fast high quality TCO/substrate preparation

n Introduce ully optimisedlight trapping schemeson large area

n

Process gas recycling/ull gas use

Target :concept or moduleswith > 15%

n Test advanced devicein pilot-reactors

n Designs or ultra-high-throughput lines/reactors

n Process simplicationn Fully integrated

production line

Basic research/undamentals

n Quantitativeunderstanding oelectronic propertieso layers andinteraces in devices

n TCO/semiconductorinteraces

n

Quantitativeunderstandingo light trapping

n Developments oimproved selected celllayers (e.g. c-SiGe,SiC, nanocrystalline-diamond)

Target :push up eciencies anddemonstrate stable cellswith > 15%

n New techniquesor very high-ratedeposition

n Incorporatequantum dots orspectrum-convertingeects in thin-lm Si

n

Combine thin-lm Siwith other PV technologyn Understand undamental

limitations o thin-lm Si

Target :concepts or stable cellswith > 17%

n higher perormancematerials

n p-type TCOsn Photonic crystals,

diraction eects,eective mediumapproaches, etc.

n

introduce new materialsn test new concepts

Target :to narrow downthe range o ideasor cost reduction


36/76

Short-term objectives:

n Manuacture o high eciency tandem (a-Si/c-Si) and triple junction devices at industrial scale

n Plasma process control and monitoring

n Quantitative understanding o the undamental limits o

c-Si solar cells incorporated into multi-junction cells

n Quantitative understanding o light trapping, and o the undamental eciency

limits o TFSi-based multi-junction devices (including SiGe alloys)

n Demonstration o TFSi cells with stabilised eciencies above 15%,

and, at the module scale, over 12%

In the long term, higher stablilised cell eciencies should be demonstrated (above 17% by 2020).

One possible route to higher eciencies could involve the incorporation o selected improved layers

into devices (e.g. c-SiGe, SiC, nanocrystalline-diamond, photonic crystals), the use o quantum dots

or spectrum-converting eects in thin-lm Si, and the combination o thin-lm Si with other absorbers

(PV technology merging).

4.3.3.3 Manuacturing and installation

Analyses suggest that within the next two or three years production costs in the range 1.3-1.6 /Wp

should be achievable or a-Si modules (eciencies o 6.5 to 7.5 %), and or micromorph modules (e-

ciencies 8 to 9%) using production equipment that has recently become available. The target or 2013

is an eciency increase to above 10%, with production costs below 1 /Wp

on rigid substrates. The

corresponding targets or fexible substrates are 9% and 0.75 /Wp

respectively. The targets assume

production lines o 100 MWp/year or glass substrates and 50 MW

p/year or fexible substrates. To

meet these goals, cost-eective deposition o microcrystalline Si on large area (> 1 m2) must be achieved.

The second priority is to ensure the availability o suitable production equipment or high quality large

area TCOs or TCO-stacks. Thirdly, to reach eciencies in the range o those o crystalline Si, improve-

ment across the manuacturing chain is required (achievable through minimising interconnection losses,

improving homogeneity and using in-line process control). The value in developing modules on both

glass and fexible substrates with reliable, lower cost packaging should be assessed.

Manuacturing and installation-related aims are summarised as ollows:

n A ull understanding o the relationship between plasma processes,

reactor geometry and layer / device properties and the eects o upscaling

n The design and construction o low-cost equipment able to deposit

c-Si and related layers over a large area at high ratesn Large area, high rate abrication o TCOs with hig

Sra Photo Voltaic En

Documents

Transcript of Sra Photo Voltaic En