PVI Lite - Vol 02 2009

New Products Solar Metrology's XRF glass panel sampling tool I Applied Materials’ ‘MaxEdge’ wire saw system I Thermotron's Solar Panel Test Chambers

Photovoltaics International Volume 02 - 2009

lite

DOWNSTREAM IS UP, UPSTREAM IS DOWN WHAT DOES THE CREDIT CRISIS MEAN FOR THE PV INDUSTRY?

THE PV-TECH BLOG: U.S. MARKET TO TAKE THE REINS

Vi si Do t us 5p n't at m fo Int on rge ers th t t ola e he r 28 t C M h e u M ll A nic ay w h , A ar A3 3. ds .64 65 a 8 8. t

THE CHANGING FACE OF PV INVERTERS K.U. LEUVEN

www.pv-tech.org

Introduction

Publisher: David Owen News Editor: Mark J. Osborne Sub-Editor: Síle Mc Mahon Senior Contributing Editor (U.S.): Tom Cheyney Production Manager: Tina Davidian Design: Andy Crisp Commissioning Editor: Adam Morrison Account Managers: Adam Morrison, Graham Davie, Daniel Ryder and Gary Kakoullis While every effort has been made to ensure the accuracy of the contents of this supplement, the publisher will accept no responsibility for any errors, or opinion expressed, or omissions, or for any loss or damage, consequential or otherwise, suffered as a result of any material here published.

Cover image: RENA CupCellPlate equipment for solar cell contact electroplating, showing Plating Cups with Ni (green), Cu (blue) and Sn (yellow).

Left of picture shows the handling system; right of picture shows inline final rinsing.

Courtesy of RENA GmbH. Printed by Ghyll Print Ltd. Photovoltaics International Lite Volume 2, 2009 ISSN: 1757-1197

EPIA leads the news at the moment (page 2) with their assessment of 2008 showing a total market of 5.5GW representing a 129% growth rate. Almost half of this came from Spain, with 2.5GW installed. With the 500MW cap on future installations in Spain now imposed, it looks like the solar industry will need to find a new market to bolster sales in 2009 (page 32).

After the bumper year in 2008 and the manifestation of the global financial crisis in 2009, many financial analysts (pages 2 & 4) are predicting a decline in growth rates or a flat result. Depending on who you speak to these days, installed capacity for 2009 is projected to be anywhere from 3.5GW to 7GW. Regardless of what your crystal ball tells you at the moment, it is clear that ASPs are coming down across the board. Furthermore, the economic climate is forcing companies up and down the supply chain to reassess their costs (pages 5 & 24).

Despite all the doom and gloom, there are many companies that are looking to future growth in the market by expanding their facilities. SolarWorld has started work on a 210,000 sq. ft. cell facility in Oregon, USA; Solar-Fabrik has begun module production at a new 200MW plant in Freiburg, Germany; and First Solar has reached 1GW of produced CdTe thin-film modules, as well as achieving US$1-per-watt manufacturing costs.

In this issue of PVI Lite we look at the future challenges for inverters with Prof. Driesen of K.U. Leuven (page 12), and provide a focus from NREL on TCOs (Transparent Conductive Oxides), used as contact layers for cells, binding agents for the module and increasingly as the second-level protection of the cells from the environment (page 15).

Remember that the technical papers you read here are just a small sampling of our full journal edition, Photovoltaics International, which is available by subscription four times a year. Subscribers can also access our full catalogue of over 100 premium articles online at www.pv-tech.org.

You can meet me and the team during Intersolar in Munich at our booth in hall A3.648.

Don’t forget to join us for a drink at 5pm on the 28th of May at the Tech Arena (A3.658) for the Cell Awards, the industry’s first manufacturing-focussed awards programme.

David Owen Photovoltaics International

The entire contents of this publication are protected by copyright, full details of which are available from the publisher. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means – electronic, mechanical, photocopying, recording or otherwise – without the prior permission of the copyright owner.

12 Photovoltaic converters: challenges for the next decade K. U. Leuven

2 News

Contents

10 Products – in brief

Courtesy: Enereco srl.

Published by: Semiconductor Media Ltd. Trans-World House, 100 City Road London EC1Y 2BP, UK Tel: +44 (0) 207 871 0123 Fax: +44 (0) 207 871 0101 E-mail: info@pv-tech.org Web: www.pv-tech.org

15 Transparent conducting oxides for advanced photovoltaic applications NREL 21 Polymer development and selection criteria for thin-film and crystalline-silicon module manufacturing Linx Consulting LLC

Photovoltaics International Lite provides access to a handful of technical papers, product reviews and opinion pieces, and serves as an introduction to the full subscription version, which features more than 20 technical papers from leading industry figures each and every quarter.

Annual subscription available for US$199.

For more information call Carlos at +44 (0) 207 871 0148 or email cnorthon@pv-tech.org

24 Downstream is up, upstream is down – what does the credit crisis mean for the PV industry?

EuPD Research

26 TRITEC represents highest quality for photovoltaic systems of all sizes TRITEC International AG

31 Jobs

28 KUKA Systems – recognizing and utilizing synergy effects KUKA Systems GmbH

32 The PV-Tech Blog

Phot ov olt aic s I nt ernat ional

1

News The European Photovoltaic Industry Association (EPIA) called it an ‘exceptional year’ for the photovoltaics industry as it reviewed the PV market’s growth for 2008 in a workshop held in Frankfurt. According to EPIA, the global PV market reached 5.5GW in 2008, up from 2.4GW installed in 2007 – a 129% increase year-on-year.

On a regional basis, Spain was the shining star. According to EPIA, Spain accounted for almost half of new installations, topping 2.5GW. However, that star has already faded with Spain imposing a 500MW cap on future installations. The most reliable and second largest market was Germany, adding 1.5GW in 2008.

Expectations are high for the U.S. to leap forward in 2009 and take up some of the slack left behind by Spain. Italy also showed growth in 2008, ranked fifth at 258MW. Some market analysts expect Italy to double installations in 2009, which could exceed 500MW. 2009 forecast and beyond The EPIA also issued a forecast for 2009 through 2013, noting that unnamed market experts still expect strong growth in 2009 and beyond, despite the current financial crisis that has impacted largescale installations in particular.

Installations are projected to increase to approximately 7GW in 2009 and could reach 22GW in 2013. The rapid growth, however, is subject to EPIA: Top 10 regional PV markets in 2008 (MW). the right policies being implemented in a growing number of countries.

“A diversification of the market is taking place with countries adopting appropriate support policies; this is very good news for the PV industry and the environment”, commented Dr. Winfried Hoffmann, President of EPIA. Market Watch

A c c o r d i n g t o I M S R e s e a r ch , a contraction in PV module shipments is anticipated for 2009 due to the significant void left by Spain, due to its new 500MW cap on installations.

Without the cap and favourable feedin-tariffs in 2008, Spain became the largest market for PV for the first time, surpassing Germany for the highest MW of installations. IMS Research expects a shortfall of some 1.5-2GW in 2009, which will not be absorbed by growth in other regions such as Italy and France.

IMS Research said that underlying demand for PV remains healthy; longterm, double-digit annual growth rates can still be expected as North America as well as new technologies will further expand the PV market in the coming years.

Gartner posts long-range forecast for photovoltaics industry According to market research firm Gartner, the PV industry has hit a wall in 2009, due to the credit crisis, ‘jittery customers’ and currency fluctuations.

The net effect will be a revenue decline of 1% for the year. Revenue from the PV market was reported to be US$16 billion in 2008.

However, putting the short-term challenges aside, the PV industry is 2

w ww. pv - t e ch . o r g

Source: Photovoltaics International

Spanish solar installation cap will cause industry contraction in 2009, says IMS Research

Gartner: PV industry revenue forecast 2008-2013 (US$ billions).

expected to grow to US$34 billion in worldwide revenue by 2013, generating an annual growth rate of 17% through to 2013. Lower ASPs on PV modules will play their part in suppressing revenue growth rather than actual demand. On a gigawatt basis, Gartner forecasts the overall PV market will grow 24%, reaching 6.4GW in 2009 and 23.4GW by 2013.

VLSI Research: photovoltaic equipment sales topped US$4.4 billion in 2008 Specialist market research firm VLSI Research has said that the photovoltaic cell and module manufacturing equipment market, which includes the thin-film equipment sector, reached US$4.4 billion in sales in 2008 as the

solar industry continued to expand and add new manufacturing capacity. The equipment market was also boosted by the emergence of various thin-film technologies that entered production for the first time in 2008, enabling suppliers to record new revenue streams.

However, John West, VLSI Research Europe’s Managing Director, cautioned that growth in the market would slow down in 2009, projecting growth of 8% compared to 2008, reaching approximately US$4.75 billion in sales. The cautionary note was due to an expected decline in conventional c-Si cell equipment sales in 2009 as overcapacity and tight financial markets curtail expansion plans. West noted that there was a halt to capacity expansions in 200 9 as PV cell and module manufacturers absorbed several years of aggressive capacity expansions.

Source: Photovoltaics International

EPIA: Photovoltaics market topped 5.5GW in 2008; 129% growth year-on-year

Overall growth in 2009 is expected to come from the continued roll-out of thin-film manufacturing due in part to the equipment lead-times and large order backlogs held by thin-film equipment suppliers.

Fab & Facilities

SolarWorld start construction on new Oregon building

Courtesy: SolarWorld AG

SolarWorld will begin construction soon on a new building adjacent to its Hillsboro, Oregon production plant.

The 210,000 sq. ft. facility, scheduled

for completion in November, will house a combination of logistics and manufacturing activities.

The green-field structure, the second phase of the company’s buildout at the site, will increase the overall plant space by 44%. SolarWorld’s main building – a converted former Komatsu semiconductor factory now home to an integrated solar-cell fab said to be the largest of its kind in North America – measures about 480,000 sq. ft. The company holds 100 acres of property at the Hillsboro location.

The move keeps the company on track to reach its goal of ramping 500MW of annual cell-making capacity in Hillsboro by 2011 and of eventually employing some 1,100 workers there.

Everbrite Solar chooses Ontario site for thin-film PV module plant Everbrite Solar has chosen Kingston, Ontario, as the site for its new amorphous/microcrystalline-silicon thin-film solar PV manufacturing facility – said to be the first of its kind in Canada. The unit of Everbrite Industries is working with several financial advisers to raise CAD$500 million for investment in the highly automated, 150MW annual capacity factory.

Everbrite Solar will use a turnkey TFPV module production line from an “unnamed overseas manufacturer.”

Scherre is also quoted as saying first modules could ship by the end of 2010, if all the financing is lined up.

Everbrite said that it would invest up to CAD$25 million to build an experimental thin-film manufacturing facility to which Queen’s researchers will have access for their studies.

GmbH, held a groundbreaking ceremony (see image above) in late March for its new ½530 million crystalline solar cell and module plant in Arnstadt, Germany.

Production is expected to start at the beginning of 2010, and the facility will be fully ramped by 2012. ersol’s crystalline-based capacity will be increased to 630MWp, compared with the approximately 200MWp reached in 2008. The new plant will be capable of producing 90 million solar cells per year, according to the company.

Solar-Fabrik begins phase I module production at 200MW automated plant Solar-Fabrik has officially opened its new state-of-the-art photovoltaics module plant (Plant III) in Freiburg. The first phase of the expansion (60MW) increases Solar-Fabrik’s total module production to 130MW, while the second phase of expansion is expected to take place in 2010. Plant III has a nominal planned capacity of 200MW and covers 15,000m2 of production floor space.

The highly-automated plant employs transverse soldering equipment and an innovative connection socket concept for the modules as well as fully automated solar wafer handling to reduce breakage.

ersol starts construction of ½530 million crystalline solar cell and module plant SolarWorld’s Hillsboro, Oregon facility.

4

From left: Holger von Hebel, Birgit Diezel, Franz Fehrenbach, Dr. Angela Merkel, Prof. Hermann Scholl, Dr. Siegfried Dais, Ludwig Reiff.

Courtesy: Solar-Fabrik

Although dominated by German-based equipment suppliers, VLSI Research noted that U.S.-based Applied Materials was the largest supplier, with sales of US$455 million in 2008. Applied took the number one spot due to its acquisitions in recent years coupled with the recognition of revenues from its first ‘Sunfab’ turnkey thin-film installations, which entered production for the first time in 2008.

The second-largest PV equipment supplier, according to VLSI Research, was Roth & Rau, with sales of US$275 million in 2008. Other German-based suppliers in the Top 10 rankings included centrotherm (No. 3, US$270 million), Manz Automation (No. 6, US$140 million), Schmid Group (No. 7, US$125 million), VON AR DE N N E Anlagentechnik GmbH (No. 8, US$120 million) and Rena GmbH (No. 9, US$85 million).

The rapid growth in thin-film solar start-ups boosted sales of not only Applied Materials but that of its largest rivals, OC Oerlikon Balzers and Ulvac, Inc. Swiss-based Oerlikon was the fourth-largest equipment supplier in VLSI Research’s rankings, with sales of US$250 million, closely followed by Ulvac of Japan with sales of US$240 million in 2008.

3S Swiss Solar, another thin-film equipment supplier also did well in the rankings, coming in at No. 10 with sales in 2008 of US$70 million.

Courtesy: ersol Solar Energy AG

Top 10 PV cell manufacturing equipment suppliers

w ww. pv - t e ch . o r g

Attended by the German Federal Chancellor Angela Merkel, ersol Solar Energy AG, a subsidiary of Robert Bosch

Solar-Fabrik Plant III Phase I module production.

Materials

Caught with dramatically lower demand for semiconductor silicon wafers and a below-seasonal demand for solar wafers, MEMC is to review its capital spending and polysilicon capacity expansion plans, executives said in response to questions from financial analysts in a quarterly conference call.

MEMC had previously announced plans to boost polysilicon production from the current 8,000MT per annum, reached at the end of 2008, to 15,000MT in 2010.

Although the company has US$1.3 billion in cash, it has not been as aggressive as other major polysilicon producers in adding capacity and has been losing market share in recent years, according to market analysts.

While executives did not reveal specifics, M E M C has experienced pricing pressure and acknowledged that recent job cuts were intended to better balance operational costs with future expected wafer demand.

The company has also had to renegotiate solar wafer prices with several major customers in light of the rapidly increasing global supply of polysilicon and the huge declines in spot market prices, even though longterm wafer contracts are well below spot price comparisons. The wafer producer also acknowledged that it would review its current outsourcing strategy for solar wafer production, which could be brought in-house via possible acquisition of some or all of its current suppliers.

China’s polysilicon industry sees multiple ramps and orders Contributing further to the worldwide polysilicon oversupply, but indicating a positivity that flies in the face of the current economic climate, the Chinese polysilicon production industry looks to be doing well with several new ramps and orders announced.

Fushun Koshuha Foundry Co. Ltd. has embarked upon a potential 10,000ton high-purity polysilicon project with U.S. company PPP and a Japanese firm. The project, which has secured an investment of US$1.2 billion, was scheduled to start construction in April and to be completed in October 2011.

Jiangsu province’s Xuzhou, Yangzhou and Lianyungang regions will be the focal points for the province’s polysilicon activity, with plans to subsidize 260MW of solar power and reach an annual polysilicon production capacity of 30,000 tons by 2011.

Tongwei Co. Ltd. is nearing the second phase of its 3,000-ton polysilicon project in Sichuan province. Groundbreaking for the second phase could happen as soon as the first half of 2009, with a third 6,000-ton phase tentatively planned

Source: Photovoltaics International

MEMC to review polysilicon expansion plans

Top 11 polysilicon producers’ capacity expansion plans (2006-12).

for before late 2010. Including the late2008 first phase of 1,000 tons, this would bring the company’s total output to 10,000 tons per annum.

In Shangyu, Zhejiang, Zhejiang ProPower Silicon has commenced construction on a solar grade polysilicon project that is expected to start production in September 2009. Annual capacity of 1,000 tons should be reached

by December 2009, with potential for expansion to 5,000 tons in 2010.

Masdar PV selects Linde Group for ‘SunFab’ thinfilm gas supply A long-term gas supply deal has been signed between Masdar PV and Linde Group, which also includes gas storage, distribution systems and on-site gas management services. Supplied gases include nitrogen (N2), hydrogen (H2), silane (SiH 4) and chamber cleaning gases, as well as argon (Ar) and helium (He). Masdar PV is currently investing US$200 million in a thin-film plant in Erfurt, Germany.

Masdar PV is using Applied Materials’ ‘SunFab’ thin-film technology and plans to build a ‘copy exact’ plant in Taweelah near Abu Dhabi. Combined capacity when fully ramped is expected to be 210MW. Lessons learnt at the Erfurt plant will be migrated to the plant in Abu Dhabi.

Air Products, Best Solar sign second thin-film PV gas supply deal Air Products and Best Solar have signed a letter of intent for the supply of liquid bulk and on-site gases to the Chinese PV manufacturer’s new amorphoussilicon thin-film photovoltaic facility in Nanchang City, Jiangxi Province.

The agreement between the two companies includes the long-term supply of hydrogen, nitrogen, helium and argon gases. When the facility comes on-stream at full capacity, it will have a module manufacturing capacity of 330MW per year. The LOI marks the second agreement between Air Products and Best Solar in recent months.

Canadian Solar notes challenges to UMG suppliers as polysilicon prices fall Canadian Solar executives said during a conference call to discuss 4Q and year-end financial results that polysilicon prices had fallen to between US$110 per kg and US$130 per kg. This had enabled the module manufacturer to renegotiate UMG silicon feedstock prices to approximately US$60 per kg, maintaining a US$50 per kg margin between the two feedstocks.

Canadian Solar uses UMG silicon in its ‘e-modules’ and has already secured contracts for these lower (15%) priced modules in 2009, equating to 120MW. The challenge for UMG suppliers, acknowledged by Dr. Shawn Qu, Chairman and CEO of Canadian Solar during questioning by financial analysts, is that he expects polysilicon prices to drop to approximately US$70 per kg, pushing UMG pricing to between US$15 per kg and US$20 per kg by the end of 2009.

LDK Solar focuses on cost reduction, not expansion, for 2009 Major solar wafer producer LDK Solar has shifted its business strategy from aggressive capacity expansions to one of production cost reductions for 2009. The continued uncertainty of the growth levels in the photovoltaics industry for the year have turned LDK S olar hyper-conservative, cutting polysilicon production targets as a consequence of delaying previously planned plant ramps and holding wafer production at 2008 final capacity levels, until demand visibility returns.

Annualized wafer production capacity increased by more than 1GW, reaching 1.46GW at the end of 2008. Wafer capacity in 2007 was 420MW. Wafer shipments reached 818MW in 2008, nearly a four-fold improvement over 2007. Broken down, the majority of planned spending will focus on the completion of ‘Train 1’ at LDK’s 15,000MT polysilicon plant, ramp Phot ov olt aic s I nt ernat ional

5

Source: Photovoltaics International

LDK Solar capacity ramp projections (March 2009).

of capacity at its smaller 1,000MT polysilicon plant that was recently completed and a small share of the CapEx budget targeting optimization of its wafer production. Due to the spending being construction focused, the majority of spending will be first-half-year loaded.

Train 2 & 3 of LDK’s 15,000MT polysilicon plant have in effect been put on hold, with manpower resources being shifted to Train 1 completion and ramp by the end of the second quarter of 2009. Executives said that any further development of the large polysilicon plant would be based on market demand.

Total polysilicon production for 2009 had previously been guided in the range of between 3,000-5,000MT in 2009, which has now been further reduced to between 2,000-3,000MT for the year.

Cell Processing

Two-tier roughness could boost solar cell efficiencies by 2%

Courtesy: Georgia Institute of Technology

A process under development at the Georgia Institute of Technology could potentially boost conventional c-Si cell efficiencies by as much as 2%.

Researchers are using two different

Suntech pushes Pluto tech to 19% efficiency on monocrystalline solar cells, 17% on multi cells

A silicon pyramid structures etched for one minute using a hydrogen fluoride/hydrogen peroxide/water solution. The resulting structure has roughness at the micron and nanometre scales.

6

types of chemical etching to create surface features at both the micron and nanometre scale that increase light absorption, reducing reflection and keeping cells clear of stray particles.

“A normal silicon surface reflects a lot of the light that comes in, but by doing this texturing, the reflection is reduced to less than 5%,” said Dennis Hess, a professor in the Georgia Tech School of Chemical and Biomolecular Engineering. “As much as 10% of the light that hits the cells is scattered because of dust and dirt of the surface. If you can keep the cells clean, in principle you can increase the efficiency. Even if you only improve this by a few percent, that could make a big difference.”

The researchers use potassium hydroxide (KOH) solution to etch the silicon along crystalline planes, creating micron-scale pyramid structures in the surface. An e-beam process is then used to apply nanometre-scale gold particles to the pyramid structures. Using a solution of hydrogen fluoride (HF) and hydrogen peroxide (H 2 O 2 ), the gold acts as the catalyst, producing controlled nanometre-scale features. The gold is removed via a potassium iodide (KI) solution and the surface coated with a fluorocarbon material, perfluorooctyl trichlorosilane (PFOS).

w ww. pv - t e ch . o r g

S u n t e c h Po w e r H o l d i n g s s a i d it is “routinely” using the Pluto technology to produce solar PV cells with conversion efficiencies of approximately 19% on monocrystalline silicon cells and 17% on multicrystalline cells. The company also confirmed that third-party test results from the Fraunhofer Institute for Solar Energy

Systems I S E in Germany show conversion efficiencies of 18.8% for monocrystalline cells and 17.2% for the multicrystalline cells manufactured on Suntech’s 34MW Pluto production line.

The patent-pending Pluto approach, based on the PE R L (passivated emitter with rear locally diffused) cell technology, has been developed by the University of New South Wales in Australia and has achieved world-record lab efficiencies of 25%. Researchers believe that the Pluto design can push power output by approximately 12% above conventional screen-printed cells.

A low-reflectivity texturing technology ensures that more photons can be absorbed throughout the day even without direct solar radiation, and thinner metal lines on the top surface reduce shading loss.

Suntech believes that ongoing enhancement of the Pluto technology will, within two years, lead to 20% conversion efficiency on monocrystalline cells and 18% on multicrystalline cells. The company’s 34MW Pluto PV cell line is fully operational; it expects to have a 100MW of installed Pluto capacity within two months.

Suntech also said it expects to receive industry certification for Plutobased modules in the 2Q and plans to ship more than 50MW of the modules in 2009.

GreenVolts, NREL join forces to commercialize advanced multijunction solar PV cells Concentrator photovoltaics company G r e e n Vo l t s a n d t h e N a t i o n a l Renewable Energy Laboratory have joined forces to develop inverted metamorphic (I M M) advanced multijunction solar cells, which have shown cell conversion efficiencies of 40.8%. GreenVolts has inked a licensing deal with N R E L to commercialize the lab’s patents, and the U.S. Department of Energy has allocated US$500,000 toward the effort.

Th e t w o - y e a r a g r e e m e n t w i l l facilitate the transfer of NREL’s IMM cell technology to GreenVolts, so that the San Francisco-based CPV company can develop a customized cell tailored to its optical system.

NREL said that it will offer the necessary technology specifications and process information as well as advice and assistance in the device optimization and technology transfer throughout the duration of the agreement. The lab will also provide reliability and other test and measurement services to qualify the results of the transfer process.

Fraunhofer ISE, Freiburg.

Thin Film

First Solar: Cost per watt & conversion efficiency milestones.

First Solar reaches 1GW of produced CdTe thin-film modules Chalking up another milestone, First Solar has said it has now produced a combined 1GW of CdTe thin-film modules since production first started in early 2002. The thin-film leader expects to have a nominal production capacity of approximately 1.2GW by the end of 2009, which equates to 23 manufacturing lines from plants on three continents.

The company noted that it took more than six years to reach 500MW of produced modules but only eight months to reach its second 500MW, testimony to its rapid capacity plans and execution.

First Solar first to US$1 per watt manufacturing cost

ENN Solar ramps on time with SunFab technology

Photon Consulting may have recently projected the PV industry to reach the US$1 per watt manufacturing cost threshold in 2012, but First Solar has ignored such projections and reached this important milestone in the 4Q08, with a cost per watt of US$0.98. First Solar, not wanting to sit back and wait for competitors to catch up, expects further reductions in the coming years that could see a cost per watt below US$0.65 by 2012 or earlier.

The progressive fall in manufacturing costs of its CdTe thin-film modules was attributed to the company’s continuous focus on cost reductions as production scaled to significant megawatt levels in only a few years.

First Solar’s annual nominal production capacity will double in 2009 to a planned 1.1GW as manufacturing plants, specifically in Malaysia ramp to full production. Plants 3&4 in Malaysia are now ramping, with capital spending expected to be between US$270 and US$300 million to cover the ramp and expansion at its existing plant in Perrysburg, Ohio. Each of the Malaysian plants or lines has a nominal capacity of approximately 190MW. First Solar also achieved its highest stable conversion efficiencies in 4Q08 of 10.8%.

ENN Solar Energy Co., Ltd. has produced its first tandem junction thin-film modules at its 60MW ‘SunFab’ line, claiming the title of first Chinese-based thin-film manufacturer to produce tandem junction cells on a production line.

The companies claim that the new milestone was achieved five months after equipment installation at EEN Solar’s facility in Langfang, China. EEN Solar has aggressive plans to ramp capacity, claiming a target of 500MW by 2011.

Sencera achieves 8.7% efficiency thin-film cell Further developments in Sencera’s cell efficiency have been ongoing since December, as the company announced it has reached an 8.7% efficiency level on its single-junction thin-film silicon solar cells. At the end of 2008, Sencera reached a conversion efficiency of 7%, both of which milestones were achieving using enhancements to its Viper manufacturing platform.

The initial 8.7% sunlight-to-electricity conversion efficiency was reached under standard test conditions, and independently confirmed by The University of Delaware’s Institute of Energy Conversion, which was designated a University Center of Excellence for

Photovoltaic Research and Education by the Department of Energy in 1992.

Such an achievement will lend to the company’s plans to produce a 7% efficient, 106W single-junction amorphous silicon module at its 35MW solar module facility, which is currently under construction in Charlotte, North Carolina.

Ascent Solar to begin CIGS thin-film module production Ascent Solar Technologies, Inc. has announced it has started regular production of monolithically integrated f l ex i b l e C I G S m o d u l e s f r o m i t s production line in Littleton, Colorado.

Ascent S olar is now the only company to begin the production of fully integrated lightweight CIGS thinfilm modules using a plastic substrate.

It plans to operate the manufacturing line with a single shift, progressing to full-scale 1.5MW rated capacity using three manufacturing shifts every day.

Courtesy: Ascent Solar

Courtesy: Fraunhofer ISE

Entech Solar research has entered a collaborative agreement with F r a u n h o f e r U S A’ s C e n t e r f o r Sustainable Energy Systems (CSE) to evaluate the possibility of incorporating high-efficiency back-contact silicon solar cells into Entech’s 20x concentrating solar systems.

Working with Freiburg, Germany’s Fraunhofer Institute for Solar Energy Systems (ISE), Fraunhofer CSE will implement research into the feasibility of this potentially efficiency-boosting addition to Entech’s current cells.

The organisations will simulate, design, and test prototype backcontact silicon solar cells and make recommendations in regard to Entech’s mass production of such products.

Source: Photovoltaics International

Entech Solar joins forces with Fraunhofer CSE for back-contact cell testing

Ascent Solar’s flexible CIGS modules.

PV Modules

PV module market to contract by 15% in worst growth year since 1994, says Greentech Media Greentech Media and the Prometheus Institute for Sustainable Development claim that 2009 will see the weakest growth year since 1994, and that the industry will be dominated by Asian multicrystalline and C I G S, with solid share for CdTe and super monocrystalline technologies, by 2012.

Using a newly formulated integrated supply-demand model, the report, ‘2009 Global PV Demand Analysis and Phot ov olt aic s I nt ernat ional

7

REC Solar to slash 2Q09 PV cell, module production by nearly 50% In response to what it calls “the challenging market environment for the solar industry” and subsequent buildup in its inventories in the first quarter, REC Solar has decided to slash its 2Q production of solar PV cells and modules by nearly 50%.

As a result of the move, the company said that approximately 180 employees will be affected through terminations of temporary employment contracts, temporary lay-offs of permanent employees, and rescheduled vacations.

As long as there is “good contract coverage,” REC also said it does not expect changes to the production of cells and modules in the 3Q and 4Q. Power Generation

Italy to more than double installed solar PV capacity by end of 2009 Incentives will lead to a more than doubling of Italy’s installed solar photovoltaic capacity – from about 440MW to 900MW – by the end of 2009. The Italian power management agency GSE said that some 338MW of PV capacity was installed in 2008 – said to be the third-largest annual rise in solar capacity in the world, on a par with the U.S. and behind Spain and Germany.

About 34,000 new PV installations with a capacity of 435MW started up 8

w ww. pv - t e ch . o r g

in the country once new government incentives took effect in mid-2007.

Those numbers compared to about 2,500 PV installations with a capacity of 22MW deployed over about two years, after the first incentive scheme was introduced in 2005. Italy could exceed 70,000 by the end of 2009 with a total capacity of 900MW.

Siemens targets dominant position in solar thermal technology Siemens has acquired a 28% share in Italian solar thermal specialist Archimede S olar Energy for an undisclosed sum, using the purchase to claim that the German engineering conglomerate was intending to become the leading global company for solar thermal power plants. Siemens believes that the solar thermal power market would experience double-digit growth rates per annum, reaching more than ½10 billion in sector revenue by 2015.

Siemens said it would provide the capital necessary for a rapid expansion of production capacity at Archimede, with the option to acquire a majority stake in the solar company in the ‘midterm.’ Archimede only had revenue in single-digit million euro range in 2008.

Archimede uses molten salt as heat transfer fluid in its solar receivers for parabolic-trough power plants.

Evergreen Solar targets utility-scale projects Evergreen Solar has entered into an agreement with RMT, Inc., a renewable energy projects firm, to boost sales of its String Ribbon modules into the utilityscale solar plant market, which is expected to grow rapidly under new renewable energy strategies, recently adopted as part of the U.S. economic stimulus pacts. The new agreement will see RMT handle all engineering, procurement and construction functions while Evergreen Solar will provide the solar panels.

The companies said that they have already submitted bids for more than 400MW of solar installations, which could be implemented over the next five years.

National Semi claims tests show 57% power pop in PV panel performance As much as 57% of the power lost because of temporary or partial shading of solar PV panels can be recouped with the use of a new optimization technology, according to test results released by National Semiconductor.

The company said that its SolarMagic power optimizers (which will be available this Spring) improve the energy harvesting of solar panels in real-world conditions, where shading and other issues can significantly reduce the performance of solar systems.

The tests, conducted at National’s

Santa Clara, California, facility, used a screen representing typical rooftop obstructions to replicate shade on a portion of a conventionally wired solar PV system. Although 8-16% of the array was shaded over the course of a day, it resulted in average power losses of 35-40%. However, an identical solar array fitted with the SolarMagic devices produced an average of 30-37% more electricity in the same conditions – effectively recouping up to 57% of the lost power, according to the company.

The test and reference arrays were each made up of two strings with 12 PV panels per string, with both strings attached to a Xantrex inverter. The performance data were collected using the inverter company’s software and had a measuring accuracy of ±5%.

Energy Innovations creates spin-off for RayTracker products Energy Innovations has announced that its RayTracker product group has been spun off to form RayTracker Inc. The new company will allow the team to take advantage of the growing demand for its RayTracker GC product line. The GC line was made to be cost-effective while boasting high efficiency and reliability.

RayTracker had been included in over 2MW of installations in 2008.

The company claims that PV panels mounted on a RayTracker system yield up to 38% more energy than fixed flat PV systems annually and up to 23% more than fixed-tilt systems.

Courtesy: RayTracker GC

Forecast: Anatomy of a Shakeout II’ (which is the companion to the groups’ ‘PV Technology, Production, and Cost, 2009 Forecast: Anatomy of a Shakeout’ report), also forecasts a contraction of the PV module market by 15% – to US$12 billion this year – on 5GW of global demand. Some key findings of the study are: s A predicted fall in module ASPs to below US$2.50 per watt in 2009 will be followed by a further decrease in 2010 to US$2.00 per watt as demand-side financing pressures force manufacturers to cut prices. s Global module capacity will grow to 27.5GW by 2012, which will be enough to produce 23GW of PV modules; thin-film will account for 34% of that total. Market share for thinfilm modules, in terms of incremental demand, will increase from 28% in 2008 to 50% by 2012. s The cost of c-Si modules is expected to decrease by around 50% to US$1.40 per watt by 2015, while the cost of CIGS modules will fall to US$0.75 per watt over the same period. s High-efficiency monocrystalline and low-cost thin-film technologies will see a 30% cost advantage over traditional multicrystalline producers as a result of efficiency adjustments.

RayTracker with mounted PV panels.

Jordan’s solar future looks bright with US$13.2 million funding from European Commission Jordan’s solar industry looks to be on the up as reports have emerged of a US$13.2 million funding package for the establishment of a solar plant and renewable energy research facility. The ~5MW plant will be located in Fujeij, near Shobak in the southwest of Jordan.

The National Energy Research Center, as it will be called, will comprise the 5MW solar power plant as well as a renewable energy training centre for local and regional workers. Construction of the plant will contribute towards the country’s national energy strategy that is striving to establish 600MW of solar power installations by the year 2020.

Products – in brief Coherent boosts laser beam quality, repetition rate and output power for productivity gains Applications: Picosecond Paladin lasers can be used for c-Si applications such as dielectric ablation, or P2/P3 patterning of thin-film CIGS layers. The AVIA family of ultraviolet lasers can be used for scribing, laser edge isolation and laser grooved buried contacts as well as emitter-wrap-through and metal-wrap-through cell designs.

Platform: The new AVIA and Paladin models are all Diode-Pumped Solid State (DPSS) lasers, which operate off single-phase electricity and require no external consumables.

Availability: Currently available.

Diffraction-limited output for minimal heat affected zones.

New lasers from Coherent, Inc. offer higher output power and increased throughput levels for solar cell manufacturing applications. The ‘AVIA’ lasers now deliver output power levels of 28W at the short ultraviolet wavelength of 355nm, and 45W in the green at 532nm. Each of these lasers represents the highest power level within their class. In addition, a new ‘Paladin’ laser now delivers ultra-fast 80MHz trains of picosecond pulses with power levels up to 16W at 355nm. The beam quality, high repetition rate and high output power of AVIA lasers combine to minimize the heat-affected zone in micromachining applications.

KLA-Tencor’s PVI-6 wafer inspection system offers up to 4x measurement accuracy improvement Applications: Optical in-line dualsided inspection of photovoltaic wafers and cells, including inspection of bare solar wafers after silicon nitride (SiN) coating, and after each metallization step and end-of-line cell classification.

Platform: The PVI-6 software includes improved ease of use and analytical tools to increase the overall yield of the solar cell production process.

Availability: Currently available.

Higher accuracy and repeatability measurements with up to 4x measurement accuracy improvement.

The newest addition to KLA-Tencor’s ICOS division's PV wafer inspection systems, the PVI-6 provides the capability to inspect solar wafers and cells at higher speed and accuracy for all stages of the production process.

These new capabilities enable solar manufacturers to achieve substantial yield improvements and more accurate product classification that includes up to 4x measurement accuracy improvement and an 80% reduction in calibration time, enabling faster product ramp during initial installation. The system also supports tool matching and central module management, and provides consistent and easily attainable results in large production environments across multiple production lines.

Pall claims 95% water re-use for new water recycling system used for wafering Applications: Reclamation of water and silicon sludge from spent grinding/ sawing water.

Platform: The system design follows a modular concept and comprises a dynamic membrane filtration (MF) unit, combined with physicochemical pretreatment of the spent process water.

The capacity of the standard systems ranges from 3m³/h to 36m³/h (up to 140MWp p.a.). Higher capacity systems are available on demand.

Availability: Currently available.

The cost of ownership of the reclaimed water ranges from US$0.50 per m³ for small systems to US$0.25 per m³ for large systems.

10

w ww. pv - t e ch . o r g

Pall Corporation has introduced a new generation of fully-automated separation systems for the reclamation of water and silicon sludge from the spent grinding/sawing water. The systems completely clarify the contaminated process water. Typically, 90 to 95% of the contaminated process water is transformed into particle-free permeate ready for re-use. The remaining concentrate may be directly discharged, mixed with other wastewater streams, or further treated to meet discharge regulations as well as to de-water the silicon debris.

Applied Materials’ ‘MaxEdge’ wire saw system first with dual-wire management Applications: c-Si ingot slicing down to 120μm.

Platform: MaxEdge is claimed to enable the of use thinner wire for larger loads at higher cutting speed than any other system, capable of handling 45% larger loads compared to the previous model, the B5.

The system's throughput and load improvements enable a reduction in the number of systems needed for a given capacity. This translates to a significant reduction in capital expense.

A reduction in the required number of operators and maintenance expense provides a lower cost of ownership.

Availability: Currently available.

Applied Materials has launched a new unified platform for slicing ingots into ultra-thin wafers that caters for both high-volume production and R&D applications. Applied’s HCT ‘MaxEdge’ wire saw system is the first in the industry to employ a dualwire management system that is claimed to offer significantly higher throughput (double) and load capacity t h a n c o m p e t i t i v e s y s t e m s . Th e compact footprint requires less factory floor space and fewer operators for equivalent megawatt output, helping to drive down the cost of manufacturing photovoltaic (PV) cells by up to US$0.18 per watt.

Dual-wire management system employs four independently-controlled direct drive motors and advanced process control.

Solar Metrology offers XRF glass panel sampling tool for CIGS and CdTe Applications: Glass substrate PV panel measurement of CIGS and CdTe photovoltaic depositions (composition and thickness) for off-line or near-line PV panel analysis.

Platform: The SMX head is bolted to a specially designed port; x-rays pass in and out of a process vessel, enabling film measurement and process control.

The tool has a full 600mm x 1200mm lateral XY range of measurement.

Availability: Currently available.

Solar Metrology has expanded its SMX XRF tool portfolio for film composition and thickness measurement of CIGS and CdTe photovoltaic depositions with the addition of the FPV (Full Panel View) SMX model. The SMX-FPV is designed

for near-line film composition and thickness control of CIGS and CdTe film stacks. The tool’s full 600mm x 1200 mm lateral XY range of measurement is designed for measurement of rigid glass substrates. The FPV provides process control of active, contact and TCO layers. Detailed analysis of full photovoltaic panels is possible including fast and repeatable copper and gallium ratio determination, while panel gradient analysis allows for yield improvement, management and conversion efficiency gains in production.

Copper and gallium ratio determination as well as panel gradient analysis.

Thermotron Solar Panel Test Chambers offer flexible module life-cycle evaluation Applications: CdTe, CIGS and c-Sibased solar modules.

Platform: Workspace dimensions will comfortably accommodate solar panels ranging in size from 2’ x 4’ up to 4’ x 6’ (1200mm x 600mm to 1200mm x 1800mm). Panels exceeding 6’ (1.8m) in length may require a walk-in-sized room for testing.

Availability: Currently available.

Thermotron Solar Panel Test Chambers are capable of performing thermal cycling, damp heat, and humidity freeze tests as detailed in IEC 61646, IEC 61215 and UL 1703. Solar panel test, measurement and data acquisition is also supported by the chamber instrumentation.

In addition to extreme temperature testing from +180°C to -70°C (356°F to -94°F), these chambers include a reliable, accurate and efficient full-range humidity system for simulating conditions from 10% RH to 98% RH. Chambers are outfitted with fixturing solutions specifically developed for easy loading and unloading of solar panels. Fixturing can adjust to safely support panels of different configurations, optimizing airflow and environmental consistency throughout the entire workspace.

Chambers cover damp heat and humidity freeze tests as detailed in IEC 61646, IEC 61215 and UL 1703.

Phot ov olt aic s I nt ernat ional

11

Photovoltaic converters: challenges for the next decade

Prof. Johan Driesen K. U. Leuven Heverlee, Belgium

ABSTRACT

Technological challenges Novel components and higher switching frequencies The vast majority of power electronic circuits in use today convert electrical power from one form to another through modulation involving switching. Such circuits also contain a considerable amount of passive c o m p o n e n t s s u ch a s i n d u c t o r s , capacitors, performing filtering, galvanic separation or energy buffering functions, as well as interconnectors.

Additionally, (digital) electronics provide control at different levels ranging from switching timing to high-level energy flow management and active protection.

Switches in use today are Si-based components such as power-MosFets of IGBTs. Novel switching components are under development; however it seems that it will be some time before they will become suitable for mass production.

These switches will be based on wide band-gap semiconductors such as SiC or GaN. SiC-based diodes are readily available, having been experimented with for quite a few years, but they still prove difficult to produce. GaN should be easier to process, but is still evolving towards relatively low switching frequencies from its original application domain of high-frequency telecommunications.

The advantages of these new components are that they are more efficient (lower losses in switching as well as conduction losses); they sustain elevated temperatures and are able to switch at considerably higher frequencies; and they are more abundant in different ranges up kHz to the MHz range, compared to the currently common lower kHz range.

Nevertheless, this increased switching 12

w ww. pv - t e ch . o r g

frequency (and its harmonics) can be regarded as a disadvantage as well, since electromagnetic compatibility problems will increase. Design of the inverter circuits will have to account for the radiative effects causing disturbances and additional losses.

Whereas power electronic circuit design has been using analogue circuit techniques, high frequency or even microwave design knowledge will now have to be applied.

The impetus to aim for still-higher frequencies is that any volume and weight reduction linked to the reduction of the characteristic values of the required inductors and capacitors would mean an easier life for many people.

Imagine thin, lightweight inverters softly buzzing in cabinets hanging somewhere on the wall, instead of the current heavy, bulky and cumbersome inverters.

An illustration of this effect is the size difference between a 50Hz system and a high-frequency galvanic separation transformer. High-frequency inductors are built around core materials such as ferrites, which will need further development when the move to the M Hz range takes place. Classical conductors can no longer be used due to troublesome skin effects.

The temperature and reliability challenge Current power electronics technology and PV inverters in particular can be developed further still, even with Si-based components and kHzrange switching frequencies. In green power conversion, the overall system efficiency of the panels and associated control algorithms including maximum power point tracking (M PPT) are paramount. The role of the inverter is often underestimated in this sense,

Courtesy: Prof. Johan Driesen.

Power generation is a rapidly changing process. Owing to evolutions in power electronics, sustainable electricity generation and consumption came to the fore and now it is nearly impossible for photovoltaics to operate without this technology. This holds true for efficient consumption such as plug-in electric and hybrid vehicles or compact efficient lighting. Power electronics need to be taken into account in relation to grids, for example in novel voltage-source HVDC connections. Photovoltaic energy conversion requires power electronics in order to adapt the floating DC-output to a fixed DC-level and typically further to a grid-compatible AC electricity. These converter (mainly inverter) technologies have evolved considerably over the past few years, in much the same way as has PV cell technology, but in a much less apparent fashion. It is, however, expected and required that the technologies will evolve even further to meet the demands of the future market and the electricity grid to which they will be connected. This article intends to give an overview of the challenges ahead for power electronics in photovoltaic energy conversion.

Figure 1. Inverters installed in the shadow of a PV array.

as traditionally this circuit is optimised alone and rated on performance, rather than for the entire conversion chain with realistic influx. Experience from other application domains (mainly electric drives) has shown that there is still a quite a bit of room for improvement, for instance at partial load performance.

Rated performances were typically not mentioned and weighted efficiencies composed of measured values at different conversion levels rarely used.

Nevertheless, in grid-connected systems, the changing grid conditions have an effect on voltage levels, which can vary between ¹10%, and can lead to voltage harmonics and imbalances and neutral conductor leakage for three-phase systems. It has been observed that these power quality issues can affect the inverter’s internal losses and decrease the efficiency by several percent. Additional functionalities such as reactive current injection also come at a price, and can lead to a need for enhanced efficiency validation techniques.

In an attempt to decrease losses, enhanced switching techniques and low-loss passives are being researched.

Techniques to optimize the operational

Figure 2. Artist’s impression of the future Smart electricity Grid.

losses of the system can be further implemented. It is important to note that lower losses can help control the internal temperature and hence simplify cooling, yielding an indirect volume and weight reduction, as well as cooling fins and fans shrink.

The evolution towards wideband gap components is important as these components can sustain higher operating temperatures, possibly up to 200-300°C, it is said. In applications such as hybrid cars where elevated temperatures occur, this is a clear advantage. But is this feature necessarily an advantage for inverters in PV applications? Classical PV inverters can have a higher internal temperature level within the switching components.

With an unchanged outside temperature, higher temperature gradients exist, yielding smaller heat exchangers to evacuate the losses. This reduced volume advantage has a downside, however, as internal PV-inverter components should eventually work at a considerably elevated temperature.

Currently available passive components and circuit assembly techniques are not yet suited for these conditions. In fact, the characteristic values of the inductors and capacitors can heavily detune when exposed to a large temperature range, causing a totally different system performance and efficiency, making it even more difficult to keep up correct functioning. Obviously, in special PV applications such as CPV, the higher temperature conditions will be welcomed.

Keeping temperatures under control is not only an efficiency-related issue, but could more importantly be considered an inverter longevity issue.

As today’s PV modules can easily last a

20-year lifetime of operation, it makes sense to be able to expect the same lifespan from inverters. In practice, however, some installations require additional insurance to cover the cost of replacing the power electronics after a couple of years. Thanks to technology evolution, the new state-ofthe-art inverter will typically be a better performing one and hence energywise it will be an improvement. From a customer’s perspective, though, this is not an ideal situation as not everybody performs regular checks on the system’s state of operation.

Inverters, though statically mounted, operate in permanently changing external conditions invoked by the daynight and seasonal cycles. This reliability challenge occurs elsewhere in many different power electronics applications, and is certainly within the focus of research. However, there are not many small-scale implementations of switching power electronics – especially not on the electricity consumption side – that are foreseen to be in quasi-continuous operation for several decades as is the case in renewable electricity generation.

When these technological evolutions are all brought together, perhaps a bright future for efficient ‘hot’ panel integrated modular inverters will come ab out. Obviously, the PV-module technology will evolve as well, imposing new challenges upon the power electronic inverter. CPV brings in power at a different scale, while multi-layer cells working in different conditions perhaps require a more enhanced MPPT. The characteristics of organic cells can be completely different from semiconductor cells, so why should the inverter be the same?

Source: European Technology Platform Smart Grids [1].

Changing role in the Smart Grid Additional functionalities for the PV inverter After discussing the ‘internal’ evolution of the PV inverter, one would almost forget that next to the PV-modules, the output side of the inverter is also changing spectacularly. Several trends force the electricity grid to rethink itself. In the frame of this article, one is inclined to think that only the move towards a more decentralised sustainable electricity production is driving this, but one should not underestimate the impact of the implementation of liberalised energy markets that demand more flexibility of the system. The consumption of electricity will increase in the coming time, despite the ever-more efficient power consumption. One of the drivers of this growth is the move towards cleaner energy forms, examples of which include electrothermal processes instead of gas combustion, or new loads such as heat pumps, charging electric vehicles, etc.

In future, the electricity system will see co-existence of large plants of a new generation, such as clean coal units, with decentralized generators and next-generation loads, probably augmented with energy storage systems. This sketch of the future poses many different technical and operational challenges to the electricity system, which will be built around an intelligently operating “Smart Grid” [1]. Making all of this work together at an affordable price is perhaps the biggest challenge of all.

For grid-coupled photovoltaic inverter systems, this synergy will have important consequences. In the past, when the Phot ov olt aic s I nt ernat ional

13

Towards aggregation of systems Finally, a higher level of control will have to be added on top of all these changes inside individual converters.

The current PV converters are in general undispatched, meaning there is no supervisory adjustment to the operation from a control centre; hence they inject power when there is solar input. Since the level of sunshine is geographically dependent in an area the size of a typical distribution grid section, the injection can be massive with destabilizing 14

w ww. pv - t e ch . o r g

Courtesy: Prof. Johan Driesen – EU-FP6-VSYNC project, [2].

penetration of such systems on the grid was low, they did not have much of an effect on the overall workability as they silently injected power without a care for voltage stability and the related reactive power exchange. The extent of the required conservative safety principles consisted of a “shut down and back off in case of an incident” philosophy. The evolution of wind power and its impact on the transmission grid in many European countries has shown that the ever-increasing share of decentralised generation systems, mostly with power electronic interfaces, should obey a more limiting grid code as soon as a critical level is reached.

They are also asked to provide grid support by delivering so-called “ancillary services”, such as stabilizing the voltage by injecting reactive power, ridethrough incidents, etc. Unfortunately, determining that critical level proves to be an extremely difficult exercise.

PV systems are often more dispersed and connected to the distribution system, which is totally different in nature from the meshed, high-voltage transmission grid. Since the distribution feeders are radial and mainly made of cables, keeping the voltage profile under control is a more complicated challenge. This necessary form of grid support can only be successful when additional functionalities are added to the PV inverters. Ride-through of transients such as voltage dips is a minimum requirement; contribution to voltage profile stabilisation is due to follow soon.

To implement this, the converter technology will have to be adapted up to a certain point. The changes to be applied are not dramatic: firstly, the power (current) rating of the gridconnected front-end will have to be increased to allow for the additional reactive currents and short-term enlarged injections. The power required to do this will have to come from internal energy storage, mainly implemented with supercapacitors and possibly batteries such as Li-types or hybrid storages.

Using the latter has additional benefits in terms of power smoothing, for example.

However, it should be noted that storage integration is still a technical challenge and a control challenge in delivering balance (long-term) and stability by creating “virtual inertia” (short-term) [2].

Figure 3. Internal view of a power electronic system coupling a distributed energy resource and storage units to the grid.

consequences when the local loads do not pick this up and the power cannot be exported to the rest of the grid. As such, a partial curtailment of solar electricity production will sometimes be unavoidable. Obviously this should stay compatible with the liberalized market and be an exception to the rule to as great an extent as possible, but these actions can be beneficial as it is a remunerable grid support service.

The coordination of solar power injection should not stay limited to emergency measures. In general, it is a good idea to aggregate distributed energy resources, including active loads such as remotely adjustable heat pumps or charging electric vehicles.

When a good portfolio is gathered, the aggregated energy balance is smoothed, becomes more predictable and may even be adjustable. Such a joint configuration can be considered a ‘Virtual Power Plant’ (VPP). With such a tool in hand, true market participation will become a reality. To make this possible, the smart grid will need dependable intelligence and communication. Within this setting, the aforementioned storage may also play a role in making the power injections controllable and time shiftable.

Conclusion Power electronics as implemented in photovoltaics will evolve significantly over the coming years. On the one hand, novel components force a rethink of the entire circuit and its components.

Classical systems need to enhance their reliability. On the other hand, additional control features have to be implemented in order to stay grid-compatible (and market-compatible). But perhaps the most important challenge has not yet been mentioned – the challenge of keeping this evolution affordable. The massive deployment of the technology scale-advantages will probably keep the price low, but is there actually an alternative?

References [1] European Technology Platform’s Smart Grids [available online at http://www.smartgrids.eu].

[2] “Virtual Inertia” research project [available online at http://www. vsync.eu].

About the Author Johan Driesen received his M.S. degree in electrotechnical engineering from the K.U. Leuven, Belgium in 1996. His Ph.D. degree, also from the K.U. Leuven, focussed on the finite element solution of coupled thermal-electromagnetic problems and related applications in electrical machines and drives, microsystems and power quality issues.

Currently, he works as an associate professor and teaches power electronics and drives at the K.U. Leuven. From 2000-2001 he was a visiting researcher in the Imperial College of Science, Technology and Medicine, London, after which he worked at the University of California, Berkeley, USA. His current area of research is in distributed energy resources, including renewable energy systems, power electronics and its applications in drives, electrical transportation and power quality.

Enquiries Prof. Johan Driesen K.U. Leuven Department of Electrical Engineering, Research Group Electrical Energy ESAT-ELECTA, Kasteelpark Arenberg 10 B-3001 Heverlee, Belgium Email: johan.driesen@esat.kuleuven.be Website: www.esat.kuleuven.be/electa

Like the look of this article?

This first appeared in Photovoltaics International’s third edition.

With an average of 20 technical papers per edition, the quarterly journal is available to subscribers – contact Carlos (details on page 1).

Transparent conducting oxides for advanced photovoltaic applications

John D. Perkins & David S. Ginley National Renewable Energy Laboratory Golden, Colorado, USA

ABSTRACT

Although simple in concept, a photovoltaic solar cell is a difficult feat of technology in execution. The challenge of balancing cell structure design, material optimization and module technology to achieve efficient, low-cost modules that perform in aggressive environments for up to a generation is huge. The modules’ structure has to support and protect a thin, fragile slice of semiconductor, while ensuring a stable environment free from contamination and moisture with little or no change in the incident light on the cell.

Key to the modules’ performance are the first-level polymeric materials that contact the cell and conductor structures, hold the module together, and in many cases form the second-level protection of the cells from the environment. In this article we explore the industry dynamics in the supply of advanced materials for module assembly, the new technology directions, and how the market will develop over the next five years.

Introduction Transparent conductors in the form of transparent conducting oxides (TCOs) are critical components in many optoelectronic applications including flat panel displays (FPD) and photovoltaics as well as organic electronics including both organic light-emitting diodes (OLE D) and organic photovoltaics (OPV) [1,2]. Unlike most metals which are opaque and most transparent materials, which are insulating, TCOs are a special class of wide band gap (~ 3eV) metal oxide semiconductors such as ZnO, SnO2 and In2O3, which can support high enough free electron concentrations (~10 21 /cm 3 ) to be effective electrical conductors [3].

Typical good transparent conductors have conductivities of 1000 – 5000S/ cm and are transparent from ~ 350 – 1500nm (thereby including the visible portion of the spectrum, 400-700nm). For comparison, we note that copper metal is about 100 times more conductive than a typical TCO.

For single-junction PV applications, transparency out to about 850nm is a requirement, which does not put too much of a constraint on TCO materials.

TCOs in photovoltaics In PV applications, transparent conductors are needed as a contact for collection of the photo-generated carriers while still allowing the light to reach the active solar absorber m a t e r i a l . A s s u ch , t r a n s p a r e n t conducting oxides are part of every thin-film photovoltaic technology. The two main PV areas where this is not the case are epitaxial multi-junction III-V solar cells (e.g. GaInP2/GaAs) [4] and conventional bulk/polycrystalline Si cells. In these cases, a thin doped top layer of either III-V semiconductor for the former or Si for the latter can provide satisfactory lateral current transport when used in conjunction with a metallic current collection grid.

This latter approach is even being integrated with a TCO as discussed later.

Moving onto PV technologies that use TC O contacts, the image on the left of Figure 1 shows the basic structure of a two-sided Sanyo HIT cell (Heterojunction with Intrinsic Thin-layer, [5]). Note the two TCO layers, one for the top and one for the bottom. The Sanyo-HIT cell is an Si Heterojunction (SHJ) structure which uses thin hydrogenated amorphous Si (a-Si:H) layers in conjunction with a crystalline Si core [6]. The image on the right in Figure 1 shows a more expanded view of a one-sided SHJ structure. Note that even for this structure, there are still two TCO layers. ZnO is part of the top of the TCO/metal multilayer top contact and TCO -coated glass forms the transparent bottom contact for this device. In these SHJ structures, the a-Si layers effectively passivate the c-Si surface, but, due to the low carrier mobility in a-Si, a TCO layer is needed for lateral current conduction to avoid resistive losses. Indium-tin-oxide is typically used in this application [6].

Figure 1. Configurations for Si heterojunction (SHJ) cells with Sanyo HIT cell on the left.

Phot ov olt aic s I nt ernat ional

15

The upper half of Figure 2 shows a cross-sectional image and schematic for a Cu-In-Ga-Se (CIGS) thin-film solar cell. The primary absorber layer is the CIGS, which has the CuInSe2 structure, but with Ga partially substituted for In to optimize the absorber band gap energy.

CdS is deposited over the CIGS to form the junction and then a TCO top layer is used as the transparent electrical contact. In the highest efficiency CIGS cells, a TCO bilayer composed of first undoped ZnO deposited directly onto the CdS layer and then a high conductivity Al-doped ZnO (AZO) on top provides the lateral current conduction [7]. The TCO layer must be deposited at ~ 200° C or lower to avoid degrading the CIGS/ CdS junction. Similarly, the lower half of Figure 2 shows a cross-sectional image and schematic for a CdTe/CdS solar cell.

In contrast to the CIGS device where the TCO layer is the last layer deposited, the CdTe layer stack begins with TCO coated glass that then becomes the substrate for the subsequent CdS, CdTe and metallic back-contact layers. This allows higher growth temperatures to be used during the TCO layer growth and the F-doped SnO2 or bilayer Cd2SnO4/ Zn 2 SnO 4 are the standard TCOs for CdTe cells [8,9].

TCO materials and thinfilm growth Figure 3 depicts the conventional five basis set of metal oxides (SnO 2, In 2 O 3 , ZnO, CdO and Ga 2 O 3 ) which form the majority of crystalline TCOs when appropriately doped. Sn-doped In2O3 (ITO), Al-doped ZnO (AZO) and F-doped SnO 2 (FTO) are by far the most commonly used TCO materials at present. CdO can have extraordinarily good electrical properties if neither the yellow colour due to lower band gap energy nor the toxicity of Cd preclude its use. On its own, Ga 2 O 3 is not conducting enough to be a practical TCO but it is included as a basis oxide for its role in more complex TCOs such as GaInO 3 [10]. Similar binary metal TCO compounds with simple stoichiometries, such as those that lie in the composition space depicted in Figure 3, include Cd2SnO4, Zn 2 SnO 4 , and In 2 Zn 2 O 5 [10,11].

When TCOs formed from three basis oxides are considered, broad ranges of compositions and structures become possible. For examples, in the Cd-InSn-O system, Cd 1+x In 2-2x Sn x O 4 over the range 0 < x < 0.75 can be made [12]. The range of possible compositions for TCOs is enormous, covering 75% of the tie line between CdIn2O4 (x = 0) and Cd2SnO4 (x=1).

With the exception of CdO, which is light yellow in colour, all of these materials just discussed are inherently transparent in the visible when made undoped and fully oxidized. To achieve practical electrical conductivities 16

w ww. pv - t e ch . o r g

o f 1 0 0 0 S / c m o r g r e a t e r, t h e s e TCO host matrix materials must be effectively doped to create free carrier concentrations on the order of 1020 – 1021/cm3. The simple single basis oxide materials are almost always made with explicit substitutional doping. Both ZnO and In 2 O 3 are usually doped on the cation site with Al to give Zn1-xAlxO and Sn to give In2-xSnxO3 respectively [13].

SnO 2, on the other hand, is generally doped on the anion site using fluorine to give SnO2-xFx but it can also be doped on the cation site using Sb [14].

The range of possible compositions for TCOs is enormous, covering 75% of the tie line between CdIn2O4 (x = 0) and Cd2SnO4 (x=1).

For Al-doped ZnO and Sn-doped In2O3, sputtering is the most common method of thin-film deposition. High quality thin films can be grown either by sputtering from ceramic metal oxide targets or by reactive sputtering from metal alloy targets. Other physical vapour deposition (PVD) methods

used to deposit these and other TCO materials include evaporation and pulsed laser deposition, a technique which is excellent for proto-typing new materials even if it is not yet practical for large-area commercial deposition.

In contrast, SnO2 is generally deposited using spray pyrolysis or various forms of chemical vapour deposition (CVD).

Spray pyrolysis deposited SnO 2:F has been a mainstay of the TCO industry for decades, especially for low-e windows and IR selective windows. We note that PVD-deposited SnO2 is generally less conducting for reasons that are not fully understood at present [14,15].

Basic opto-electronic properties Figure 4 shows the optical reflection, transmission and absorption spectra for a typical commercial ZnO TCO on glass. Collectively, these show the key spectral features of a TCO material.

First, the material is quite transparent, ~ 80%, in the visible portion of the spectrum, 400 – 700nm. Across this spectral region where the sample is transparent, oscillations due to thinfilm interference effects can be seen in both the transmission and reflection spectra. The short wavelength cut off in the transmission at ~ 300nm is due to the fundamental band gap excitation from the valence band to the conduction band of the basis semiconductor, ZnO in this case.

Figure 2. CIGS (top) and CdTe (bottom) PV structures in cross-section.

Figure 3. Composition space for conventional TCO materials.

Figure 4. Optical spectra of typical (ZnO) transparent conductor.

The gradual long wavelength decrease in the transmission starting at ~ 1000nm and the corresponding increase in the reflection starting at ~ 1500nm are due to oscillations of the conduction band electrons known as plasma oscillations, or plasmons for short. The corresponding schematic electronic structure for a heavily doped semiconductor with a completely filled lower valence band and significant free electron density in the upper conduction band states is shown in Figure 5.

What distinguishes TCOs from conventional semiconductors is that the valence band to conduction band (band gap) energy is very large, 3eV or more, which makes TCO materials transparent in the visible spectrum. Furthermore, TCO materials allow for conduction band free carrier densities as high as 10 21 electrons/cm 3 , which enables the high conductivities.

Returning to the optical spectra in Figure 5, there can also be substantial absorption due to these plasma oscillations as is the case for this particular sample with the maximum absorption occurring at the characteristic plasma wavelength, λ p. As the number of electrons in the conduction band, N, is increased, such as by substitutional doping, the plasma wavelength shifts to shorter wavelengths as λ p ∞ 1/√N which also effects the electrical conductivity (σ) since σ= Neμ where e is the electron charge and μ the electron mobility [16]. Hence, one fundamental reality of TCO materials is that there is an inherent tradeoff between conductivity and the long wavelength transparency limit. At very high electron concentrations, this can even decrease the visible wavelength transparency.

Phot ov olt aic s I nt ernat ional

17

Figure 5. Schematic electronic structure of TCO materials.

For example, Figure 6 shows how the infrared transparency increases f o r S n O 2 TC O s a s t h e s h e e t resistance is increased from 5Ω/ sq. to 100Ω/sq. Even though both of these SnO 2 samples have similar visible wavelength transparency, the 5Ω/sq. sample would be unusable as a transparent conductor for telecom applications at 1500nm.

TCO optimization includes not just the inherent trade-off between conductivity and transparency just discussed, but also many other application-specific constraints such as chemical compatibility, required deposition temperature, stability at the operating temperature and surface roughness, among others. Collectively, these examples should make it clear that there is no such thing as a single ‘best’ TCO and that TCOs must be tailored to the constraints of the specific application. This includes not just the broad distinction between TCOs for flat panel displays, PV and telecom, but also the distinct TCO requirements for the different thinfilm PV technologies. This ‘set’ of requirements is what is driving the development of new TCO materials in a number of application areas.

Figure 6. Optical transmission spectra of SnO2 TCOs with different sheet resistances.

crystalline ITO with 10wt.% SnO 2 in In2O3 deposited at 250 – 350°C which typically has electron mobilities of ~ 30cm2/V-sec.

Figure 7 shows the electrical conductivity, carrier concentration and electron mobility for Ti-doped In 2 O 3 on glass grown by sputtering

[17]. Of particular note are the high mobilities, > 80 cm2/V-sec, observed for films with Ti concentration of 1-2%. This is attributed primarily to the high doping efficiency of Ti in In2O3, which is found to be near unity for Ti concentrations near 2%. For comparison, in most ITO the doping

Recent materials developments Three recent trends in new TCO m a t e r i a l s r e s e a r ch a r e : 1 ) Th e development of high electron mobility materials; 2) The use of amorphous mixed metal oxide TCOs which can be deposited at low or even ambient temperature, and 3) The discovery of TiO 2 -based TCOs. The practical industry standard reference point for this new materials development is 18

w ww. pv - t e ch . o r g

Figure 7. Electrical conductivity (σ), carrier concentration (N) and mobility (μ) of Ti-doped In2O3.

easily deposited onto glass or plastic substrates by ambient temperature sputtering, have excellent optical and electrical properties, are very smooth (R RMS < 0.5nm) and once deposited, do not crystallize until heated to 500°C or higher [22]. Figure 8 shows the electrical conductivity, carrier concentration and electron mobility as a function of metals composition for IZO films grown by composition gradient combinatorial sputtering [23]. Note that broad conductivity maximum with σ ≈ 3000S/cm occurs in the composition range which is amorphous, ~ 55 to 80 at. % indium in this case. While for transparent conductor applications, a high carrier concentration is needed, we note that when grown in the presence of 5 – 10% oxygen, these amorphous mixed metal oxides have carrier concentrations of order 1016 – 1018/ cm 3 and are being developed as channel layers for transparent thinfilm transistors (TTFTs) [24,25]. At present, amorphous In-Ga-Zn-O is the primary material of interest for TTFT applications.

Figure 8. Electrical conductivity (σ), carrier concentration (N) and mobility (μ) of In-Zn-O.

Figure 9. Schematic cross-section of an organic photovoltaic (OPV).

efficiency of Sn is about 1 electron per 4 Sn atoms. Mo-doped In2O3 is also a high mobility variant on doped indium oxide with mobilities of 60cm2/V-sec for sputtered films on glass [18] and as high as 125cm 2 /V-sec for films deposited by PLD onto single crystal substrates [19]. Interestingly, in this case the doping efficiency is less than half of that for the Ti. While traditional TCOs are highly crystalline, recently, a new class of amorphous TCO materials, typified by amorphous In-Zn- O (a-I ZO) and based on double (or triple)

oxides of heavy metal cations with ionic electronic configuration (n-1) d 10 ns 0 has emerged [20,21]. These materials typically exhibit an electron mobility of 30–60cm 2 /V-sec, which is unusually high and is thought to arise from the direct spatial overlap of the large and spherical heavy metal cation ns0 orbitals and the lack of grain boundaries.

For comparison, typical thin-film amorphous Si has mobilities of less than 1cm 2 /V-sec. Technologically, a-I ZO and similar materials are of great interest because they can be

One fundamental reality of TCO materials is that there is an inherent tradeoff between conductivity and the long wavelength transparency limit.

Recently, Nb and Ta-doped TiO2 have been shown to be good transparent conductors with conductivities of ~ 3000S/cm on single crystal substrates and ~ 2000S/cm on glass [26-28].

The highest conductivity has been observed for the Nb-doped materials and some recent theoretical results support this [29]. In a simple electron counting model for TiO2, the Ti4+ has no electrons in the 3d states, which contradicts the conventional wisdom that metal oxide materials for TCOs should always have filled d-shell states. To obtain these good TCO properties, the TiO 2 must have the anatase structure perhaps due to a much lower effective mass than rutile.

The more common rutile TiO2 is not a good TCO. At present, it is not known whether anatase TiO 2 is the first member of a new general class of TCOs or a singular exception to the filled d-states rule.

Phot ov olt aic s I nt ernat ional

19

industry standard with Al-doped ZnO and F-doped SnO 2 also being used commercially. However, due to factors such as cost, reactivity, deposition temperature and stability, these three materials are not sufficient to meet the TCO requirements for emerging high performance applications both in PV or other opto-electronic a p p l i c a t i o n s . C o n s e q u e n t i a l l y, new materials are being actively developed. A few important areas of current TCO materials research include compositionally complex binary, ternary and beyond materials, amorphous mixed metal oxide TCOs, early transition metal series TCOs and higher performance dopants. The past decade has seen a resurgence in TCO research and significant advances are likely in the next five years.

Figure 10. Optical absorption of Mg substituted ZnO. Inset: Effect of TCO band gap on OPV open circuit voltage.

Organic PV applications TCO materials are also critical in organic photovoltaics (OPV), where they can be used in a conventional geometry as shown in Figure 9 or in an inverted geometry where the oxide becomes the electron acceptor.

Clearly, these two configurations require very different characteristics for the TCO in terms of electron affinity, surface chemistry and even doping level. A typical layered structure for the conventional bulk heterojunction is shown in Figure 9 where the TCO acts as the hole extraction contact in conjunction with either an organic or inorganic hole transport layer (HTL).

In this application, reducing the energy level mismatch between the Fermi level of the TCO and the highest occupied molecular level (HOMO) of the organic semiconductor is believed to be central to improving the current collection and operating voltage.

In general, for TCOs used in typical bulk heterojunction OPV (Figure 9), higher work function TCOs are needed to improve the energy level matching [30].

A l t e r n a t i v e l y, TC O s o r s i m i l a r materials may also have applications as the electron acceptors in an inverted OPV device. In this case, lower work functions TCOs are needed to move the conduction band higher and thus increase the open circuit potential.

In these devices, due to the strong exciton binding energy and short exciton diffusion length in organic semiconductors, the photo-excited excitons (electron-hole pairs) can only be effectively split by an interfaceinduced charge separation, where 20

w ww. pv - t e ch . o r g

the electron is transferred from the organic absorber to the acceptor. In this case, TCOs with a higher band gap energy (lower work function) are desired to avoid voltage loss when the electron transfers from the lowest unoccupied molecular orbital (LUMO) of the organic to the TCO conduction band. Figure 10 shows both the band gap energy increase in ZnO (shorter wavelength absorption edge) due to substituting Mg in place of Zn and the corresponding increased open circuit voltage in a simple planer OPV device where Zn(Mg)O is used as the electron acceptor [31]. Overall, the substitution of Mg for Zn in ZnO decreases the band offset between the LUMO of the polymer donor and the conduction band of the acceptor metal oxide material.

As shown in the inset, the V OC was increased from 500 to 900mV in the P3HT–Zn 1–x Mg x O devices through the substitution of Mg into ZnO over a range of 0 to 30% Mg for Zn. However, to be effectively incorporated into a practical OPV device, this basic materials result would have to be transferable to an intercalated metaloxide/organic active layer structure with nanometre-length scales. This remains an active area of research.

Summary Thin film photovoltaics require an optically-transparent electricallyconducting contact layer to enable current extraction while allowing sunlight to reach the active PV junction. For most PV applications, this need is met with transparent conducting oxides that are heavily doped wide band gap semiconductors.

Crystalline Sn-doped In 2 O 3 is the

Acknowledgements This work was supported by the National Center for Photovoltaics (NCPV) at National Renewable Energy Laboratory through the U.S. Department of Energy under Contract No. DE-AC3699G010337.

References All references are available in the Third Edition of Photovoltaics International journal (p. 95).

About the Authors John Perkins is a Senior Scientist in the National Center for Photovoltaics at the National Renewable Energy Laboratory in Golden, Colorado, USA. He received his Ph.D. in physics in 1994 from MIT. His current research focuses on thin-film transparent conductors and combinatorial approaches to material science research.

David Ginley is a research fellow and group manager in the National Center for Photovoltaics at the National Renewable Energy Laboratory. He received his Ph.D. in inorganic chemistry in 1976 from MIT. Research interests include nanomaterials, transparent conductive oxides, organic photovoltaics, combinatorial materials science and new process technologies for solar energy conversion.

Enquiries John Perkins National Renewable Energy Laboratory 1617 Cole Blvd., Golden, CO 80401 USA Email: john.perkins@nrel.gov

Like the look of this article?

This first appeared in Photovoltaics International’s third edition.

With an average of 20 technical papers per edition, the quarterly journal is available to subscribers – contact Carlos (details on page 1).

Polymer development and selection criteria for thinfilm and crystalline-silicon module manufacturing

Mark Thirsk Linx Consulting LLC Mendon, MA, USA

ABSTRACT

Although simple in concept, a photovoltaic solar cell is a difficult feat of technology in execution. The challenge of balancing cell structure design, material optimization and module technology to achieve efficient, low-cost modules that perform in aggressive environments for up to a generation is huge. The modules’ structure has to support and protect a thin, fragile slice of semiconductor, while ensuring a stable environment free from contamination and moisture with little or no change in the incident light on the cell.

Key to the modules’ performance are the first-level polymeric materials that contact the cell and conductor structures, hold the module together, and in many cases form the second-level protection of the cells from the environment. In this article we explore the industry dynamics in the supply of advanced materials for module assembly, the new technology directions, and how the market will develop over the next five years.

Module manufacturing needs Modules are critical in protecting the various types of photovoltaic cells during the transport, mounting and life of the cells. With ever-decreasing thickness of the absorbers in the overall module build, the module is an integral part of the final electricity generating effort, and must be designed to protect the absorber while optimizing its efficiency. Furthermore, the module must meet the overall needs of cost, stability and weight required in the final installation.

Cell architectures vary in their choices for substrate material, and whether the substrate used for manufacture performs as a substrate or superstrate in the final module.

The market for cells in 2008 was still dominated by crystalline silicon cells of both the monocrystalline and multicrystalline wafer type. While there are multiple methods of creating these wafers (sawing, string ribbon growth, meniscus growth, cleaving, etc.), the methods used for these wafer-based modules is generally common in that a glass front panel provides the stability and the face to the sun, and the cells are embedded in layers of polymeric sheets to provide the cell insulation, protection and fixturing. Differences in

the cells’ interconnection from module to module derive mostly from the emitter architecture chosen by the cell maker, and whether the cells need to be contacted on both front and back, or just on the back. This article discusses this market segment as one despite these small differences in process and design, while focusing on the choices of the polymer films used.

The landscape of the thin-film segment is significantly different.

There are many thin-film absorbers in production, and many more being developed. Cell architectures vary in their choices for substrate material, and whether the substrate used for manufacture performs as a substrate or superstrate in the final module.

Additionally, the thin-film technology in question may be produced on a rigid substrate or on a flexible substrate – again, influencing the needs for any moduling materials. Finally, the thinfilm absorbers vary from being stable, passivated inorganic materials that are tolerant to moisture and temperature, to being sensitive organic materials that require high degrees of protection from damaging radiation, moisture, oxygen and other environmental factors.

To put the moduling challenge in sharper perspective it is worth bearing in mind that the vast majority of modules are expected to survive in very harsh environments with a minimum of maintenance. By definition, the modules will be in direct sunlight with many years of high intensity UV exposure. Added to this, the modules will generally be exposed to all the variables of weather throughout the year, including rain, high and low temperatures, hail, snow, wind, sand and

debris. Many modules are expected to withstand 20 to 25 years of exposure with precisely limited changes in performance detailed in the module delivery specification.

The only materials that regularly meet such stringent environmental challenges are ceramic (concrete, brick, tile, etc) or self-passivating metals (copper, aluminum, etc). Very few organic surfaces endure such high performance goals (perhaps with the exception of some fiberglass, bitumen, and a few speciality paints). A generic list of overall module requirements is given in Table 1.

C-Si modules The generic crystalline cell module uses a well-proven design and material set to encapsulate and protect the cells.

Cells typically consist of the following components and films: Physical support of the cells Mechanical protection Radiation protection Moisture protection Electrical insulation Oxygen protection Shock protection Sand and dirt protection Manufacturable Low cost Architecturally attractive Physical properties consistent with the cell (flexible or rigid) Table 1. Module performance requirements.

Phot ov olt aic s I nt ernat ional

21

Glass panel. Usually molded, structured low-iron soda-lime glass, sometimes tempered for strength. This glass is sometimes coated with an antireflecting layer, or a self-cleaning layer to improve the module performance.

Adhesive/encapsulant (front). An organic layer that adheres the glass to the front side of the cell, as well as physically encapsulating the cells and front-side interconnect metals. This film must be close to optically transparent at absorbed wavelengths, and must not degrade with exposure. Additionally, the layer must become liquid during module lamination to completely contact the cells.

Cells. Finished, tested wafers assembled into strings of cells with tinned copper strips and connections that will lead to the junction box.

Adhesive/encapsulant (back). An organic layer that adheres to the back side of the wafers and the backsheet material. It also physically encapsulates t h e s i l i c o n c e l l s a n d b a ck - s i d e interconnect metals.

Backsheet. A high-performance (and consequentially often high value) laminated polymer sheet that combines multiple properties. This layer must provide electrical insulation of the cells, withstand UV exposure from the exposed front side as well as the backside, and protect the cells from moisture and contaminants.

Backsheet Approximately 90% of C-Si modules use backsheets. The backsheet material choice is highly dependent on the final module performance expectations.

Typically the insulation is provided by polyethylene terepthalate (PET), and high breakdown voltages will be achieved with thicker PET films. As PET is sensitive to degradation by UV, modules designed with longer lifetimes will typically protect the PET with layers of fluoropolymer. The most commonly used fluoropolymer today is PVF (polyvinyl fluoride), although PVDF (polyvinylidene fluoride) is gaining in acceptance. In some cases the fluoropolymer is only applied to the back of the PET, but for increased protection it is applied to the front and back of the PET. Contrary to popular belief, the moisture protection of the PVF/PET/ PVF laminate is poor, and backsheet laminators offer the option of including an aluminum layer in the backsheet to reduce moisture ingress.

Variants of this relatively simple PVF/ PET/PVF sandwich are common. In some cases an adhesive layer may be included on the inside of the laminate to reduce the complexity of lay-up of the final module. PVF layers may be colored to offer better reflectivity, or more pleasing aesthetics. A notable exception in the global market to these backsheet lamination schemes is in 22

w ww. pv - t e ch . o r g

Japan, where many modules have been designed for shorter overall lifetimes.

These modules use several (2 to 3) layers of PET and do not commonly incorporate a fluoropolymer layer. To mitigate UV damage the outer layers of PET will often be of a differing composition, and include UV absorbing dyes or pigments to shield the inner layers from damage.

The backsheets are laminated by specialist manufacturers, and require adhesives to ensure good adhesion of each component sheet within the laminate. These laminates are typically epoxies, but other compositions such as acrylics have been evaluated.

Adhesive/encapsulant The most commonly used adhesive in C-Si module manufacture today is Ethyl Vinyl Acetate. EVA is a crosslinking elastomeric and is supplied as a partially crosslinked sheet that is rubbery and tacky in feel. The sheet is often supplied with a backing sheet, which is removed just prior to use. The elastomeric nature of the film allows conformal coating of the cells, a process that is completed in the curing operation and conducted in a vacuum laminator. Since the reaction occurring in the laminator is a crosslinking reaction that is time- and temperature-dependant, there is an inherent cycle time in the process that can be in the order of 10 to 15 minutes per module. For this reason, EVA curing of modules is often carried out on parallel laminators, or, increasingly, in stacked vacuum laminators which process multiple modules at a time.

A concern with the use of EVA is the outgassing of relatively corrosive byproducts on curing.

A concern with the use of EVA is the outgassing of relatively corrosive byproducts on curing. These byproducts can limit the life of the laminator diaphragms, and require increased pump maintenance.

An emerging alternative material for use as an adhesive is polyvinyl butyral (PVB), a thermoplastic. Having been used in thin-film glass-on-glass modules for some time, improved PVB materials are coming to market that show equivalence with EVA, without the requirement for long cure times. The PVB becomes liquid above the glass transition temperature, and care needs to be taken that the module is physically supported during lamination and cooling until the adhesive has solidified. PVB has the advantage of not needing the long cure, and can be applied in roll lamination.

Challenging these film-based solutions is the promotion of a recent development of silicone-based encapsulants. These materials are applied as a liquid and cured in situ, resulting in encapsulants that have excellent stability, high transparency, and good moisture blocking characteristics.

Module quality In the final analysis, the moduling approach for C-Si cells relies as much on proven technology as on the pros and cons of each technology, either established or novel. Accelerated testing of the finished modules is not perfectly predictive of their performance in use, and real-life testing is not an option for items with such long life expectancy.

Thus, there is always hesitancy in introducing new processes and replacing known quantities with proven track records.

Thin film The dominant moduling technology for thin-film absorbers is glass-on-glass. In this technology a glass sheet is used as the manufacturing substrate, and once complete, a second glass sheet (either a superstrate or substrate) is laminated to the completed cell on the glass sheet, providing front and back protection for the cell, and requiring only an edge seal to make the final module hermetic.

About 20% of thin-film modules use polymer-based encapsulants, either in the form of backsheets similar to c-Si technology, or as front-side protection for cells fabricated on substrates like steel as a transparent protection layer over the cell.

Glass on glass The most common approach to thinfilm modules uses a PVB film between two float glass sheets. The inherent strength and protection afforded by the two glass layers provides excellent protection for the absorber.

In fact, in two common embodiments (tandem junction silicon and CdTe modules) there is no specified frame, and in some cases no edge sealant is used. These simple modules are laminated with niprollers and cured in an autoclave. This technique has been demonstrated in very large module dimensions (up to 2.5m on a side), showing savings in support frames and Balance of Systems.

Backsheet on glass Dual glass sheets incur a penalty in weight, which may be unacceptable on rooftop applications. An alternative approach is to laminate the cell and glass front with a polymer backsheet.

On the face of it, this should not cause major concerns due to the generally unreactive nature of the semiconductor absorber; however, this technique is likely better suited to Si-based cells.

Figure 1. Moduling material demand (in millions of square metres).

Figure 2. Moduling material market forecast.

Frontsheet on glass or steel In the case of some CIS/CIGS designs, and some specific Si-based dual and triple junction designs, the front side of the absorber must be covered by a transparent encapsulant. Commonly used encapsulants are fluoropolymerbased films such as ETFE, which offer good environmental resistance, as well as flexibility in the case of cells built on steel.

Market outlook Linx Consulting, together with Alternative Energy Investing, have developed forecast models for the materials demand in the PV industry. Despite a poor prognosis for 2009, we anticipate that the cost competition in the down year will act to further spur growth in 2010 and beyond, and increased concern for the provision of alternative energy will drive subsidies and feed-in tariffs to support adoption of PV projects.

Although overall growth may not parallel that seen in the past few years, we still expect double-digit compound rates and strong demand for moduling sheets and materials from 2010 onwards.

Figure 1 shows a forecast for the various moduling materials in square

metres. The largest segment by module area is that of the c-Si adhesive, since each module uses two sheets. Figure 2 is a forecast of the market growth for the materials shown in Figure 1. While dominated by the cost of glass, the predictions still show a US$220 million polymer market growing to more than US$1 billion in 2013.

New challenges An underlying principle of the whole photovoltaic market is the continual need to reduce the cost per watt of power generation capability. Integral to this goal is the reduction of the purchase cost of modules. Since most of the components of the module are already based on materials that are largely commoditized (glass, aluminium, stainless steel, polymer sheet, etc.) the prices are largely market driven, and scale effects are mostly irrelevant. However, the cost of specialization of these materials will reduce as volumes increase, and the market prices will approach the commodity levels.

Despite this potential route to cost reduction, any competitor in this market must include a continuous,

relentless approach to cost reduction; at least until grid parity is reached to a broad extent, but also to ensure c ompet it iv eness ev en aft er t hi s juncture.

Product differentiation through technical performance is critical to developing improved module and cell efficiency, and thus the aggressive Levelized Cost of Energy targets needed to achieve grid parity must be met. Incremental improvements in technology will lead to cells and modules that are better suited to the segment needs, whether on-grid or off. These changes will take time to implement due to the inherent difficulty of knowing if they are durable for the intended life cycle of the modules, but the current module designs will likely not survive unchanged in the relentless quest for grid parity.

About the Author Mark Thirsk is Managing Partner of Linx Consulting. Mark has over 20 yearsâ&#x20AC;&#x2122; experience spanning many materials and processes in wafer fabrication. He has served on the SEMI Chemicals and Gases Manufacturers Group (CGMG) since 1999, acting as Chairman between 2001 and 2003.

Enquiries Email: mthirsk@linx-consulting.com Tel: +1 617 273 8837

Like the look of this article?

This first appeared in Photovoltaics Internationalâ&#x20AC;&#x2122;s fourth edition.

With an average of 20 technical papers per edition, the quarterly journal is available to subscribers â&#x20AC;&#x201C; contact Carlos (details on page 1).

Phot ov olt aic s I nt ernat ional

23

Downstream is up, upstream is down – what does the credit crisis mean for the PV industry?

Daniel Pohl EuPD Research 360Consult, Bonn, Germany

ABSTRACT

News of credit crunch woes filtering down the lines over the past few months has instilled a sense of frugality in all industry sectors. While the credit crisis has indeed affected the PV industry, German banks, investors and creditors have claimed that the financing of small PV systems in the private sector seems not to be endangered. Downstream players are still optimistic, but the upstream sector is anticipating severe damage as a result of the economic situation. An interactive workshop-style discussion, hosted by market researcher EuPD Research and the CleanTech consultancy 360Consult, invited top-level executives to contribute their experiences of the current financial situation, as discussed in this paper.

comment leads one to the assumption that it is mainly the upstream players – the manufacturers – that are going to be affected by the current situation, whereas the downstream players – such as installers or wholesalers – might get off cheaply. “According to our clients, investors and planning companies in particular in the league of megawatt plants are expecting difficulties when it comes to financing”, Wackerbeck states, adding that the same applies for all photovoltaic companies with ambitious expansion plans in 2009, and, as a result, high cash needs.

Where might the crisis lead?

Within the framework of a recently held workshop in Germany’s finance metropolis of Frankfurt on the Main, more than 80 top-level executives from the photovoltaics sector exchanged their thoughts on the issue of the outlook for the future of the industry. One main concern of the delegates from various

Source: BSW 02/2009* *BSW-Solar-Index, conducted by EuPD Research commissioned by BSW.

The effects of the international financial crisis clearly affect all industrial sectors around the globe, including greentech companies and the photovoltaics sector.

Whereas the segment of small systems for the private end-customer seems not to have been affected by the credit crunch, the cost-intensive segment of large-scale PV plants is expecting stormy weather.

“Due to a very comfortable situation in the German banking sector, financing of small systems for private end-customers should not be affected by these global turbulences in the first place. Besides the KfW (Kreditanstalt für Wiederaufbau), comparable to a state bank who [sic] is granting private loans for reconstruction and modernization in the housing sector, the Umweltbank as well as the cooperative banks have not withdrawn funds to finance these projects”, explained Markus Wackerbeck, Senior Consultant at the German cleantech consultancy 360Consult, part of Hoehner Research & Consulting Group. This

‘Downstream is up, upstream is down’

Figure 1. German module and system prices point downward.

24

nations was the potential for difficulties emerging in taking up dept capital in the forefront of project planning. Due to the current financial situation, venture capital is becoming more and more expensive.

But on the other hand, the crisis may even lead to desperately needed market consolidation, which might turn out to be very healthy for the PV industry as a whole. Other factors that need consideration in this discussion are the market stimulation packages that are being adopted by governments all across the globe. “These enormous financial aids granted by most international administrations will clearly show a positive effect on the whole cleantech industry, including the PV sector”, Markus Wackerbeck points out.

In addition to these ad hoc measures, the solar industry benefits from longterm promotion schemes such as feed-in-tariffs (FIT), which should keep the demand from plummeting in the short term. “Nevertheless, our consultants presume that the recession at hand clearly has a negative impact on purchase power and investment abilities as well as the energy demand in general.

In the scope of an international downturn, the overall energy demand is going to decrease and the whole climate debate could partly become a luxury problem compared to accumulated insolvencies, unemployment and political instability”, warns Wackerbeck.

w ww. pv - t e ch . o r g

Based on recent surveys conducted by market researcher EuPD Research, 360Consult concludes that downstream players such as wholesalers and installers will profit slightly from the upcoming changes in the market. While in the past photovoltaic products and

Source: EuPD Research 01/2009.

modules in Germany, one of the leading PV markets worldwide, showed a neardouble-digit fall. Consequently, investment in PV installations has become increasingly attractive for private investors – the end of the year might even see the surviving downstream players stronger than they were before this global issue.

Point of view: different perceptions of the crisis

Source: PEREP Analytics.

Figure 2. Different perceptions of the crisis.

Source: EuPD Research 01/2009.

Figure 3. Financing from venture capital – indicators point downward.

Figure 4. Responses to capital procurement concerns, gathered from various sectors of the PV industry.

PV components have practically sold themselves due to robust subsidies and incentives as well as a level of demand far above the available supply, the future will bring more challenges and competition.

“The oversupply scenario has come true”, states market expert Markus A.W. Hoehner, CEO of Hoehner Research & Consulting Group, which has been actively surveying the global energy markets for nearly a decade now. “The industry finds itself in the considerable shadow of the global financial crisis, and a worldwide recession is hindering the investment plans of some upstream

companies. According to our analysts from 360Consult, the industry is facing a time of thorough upheaval unlike anything it has faced in the past,” said Mr. Hoehner.

Such a hindrance for the producers may come in handy for the installers and wholesalers and, last but not least, for the end customer. Within the ‘Price Index for PV Modules’, conducted by EuPD Research on behalf of the German solar industry association BSW-Solar, the market experts from Bonn are clearly spotting a significant decline in prices, illustrated in Figure 2. Compared to the previous quarter, prices for solar

The perception of the PV manufacturers of the crisis differs from the perception of the installers and wholesalers (see Figure 3). While producers and equipment vendors in the field of crystalline and thin-film modules can expect mostly negative capital procurement outcomes from the financial crisis, the downstream players – German installers in particular – tend to lean more towards a neutral or positive assessment – a tendency Markus Wackerbeck strongly supports.

“Such an economic turndown as this we are currently facing also provides the opportunity to boost the transformation process from a seller’s into a buyer’s market – in the long-term, a really positive development.”

About the Author Daniel Pohl graduated with an M.A. in North American studies, literature and political science from the University of Bonn and the University ParisSorbonne. He has been working as an editor and media consultant in the field of economics and renewable energies and is now heading the corporate communications department at EuPD Research and 360Consult in Bonn.

Throughout his career, he has published numerous articles on diverse energy topics in national and international special-interest magazines. He has also worked for national newspapers, broadcasting stations and a TV production company.

Enquiries EuPD Research | 360Consult Adenauerallee 134, Bonn D 53113 Germany Email: d.pohl@eupd-research.com Websites: w w w. e u p d - r e s e a r c h . c o m www.360consult.com

Like the look of this article?

This first appeared in Photovoltaics International’s third edition.

With an average of 20 technical papers per edition, the quarterly journal is available to subscribers – contact Carlos (details on page 1).

Phot ov olt aic s I nt ernat ional

25

TRITEC represents highest quality for photovoltaic systems of all sizes TRITEC International AG The international head office of TRITEC International AG is located in Basle. For 22 years, TRITEC has been an international photovoltaic system supplier and centre of competence in the area of renewable energies. All components of photovoltaic systems, be they intended for grid connection or as stand-alone systems, must be perfectly coordinated in order to achieve optimal efficiency.

Therefore, TRITEC advises specialists during the planning, design and implementation of photovoltaic systems.

Besides being a wholesaler, TRITEC is also a system integrator and manufacturer of proprietary products.

As a system integrator located in Aarberg close to Bern in Switzerland, TRITEC implements large international photovoltaic projects. One of the major projects of the past few years was the world’s largest roof-integrated solar power plant on a football stadium – the stade de Suisse in Bern, Switzerland – with an installed power of 1.35MWp, as shown above.

TRI-SEN (see below), the irradiation and temperature sensor, measures cell temperature and irradiation continuously, with an accuracy of ±2%. Throughout this process, there is no need for the solar cells to be handled.

Information is transmitted via infrared to the characteristic curve analyzer (TRI-KA). The results are immediately visible on the clearly arranged display, and thus faults, malfunctions or output deficits can be recognized instantly in the strings. Up to 125 non-transient measured curves remain stored in the TRI-KA. Thanks to specially developed software, the measurements can be read on a PC via USB interface.

As a manufacturer, TRITEC is active in the Measuring Technology area. TRITEC’s own-brand products include the characteristic curve-measuring instrument TRI-KA and the irradiation sensor Spektron.

TRITEC pays great attention to the quality of their products, and their portfolio only contains A-brands from established manufacturers guaranteeing the highest quality. On the one hand, for TRITEC, clean energy and sustainability belong together, and on the other, systems generating solar energy need to be capable of being exposed to the harshest weather conditions for many years.

Performance control of photovoltaic systems with the solar measuring instrument TRI-KA TRI-KA and TRI-SEN Controlled solar energy: how efficient is your photovoltaic system?

Using the characteristic curve-measuring instrument TRI-KA (see right), a qualitative analysis of the output of photovoltaic systems is easily available.

The measuring process of the TRI-KA, coupled with the TRI-SEN, allows measurement of the output, short-circuit current and off-load voltage directly at the solar system with the touch of a button. With a total weight of only 650g and operating software available in five languages, the TRI-KA simplifies the system performance control for specialists and laypersons alike.

26

w ww. pv - t e ch . o r g

TRITEC’s origins can be found in the implementation of the first megawatt photovoltaic project in Switzerland in 1987. Today, TRITEC is the leading photovoltaics company in Switzerland with its international head office in Basle, and also has company bases in Switzerland, Germany, France and Spain as well as partner companies across Europe.

KUKA Systems – recognizing and utilizing synergy effects KUKA Systems successfully transfers technology from the automotive industry to photovoltaic production KUKA Systems offers its customers comprehensive automation solutions in the field of systems engineering.

With automotive manufacturers and their suppliers, KUKA has for many years now been the market and technology leader, with a market share of around 25%. In this field, KUKA Systems concentrates on the engineering, supply and commissioning of flexible systems for vehicle body production – on which a number of different models or model variants can be produced – as well as on its core competencies in robot-based technologies such as welding, cutting, joining, forming, press linking, bonding and sealing, together with logistics and transfer technology.

KUKA Systems is now putting this know-how to use in other industries, where the level of automation is considerably lower. Given the scarcity of raw materials and the current discussions about climate change, KUKA Systems is strongly committed to the solar industry, as here it can use its automation expertise to streamline production at every stage of the value creation chain.

Just as for customers in the automotive industry, systems for the solar sector are also planned first as

28

w ww. pv - t e ch . o r g

a “virtual factory”, in which the viability is tested digitally before the system is implemented. In this process (see Figure 1), all production sequences are simulated and finetuned until the maximum possible technical availability, the optimal output and best quality are achieved. The KUKA engineers set themselves high standards, and achieve a technical availability of over 90%.

The use of 3D layouts and animations allows maximum planning detail for the entire logistics section. This enables systems to be integrated into existing production shops with the best possible fit and the sequence of operations to be optimized.

A material flow simulation lasting over 30 days identifies bottlenecks and serves as the basis for a buffer strategy.

Furthermore, concepts can be developed in advance with regard to redundancy, the separation of individual processes, and buffers.

Figure 1. Application of vapour sealant tape on thin-film modules using the newest evolution of the KS Tapetec Series.

Figure 2. Application of flange tape for sealing preparation on an automotive body in white using the KS Tapetec Series.

As in the automotive industry, value creation can also be increased in the production and assembly of solar modules by using robots to combine complex tasks. In this way, a number of operations can be performed in direct sequence using an industrial robot and suitable peripheral equipment (see Figure 2), leading to increased efficiency and reduced costs.

The greatest advantage of automation is the constant high level of quality in production. The use of automated processes reduces the extent of the manual work being carried out on the module, enabling a high degree of reproducibility with regard to the required quality standards.

One example of this is the ROBO FRAME module from KUKA Systems, which uses an industrial robot with a high payload capacity for the automatic framing of solar modules. The main advantages of this automated process are the prevention of deformation and scratching, improved quality of the end product, and greater throughput of the production system as a result of precision and high availability.

A high level of standardization of individual components and control software simplifies not only the maintenance of a production system, but also the spare parts management and planning, as well as the training of the operating personnel; this in turn leads to a reduction in costs. Moreover, standardized processes and sequences maximize operator safety.

In process technology, KUKA Systems offers its automotive customers a wide range of services. With joining processes, for example, consideration is given to the entire range of available and feasible processes – a strategy that can also be transferred to PV production. To this end, KUKA Systems develops its own solutions. The KUKA Solar product group has already generated several patent applications and around 10 new robot-based process technologies. The main target markets in this field are the US, Asia and Europe.

KUKA sees further automation potential in the area of wafer and cell production, in which sector the company draws on its wealth of experience in designing complex systems for the automotive industry.

A further example of an automotive application that served as a model for the solar sector is window installation in vehicles. This process demands mastery of complex and highly sensitive operations: perfect timing, tailored measuring technology and the right robot programming are just as essential here as in the automated assembly of solar modules. The ROBO TAPE component supplements the ROBO FRAME module described above, building once again on experience from the automotive industry.

Turnkey solutions for the solar sector KUKA Thin-Film Automation: for thin-film production featuring necessary handling tasks in the cleanroom.

KUKA Wafer Automation: offers automation solutions ranging from the ingot to the wafer, and from simple handling tasks for crucibles right up to wafer separation.

In addition to automation of handling and quality control tasks, process machines such as wire saws or grinding machines can also be integrated.

KUKA Module Line: an application for the back end, this system encompasses all types of PV modules, from thinfilm, glass/sheet, glass/glass or concentrator solutions, right up to the flexible module.

KUKA Collector Automation: offering solutions for manufacturing solar collectors and ensuring a costeffective and ergonomically balanced production sequence.

KUKA Systems Energy Solutions KUKA Systems Energy is the worldwide network of KUKA Systems competence centres. The group specializes in the automation of processes in the solar industry and offers manufacturers a range of products and services for the implementation of production processes with utmost precision and efficiency.

KUKA Systems Energy is specialized in the supply of turnkey systems. Our core competencies are the analysis, modelling and optimization of complex production processes that are integrated and supported on site.

All the automation solutions can be flexibly scaled – from the robotic cell as an automated production island, right up to complete production lines.

Main areas of activity: Planning and engineering Thin-film handling Highly automated brick and wafer line Thermal collector solutions Automated module manufacturing

KUKA Systems KUKA Systems is an international supplier of flexible robot-based automation systems for the automotive, aerospace, energy and general industries. Some 3,780 employees worldwide work on ideas, concepts and solutions for automated production and the provision of products and services for virtually all tasks in the industrial processing of metallic and non-metallic materials.

The product range is presented and marketed internationally via the company’s subsidiaries and sales offices in Europe, America and Asia.

KUKA Systems recorded an order volume of around €855 million for the 2008 business year.

For more information, please contact: KUKA Systems GmbH Markus Meier, Head of Marketing/Communications Tel: +49 821 797 2483 Fax: +49 821 797 1951 Email: markus.meier@kuka.de Phot ov olt aic s I nt ernat ional

29

Subscribe to the full journal today!

Articles presented in Photovoltaics International are independent and depict actual data and findinggs for the education of facilities managers, executives and engineers. We will provide coverage of the best and latest developments in the PV manufacturing industry, and present these in a clear, easiily navigable format. The next edition will be published in May 2009. Annual subscriptions are US$1999 and will guarantee that you receive business-critical information every quarter.

I would like to purchase a subscription. $59.00 USD Price: 1 x Issue $199.00 USD 4 x Issue Method of payment:

❑ ❑ Please start my subscription with edition ............

(includes international delivery) (includes international delivery)

❑ Credit Card

❑ Bank Transfer

Name: ..............................................................................................................................................................................................................................................................................

Job Title: ...............................................................................................

Job Function: .....................................................................................................................

Company: ...........................................................................................

Div/Dept: ............................................................................................................................

Street Address: .............................................................................................................................................................................................................................................................

City: ........................................................................................................

Post Code/Zip: ...................................................................................................................

Country: ..........................................................................................................................................................................................................................................................................

Telephone Number: ........................................................................

Web URL: ............................................................................................................................

E-mail: ..................................................................................................

For the purposes of our circulation audit,please indicate the last digit of your birth year (YYYY):.......................................................................................

Company Activity (tick where appropriate):

❑ PV Manufacturer (inc.Thin Film & Module) ❑ ❑

❑ Materials Supplier ❑ Equipment Supplier If you ticked any of the above three options, please indicate the company technology type by ticking one or more of the boxes below: ❑ Si Cell ❑ Thin Film ❑ Module ❑ Concentrator ❑ Emerging ❑ Polysilicon ❑ Ingot/Wafer University ❑ Energy Utility Supplier ❑ Government Agency R&D Facility ❑ Financial Community ❑ Other (please specify..................)

In order to continually improve Photovoltaics International we require your feedback. We would be very grateful if you would answer the following questions: (Q) Which section(s) of the publication are of interest to you? (please tick) Fab + Facilities ❑

Materials ❑

Cell Processing ❑

Thin Film ❑

PV Modules ❑

Power Generation ❑

Market Watch ❑

(Q) What technical subjects do you wish to see in future editions?........................................................................................................................................................... ..............................................................................................................................................................................................................................................................................................

Signature:....................................................................................................................................

Payment Details: Fax on +44 (0) 20 7871 0101 or email info@pv-tech.org: Type of credit card: Card Number: Expiry Date: 3 Digit CVV Code (back of card): Cardholder’s name: ...........................................................................................

Bank Transfer Details: Account Name: Semiconductor Media Limited Account Number: 80686832 Sort Code: 20-39-53 Swift Code: BARC GB 22 IBAN Number: GB 42 BARC 203953 80 68 68 32 Bank:

Post: Make cheques payable to “Semiconductor Media Ltd.” Photovoltaics International, Trans-World House, 100 City Road, London EC1Y 2BP, UK Online: PayPal - visit www.pv-tech.org/shop To request an invoice please call on 020 7871 0148

Date ..................................................................................

Barclays Bank Plc, 10 Hart Street, Henley-on-Thames, Oxon, RG9 2AX.

All invoices are calculated in Pounds Sterling.

Any payments made in US$ must be made according to the appropriate exchange rate at the time of payment.

Apply for the positions below and many more at www.JobsinPV.com

Applied Materials Engineering Manager

GCL-Silicon General Manager, US/Europe Sales and Marketing

Location: Santa Clara, California.

Experience: 5-15 years’ industrial experience in solar photovoltaics or closely related manufacturing fields is required. Ph.D./M.Sc. in Electrical Engineering, Physics, Chemistry, or Materials Science required.

Location: U.S. and Germany.

Experience: 7+ years of sales and marketing experience in relevant industries with a proven track record of success. Bachelor’s degree from a leading academic institution required.

This position is in the cSi Solar technology group.The successful candidate will be responsible for managing a technology group to develop leading edge and nextgeneration high-efficiency crystal solar cell technologies using proprietary process, device, and tool technologies. Responsibilities include development of industry leading-edge high-efficiency new cell design, yield, reliability, metrology, and optimization of process steps with a view to minimization of manufacturing cost.This requires deep understanding of the full sequence of production steps and the related process flow.

GCL-Silicon are one of the world’s leading solar companies and are seeking a General Manager of Sales and Marketing for the US/Europe regions.The role will involve management of existing relationships and development of new account relationships, while achieving revenue generation targets of the region.The successful applicant will lead marketing efforts in the region, including building and maintaining of GCL Silicon’s brand, as well as formulating and execution of the company’s sales and marketing strategies and targets in the region in conjunction with the head office in China.

REQUIRED SKILLS • Strong background of leading-edge crystal silicon solar cell process technology and factory experience with hands-on knowledge are required.

• Proficiency in solar cell device physics and optical properties of materials is also required.

• Practical knowledge of various equipment platforms and tool design trade-off considerations is a big plus.

• Strong oral and written communication and teamwork skills are mandatory.

REQUIRED SKILLS • In-depth knowledge of solar and strong relationships with key solar industry players in the US/Europe.

• Ability to think and plan strategically.

• Team player with strong leadership, interpersonal, presentation and communication skills.

• Fluency in English is essential; any European language skill is preferable (for the Europe role) but not a must.

• Frequent travel to clients’ locations required.

Going Places XsunX promotes Grimes to President, Wendt to CTO Thin-film solar PV module manufacturer XsunX has promoted two of its executives to positions of greater responsibility. Joseph Grimes, current company COO, has been named President, while VP of engineering Robert Wendt will take on the CTO role.

Grimes (pictured) will have oversight responsibilities in helping to define XsunX's strategies, while continuing to execute the company's business development objectives. TFPV veteran Wendt will oversee thin-film technology development, product design, and technology commercialization efforts.

Frank Faller named Managing Director at centrotherm photovoltaics centrotherm photovoltaics AG has named Dr. Frank Faller (pictured) as Managing Director at the company’s US subsidiary, centrotherm photovoltaics, Inc. in Atlanta, Georgia.

Frank Faller, who has been in the solar industry for over 15 years, had previously been Head of Sales & Marketing at Deutsche Solar AG, SolarWorld’s wafer company, in addition to spending time at the Fraunhofer Institute for Solar Energy Systems.

Gerhard Stryi-Hipp named Head of Energy Policy at Fraunhofer ISE Fraunhofer ISE has announced that Gerhard Stryi-Hipp (pictured) will take over as Head of Energy Policy and Group Leader of Thermal Collectors and Applications in beginning in July. Stryi-Hipp takes over from Matthias Rommel,who is leaving the company to take up the role of Professor and Director of the Institute for Solar Technologies SPF of the University of Applied Sciences, Rapperswil.

Stryi-Hipp had previously been Chairman of the German Solar Industry Association BSW-Solar for 15 years, in addition to being Chairman of the Heating & Cooling European Technology Platform RHC-ETP. He has also been a Board Member at the European Solar Thermal Industry Association ESTIF.

Yingli reshuffles executive board Yingli Green Energy Holding Company Limited announced Zongwei Li will join the company’s Board of Directors and Professor Ming Huang (pictured) will join the Board's Audit Committee. George Jian Chuang has resigned from the Board of Directors to pursue other opportunities.

CEO Zongwei Li joined Yingli in 2006 after serving as Senior Audit Manager and Audit Manager at the accounting firm PricewaterhouseCoopers for 11 years.

Professor Ming Huang became Director of Yingli in August of 2008.

REC bids farewell to CEO Erik Thorsen

Tan Wee Seng named Independent Director of Board at ReneSola

Departing the company following a four-year tenure, President and CEO Erik Thorsen has agreed with REC Group’s Board of Directors to depart and make way for a new Group leader. Fitting this bill is Board Chairman Ole Enger,former CEO of SAPA AB, who took over the President and CEO role on April 4th.

ReneSola Ltd. has named Tan Wee Seng as Independent Director of its Board, replacing Robert Naii Lee, who had resigned for personal reasons. Prior to ReneSola,Tan Wee Seng had been with Li Ning Company Limited (LNCL), where he served as Executive Director, CFO and in other positions.

Mr. Enger’s most recent role was as CEO of SAPA AB and Executive Vice President for Orkla ASA. Previously, he served a 13-year period as President and CEO of Elkem ASA, prior to which he worked in managerial roles for six years at Norsk Hydro ASA. Mr. Enger has had a seat on the REC Board since 2005 and has been Chairman of the Board since 2007,a role that he will now be filled by Tore Schiøtz on an interim basis.

In addition, he has also served as Senior VP at Reuters Limited, where he handled the business management in China and other parts of Asia and as Finance Manager and Managing Director of AFE Computer Services Limited. He has more than three decades of experience in business and post-acquisition management.

The PV-Tech Blog

By Tom Cheyney

Shakin’ all over: Move over world, the solar leader may soon be the USA

Photo: Tom Cheyney

What country is the “great hope for the global solar industry”?

Not Germany, although it still has something left in the tank before its recent torrid – and subsidized – growth rate hits the wall. Certainly not policy-challenged Spain, whose incentive program about-face has left a gigawatt or so of modules originally designated for Iberian installations sitting in warehouses with no pipeline to fill. Italy or Greece? Sure, they’ll grow at a nice clip, but can’t be counted on for any drastic scaling or consistency of governmental attention. China perhaps? Although its manufacturing chops will continue to feed the international market, the PRC’s own domestic demand curve isn’t steep enough to gobble up all of that capacity, despite the leadership’s helping-hands policies.

No, the saviour of solar is the USA, opined the Prometheus Institute’s Travis Bradford during Greentech Media’s recent “Surviving the Shakeout” solar industry summit.

After going over the state of the industry – which has flipped from a supply-constrained sector to one where demand contingencies call the shots and PV module ASPs have plummeted since late 2008 – Bradford focused on the current state and future promise of the U.S. market. Given the massive insolation enjoyed in the States, solar patriots have long touted the country’s great potential for PV and thermal.

If Bradford is correct, that potential will become reality over the next decade.

He showed data comparing the total annual sunlight measured in kilowatt-hour per square metre compared to average electricity rates tabbed in cents per kilowatt-hour, which revealed that many U.S. population centres are or will soon become cost-effective for solar and reach grid parity – even without subsidies. The U.S. numbers are better than any other market in the industrialized world, Bradford says.

Given the recent about-face of U.S. policy regarding solar and other renewables, he characterized the combination of healthy federal and state incentives – including their positive impact on the equity and debt parts of the financial equation – with the drop in module prices, as a nirvana-like situation.

Bradford showed NREL map charts that depict electrical rate differences in various parts of the country. Two projections for

32

w ww. pv - t e ch . o r g

2015 – one factoring in a 0.5% annual rate hike, the other a 1% raise – predict that either half or two-thirds of the U.S. market can get below grid parity with solar power. Put another way, 98% or 99% of the country will be within 5 cents per kilowatt-hour.

This confluence of factors leads to what he calls a “near-term windfall in the U.S. industry,” one that will strengthen when the “financing lynchpin” is ameliorated.

Another photovoltaic patriot, Spire’s Roger Little, echoed Bradford’s bullishness, claiming the U.S. “will become the world’s largest solar energy market.” Citing positives like the federal stimulus package, the growing call for huge utility projects, state incentives that could be worth up to 250GW in 10 years, the enactment of renewable portfolio standards in most states and a federal one in the works, and the push for creating green jobs, Little said it’s time “to get into this business and take advantage of the opportunities.”

He believes the best way to kick-start the lagging U.S. PV manufacturing base is not to build vertically-integrated gigawatt factories but to adopt what he calls “distributed module assembly,” a kinder, gentler plan for building relatively modest 50MW automated, mostly crystalline-silicon-based panel plants throughout the country, positioning them to feed the growing solar demand.

If one of these plants were built in each state, the cumulative production would cover what Little believes will be a nearly 2.5GW shortfall in the U.S. capacity to meet its own demand by 2012.

The Little Plan’s basic assumptions include a 219W module with 15.5% efficient cells costing US$1.80 per watt, seven-year depreciation, and a 20,000 square-foot factory at US$10 per square foot per year. He puts the non-cell-related module costs at 54 cents per watt and the total module cost, including the cell, at US$2.34.

He offered an example that showed an upfront investment of US$10 million, from which the federal ITC slices out US$3 million and the state tax break removes another US$5 million, resulting in a net cost for a 50MW module fab of a mere US$2 million.

The annual profit and loss for such a factory? With annual revenues at US$140 million (at US$2.80 per watt times 50MW), after working through the costs of goods sold and general/administrative expenses, the EDITDA is a not-measly US$14.9 million.

Little’s model also posits the creation of more than 10,000 jobs in the module assembly and module tool businesses by 2012.

He thinks that building “utility solar farm factories,” where supersize, 1kW modules are fabricated close to where they will be deployed, could maximize cost reductions to the tune of US$2.96 per watt in utility system costs.

Although his plan is a tad self-serving given Spire’s business model to sell turnkey module production lines, the Little Plan is intriguing – and well worth including in the conversation about approaches to bulking up the U.S. solar manufacturing base to service burgeoning domestic demand and be more competitive globally.

Tom Cheyney is Senior Contributing Editor (U.S.) for the Photovoltaics International journal and writes blogs for PV-Tech.org.