Page 1

New Products Manz Automation’s OneStep SE system


TRUMPF’s ultra-short pulsed lasers


Solland Solar’s new Sunweb modules

Photovoltaics International Volume 05 - 2010

Atmospheric deposition for thin films Exploring the burgeoning UK solar market The PV-Tech Blog Commercializing CI(G)S with Sulfurcell, Ascent & Solar Frontier

Ha Visi ll 3 t u , L s at ev E el U 2, PV B o SE ot C hD 1

NREL investigates

Introduction 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: Web: Publisher: David Owen Managing Editor: Síle Mc Mahon Senior Editor, North America: Tom Cheyney Senior News Editor: Mark Osborne News & Features Editor: Emma Hughes Production Manager: Tina Davidian Design: Daniel Brown Commissioning Editor: Adam Morrison Sales Manager: Neill Wightman Account Managers: Adam Morrison, Graham Davie, Daniel Ryder, Gary Kakoullis, David Evans, Nick Richardson & James Park Marketing Manager: Joy-Fleur Brettschneider 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: Glass sheets awaiting CdTe deposition at Abound Solar. Photo courtesy of Julian Hawkins, Abound Solar.

Printed by Quentin Press Photovoltaics International Lite Volume 5, 2010 ISSN: 1757-1197 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.

A booming PV market has seen a flurry of market forecasters make upward revisions to PV installation figures for 2010. Across the supply chain, capacity expansion plans from polysilicon solar cells and modules have in some cases doubled, yet many manufacturers are in ‘sold-out’ mode for the rest of the year. Bottlenecks have also returned to the fore, with supply constraints being felt at cell manufacturers and solar inverter assemblers as a result of component shortages. It seems like only yesterday that the key supply concern was polysilicon, but with significant investments from traditional suppliers and new entrants, such shortages have not occurred during this upturn. As featured in this edition of Photovoltaics International Lite, Wacker Chemie offers an insight into the future sustainability of polysilicon production to meet the both the purity requirements and environmental footprint of such a vital material for the PV industry. Despite the macro-industry headlines over the FiT regressions in markets such as Germany, Italy and Spain, there is no hiding from the fact that business is thriving and should continue to do so, albeit at a slower pace through 2010. One of many new regional markets set for expansion over the next few years is the UK, having introduced its first FiT in April 2010. The UK’s market potential shouldn’t be underestimated, as Emma Hughes, our news and features editor, elaborates in a special article in this edition. Although residential rooftop installations dominate key markets such as Germany, greater emphasis on utility-scale projects in Spain and the U.S. can encourage the adoption of different technologies. PV Resources takes a look at the growing market for power plants, especially in new regions such as Italy and the Czech Republic. It’s good to see that the backdrop to this year’s EU PVSEC in Valencia isn’t being dominated by doom, gloom and market uncertainties. On the contrary, there seems to be an air of celebration (for want of a better word) of the technology advancements being rushed into volume production, developments that are promising significant improvements in solar cell efficiencies, driving cost per Watt ever closer to the ultimate goal of grid parity. Some of these technologies are touched upon in a paper by NREL, which provides an overview of some of the key atmospheric deposition techniques that could enable much higher cell efficiencies without impacting manufacturing costs. In thin film we have already witnessed further manufacturing cost-per-Watt records and manufacturing scaling to over 1GW. CIGS technology is rapidly becoming CdTe’s next competitor, with the potential for c-Si conversion efficiency comparability, but without the associated manufacturing costs. We hope you enjoy this sample of what Photovoltaics International has to offer. Don’t forget that we will be covering all the latest announcements and technology developments from EU PVSEC online at Feel free to drop by our booth in Hall 3, Level 2, Booth D1. Looking forward to meeting you at the show!

Mark Osborne Photovoltaics International

Contents Photo: Acciona Solar

2 News 14 Products

Photovoltaics International Lite EU PVSEC edition has been produced for exclusive distribution at EU PVSEC 2010 in Valencia, Spain, providing 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 Nick at +44 (0) 207 871 0123 or email

17 Integrated loops: a prerequisite for sustainable and environmentally-friendly polysilicon production Wacker Chemie AG 24 Tabbing-stringing quality control challenges Fraunhofer Institut für Solare Energiesysteme 32 Large-scale PV power plants – new markets and challenges PV Resources

38 Atmospheric deposition techniques for photovoltaics National Renewable Energy Laboratory 48 Expectations for the UK solar market Photovoltaics International 56 The PV-Tech Blog: CI(G)S spells success Tom Cheyney P hot ovol t ai cs Int er nat i onal


News ZSW sets another new CIGS solar cell record

Photo: ZSW

The Stuttgart-based researchers at Zentrum für Sonnenenergie- und WasserstoffForschung Baden-Württemberg, Germany (Centre for Solar Energy and Hydrogen Research, ZSW) have demonstrated a CIGS solar cell conversion efficiency of 20.3%. The area of the world record cell is 0.5 square centimetres and surpasses their previous record of 20.1%, established in April 2010. “Our researchers have made the cells in a CIGS laboratory coating plant using a modified co-evaporation process, which in principle can be scaled up to commercial production processes,” noted Dr. Michael Powalla, member of the board and head ZSW’s world record-holding solar of the photovoltaics division at ZSW. “The Fraunhofer ISE in Freiburg, Germany cell. has confirmed the new results. However, it would take a while before the increased efficiency of CIGS solar cells can be commercially utilized.” Manz Automation has recently signed a licensing and strategic alliance agreement with Würth Solar, which partnered with ZSW on CIGS technology. Manz is now responsible for introducing advances and turnkey solutions based on the R&D at ZSW.

Thin-Film News

CIGS developer AQT Solar opens 15MW cellproduction line, signs first major customer installation CIGS cell developer AQT Solar has opened it first manufacturing facility in Sunnyvale, CA. The company said that it took less than eight weeks to go from preparation, build-out, line implementation and qualification, and production initiation at the plant. The factory has already begun to fill current customer orders of 20MW, with substantial purchase orders in the pipeline. AQT also announced its first customer installation, Sol Pacifico, a large property development, which has ordered 2MW of PV systems to power its gated resort community on the Pacific Coast in Baja, Mexico. The installation is scheduled to break ground in 2011, starting as a 2MW project with the potential to grow to 9MW. The new production facility, benefiting from a recent US$10 million round of funding, houses a 15MW CIGS cell manufacturing line and can be easily scaled up to 60MW capacity, according to AQT. The company attributes the rapid ramp-up of the manufacturing line to its modular design, which allowed for quick onsite deployment. AQT also credited the company’s partnership with equipment firm Intevac.

Honda Soltec to market CIGS module line-up overseas; conversion efficiencies reach 11.6% With a CIGS thin-film production capacity of 27.5MW after initiating production in 2007, Honda Soltec has until now remained focused on the Japanese domestic market. However, the company has reported that it is marketing two new residential-use modules with maximum


www. pv -t e ch .o rg

outputs of 120W and 130W and conversion efficiencies of 11.6%, claiming the highest figures among CIGS-based solar cells being sold in Japan. Before the new addition, Honda Soltec’s product lineup included two types of residential-use solar cell modules with maximum output of 125W and 115W, respectively, and one type of power conditioner with a rated output of 4.0kW. As a result of the new modules, a large-capacity power conditioner with rated output of 5.5kW has also been introduced.

Honda Soltec’s 130W CIGSbased module has a solar energy conversion efficiency of 11.6%.

Suntech stops thin-film production; 2Q revenue stands at US$620 million plus The increasing cost competitiveness of crystalline silicon technology has seen another thin-film victim, this time in the form of Suntech. Although comprising only a small part of the leading Chinabased c-Si module manufacturer’s product armoury, its amorphous silicon thin-film solar module line has ceased production. Suntech is currently restructuring its Shanghai facility to focus on the manufacture of crystalline silicon solar cells at a time when the company is at full capacity and sold-out in the first-half of the year. Suntech was a customer of Applied Materials’ SunFab turnkey technology and had installed a 50MW

Suntech’s Goodyear, Arizona production facility. line in 2009. Applied Materials recently announced it would be stop selling the SunFab technology to new customers and reallocated staff and resources to c-Si technology. “While the thin-film and Shunda related charges will significantly impact our second quarter financial results, they have no bearing on our core manufacturing operations which are performing very well,” remarked Dr. Zhengrong Shi, Suntech’s chairman and CEO. “Going forward, we will continue to focus on our primary mission of supplying the most reliable and high performance solar panels in the industry.”

Specialty glass, ceramics company to enter thin-film market Cor ning’s board of directors has approved a capital expenditure of approximately US$180 million to expand the company’s Harrodsburg, Kentucky, manufacturing facility. The expansion will provide additional capacity for Corning’s Gorilla glass business and entry into the growing thin-film photovoltaic glass market. The Harrodsburg investments are included in Corning’s previously announced capital expenditure expectations for 2010 and 2011. Assistance for this expansion project has been offered by the state of Kentucky and Governor Steven Beshear’s office t h ro u g h t h e K e n t u c k y E c o n o m i c Development Finance Authority in up to US$4.5 million in Kentucky Business

Ascent Solar receives external certification for module encapsulation materials Ascent Solar’s module encapsulation packaging solution for flexible monolithically integrated CIGS modules has successfully passed a critical environmental testing milestone. An u n n amed in de p e nd e nt l a b ora t ory conducted a series of tests on the Ascent Solar modules under the re q u i re m e n t s o f t h e I E C 6 1 6 4 6 standards.

Ascent Solar’s flexible thin-film module. The CIGS packaging solution successfully passed all of the test requirements including the rigorous standard of 1,000 hours of damp heat testing (85% relative humidity and 85°C temperature) required by IEC for performance and long-term reliability.

Materials News

centrotherm SiTec claims improved crystallization furnace process centrotherm photovoltaics’ subsidiary centrotherm SiTec has achieved a new cell efficiency record through an improved process in its crystallization furnace to produce multicrystalline ingots. Further progress has also been claimed in boosting furnace capacity to 650kg without major modifications. Production conditions in a pilot line operation have produced average efficiencies of 16.6%, and a peak result of up to 17.0%, according to the company. centrotherm SiTec also said it had reduced the manufacturing costs

Silicon production at centrotherm SiTec.


www. pv -t e ch .o rg

by 8% with the new crystallization process compared with the previous process sequence. An optimized hot zone for a quartz crucible of 880mm x 880mm x 420mm is employed, which enables more efficient silicon smelting. An optimized crystallization process management system is also used.

SiC Processing starts construction on two more slurry recycling lines for Yingli Solar SiC Processing has started construction on two additional 15,000tn slurry recycling lines at Yingli Green’s solar PV manufacturing campus in Baoding, China. The new lines are scheduled to begin operations in the second quarter of 2011. Once the new slurry recycling capability comes online, the combined capacity of the new and existing lines will be 60,000tn. About 20 million will be invested in the construction of the additional recycling lines in Baoding, according to SiC. Nordic Capital Fund VII recently acquired 70% of SiC Processing.

polysilicon production technology,” said Tom Gutierrez, GT Solar’s president and CEO. “The additional throughput and cost savings of the SDR 400 CVD reactor allows OCI to further enhance their cost competitiveness.”

Meyer Burger signs CHF60m supply contract with Nexolon Meyer Burger has signed a CHF60 million contract with Korea-based Nexolon for the supply of wire saws, ID-saws and wafer inspection systems. The order comes as Nexolon prepares to expand its production facility from 850MW to 1GW by the end of 2011. Meyer Burger and Hennecke Systems have also completed the contract held with Nexelon for the supply of slicingsand wafer inspection-systems.

ReneSola signs 836MW wafer supply deal with two Taiwan-based companies

Meyer Burger subsidiary Hennecke Systems’ wafer inspection tool.

ReneSola has signed two wafer supply agreements to provide Taiwanbased Neo Solar Power and Solartech Energy with approximately 836MW of monocrystalline and multicrystalline solar wafers through December 2013. Under the terms of the contract signed with Neo Solar, ReneSola will supply approximately 293MW of multicrystalline wafers from July 2010 to December 2013 and approximately 141MW of monocrystalline wafers from October 2010 to December 2013. The company will also supply Solartech with approximately 402MW of multicrystalline wafers from July 2010 to December 2013.

Cell Processing News

GT Solar gains US$23.4 million order from OCI Company OCI Company has placed a US$23.4 million follow-on order from GT Solar International. The order calls for delivery of GT Solar’s SDR 400 CVD, which will be used in GT Solar’s CVD OCI’s phase 3.5 reactor. polysilicon plant expansion that will be completed in October 2011. “We are pleased by the confidence OCI has demonstrated with our

SunPower pushes c-Si solar cell efficiency record to 24.2% Produced at its solar cell plant in the Philippines, SunPower has had a full-scale solar cell verified by NREL (U.S. Department of Energy’s National Renewable Energy Lab) with a conversion efficiency of 24.2%. This is a new world record for c-Si cells. SunPower did not disclose any processing details that enabled the new record efficiencies or when such cells would enter volume production. SunPower also noted that it had increased its cell efficiencies a full four percentage points over the last five years. However, although the company touted it had radically reduced manufacturing costs during the same period, SunPower has entered into a joint manufacturing partnership with Taiwan-

Source: SunPower Corp.

Investment incentives, and up to US$1 million in tax rebates related to construction costs.

SunPower’s proprietary all-back contact cell.

based AUO to drive down manufacturing costs of its next-generation cells.

silicon ink. Development work will be performed at both Innovalight’s headquarters in California and at JA Solar’s R&D center in China with the new agreement lasting three years.

Dow Corning, University of Toledo join forces in solar R&D effort

Hyundai Heavy Industries to nearly double cell, module capacity to 600MW by 2011

Researchers at Dow Corning and The University of Toledo (UT) have signed an MOU to engage in collaborative discussions on photovoltaic solar research and development efforts to help bring down the cost of solar energy. “Both Dow Corning and UT want homes and businesses throughout the world to take advantage of clean, renewable energy from the sun,” said Gregg Zank, senior vice president and chief technology officer of Dow Corning. “It is essential that businesses, academics and the government collaborate in order to accelerate the advancement of solar technologies.”

Hyundai Heavy Industries said it plans to double its annual module and cell production capacity from the current levels of 320MW and 370MW to 600MW, respectively, by early 2011. The company will complete the near-doubling of its capacity through the expansion of its solar power factory in Eumseong, north Choong-cheong province, by early 2011 and start full-scale production from the second quarter of that year. Hyundai Heavy has also been processing 3000 tons of polysilicon prototypes at Korea Advanced Materials, a company jointly established with KCC.

VLSI Standards adds ISO-17025 to its solar cell certification services In a move to help solar cell manufacturers meet metrology quality requirements, VLSI Standards has enhanced its portfolio to include accreditation to the ISO-17025 standard for cells of up to 156mm x 156mm in size. Production cell measurements will continue to be serviced from VLSI’s calibration laboratory in California. VLSI’s solar cell certification service offers traceable measurements of production cells of short circuit current (Isc), open circuit voltage (Voc), maximum power (Pmax), cell area, spectral response (or quantum efficiency), fill factor and cell efficiency. Measurements will be performed under STC with an option to have temperature dependence measurements.

JA Solar to push for 20% cell efficiencies with extended collaboration with Innovalight A joint development agreement has been signed between JA Solar and Innovalight to push cell conversion efficiencies past 20% for JA Solar’s recently introduced Secium high efficiency solar cells. The cell manufacturer has already achieved conversion efficiency of 18.9% using Innovalight’s nanotechnology-based


www. pv -t e ch .o rg

Bosch Solar officially opens Arnstadt crystalline solar cell production plant Energy has formally opened its new Arnstadt, Thuringia-based crystalline solar cell production plant. The 12,000m2 facility provides the company with extra capacity of 10,000 cells an hour, or 90 million solar cells a year. With the completion of the Arnstadt plant, Bosch aims to almost triple its nominal capacity to around 630MWp and thus provide a major boost to its growth strategy. The expansion of its facilities will be completed in 2011, when the R&D centre and the new main administrative building are also finalized. The opening ceremony for the whole complex will be held in summer 2011. By 2012, Bosch will have invested around 530 million in this location, creating a total of 1,100 jobs.

China Sunergy, REC Wafer dispute continues The ongoing dispute between China Sunergy and REC Wafer continues as the company reveals that it is considering appealing against the decision in favour of the Norway-based wafering company. The ruling by the Norwegian District Court on July 5th, which took the side of REC, is now set to be contested by China Sunergy, meaning that the US$50 million bank guarantee is currently still blocked from being withdrawn by REC Wafer.

SunPower taps GE to provide ultrapure water system for new Malaysian joint-venture solar-cell fab

Source: REC

Source: Dow Corning

Dow Corning solar cell research.

started up its 5,000tn manufacturing facility. The factory, which can be upgraded to 8,000tn, is said to be the largest single poly production site in Taiwan. TPSI plans to increase its manufacturing capacity to as much as 12,000tn in the coming years. For this project, centrotherm SiTec provided the development, technology, and equipment know-how. The German company was responsible for all of the factory planning, and provided the basic engineering, process expertise, and reactors and converters required. A vent gas recovery system developed by centrotherm SiTec was also deployed, which faciliates better exploitation of process gases, thereby significantly cutting both manufacturing costs and energy requirements as well as eliminating harmful effluents.

Cell production in Narvik, Norway.

Fab and Facilities News

Taiwan PolySilicon starts up 5000tn manufacturing plant developed by centrotherm SiTec Taiwan PolySilicon, a turnkey customer of centrotherm SiTec, has successfully passed its initial production phase and

SunPower has chosen GE to provide the ultrapure water systems for the new SunPower-AUO joint-venture Fab 3 solar-cell fabrication plant under construction in Malaysia. The UPW equipment will save more than 230 million gallons (>870 million litres) of water relative to other technologies, the companies said. The facility will be one of the largest silicon solar manufacturing factories in the world and is expected to begin operations near the end of 2010 and ramp production during the following two years. The facility is located 20km north of Melaka, Malaysia, in a region that has experienced prior water shortages and drought. GE will design, supply, and install the advanced UPW system featuring the internationally patented high-efficiency

Let the Sun Turn Our Silver into Your Gold

Heraeus silver paste is proven to increase the performance, value and output of your solar cells. Our frontside pastes offer higher efficiencies and greater aspect ratios to create superior electrical performance. This increased output means more value to your customers and lower cost per watt for you deliver you. Something the competition can’t deliver. Our backside bus bar silver pastes are lead-free and offer industry-leading adhesion and coverage. And we can customize formulations to meet your exact requirements. At production plants and application labs in China, Taiwan, Europe and North America, our dedicated personnel provide unmatched local technical and sales support to respond quickly to your business needs. Call us, we can help you quickly realize the benefits of working with Heraeus. Learn more. Visit our website or contact us today.

Heraeus Materials Technology LLC Photovoltaic Group West Conshohocken, PA 19428 S USA +1 610-825-6050

Courtesy: Solyndra

AMAT reports strong Q3 results but warns that solar sector will weaken in Q4

Solyndra’s Fab2 facility. reverse-osmosis (HERO) process. The system, operating on a challenging and variable feedwater source, will provide 2400 gal/min (9085L/min) of ultrapure water for manufacturing.

Eyelit revenue boosted in the quarter by Solyndra and SolarWorld plant expansions Privately held manufacturing software firm Eyelit said it had tripled its first-quarter license revenue and recorded a 75% total company revenue growth since the same quarter last year, which was led by CIGS company Solyndra for its 500MW Fab2 project. Also of note was SolarWorld’s new solar wafer production plant in Freiberg, Germany. Eyelit also said that its overall company staffing grew by more than 25%, with a 60% increase in consulting services during the first quarter to support the growing global demand for Eyelit products.

Initial production of Yingli Green 400MW capacity expansion begins Yingli Green has begun the initial production of its latest 400MW capacity expansion, including 300MW of Panda monocrystalline silicon-based production capacity at the Baoding headquarters, and 100MW of multicrystalline siliconbased production capacity in Haikou, Hainan Province.

to reach 14.6GW, a 95% increase from 2009 and nearly three times size of the market back in 2008. Increased demand from Germany and other European countries is fuelling demand but growth in other regions such as the U.S. is playing a part, IMS Research said. “Basing our forecast on inverter production is incredibly important this year as it’s well documented that inverter supply is limiting the PV market to a massive extent,” noted Ash Sharma, PV research director at IMS Research. “Although demand may be higher than this 14.6GW, if customers are not able to secure inverters then installations will not be completed.” According to the market research firm, the top three markets in 2010 will be Germany, Italy and Czech Republic, which are predicted to install a combined 9.8GW of new PV capacity. Germany is expected to account for some 47% of new capacity, IMS Research said. However, emerging markets in Asia and North America will gain share over Europe, leading to a slight share fall for EMEA countries to 78% of the market. With acute shortages in top-brand inverters, IMS Research noted that double ordering is taking place.

Applied Materials posted strong quarterly results that beat the analysts estimates, reporting net sales of US$2.52 billion, operating profit of US$183 million, and net income of US$123 million or US$0.09 per share. Non-GAAP net income was US$234 million or US$0.17 per share. However the world’s top supplier of semiconductor manufacturing equipment warned that sales of solar and energy equipment would decrease from 15% to make up 10%–20% of company revenue this quarter due to strong Q3 results and looming feed-in tariff cuts. “Applied had strong results across our semiconductor, display and crystalline silicon solar businesses, and we now expect Silicon Systems Group net sales to be up by more than 160% over fiscal 2009,” said Mike Splinter, chairman and chief executive officer. “During the quarter, we took actions that focus our Energy and Environmental Solutions segment on our most promising opportunities in solar and advanced energy, and strengthen our company’s financial outlook.” The company’s fast-growing solar division saw net sales more than double from the fiscal second quarter to US$387 million, led by record demand for its crystalline silicon solar equipment.

German PV market remains on top in 2009; set to shift in 2011 According to the latest report from Solarbuzz, Germany remains market leader in the European solar photovoltaic sector, despite feed-in tariff cuts and module pricing adjustments. The German PV market reached 3.87GW in 2009, while Italy came in second place with 770MW and the Czech Republic, France and Belgium combined to add 933MW of newly installed capacity. Even though Germany came out

Market Watch News

Despite only a small improvement in solar inverter component supply expected in the second-half of the year, IMS Research has raised its forecast for global photovoltaics system installations. The market research firm expects installs


www. pv -t e ch .o rg

Source: IMS Research

Despite inverter shortages, IMS Research raises 2010 solar market to 14.6GW

Forecast PV growth in 2010 by installation type.

«Competition is not about winning. It’s about being better than the rest.» Robby Naish Youngest windsurfing world champion ever at the age of 13; the world speed record holder; continuously leading the world championship for more than 25 years

We at Oerlikon Solar are proud to provide a new revolutionary solar technology that is ahead of its time. Be a part of the revolution and find out more about the new THINFAB and its world record breaking non-toxic, environmentally friendly solar technology at /solar/thinfab.

on top in 2009, with a recorded 109% growth, Solarbuzz reports that this expansion could have been even larger if not for the shortage of inverters that has curbed the market since the September of that year. Growth of the total European market was just 16% in 2009, while growth excluding Spain was 126%. This figure highlights the vulnerability of the overall market to policy review in the larger markets balanced against the growth of emerging markets, claims Solarbuzz.

Power Generation News

DRI Energy completes 1MW installation at Mercer Foods’ CA facility DRI Energy has completed a 1MW ground-mounted solar photovoltaic installation at the Mercer Foods facility in Modesto, CA. The project was accomplished in 33 days, meeting the deadline required in order to preserve Mercer Foods’ solar rebate from Modesto Irrigation District (MID). The installation includes 3,582 Suntech 260W modules, a Satcon 500kW inverter and a Satcon 375kW inverter.

DRI Energy Mercer Foods installation.

Solon, UK developer 35 Degrees to build 1.3MW ground-mounted system in Cornwall Solon has signed a deal with British project developer 35 Degrees to build a 1.3MWp solar power plant on the grounds of a former tin-ore processing facility near Bissoe in Cornwall. The companies said that the system, to be deployed on three hectares of land, could be the first ground-mounted PV installation built in the UK. The proposed agreement includes the planning, engineering, procurement,

Solaire Generation’s 3MW-plus canopy installation.


www. pv -t e ch .o rg

and construction for the turnkey delivery of the monocrystalline-silicon modulebased system to 35 Degrees. Both parties said they intend to enter into a long-term collaboration in this growth market in the UK. 35 Degrees was founded as a project development company in February 2010 and is planning to install 100MWp of solar energy over the next five years in the UK, concentrating in the southwest of England. Solon will work with an experienced, locally based installation company to build the system, which, once completed, will generate about 1.25 million kWh of solar electricity annually. Finalization of the project is dependent on obtaining the necessary planning permits; financial terms of the deal were not disclosed.


Solaire to supply multimegawatt PV parking canopy system at New Jersey site Solaire Generation has been tapped to supply a system of its photovoltaic parking structures for a large commercial solar power plant at an unnamed Fortune 100 corporate office campus in central New Jersey. The 3MW-plus canopy installation, made up of 25 rows of more than 11,350 crystalline-silicon panels covering over 1,350 parking spaces, was begun in July and will be completed by early 2011, according to the company. Once commissioned, the PV canopy s y s t e m s — w h i c h f e a t u re S o l a i re ’s patented dual-incline, integrated decking/gutter design—will generate more than 4 million kWh annually. The financial terms of the deal were not disclosed. The company said it expects to build canopies supporting at least 5MW of installed PV capacity by the end of this year.

Solon, CBS Energy to supply PV modules for Bright Generation’s SolarCombi Solon and CBD Energy, its Australian representative, have formed a new partnership with Murdoch Universitybased start-up Bright Generation,

Solon Black 130/04 solar modules.

for the supply of Solon Black 130/04 solar modules. Bright Generation will integrate the modules into its product, SolarCombi, which combines photovoltaics and solar thermal power in the single streamlined system and is marketed to home owners in Australia.

BIPV installation in China arrives on time China Energy Conservation and Environmental Protection group has now completed one of the largest building integrated photovoltaic (BIPV) installations in the world. The 6.68MW system is fully integrated into the awning at the Hongqiao high-speed railway station in Beijing, Shanghai. The 160 million yuan (US$23.6m), system covers a total roof area of 61,000m 2 with 20,000 solar panels, which have produced 300,000kWh of power since the project was completed in early July. The Chinese government is also tendering for bids to develop 13 solar projects with a combined capacity of 280MW in the western regions. The move follows last year’s bidding for a 10MW solar power plant in Dunhuang, Gansu Province. The government has set a target to install 20GW of solar energy capacity by 2020.

PV Modules News

Dow Chemical to begin commercial-scale production of new PV module encapsulant film in Q4 2010 Dow Chemical is ramping production of a new polyolefin-based PV module encapsulant film at its manufacturing plant in Findlay, OH, where a state-ofthe-art production line has been added to address the growing demand for specialty films for use in crystalline-silicon and thin-film solar panels. The company said that commercialscale manufacturing of the new UL-listed film, called Enlight, will begin on the Findlay line in the fourth quarter of 2010, as part of a phase-in plan to upgrade production and increase capacity across the globe to address increasing demand

Energy Saving[≠50%CUT] High productivity[60MW]

Location Briefing SOLAR VALLEY, Germany

for this and other differentiated encapsulants. Dow claims the new encapsulants can enhance efficiencies in PV module production and lead to lower conversion costs, as well as offer greater module stability and improved electrical performance compared to EVA-based traditional encapsulant, thus improving the reliability and extending the service life of modules.

AUO gains carbon footprint verification for EcoDuo PM220P00 PV module

Location: The SOLAR VALLEY Saxony-Anhalt, located in the centrally suited state of Saxony-Anhalt bordering the states of Saxony and Thuringia, is one of the leading Solar regions in the world and is comprised of a network of 29 international operating firms, nine leading research institutes, and four universities. The SOLAR VALLEY Centre for Silicon Photovoltaics has the highest density of companies involved in the photovoltaics industry. Introduction: The SOLAR VALLEY also lives up to its superior production of photovoltaics and uses its own products. In order to initiate private investments in solar electricity systems, the town of Bitterfeld-Wolfen and the district of Anhalt-Bitterfeld launched the “1000 Roofs Programme” initiative. Manufacturers like Q-Cells and Sovello have signed up for the project, providing a quota of solar modules with special conditions for the first 1000 roof projects and is offering local installation companies special training courses. As for the financing part, the local savings bank will provide support as a partner. The project is aimed at all citizens in the region who are willing to contribute actively to the protection of the environment by installing a PV system on their roof and at the same time want to achieve secure returns over a period of 20 years. A new trend within the project is the idea to even put up solar modules on the roofs of churches and historical buildings of Saxony-Anhalt. Key features/incentives: y Cell Award Winner for “Best Region for Manufacturing Solar Technology” in 2009 y Around 3,500 jobs in the solar industry attest to the area’s economic strength y Generous investment incentives cover a high percentage of capex y Fast-track project realization, due to the close proximity of the world’s leading PV equipment suppliers and superior engineering as well as local authority services and support y Leading glass producers and glass suppliers located in Saxony-Anhalt offer best siting conditions y Close R&D cooperation with Germany’s leading PV research institutes and four universities y Shorter time to markets, via state-of-the-art infrastructure for lower rate of long-term transport inventories y Skilled and flexible workforces, low labour costs y Nearby international schools to join the solar family life y A modern and environmentally friendly place to live and work

Key tenants Q-Cells, Sovello, PV-Crystalox, Solibro and Malibu


www. pv -t e ch .o rg

Trina Solar to supply 35MW of PV modules to SunEdison in 2010 Trina Solar’s U.S. unit has signed a sales agreement with MEMC subsidiary SunEdison to supply 35MW of crystalline-silicon solar modules for the remainder of 2010. SunEdison said it expects to use the panels for projects in North America and Europe. Financial terms of the Trina Solar commercial deal were not disclosed. installation in Todi, Italy. During MEMC’s recent second-quarter earnings call, the SunEdison unit was reported to have 111.5MW of solar power projects under construction that it expected to complete by year’s end.

Solaria’s Shugar says company will ship ‘multiple MWs’ to enXco over next two quarters More details have emerged about the five-year global module supply deal signed between LCPV module maker Solaria and enXco, which also included a small equity investment by the EDF Energies Nouvelles unit in the solar firm. Solaria CEO Dan Shugar told PV-Tech that his company “will be shipping several megawatts over the next two quarters to specific projects in North America and Europe,” adding that “specific projects announcements will be forthcoming.” The agreement, which was described as a combination of a firm order and significant options, will enable enXco to execute on its growing portfolio of utility-scale solar PV projects in the United States and Canada, the companies said. The financial terms of the deal have not been disclosed. S o l a r i a ’s p r o p r i e t a r y low-concentration monocrystalline-silicon PV modules, designed specifically for groundmounted tracking systems, have been certified to UL1703 and IEC61215 Solaria PV modules. standards.

Source: Trina Solar

SOLAR VALLEY Saxony-Anhalt – a sustainable region in the heart of Germany

AU Optronics (AUO) has become the first company to achieve carbon footprint verification on EcoDuo PM220P00, the multicrystalline photovoltaic module, according to the international carbon footprint standard PAS2050. The product carbon footprint of EcoDuo PM220P00 refers to the total amount of CO 2 emissions, including the raw materials, manufacturing and distribution stages. By tracing the carbon footprint, AUO verified that raw materials took up around 82% of total carbon emission, manufacturing about 17% and distribution approximately 1%, among which raw materials led to the most emissions.

Visit us at Solarpeq in Dusseldorf (D) 28 Sept - 1 Oct 2010 Hall 14 / Stand D49

Or visit us at the 25th EU PVSEC in Valencia (ES) 6-9 September 2010 Stand L3/H4/D29

Analyse solar simulators with SolarSpec The SolarSpec series allow you to do an evaluation and qualification of solar flashers and simulators after the norm IEC60904-9 for Class AAA simulators.

Inline analysis of thin fi fillms & coatings with the NanoCalc The NanoCalc-XR is a versatile and configurable reflectometer system designed for accurate single or inline thin film measurements. nttts. s.

Analysis from 200-2500 nm with the QE65000 & NIRQuest The QE65000 & NIRQuest are reliable systems for transmission and reflection measurements. Measure UV-VIS or VIS-NIR, or combine the two for a full analysis.

Measure solar irradiance with the Jaz Light Meter The Jaz Light Meter is a pre-configured, battery-operated spectrophotometer for absolute spectral solar irradiance measurements.

Find out what we can do for you: mail us at! O ce a n O p t i c s E M E A Tel: +31 26 3190500

Lo c al Supp or t D: +49 711 3416960 UK: +44 1865 263180 FR: +33 148 576 136

Products Solland Solar’s new Sunweb modules use MWT cells for 10% higher yields Applications: Sunweb modules are especially designed for the European residential rooftop and BIPV markets, covering also middle-sized commercial rooftop projects. Platform: High efficient, silicon based MWT back-contact cells with narrow spacing, low serial resistance and low surface shadowing due to the optimized metallization pattern. Availability: Early 2011.

Solland Solar’s Sunweb modules are claimed to yield 10% higher power output than its competitors per m 2 on the module level. The modules are designed using the company’s backcontact Sunweb cells, and use Solland’s patented In-Laminated Soldering technology (ILS). The Sunweb cell is based on the metal wrap-through (MWT) concept and has a unique metallization pattern on the front side, making it ideal for residential rooftop installations and BIPV markets.

CoreFlow’s wafer singulation system eliminates manual labour Applications: Crystalline solar wafer singulation process. Wafer thickness >120μm (± 30μm). Wafer size (rectangular) 156mm; 125mm (optional). Platform: SingFlow features stressfree wafer singulation with a unique control that ensures minimal contact force on the wafer. Availability: Currently available.

CoreFlow’s ‘SingFlow’ wafer singulation system is claimed to be the first automatic system that successfully deals with the delicate separation of wafers from blocks, thus eliminating manual labour. Based on a unique, stress-free, pure-shear singulation concept, this patent-pending approach delivers the first field-proven fully-automated production system, and improves production efficiency, throughput and yield of the manufacturing process. The modular system design supports the output to different configurations of lane-based cleaning systems, supporting five to six, or cassette loading for batch cleaning. System throughput with a two-head singulation configuration is 3,000wph with less than 0.25% breakage rate.

Solamet PV16x series metallization pastes from DuPont Applications: c-Si solar cell metallization. Platform: Available in a large range of options for a variety of printing requirements. Availability: Currently available.


www. pv -t e c h .o rg

DuPont Microcircuit Materials (MCM) has introduced the Solamet PV16x series photovoltaic metallization paste range, which are claimed to deliver up to 0.4% greater conversion efficiency for solar cells. Suited to a wide range of printing line widths and processes, the pastes feature low-contact resistance, achieved through advanced chemistry providing lower contact resistance to the Si, as well as reduced bulk resistivity of the paste which yields higher gridline conductivity. Capable of contacting shallow emitters of up to 85:/sq and deep emitters, the Solamet PV16x metallization pastes have undergone extensive customer testing.

TCB offers global rotatable target bonding solution A p pl i cati on s : All t hi n-fi l m P V sputtering materials, including TCO, buffer, absorber and back contact. Platform: Supports all current deposition platforms. Availability: Currently available at all TCB locations.

Thermal Conductive Bonding (TCB) has made available its rotatable target bonding process for lengths up to 4m in the U.S., Asia and Europe. TCB has implemented specialized Ultrasonic Testing equipment for this configuration in order to assure bond quality and performance. This equipment provides detailed imagery of the internal bond to identify defects, and is essential in assuring target assembly functionality and reliability. TCB has bonded over 1000 rotatable targets globally from various metals to ceramics and from lengths of 550mm to 4000mm.

Manz Automation’s OneStep selective emitter system boosts efficiencies by 0.5% Applications: c-Si production applying n-type emitters and frontside metallization as well as existing lines (retrofit). Platform: Throughput: 1200 or 2400 wafers per hour (configurable); accuracy: ±10μm. Footprint (including automation): 4.7 x 2.7m2. Efficiency gain up to 0.5% absolute. Availability: Currently available.

Manz Automation has developed the OneStep selective emitter (SE) system for c-Si solar cells. This process consists of only one single process step, without any consumable usage, and is claimed to enable cell efficiency gains of up to 0.5%. Introduced between emitter diffusion and phosphorous glass (PSG) etch, the single step uses pulsed laser irradiation to locally scan the wafer surface, forming highly-doped areas by local liquid-state diffusion of phosphorous from the PSG layer. After anti-reflection coating, the metallization grid is deposited on top of the highly doped areas. The local doping leads to a reduction of the specific contact resistance from silicon to metal.

TRUMPF offers ultra-short pulsed lasers for high-precision thin-film applications Applications: Wide range of thin-film processing steps. Platform: TruMicro series 3000, 5000 and 7000 lasers have average outputs between 8 and 750W with pulse durations from the pico- to the microsecond range. Availability: Currently available.

TRUMPF’s user-orientated microprocessing lasers from the TruMicro Series 3000, 5000 and 7000 range are suited to an increasing number of applications for thin-film processes including CIGS patterning steps. For TCO film ablation, the TruMicro Series 3000 offers a range of small, compact units with wavelengths of 1064nm and 532nm, ideal for P1, P2 and P3 patterning. Patterning of CIGS thin-film cells can be achieved using picosecond lasers, which feature ultrashort pulses t ha t c a n abl a te m a te ri a l w i tho ut significant heating of the marginal zone of the process, thus preventing cracking, melting or exfoliation of the layers.

P hot ovol t ai cs Int er nat i onal


Products SunSil’s Integra module integrates electronics, micro-inverter and software Applications: A wide range of potential applications from commercial to residential markets. P l a t f o r m : S u n S i l ’s i n t e g r a t e d architecture means that the Integra can be assembled in a few steps with the use of their fully-automated and high-throughput manufacturing line in Denmark. Availability: Initial production of the Integra starts in Q4 2010 with volume in Q1 2011. The units, which cost around €900 each, are currently undergoing certification by Intertek.

SunSil’s Integra solar module is a fully integrated system that is claimed to increase yields by 30% or more while also reducing initial costs. It integrates all of the components of a standard PV system into one 230V 300W AC module using embedded electronics, micro-inverter and software to harvest the maximum amount of electricity from the sun under any conditions. The laser cuts each six-inch square cell into microcells; each of these is monitored and dynamically controlled by SunSil’s patented ‘Dynamic Microcell Optimisation’ technology to provide the optimum output for the whole module.

Robert Bürkle offers flexible back-end line system for crystalline cell modules Applications: Back-end line for the production of crystalline solar cell modules. Platform: Module sizes of 1 x 1.7mm and 1 x 2mm can be accommodated. Cell matrix: 60 or 72 cells per module. Systems can be supplied in 25/50MW sizes. Availability: Currently available.

Robert Bürkle has introduced a complete flexible back-end line for the production of crystalline cell modules, which, it claims, yields the highest throughput per square metre. The line accommodates straight and U-shaped line layouts and can be easily extended to a higher capacity and integrated into an existing line. Different levels of automation are available. The lamination line concept can vary from two to three presses, and can also incorporate another stringer.

Huntsman Advanced Materials’ new encapsulating materials for hot climates Applications: Enables an energysaving encapsulation process at low temperatures and eliminates the critical lamination process with EVA. Platform: Ideal for installation of PV plants in southern areas, which could be combined with solar fuel generators where electrical energy is used to convert CO2 and H2O into natural gas. Availability: Early 2011.


www. pv -t e c h .o rg

Huntsman has developed new materials and processing technologies that will enable actively temperaturecontrolled PV modules with increased efficiency. The ‘Araldite’ AY 4583/ HY 4583 liquid system enables an energy-saving encapsulation process at low temperatures and eliminates the critical lamination process with EVA. Heat conductive, electrical insulating adhesives enable a new module design and construction with improved heat transfer and efficiency. The system claims excellent strength and bonding properties, while the flexible and thermal conductive adhesives protect the flexibility of solar cells.

Integrated loops: a prerequisite for sustainable and environmentally-friendly polysilicon production

Sebastian Liebischer, Dieter Weidhaus & Dr. Tobias Weiss, Wacker Chemie AG, Munich, Germany

ABSTRACT The photovoltaic market, which is dominated by polysilicon-based crystalline solar cells, has been developing rapidly, with growth rates in the double-digit range for several years. In order to meet increasing demand for hyperpure polysilicon, manufacturers need to adhere to environmentally-friendly production processes with low energy consumption. This article highlights the key processes needed to manufacture hyperpure polycrystalline silicon and explores the related challenges and solutions for sustainable polysilicon production. Our findings prove that only an intelligent interaction of all necessary process steps fulfils the requirements for minimized production residue volumes and low energy consumption. Totally integrated production loops for all essential media are prerequisite to reach these targets. Once implemented, these highly efficient production processes serve as an excellent platform technology for the continued healthy growth of the PV industry.

Introduction Global warming and limited availability of fossil fuels have been driving the need for increased and more reliable sources of renewable energy. Due to its increasing competitiveness, photovoltaics has become the strongest-growing

technology in the renewable energy sector with an impressive compound annual growth rate of 43% (2007–2014; see Fig. 1). Although a variety of PV technologies is being explored, only a limited number is suitable for and has reached mass-

production scale. Of those, crystallinebased PV is the clearly dominating technology with a share of more than 80% in 2009’s PV market [2]. The basic raw material for crystalline PV cells is polysilicon. Its semiconducting properties are used to convert sunlight

state-of-the-art cost structures and an environmentally-benign polysilicon production process.

35 30,0 30.0 30

24.6 24,6

Gigawatt [GW]



19,1 19.1


15.5 15,5


15,4 15.4



Japan Rest of EU


6.1 6,1 5

1.3 1,3

1.6 1,6



7,2 7.2

France Italy

2,6 2.6

Spain Germany

0 2007








Source: EPIA 05/2010

Figure 1. Historical annual PV market development and forecast from 2005 to 2014 [1].

Integration levels of polysilicon production By far the most common approach to producing polysilicon globally is the so-called ‘Siemens process’, which involves the deposition of polysilicon from trichlorosilane (TCS). TCS and hydrogen are fed to heated silicon rods within special reactors. A pyrolysis reaction takes place at the hot silicon surface and elemental silicon is deposited on the rod’s surface. In this process, two main reactions of TCS have to be considered: 4 SiHCl3 — > Si + 3 SiCl4 + 2 H2 —


SiHCl3 + H2 — > Si + 3 HCl —


TCS as feedstock is obtained by chlorination of metallurgical silicon. The corresponding process is carried out in fluidized bed reactors, in which finelyground metallurgical silicon reacts with hydrogen chloride (HCl): Simg + 3 HCl — > SiHCl3 + H2 —

Figure 2. Global polysilicon supply and demand [3]. into electricity. Silicon is the second most abundant element in the earth’s crust. Therefore, its availability is basically unlimited. Large-scale metallurgical processes are employed to convert quartz (SiO 2 ) into raw silicon (9899% purity) for numerous technical applications. However, sophisticated production technology is needed to convert raw silicon into the hyperpure polysilicon needed for photovoltaic applications.

market entrants. However, only suppliers with state-of-the-art technology and cost structures will be competitive in the long run. Consequently, the deflation of the recent polysilicon price bubble has rendered many new projects uncompetitive in terms of quality and cost. A highly efficient use of energy and raw materials is vital to achieving

This crude TCS is purified by multiple distillation steps to reach the required purity for deposition feedstock. Following Equations 1 and 2, up to three-quarters of the TCS is converted to tetrachlorosilane (STC) and does not contribute to polysilicon deposition. Consequently, a maximum of about 18kg of STC is generated per kg of deposited polysilicon. It goes without saying that an economic recovery method is necessary for this by-product – a fact that unfortunately has not been taken properly into consideration by all newcomers in the past [4]. Established polysilicon producers are utilizing several methods of STC recovery. One option is to use STC as feedstock for the value-adding

“The deflation of the recent polysilicon price bubble has rendered many new projects uncompetitive.” Established polysilicon producers support this strong PV market growth by making huge investments to increase their capacities accordingly (Fig. 2). Unfortunately, long lead times for new capacities and exploding short-term demand may result in severe undersupply situations like that experienced in 2007/2008. The resulting price bubble has attracted many new


www. pv -t e ch .o rg


Figure 3. Integrated product flow at WACKER POLYSILICON’s Burghausen site.

Figure 4. Standalone polysilicon production. production of pyrogenic silica: SiCl4 + 2 H2 + O2 — > SiO2 + 4 HCl —


If this is the only possible route for STC, the production of polysilicon and pyrogenic silica are coupled, which can quickly lead to restrictions. Therefore, the conversion of STC back to TCS by hydrogenation is mainly used to recover the STC for the production of polysilicon: SiCl4 + H2 — > SiHCl3 + HCl —


In either case, production of pyrogenic silica or STC conversion, HCl is

generated as a by-product which can be fed back into the TCS synthesis. Besides STC, there are further chlorosilane by-products generated during TCS synthesis and polysilicon deposition. The expense of processing these by-products can be minimized by linking the polysilicon production with that of organofunctional silanes, silicates and silicones. Predetermined loss of material can be converted into value-adding products if this is done properly. Fig. 3 shows a corresponding closed-loop production system in operation at WACKER POLYSILICON’s Burghausen site.

In a standalone polysilicon production process as shown in Fig. 4, approximately 100% of the STC can be recycled to TCS for polysilicon production. The major challenge of such a plant is the optimization of each single process step and the steps’ interaction to minimize all kinds of chlorosilane by-products which are not usable for polysilicon deposition. In any case, a highly integrated closedloop production process is a prerequisite for a sustainable and economical production of polysilicon. Energyefficient processes and the minimization of residue are further technological factors that can contribute to a highly efficient polysilicon production process. The following section discusses some aspects of this task in detail.

Challenges in production of trichlorosilane As noted above, metallurgical-grade silicon is used as a feedstock to synthesize solar-grade TCS. Impurities in the silicon metal significantly impact TCS quality and the productivity of the production process. Many earlier investigations were generally limited to the influence of metallurgical-grade silicon on TCS selectivity. However, in addition to this very important aspect, there are still many other issues that influence cost, quality, safety and environmental impact. Fig. 5 shows in detail the process for manufacturing TCS.

AlCl 3 from the process have been described in different publications [6,7]. Despite the problems associated with AlCl3, a certain level of aluminium in the bulk material is necessary to achieve a high reactivity and TCS selectivity. However, an excessive amount of aluminium quickly increases the costs of AlCl 3 workup and disposal. The challenge therefore is to optimize the aluminium content and the operation mode of the reactor in such a way that the reactivity does not decrease, but less AlCl 3 is produced. The removed AlCl 3 can be treated in the same way as silicon metal residues. Due to the potential high purity of re-sublimated anhydrous AlCl3, this by-product can also be recovered and sold, e.g. as a catalyst.

Figure 5. TCS synthesis. Chlorosilanes are obtained by reacting metallurgical-grade silicon with HCl. This reaction can be carried out in a fluidizedbed reactor at temperatures of 300400°C. The reaction conditions in the fluidized bed are influenced by reactor design, particle size distribution of the silicon feed, and HCl flow. At an integrated closed loop production site, the TCS production should be continuously optimized towards minimizing the amount of by-products. The HCl first-pass yield should reach almost 100% and the TCS selectivity might be up to 90%. The result is a small loop of HCl and only a small formation of STC which saves energy in vent-gas recovery and in STC conversion. Furthermore the use of metallurgical-grade silicon with a well-selected composition of non-silicon metals (Fe, Al) reduces the production of undesired high and low-boiling chlorosilanes and metal chlorides.

Silicon metal wastes Together with the metallurgical-grade silicon, impurities such as slags, calcium, aluminium and iron compounds are introduced into the reactor. Some of these are elutriated from the reactor as fines with the product gas stream; others remain with the bulk material in the reactor where they accumulate. These are two methods of producing siliconcontaining wastes.

“If a particular impurity level is exceeded, the reactivity of the bulk material decreases.” On the one hand, dust particles are continuously elutriated from the reactor. A first dust fraction is separated by a cyclone in order to be fed back to the reactor. A second dust fraction, which is collected by filters, is even finer than the first fraction and contains a high concentration of impurities. On the


www. pv -t e ch .o rg

other hand, impurities accumulate in the reactor during the continuous TCS synthesis. If a particular impurity level is exceeded, the reactivity of the bulk material decreases, which can even lead to a shutdown of the reactor. At this point, the bulk material has to be removed from the reactor. An important feature of a properly designed reactor system is the selective control of bulk material impurity levels and silicon loss minimization, both controlled by fines discharge [5]. Benefits are high reactor productivity and low amounts of silicon-containing residues. There are two methods of disposal of such residues, the first of which is treatment with lime, which transforms the residues into innocuous substances that can easily be land-filled. However, this procedure entails a certain amount of effort and cost. The second option is to sell the previous by-products to metallurgic and cement industries – an economically and ecologically sound and reasonable option as the substances can then be used for other processing purposes.

AlCl3 removal Under TCS synthesis conditions, most of the aluminium contained in metallurgical-grade silicon reacts with hydrogen chloride. The resulting AlCl3 sublimates at temperatures over 150°C and passes through the filter system along with the gaseous silanes. However, if the gas temperature drops below 150°C, AlCl3 desublimates and is deposited on the inner walls of the equipment or the piping. Therefore, it is extremely important to have an effective method for separating AlCl3 completely from the product gas and then removing it from the process; otherwise the AlCl3 would be deposited on downstream equipment like pipes, tanks and even distillation columns. Cleaning of contaminated equipment can lead to a severe safety problem due to the high reactivity and exothermic potential of AlCl3. Methods for removing

Condensation and ventgas recovery Downstream of TCS synthesis, dust separation and AlCl3 removal, the gas stream is condensed and separated into a liquid fraction (crude silane) and a gaseous fraction that contains mainly hydrogen and a small amount of HCl and non-condensed chlorosilanes. The liquid chlorosilane fraction is stored in tanks and is subsequently distilled. The gaseous fraction is purified in a gas recovery unit. Hydrogen chloride is then fed back to the TCS synthesis, while chlorosilanes are condensed and recovered for distillation. Hydrogen can be recovered with high purity, suitable for use in STC hydrogenation. By using an appropriate vent-gas recovery, the disposal of residues can be avoided.

Chlorosilane distillation residues The distillation of hyperpure TCS separates chlorosilane fractions that are enriched with certain impurities, for example, boron and carbon-containing compounds, which can be found in low and high-boiling fractions. As a matter of principle, these fractions enriched with impurities have to be discharged from the distillation system to achieve the high quality standards required for hyperpure T C S . I f p re c a u t i o n a r y m e a s u re s are not taken, the disposal of these contaminated fractions is associated with high silicon and chloride losses.

“Small fractions with impurities have to be discarded for achieving the high purity of TCS.” Within an integrated chemical production site, all these fractions can be recycled and thus do not require disposal measures. Depending on the main component of the respective fraction – dichlorosilane, TCS or STC – and the concentration of the impurities,

customized solutions can be developed to utilize these fractions. This significantly reduces the loss of silicon and chloride equivalent, which in turn means lower disposal costs, low environmental impact and low production costs of the main product, hyperpure TCS. In a standalone polysilicon plant, the by-products formed in the TCS reaction have to be avoided. This can be done as noted by using appropriate metallurgicalgrade silicon and an optimized operation mode of the TCS reactor. Secondly, the amount of discharged chlorosilane fractions from the distillation has to be reduced to a minimum by concentrating the impurities in the discarded streams to a maximum level. The high-boiling f r a c t i o n o f c h l o ro s i l a n e s c a n b e recovered in several ways, for example by feeding it back to the fluidized bed reactor [8]. Low-boiling chlorosilanes can be chlorinated to TCS, which can be recycled and used for polysilicon deposition [9].

colleagues in R&D and the corporate departments of WACKER POLYSILICON for their suggestions and assistance in the preparation of this article.



The authors would like to thank their


References [1]

[2] [3]



European Photovoltaic Industry Association (EPIA) 2010, “Global Market Outlook for Photovoltaics until 2014”, p. 8. Photon International, March 2010 edition, p. 195. Schindlbeck, E. 2010, “Leadership in Polysilicon”, Capital Market Day Wacker Chemie AG. Cha, A.E. March 9, 2008, “Solar Energy Firms Leave Waste Behind in China”, Washington Post Foreign Service [available online at http:// content/article/2008/03/08/ AR2008030802595.html]. Hesse, K. & Pätzold, U. 2006, “Survey over the TCS process”, Proc. Silicon for the Chemical Industry VIII, pp. 157-166. Kroupa, M.G. 2002, “High Boiling Residue Recovery Process for the



Synthesis of Trichlorosilane”, Proc. Silicon for the Chemical Industry VI, pp. 201-207. Lord, S.M., “Process for removing aluminum and other metal chlorides from chlorosilanes”, US Patent 7 736 614, June 15, 2010. Fabry, L., Stepp, M. & Pätzold, U., “Wiederverwertung von hochsiedenden Verbindungen innerhalb eines Chlorsilanverbunds”, Europe Patent 1 991 501/ DE102006009954, September 9, 2007. Burgie, E.A., Heng, O.A. & Lange, T.E., “Treatment of vent gas to remove hydrogen chloride”, US Patent 5 401 872, March 28, 1995.

Want to read more? Download part two of this article at the Photovoltaics International journal archive online at Author information and contact details are also available online












 ! " # $ 

$ $

   % #        



 &  '

$' &  (   "))*+  ,



+ ,'%)$'  








     7   0  


  &    ,'   &

      & 6   & 



   ,0  '

, &




  &       ,  &


       '  &   

  ` .  '   ` /

&   0 &  ` 1     ` 2 &  &  ` 3         ` 4    `   '      &   

&  ` 5              

   $  ,   



www. pv -t e ch .o rg







    &  $    

$ $




         ,'               &   5:3 ;))<5              

   54   0



      &     #  


'   $       ,   =            







  &   $      >





 &&                             &       

,        '       

      &   & ,        


&         ( 

 &  ??*5         

      "))@5 ,( 




> ,    ( 




"   #"$%!&'(' ()*+ , ,((-*+.(.,((/ 0    1 " 2)34 5 16+(&7+78((79 1 2:   ;   .   89

$ $


P hot ovol t ai cs Int er nat i onal


Tabbing-stringing quality control challenges

Dr. Harry Wirth, Dept. Materials Research and Applied Optics, Fraunhofer Institut für Solare Energiesysteme (ISE), Freiburg, Germany

ABSTRACT Cell interconnection is recognized as the most critical process with respect to module production yield. If the process is not carefully controlled, cell cracking and subsequent breakage may occur. Many manufacturers promise breakage rates below 0.3-0.5% on their tabber-stringers, which applies for cells above 160-180μm thickness that are free from initial cracks. In real production, this figure strongly depends on materials, process parameters and throughput. This paper outlines some approaches that should be taken to avoid high levels of breakage in the cell interconnection process.

Automated cell interconnection on tabber-stringers Solar cells are interconnected to a string (Fig. 1) using flat copper wires coated with solder. These ribbons are approximately 120-180μm thick and 1.5-2.4mm wide. Two-busbar cells require higher crosssections to limit serial resistance losses than three-busbar cells. Most manufacturers still use leaded solder. Research backing the switching to leadfree options is ongoing, with the aim of preparing for compliance with future RoHS requirements for solar modules. The ribbons are applied on both sides of the solar cell, covering the silver busbars. Flux is applied to the ribbons or the cell busbars. The solvent is then dried out. The solid flux melts during preheating, somewhat below the solder melting temperature. In its liquid,

Figure 3. Schematic SnPb soldering temperature profile with preheating, soldering and cooling phase; solder is in liquid phase above dotted line. activated state, flux temporarily improves wetting by reducing oxidized top layers and by protecting the surface against new oxidation. During soldering, additional heat is applied from one side of the cell to achieve a temperature that exceeds the solder melting point by 35-40°C (Fig. 3). The molten solder must wet the busbar and completely fill the gap between ribbon and busbar. All solder joints of one cell are formed simultaneously or in a quick sequence. During cooling, the thermal mismatch in CTE (coefficient of thermal expansion) between ribbon and cell leads to

Figure 1. Cell interconnection scheme.

melting temp./range [°C]

240 220 200 180

mechanical stress. The copper with a CTE of 16.5 e-6/°C strives to contract much stronger than silicon (2.6e-6/°C). If the solder solidification point, the cooling speed, the material crosssections and the ribbon ductility are not chosen carefully, cells may crack. Decreasing cell thickness will increase the cell stress. In cells soldered on one side only, the thermomechanical stress after cooling becomes visible as bimetallic effect (Fig. 4). With finite element modelling, the stress on a silicon wafer induced by cooling can be calculated. In Fig. 5, the red areas indicated represent tensile stress, while the blue represents compressive stress. The deep blue areas indicate compressive stress caused by the joints between the copper wire and the wafer, a result of copper’s larger CTE which causes more extreme contraction than silicon. For the brittle silicon, tensile stress is critical.

160 140

Sn A

Figure 2. Melting points of eutectic solders.


www. pv -t e ch .o rg

2 Bi Sn 4

7 b3 Sn P






.5 Sn Ag 3



Figure 4. Stress visualization on unilateral soldered 160μm cell strips caused by CTE difference.

state-of-the-art machines. Automated cell interconnection is implemented as a onestep process or in two steps, with separate tabbing and stringing.

Initial cell inspection

Figure 5. Calculated normal x- and y-directed stress in cell joints after cooling.

Figure 6. Tabbing unit from Schmid (left) and combined tabber-stringer from Somont.

033- ! ! &

The solder joint over the busbar length may be continuous or interrupted. Interruptions longer than 15-20mm may generate series resistance losses, depending on the resistivity of the busbar. The tabber-stringers used for fully

automated cell interconnection work with a variety of heating technologies including contact, induction, resistive, lamp, hot air or laser soldering. Soldering is usually performed within a cycle time of approx. three seconds per cell on

The electric parameters of solar cells are always measured after their production, not least because STC power will eventually determine cell price. Module manufacturers usually check electric properties of incoming cells only randomly in an off-line procedure before string production. F o r t h e m o d u l e m a n u f a c t u re r, mechanical damage recognition in cells is a crucial task. In order to achieve safe automated handling and soldering, it is important to ensure edge integrity of the cells. Damaged edges are very likely to cause cell breakage during string and module production, leading to rejected strings, string repair or even machine down-time for cleaning. Many edge cracks will also cause power loss in the cell and ultimately in the string and module. Finally, clients may reject modules due to visible cell defects, even when they are getting the specified module power. Visual inspection systems are used to check cell edges and busbars and to align cells for stringing (Fig. 7). A complete busbar print is required for proper soldering. Aligning cells through busbar recognition gives higher precision and handling safety than using


Source: GP Solar

Source: RUV Systems

Figure 9. Ultrasonic response spectrum of an intact and a cracked cell as recorded by the RUV crack detection system.

Figure 10. EL image and contactless IR scanned image of the same cell showing several cracks.

mechanical edge contact. In Somontâ&#x20AC;&#x2122;s Rapid and Certus stringers, the cells are aligned by servo motors during the transfer to the soldering belt, based on the vision system information. Aside from broken edges and corners, less visible cracks inside the cell perimeter are also critical. Cracks usually extend into the cell surface, either on the emitter side, on the base, or both. A crack will nearly always increase the series resistance of a cell, depending on its length, its orientation and position with respect to the design current flow on the cell. If the crack occurred in wafer or cell production, and did not merit cell rejection, the electric damage will lead to a lower cell power classification. Cracks may also originate from transport and handling before cells enter the tabberstringer. Even if cracks do not immediately weaken electric cell performance as specified by the cell supplier, they threaten future cell performance and durability. Mechanical loads during module production, transport, installation and operation can lead to crack growth. As soon as cell metallization is severely affected by a crack, it will increase cell series resistance and weaken cell performance. Due to the series interconnection of cells and common diode protection schemes, the weak cell will affect two strings, which usually

account for one third of the PV module. Therefore, the module producer must prevent cracked cells entering production and must ensure that the processes, especially the cell interconnection, will not produce cracks. If cracks are detected, their origin must be determined, as the processes being used may not only be revealing cell cracks, but causing them. Small cracks pose a serious challenge to inline metrology, which has led to a multitude of technologies for initial crack detection (prior to soldering) being proposed. Crack length may grow if cells are bent. Longer cracks will then increase the bow of a cell under the same bending force. The approach of comparing the elastic response from two consecutive stress tests was followed in a crack testing device developed by Solarwatt; however, the device did not find its way into production lines. A n o t h e r a p p ro a c h c o m e s f ro m acoustics. Cracks may change the resonance frequency of cells subjected to sound excitation or they may introduce new frequencies in the response spectrum. Fig. 8 shows a reduction of the main resonance frequency by nearly 20Hz due to cell cracks, as recorded by a vibrometer. Some cracks also generate lowfrequency noise. The measurements displayed in Fig. 8 have been performed

on single tabbed cells. RUV Systems offers a system for inline or offline crack detection based on acoustic resonance measurement [1], which uses ultrasonic frequencies, thus avoiding interference with other sound sources in the production environment. The inline implementation achieves a cycle time of three seconds. Electroluminescence (EL) imaging has become a widely used tool for the recognition of defects in solar cells [2]. When a current flows through the cell, the recombination of electrons and holes emits radiation between 900â&#x20AC;&#x201C;1400nm, peaking around 1150nm. This radiation can be registered by CCD detectors with spectral response ranging from approximately 300 to 1100nm or by InGaAs detectors ranging from 900 to 1700nm. The latter offer better signal amplitudes, but have drawbacks in terms of lower image resolution and higher price. Cracks appear as fine dark lines in EL images. Cracks that interrupt metallization and disconnect a region of the solar cell will appear as extended dark areas. In monocrystalline cells, cracks usually follow the crystal axis and are easily detected due to the homogeneous EL activity. In multicrystalline cells, grain boundaries and impurities render the automated detection more difficult (Fig. 10 (left)).

60 intact cell cracked cell

amplitude [Îźm].

50 40 30 20

Source: Somont


Figure 7. Vision system for initial cell alignment and control.


www. pv -t e ch .o rg

0 50


150 frequency [Hz]


Figure 8. Response spectrum of an intact cell and a cracked cell.



stress [N/mm2] .

200 150 100 50 0 0%









strain [%] Figure 11. Stress-strain diagram of a ribbon. At present, initial EL imaging is not a common feature in automated tabberstringers, possibly a result of the cost of inline EL implementation and the requirement of cell contacting. EL also requires a dark environment. GP Solar recently presented a contactless IR inspection system for cells, the GP MICROD Vision, which is based on a line scanner that measures reflected IR radiation. In the reflected signal, material defects of mc silicon wafers do not show up, making the detection of cracks somewhat easier. The system measures on the fly with a belt speed of up to 400mm/s. GP Solar’s crackdetection software claims very stable recognition due to the exclusion of grain boundaries and other bulk material defects.

Materials Ribbon The choice of an appropriate ribbon crosssection is influenced by contradicting requirements. A high cross-section is required for low serial resistance power losses, but will increase the mechanical stress on the tabbed cell; moreover, if the ribbon width exceeds the width of the cell busbar, additional shading losses are generated. The use of ETP quality copper ensures high electric conductivity. Tinbased solder coating materials have a much lower electric conductivity than copper and will contribute little to the total conductivity. The mechanical properties of the ribbon are of huge importance to string durability from the string manufacturing all the way to module operation.

The yield strength (Rp0.2) has a direct influence on the thermomechanical stress, which is brought about by a mismatch in the coefficients of thermal expansion (CTE) of copper and silicon. The stress occurs immediately after soldering takes place, starting at the solidus temperature of the solder down to room temperature. It also occurs during module operation due to temperature variations, and, of course, in the temperature cycle test from -40°C to +85°C according to the IEC standard. Low yield strength will relieve this stress by supporting plastic deformation of the ribbon. At present, manufacturers are struggling to decrease Rp0.2 to values below 100N/mm 2 , a ‘softness’ that is achieved by thermal annealing. In the stringer, the ribbon is cut and sometimes stretched to ensure straight alignment. This ribbon handling must ensure that yield strength is not significantly increased. A second important ribbon specification is the elongation at fracture. Modules in operation will face deflections due to wind and snow loads, combined with temperature changes. As a result, the ribbon sections between the cells have to resist elongation. Typical specifications can guarantee elongation values at fracture above 25%. The measured stressstrain data in Fig. 11 indicates an Rp0.2 point at 145N/mm2. Tensile strength Rm is just below 250N/mm2 (without correction for reduced cross-section) and elongation at fracture only reaches 7%. Additional parameters influencing stress behaviour include ribbon aspect ratio (height/width) and ribbon design. Current developments are focussing on

Figure 12. Wetting angle measurement for molten solder balls on cell busbar.


www. pv -t e c h .o rg

ribbons with reduced CTE and ribbons with a layer structure. The copper core of ribbons is coated by hot dipping or plating with tinbased solders. Leaded qualities include SnPb 63/37 or SnPbAg 62/36/2; leadfree qualities include Sn100, SnAg 96.5/3.5 or SnAgCu 96/3.5/0.5. Lowmelting lead-free solders show various compositions of tin and bismuth. Noneutectic alloys display a melting temperature range between their solidus and liquidus points instead of a singular melting point. Solder coating should have a uniform thickness over the ribbon length that is similar on both sides and research has yielded best results with solder layers of 10-20μm.

Solderability During soldering, the molten material must spread over the busbar surface and fill the air gap between busbar and ribbon, which requires good wetting properties to ensure occurrence at temperatures not higher than 40°C above the melting point. The interaction between busbar surface, flux and solder is responsible for the wetting process. Wetting quality is measured by specifying the wetting angle, which can be observed by applying solder balls on cell busbars in a chamber with inert gas, as shown in Fig. 12. Small values below 30° are considered good wetting in the electronics industry, but they are difficult to achieve on solar cell metallization. Good wetting will lead to high peel forces. For good solderability, cell storage must ensure that the top layer of oxidized silver stays thin. Otherwise, soldering temperature and flux activity have to be increased and may lead to secondary problems.

Fluxing Cell soldering requires a no-clean, halogen-free flux that is deposited on the ribbon or on the cell busbars, usually occurring in the tabber-stringer just before soldering by using dilutions with very low solid content. The amount of flux needs to be carefully controlled, as excessive flux application contaminates the machine and the strings and may also cause adhesion problems in the laminated module. Insufficient flux may cause wetting problems during soldering, leading to incomplete joints. Flux can be applied on the wire by transporting it through a flux bath. The Somont Certus stringer uses a flux




Figure 13. Cell stress parameters after solder cooling. jet technology, with optional selective fluxing on just the contact side of the ribbon. Other manufacturers favour the application of flux directly to the cell busbars.

Critical parameters for cell stress As a result of CTE mismatch, soldered joints start to stress the cell immediately after cooling below the solidus point of the solder [3-7]. After the stringer step, this stress may lead to easily audible cell cracking. A multitude of parameters influence the magnitude of the thermomechanical stress and its evolution in time, depicted by the schematic in Fig. 13. Solder and ribbon

are responsible for stress relaxation by plastic deformation. Since this relaxation takes some time, the joints are at their lowest strength immediately after soldering. This is also reflected by the fact that pull tests show better results when performed some time after soldering.



cells by electroluminescence”, Appl. Phys. Lett., Vol. 86, 262108. Gabor, A.M. et al. 2006, “Soldering induced damage to thin Si solar cells and detection of cracked cells in modules”, 21st EU PVSEC, Dresden, Germany, pp. 2042-2047. Wiese, S. et al. 2009, “Constitutive Behaviour of Copper Ribbons used in Solar Cell Assembly Processes”, 10th. Int. EuroSimE, Delft, The Netherlands. Lalaguna, B. et al. 2008, “Evaluation of stress on cells during different interconnection processes”, 23rd EU PVSEC, Valencia, Spain. Micciche, B. & Dingle, B. 2006, “Understanding the causes for cell breakage during the cell interconnecting process – part I”, 21st EU PVSEC, Dresden, Germany. Micciche, B., Dingle, B. & Sidelinger, S. 2007, “Understanding the causes for cell breakage during the cell interconnecting process – part II”, 22nd EU PVSEC, Milan, Italy.

References [1]


Dallas, W., Polupan, O. & Ostapenko, S. 2007, “Resonance ultrasonic vibrations for crack detection in photovoltaic silicon wafers”, Meas. Sci. Technol., Vol. 18, pp. 852-858. Fuyuki, T. et al. 2005, “Photographic surveying of minority carrier diffusion length in polycrystalline silicon solar

Want to read more? Download part two of this article at the Photovoltaics International journal archive online at Author information and contact details are also available online


KYNAR PVDF Blown Films, a valuable proposal for the photovoltaic market By Dr. Anthony Bonnet, ARKEMA

A new type of durable fluoropolymer film Developed by the R&D department of ARKEMA, a new range of fluoropolymer films have been introduced and are now being widely used in the photovoltaic industry. Based on polyvinylidene difluoride (PVDF), this film is produced using a very flexible and cost-effective blown film technology. Patented by ARKEMA, this proprietary technology allows a mass production of quality and high durability films.

KYNAR® Film offers durable and cost-effective protection Due to its complete integration from resin to film, ARKEMA can offer a very cost-effective product. From 2011 onwards, KYNAR® resin will be produced on three continents.

Built to last! The tremendous durability of the KYNAR® Film is linked to the chemical nature of PVDF. The high fluorine content as well as the


www. pv -t e c h .o rg

strong energy between carbon and fluorine atoms induce a total non-sensitivity to UV rays and harsh weathering conditions.

Major players love it The KYNAR® Film is now a recognized material in the PV domain, serving the major players of the industry. The KYNAR ® Film comes in a range of different grades with widths ranging from 600mm to more than two meters, and is available in colors such as black or white. This product has been designed to fit perfectly with the regular roll-to-roll lamination process used nowadays in the PV industry. The KYNAR® Film is easily handled and can be used with designed adhesives.

Main features of the KYNAR® Film Based on a multilayer structure, the KYNAR® Film brings with it many advantages. Firstly, the film has permanent resistance to sunlight as well as a complete UV opacity, which means that the PET cores of the backsheets are protected for many decades from any UV degradation. KYNAR® Film also has a good barrier to moisture and delays the hydrolysis of the PET substrate, which leads to a higher lifetime expectancy of the complete backsheet. This film features other advantages like a VTM-0 rating, as well as an excellent sand abrasion resistance, which makes this film a material of choice for stand-alone panels, particularly in windy or dry areas. Another feature of note is the film’s very high total solar reflectivity of greater than 83%. Adhesion of the KYNAR® film onto suitable PET substrates is obtained using specific adhesives for the PV industry. The

KYNAR® Film requires a surface treatment in order to obtain a good and durable adhesion with both the adhesives and with encapsulating EVA resins. Corona or plasma treatment can also be used. A high level of adhesion can be obtained on PET with values higher than 1N/mm and more than 10N/mm onto EVA.

Kynar® Film ` Provide long-lasting protection from UV light; even with transparent film ` Provide outstanding gloss retention and color stability ` Resist chemicals and corrosives ` Resist stains and graffiti ` Shed dirt and resist bacterial growth ` Resist abrasion and clean easily ` Show excellent tensile and elongation properties, allowing thermoforming at high draw ratios ` Provide tough, flexible, printable surfaces.

production of resin for the coating and PV industries. ARKEMA is investing heavily to follow the growth of the PV industry.

Performance and cost-effective material Recycling is possible The KYNAR® Film is recyclable. Based on a thermoplastic resin, this film exhibits a very high resistance to thermo oxidation and can be ground jointly with a PET core layer and turned into pellets. These pellets can then be melt processed for use in other applications.

ARKEMA invests in the PV industry ARKEMA has decided to have a global presence to better serve its customers. Our new site in China will be dedicated to the

Among all other available fluoropolymers, the KYNAR® PVDF resin and the KYNAR® Film bring a cost-effective solution which can be adjusted to customers’ needs. Backsheets with KYNAR® on one or on both sides of the PET core can be found on the market under the trademarks KPE™ or KPK™.

For more information, please visit us at the 25th EU PVSEC in Hall 4 / Level 2 / A25 or contact Anthony Bonnet at

ARKEMA’s worldwide presence includes three sites of production of the KYNAR® Film, with a brand new production site opening at the beginning of 2011 in China.

P hot ovol t ai cs Int er nat i onal


Large-scale PV power plants – new markets and challenges

Denis Lenardic, PV Resources, Jesenice, Slovenia


Earlier this year, worldwide cumulative power capacity surpassed 6GW (taking into consideration only PV power plants >200kW DC). Europe still holds the largest market share at 87%, but in comparison with the figures scored in 2008 [1, 2], the U.S. increased its market share to 7%. Estimated annual installed power capacity worldwide for the period from 2000 to 2009 [3] is presented in Table 1. The data are based on detailed figures taken from more than 3,400 large-scale PV plants put into service in recent years that had cumulative peak power capacity of more than 6GWp. D u e t o a s h o r t a g e o f re l i a b l e databases and national or international sources of statistical information concerning large-scale photovoltaic power plants, the statistical data presented here should be considered “conservative”. However, some national statistical sources of note include those provided by some Italian [4] and German [5] agencies.

Annual and cumulative installed power output capacity It is estimated that more than 1.7GW large-scale power plants were constructed and put into service in 2009 (Fig. 1) [3]. Market leader Germany has more than 650MW installed (see Fig. 5 for more details), followed by Italy with about 250MW and the Czech Republic with about 200MW installed power capacity. Fig. 2 shows a Europe-wide breakdown of installed cumulative power capacity as at December 2009.

Annual installed Cumulative installed


Despite the collapsed Spanish market and the general state of the world’s economy, the past year was not a bad year at all from the perspective of installed power capacity of large-scale PV power plants. Installed power capacity surpassed expectations while also bringing new markets into the spotlight, which means that the traditional market leaders of Spain, Germany and the U.S. are no longer the only ‘key’ markets. With the exception of Germany, the past year’s most noteworthy market boost was seen in the Czech Republic and Italy, with similar shake-ups seen in the Asian tiger countries of China and India. With many large-scale PV power plants recently brought into commission in these countries, China and India are brimming with potential for the near future.

Figure 1. Annual installed power output capacity (MWp) from 2005 to 2009, taking into consideration PV power plants >200kWp. The market share of large-scale gridconnected PV power plants has been increasing continuously in recent years (see Fig. 3) in comparison with total annual installed PV power capacity [6]. In 2005, market share comprised less than 10% of the annual installed PV power capacity, while in 2007 market share of large-scale PV power plants reached almost 25% of the annual installed power capacity. In 2009, this figure was close to 30%.

New markets Significant progress was not restricted to the key markets of Germany and Italy, however. It is now estimated that the Czech Republic has about 200MW of new power capacity installed (large-scale PV power plants), with a resulting third-

place ranking after Germany and Italy. Data for the most important new markets is presented in Table 2. Belgium, one of the most interesting new markets, is a prime example of regionally-driven policy. While the Flemish region has seen more than 80MW (rough estimate) of large-scale PV power plants commissioned in 2009, other parts of the country have thus far only shown slow progress. As expected, developments in Bulgaria and Greece have resulted in the commissioning of some large-scale PV power plants. Small markets like Slovakia and Slovenia have also made first steps toward joining the prestigious “MW-range Club”. Progress in the French markets was lower than expected, but with the recent











4.2 29.1

9.4 38.5

20.3 58.8

29.8 88.6

81.9 171

111 282

211 493

626 1119

2957 4076

1748 5824

Table 1. Estimated annual and cumulative installed power output capacity [3] worldwide of large-scale photovoltaic power plants (>200kWp) from 2000-2009. 32

www. pv -t e ch .o rg

Map courtesy of U.Dewald, H.-J.Ehrig, RWTH Aachen

Figure 2. PV power capacity in MW (>200kW) installed in Europe as at December 2009. commissioning of some large-scale power plants, cumulative power capacity in France (including French overseas territories) is estimated to be more than 100MW. Some of these French

overseas territories have seen significant development in this vein, with plants like La Roseraye and La Mangassaye on Reunion Island completed this year; Guadeloupe and Mayotte are

also broaching the MW range in recent projects. PV market share figures for large-scale PV power plants in France and the French overseas territories are presented in Fig. 4. Although Europe currently retains the number one region position, other regions and countries such as Canada, China and India have also been showing quite rapid developments. Canada has recently commissioned some very largescale PV power plants. The 20MW Sarnia project has already undergone expansion, and following completion of the second stage at the end of the year, this particular plant will take pole position on the PV ranking list. Asiaâ&#x20AC;&#x2122;s most important new market is China with some very large-scale power plants under construction, while some MW-range PV power plants were also commissioned in recent months in India and Thailand. From a short- and mid-term perspective, these Asian markets are the most promising markets worldwide; however, shortterm opportunities are more plentiful in the European countries of Italy, France, Greece and Bulgaria.

Future challenges With ongoing discussions in the EU in regard to the use of agricultural land (and similar areas) for PV power plants, the industry needs to always be on the lookout for new suitable locations. Large roofs hold significant potential for roof-

Annual installed 2009 Cumulative installed








Thailand ***

49 68

35 55

14 74

3.2 2.0

14 18.2

70 75

4 4.3

1.1 8.5

* Very promising new markets for the near future; significant progress expected this year. ** Including overseas territories. *** Significant progress observed in the first half of 2010, valid also for China and India.

Courtesy of EPIA,

Table 2. Large-scale PV power plants: annual and cumulative power capacity (MW) installed in 2009 for some new markets [3].

Figure 3. Large-scale PV power capacity as a percentage share of total PV power capacity installed.

mounted PV power plants, in particular the current state of play in Belgium where about 75% of PV power capacity is delivered by roof-mounted PV power plants. California’s market share of roofmounted PV power plants is also quite high. As shown in Fig. 5, Germany’s share of roof-mounted PV power plants remains quite low – this is also the case in many other developed countries. Roof-mounted power plants will undoubtedly have a much higher market share in the future in developed countries, whereas ground-mounted PV power plants will still dominate in desert areas and countries where scope is not an issue. For ground-mounted PV power plants, sites such as abandoned wastetreatment facilities or waste-water treatment facilities should be used more extensively: it is currently estimated that about 1% of the world’s large-scale PV power plants is located on such areas [3]. Germany and some other countries have adapted abandoned military areas for the construction of large PV plants, such as Lieberose (53MW), Finsterwalde (42MW) and Brandis (40MW) in Germany, and the 1.2MW Sault plant in France and the Czech Republic’s 13.6MW St|íbro plant (see Fig. 6).

“Other very important but often ignored issues are reliable and precise yield prediction and effective power plant monitoring.”

Figure 4. Market shares for France and French overseas territories as at June 2010 (large-scale PV power plants only).


www. pv -t e ch .o rg

Several key issues will play major roles in construction and operation of PV power plants in the coming years. New markets for MW-range plants are completely different from those established European and U.S. markets that claim close to two decades of experience in the construction and operation of PV power plants. The first few large-scale PV plants in Europe commissioned about two decades ago had typical power outputs of 500kW and were usually the result of research projects. Those plants commissioned in new markets – China and India in particular – are very large-scale

to these issues is essential for accurate electricity price calculation. Possible investment price drops may end up lowering feed-in tariffs (or subsidies), and for owners and operators of PV power plants, the electricity price – especially once grid parity has been reached – will be the most important parameter for ensuring an economical plant operation. PV power plants that have implemented more precise monitoring and electricity price calculation measures will be more compatible in the long term with other renewable (and non-renewable) sources of energy.

Acknowledgements The author would like to express his very special thanks to Ulrich Dewald and Hans-Joachim Ehrig from RWTH University of Aachen for preparing the chart, and to S.A.G. Solarstrom for providing the photo of the St|íbro PV power plant.

References [1]


[3] Figure 5. Ground-mounted vs. roof-mounted (large-scale) PV power plants: estimated market share in German states (data correct as of June 2010) [3]. MW-range facilities that are anticipated to function reliably for at least the next 20 years. New markets require top-ofthe-range project planning, financing and construction quality, as failed projects could essentially lead to significant financial losses. In the worst-case scenario, they could also have negative


impact on regional renewable energy policies. Other very important but often ignored issues are reliable and precise yield prediction and effective power plant monitoring. In conjunction with some other activities such as precise maintenance cost estimation, attention



Solarpraxis 2010, PV Power Plants 2010 Industry Guide [available online at]. Lenardi|, D., Petrak, S. & Dewald, U. 2009, Large-Scale Photovoltaic Power Plants: Annual Review 2008 (Extended Edition), ISBN 978-961245-739-6. PV Resources database [(partially) available online at www.]. Gestore dei Servizi Energetici (GSE S.p.A) [available online at http://;]. German Federal Network Agency/ B u n d e s n e t z a g e n t u r, E E G Statistikbericht 2008 [available online at]. EPIA 2010, Global market outlook for photovoltaics until 2014; May 2010 update [available online at].

Ccourtesy of S.A.G. Solarstrom

About the Author

Figure 6. The StŐíbro 13.6MW project, built on an abandoned military surface, is the largest Czech power plant put into service in 2009.


www. pv -t e ch .o rg

Denis Lenardi ÿ holds a degree in electrical engineering from the University of Ljubljana, Slovenia. From 2004 to 2008 he served as chairman of the Slovene national section of the IEC »TC82« Technical Committee. He has been systematically collecting data regarding large-scale photovoltaic power plants for several years. The data that forms the basis of this article is available to the public free of charge at http://

Enquiries Cesta Revoucije 3 SI-4270 Jesenice Slovenia Email: Web:

Atmospheric deposition techniques for photovoltaics

Heather A. S. Platt & Maikel F. A. M. van Hest, National Renewable Energy Laboratory, Golden, Colorado, USA

ABSTRACT With the never-ending need to reduce production costs, interest in atmospheric deposition techniques is steadily increasing. Even though atmospheric deposition is not new to photovoltaics, and in some cases is actually required to get the best cell performance, many of the fabrication processes for photovoltaic cells are vacuum-based. Due to the diversity in atmospheric deposition techniques available, there are opportunities for applications in thin film and patterned deposition. This paper discusses some of the deposition techniques and their applications, benefits and drawbacks.

Introduction Solar cell manufacturing involves a myriad of simple and complex thinfilm deposition techniques that include sputtering, evaporation, printing and chemical bath. Solid sources such as targets or powders provide the required elements and compounds to form films via the vacuum-based techniques, and there is a broad base of knowledge available for processing materials like ZnO or Mo into the appropriate precursor shapes. Vacuum-based techniques utilizing such precursors have traditionally dominated thin-film production, and steady improvements in material handling and processing conditions have led to dramatic decreases in the cost of photovoltaic-generated electricity to the current lowest prices of US$1.74/Wp for multicrystalline modules and US$1.07/ Wp for thin-film modules [1].

â&#x20AC;&#x153;Vacuum-based techniques utilizing such precursors have traditionally dominated thin-film production.â&#x20AC;? Also key to the vitally important decrease in the price of solar modules has been the more recent integration of solution-based thin-film deposition techniques like screen-printing and electro-deposition. While appropriate chemistries for the inks and baths are not fully established and often take time to develop, many of these liquid precursors are composed of low-cost materials such as water and abundant metal salts. Developing processes like aerosol jetprinting can produce patterns directly, so the inks are used efficiently and little waste is produced. Many of these techniques can also be used to deposit inks at atmospheric pressure â&#x20AC;&#x201C; often simply in air, so there is no need for a vacuum chamber and the accompanying infrastructure. There are clearly some


www. pv -t e ch .o rg

Figure 1. Atmospheric deposition processes that have the potential to contribute to a) a standard Si solar cell, and b) a standard CIGS solar cell. The techniques already utilized in production are indicated by italicized text. attractive aspects of solution-based thinfilm deposition techniques. We will review the atmospheric processing techniques that are already contributing to photovoltaic module production, along with other promising methods that are under development. These techniques are summarized for wafer Si and thin-film CuIn 1-x GaxSe 2 (CIGS)-based cells in Fig. 1. The basic mechanics of each deposition

methodology are described and further illustrated with specific examples and references for those interested in more details.

Direct patterning techniques Screen-printing For conventional wafer silicon photovoltaics, the most commonly-used

Figure 2. A research multi-material inkjet system with Dimatix Spectra piezo inkjet heads mounted on a universal X-Y platform. This research system is part of the Atmospheric Processing Platform in the Process Development and Integration Laboratory at NREL [5]. technique to deposit contacts is screenprinting, which utilizes a woven mesh to support a mask with the desired pattern to deposit both of the contacts on the front and the back. The screen is placed in direct contact with the substrate, and paste is extruded through the screen onto the substrate. This process directly

Figure 3. An Optomec research aerosol jet deposition system mounted on a universal X-Y platform. This research system is part of the Atmospheric Processing Platform in the Process Development and Integration Laboratory at NREL [5]. exposes the substrate to a force, which can result in breakage of the silicon wafer. Typical line widths that can be obtained are in the 100 to 125Îźm range; however, recent developments in screens and pastes have enabled line widths as low as 50Îźm [2]. Overall, screen-printing is a well-established deposition method

in the photovoltaics industry. Silver pastes containing glass frit for hightemperature contact formation to silicon have been around since the 1970s. A close alternative to screen-printing is gravure printing in which the paste is transferred using a printing plate. Screen-printing can also be used



L2/H4 h


2 5t



Stand PVSEC / WC


           ! " # $  #    &   





➼âĄ?ăŚ&#x2026;á&#x2014;ąâžĽ Ď&#x17E;â?&#x2039; ŕŹ&#x161;ä&#x152;&#x152;áł?ä°¤Ý&#x20AC;ŕ§&#x152; $VLD  VDOHVFKLQD#OHVNHUFRP

Figure 4. A single crystal silicon wafer with silicon nitride anti-reflection coating and a metallization grid deposited by aerosol jet. for alternatives to the normal front grid contact structures, i.e. interdigitated contacts, although the highest efficiency has been obtained using vacuumbased deposition approaches. Screenprinting is also used to deposit contacts on other types of photovoltaics, where the problem of breakage is limited since the substrate is either rigid, i.e. glass, or flexible, i.e. metal or plastic foil. Alternatively, screen-printing can be used to deposit films with thicknesses of several microns, i.e. CIGS or CdTe absorber layers.

Inkjet and aerosol jet printing In an effort to reduce the contact grid line width, and therefore the shadow losses,

inkjet printing (Fig. 2) and aerosol jetting (Fig. 3) are being developed as noncontact replacement methods for screenprinting. In the case of inkjet printing, small droplets (1-30pL) are created in a controlled fashion within the inkjet nozzle. These droplets can be deposited onto a substrate with great precision and reproducibility. In the case of aerosol jetting, a cloud of small droplets (~1pL) is created and then transported to a nozzle via a gas stream. In the nozzle the aerosol is surrounded by a second gas stream, which confines the aerosol so that it can be deposited in a narrow area. Droplet position is random within the deposition area. Inkjet printing can produce metal grid lines with widths of less than 25Îźm

[3], and aerosol jetting can even push below 10Îźm line widths [4]. Besides the reduced shadow losses of a front metallization grid, benefits of using non-contact printing methods include thinner wafers (in the case of wafer silicon), therefore reducing the material usage. It also enables the development of alternative contact structures, i.e. doped contacts and multilayer contacts. The printing technologies permit changes to the printed pattern on the fly, which is ideal for prototyping. Multi-material printing systems enable deposition of interdigitated back contacts without the use of lithography. An example of a cell with aerosol-jetted contacts is shown in Fig. 4. The key to the success of non-contact printing methods in photovoltaics is the development of deposition tools with sufficient throughput and inks that work well with these systems and produce the desired material properties, i.e. conductivity, adhesion, line width, line thickness, contact resistance, etc. Both methods are under development and in their initial production testing phases. Several companies are developing noncontact printing tools for the photovoltaic industry that work at production throughput [6], while smaller tabletop printing systems for initial rapid solar cell application development are also available. Most of the screen-printing paste suppliers and some chemical companies are actively developing noncontact printable inks. The majority of metallization inks use nanoparticles as a precursor component; however, alternative inks exist that use metal organic decomposition precursors [3]. Non-contact printing methods can also be used for thin-film deposition, but their real strength is in the field of patterned deposits.

Bath-based thin-film deposition techniques Bath-based deposition techniques re q u i re l i t t l e u p - f ro n t e q u i p m e n t investment in that they can be carried out in almost any container that will hold liquid. In addition, many useful films and patterns can be deposited from baths that are composed of abundant and low-cost reactants in aqueous solution. Bath-based deposition techniques have broad application, and as a result a dizzying number of variations exist. Here the focus will be on electro-, electroless, and chemical bath deposition to provide an overview of how such techniques are already contributing to photovoltaic cell production and where they may be utilized in the future.

Electro-deposition Figure 5. A Sono-Tek spray deposition nozzle. This nozzle is part of a research deposition system at NREL.


www. pv -t e ch .o rg

In addition to the bath itself, the electrodeposition process requires an external current source [7]. This current is

applied to the bath via two electrodes: the negative cathode, where positive cations migrate, and the positive anode, where negative anions collect. The most common use of this technique is to reduce metal cations to metal atoms at the cathode and thus form a coating. This cathode can be anything from a metal film or TCO-coated substrate to printed metal lines. For contact formation, electro-deposition can be used in a seed and plate approach to thicken seed lines deposited by a printing technique on a solar cell. While deposition of metals by means of this process is well established, a more recent and innovative application is the deposition of CIGS thin-film absorbers [8].

Electroless deposition The very name of this approach provides a clear indication of the difference between this technique and its electrodeposition cousin. Instead of electrodes providing external current, chemical reducing agents are added to the bath to provide the driving force to produce individual metal or alloy coatings [7,9]. Not every metal or alloy can be deposited in an electroless fashion because a catalytic surface is required to drive the reduction process, so both the initial surface and the growing film must be able to fill this role. This technique can also be used to thicken previously deposited metal seed lines.

Chemical bath deposition The compound films produced by chemical bath deposition proceed via a different mechanism than the electron-based electro- and electroless techniques. The final compound should be rather insoluble and also very stable

in the typically aqueous bath, and the necessary ions must be generated slowly or the compound will simply precipitate from solution as largegrained powder [10]. The chemistry required to accomplish these stringent goals varies with the desired film, but once it is established the compound may be deposited on metals, plastics or ceramics. CdS is a widely-studied material that can be deposited via chemical bath, a technique that still produces the best quality films for buffers in CIGS solar cells [11]. CIGS layers have also been produced from chemical baths [8], and the deposition of metal oxide insulators and conductors have been widely explored [12].

Ink-based thin-film deposition techniques Deposition techniques such as screenprinting and non-contact printing can be used for the deposition of thin films, but might not be the most costeffective approach when a uniform coating is required on a large-area substrate. Alternative techniques that also have high material utilization used in production of photovoltaic devices are based on spray deposition and blade coating. Obviously, the main use for these techniques is in the deposition of thin film.

Spray coating Spray coating is a technique that can be used to deposit homogeneous films on large substrates. During spray deposition, precursor ink is pumped toward the tip of a spray nozzle, where the liquid is agitated (ultrasonically or otherwise), creating a mist of small

droplets. A gas flow can be used to carry this mist toward a substrate (Fig. 5). Many different spray nozzle configurations, shapes and sizes are available. Spray deposition is extensively used to deposit fluorine-doped tin oxide (FTO), a TCO commonly used in CdTe photovoltaics. In order to spray deposit FTO, a precursor solution is sprayed onto a hot substrate, and the desired material is acquired by means of pyrolysis on the surface. Due to the high temperature, this method can only be used to deposit a thin film on top of materials that can sustain such aggressive heating, such as glass.

“During spray deposition, precursor ink is pumped toward the tip of a spray nozzle, where the liquid is agitated.” An alternative and less aggressive method using a spray-coating technique is to deposit the inks at a lower substrate temperature, and subsequently use moderate heating to drive off solvent. When using this method, in most cases, a post-deposition processing step will be required to obtain the desired material. If this processing step includes high temperatures, rapid thermal processing (RTP) can be used to keep the exposure time of underlying layers to a minimum. Spray deposition of films other than TCOs for photovoltaics is currently predominantly used on a research level, although it can be readily scaled to production-size substrates and throughput.

Blade coating

Figure 6. A Coatema easycoater with a 6” slot-die coating blade.


www. pv -t e ch .o rg

Blade-coating techniques come in numerous variations, and slot die coating is one of the most widely used in photovoltaic production. In slot die coating, two metal or plastic plates with controlled spacing are used to create a small bead of liquid at the open end. The slot die is positioned in such a way that the bead touches the substrate surface but the applicator itself does not. The substrate is moved perpendicular to the opening in the slot die. Ink must be pumped through the slot die at a rate equal to the consumption at the substrate in order to get a homogeneous coating (Fig. 6). Slot die coating is extensively used in the field of organic photovoltaics as it allows for the deposition of extremely thin films (~100nm). Slot die coating is very versatile as it can also be used to deposit film with thicknesses of several μm, which makes it suitable for deposition of inorganic thin-film photovoltaics such as CIGS and CdTe. The thickness

of the deposited films is controlled by the properties of the ink, i.e., viscosity and wetability, as well as the slot die deposition parameters, i.e., substrate draw speed, slot width, slot-to-substrate distance and angle. As an alternative to slot die coating, knife coating can be used. Knife coating is less suitable for some applications since it is a contact deposition method, whereas slot die coating is a non-contact method.

Spin coating Like the other ink-based thin-film deposition techniques, spin coating starts with liquid ink containing carefully chosen components. This ink is applied to the substrate, and the substrate is rotated rapidly. Solvent is lost as the ink flows radially outward and thins, which increases the concentration of the remaining solids and the overall viscosity of the ink [13]. Ultimately, the ink viscosity reaches a critical point and the final solid film forms, although in many cases postdeposition heat treatment is required to decompose undesirable components. Spin coating was developed to d e p o s i t o r g a n i c p h o t o re s i s t s f o r patterning electronic components, and it is a favourite technique of the organic photovoltaics research community for depositing conducting or semiconducting polymers. Inorganic material scientists have also developed inks that can be spin coated to deposit high-quality films of materials such as ZnO [14] and CIGS [15]. While this technique is quite versatile and useful for small-scale precursor ink development, it may not be the best choice for the large-scale manufacturing regime where fast processing and high throughput are required.

Conclusions We have reviewed atmospheric thin-film deposition techniques that are already used in solar cell production, such as screen-printing, electro-deposition, and chemical bath deposition, as well as other techniques that are under development for future use. Many of these methods are simple enough that they can be implemented rapidly with low capital investment, although development of precursor inks is not always as straightforward. An excellent overview of chemical strategies for preparing such solutions for inorganic materials is available in a recent publication [16]. The ultimate goal of any efforts made toward precursor development and thinfilm or pattern deposition is to decrease cost, and based on this criterion the atmospheric processing techniques described here are not equivalent. While low up-front equipment costs are attractive, materials utilization over the lifetime of a manufacturing plant will have a more significant impact on the final


www. pv -t e ch .o rg

module cost. Spray and slot die coaters are able to deposit the majority of their inks as blanket-coated films, and the various printing techniques discussed utilize an even greater percentage of the precursor solution to create thinfilm patterns. By contrast, bath-based techniques convert a small fraction of the initial solution into thin films or lines. The photovoltaics community will undoubtedly continue to develop innovative approaches to reducing the cost of solar cells, and atmospheric techniques will play their part along with more traditional vacuum-based thin-film deposition methods. While it cannot be predicted which atmospheric processing techniques will gain wide acceptance within the industry, we are confident that at least some of them will provide stepping-stones toward the goal of grid parity.





References [1]

Solarbuzz, “July 2010 Retail Price Survey”, [available online at http://]. [2] Essemtec press release May 3rd 2010, “Micro Structures with Nano Pastes”, [available online at asp?id=10353]. [3] v a n H e s t , M . F. A . M . e t a l . 2008, “Direct-Write Contacts: Metallization and Contact Formation”, Proc. 33rd IEEE PVSC, San Diego, California. [4] Aerosol jet applications [available online at site/aerosol_jet_home]. [5] Atmospheric processing platform capabilities [available online at capabilities.html]. [6] C h u n d u r i , S . K . 2 0 1 0 , “ J e t Speed, High Accuracy”, Photon International, Vol. 1, pp. 110-121. [7] L o w e n h e i m , F. A . 1 9 7 8 , El e c tro pl a ti ng : Funda m e nta l s of Surface Finishing, New York: McGraw-Hill. [8] Hibberd, C.J. et al. 2009, “Nonvacuum methods for formation of Cu(In, Ga)(Se, S)2 thin film photovoltaic absorbers”, Progress in Photovoltaics: Research and Applications, [Early view – available online at http://www3.interscience. nal/122607261/ abstract]. [9] Shacham-Diamand, Y. & Sverdlov, Y. 2 0 0 0 , “ E l e c t r o c h e m i c a l l y deposited thin film alloys for ULSI and MEMS applications”, Microelectronic Engineering, Vol. 50, pp. 525-531. [10] Hodes, G. 2003, Chemical Solution Deposition of Semiconductor Films, New York: Marcel Dekker, Inc. [11] Bhattacharya, R. 2009, “Chemical Bath Deposition, Electrodeposition, and Electroless Deposition of


Semiconductors, Superconductors, and Oxide Materials”, in Solution Processing of Inorganic Materials, D.B. Mitzi (ed.), Hoboken, New Jersey: John Wiley & Sons, Inc. Hodes, G. 2007, “Semiconductor and ceramic nanoparticle films deposited by chemical bath deposition”, Physical Chemistry Chemical Physics, Vol. 9, pp. 21812196. M e y e r h o f e r, D. 1978, “Characteristics of resist films produced by spinning”, Journal of Applied Physics, Vol. 49, pp. 39933997. Meyers, S.T. et al. 2008, “Aqueous Inorganic Inks for Low-Temperature Fabrication of ZnO TFTs”, Journal of the American Chemical Society, Vol. 130, pp. 17603-17609. Mitzi, D.B. 2009, “Solution P ro c e s s i n g o f C h a l c o g e n i d e Semiconductors via Dimensional Reduction”, Advanced Materials, Vol. 21, pp. 3141-3158. Solution Processing of Inorganic Materials, D.B. Mitzi (ed.), Hoboken, New Jersey: John Wiley & Sons, Inc.

About the Authors Dr. Heather Platt studied the solution deposition of metal oxide thin films during her Ph.D. work with Prof. Douglas Keszler at Oregon State University. She is now developing inks to deposit metal films and lines on silicon wafers for photovoltaic cells at the National Renewable Energy Laboratory (NREL). She is extensively using direct-write deposition, spray deposition and RTP. Dr. Maikel van Hest is currently a senior scientist in the process technology and advanced concepts group in the National Center for Photovoltaics at NREL. His principal areas of research and expertise include: solution-processing for silicon and thin-film solar cell application, vacuum deposition and transparent conductive oxides. His current work focuses on the development and basic science of solution-deposited materials (metals, CIGS, CdTe and TCOs) and the development of next-generation process technology for materials and device development (direct write deposition, spray deposition and RTP).

Enquiries Maikel van Hest, Ph.D. National Renewable Energy Laboratory 1617 Cole Blvd. Golden CO 80401-3393 USA Tel: +1 303 384 6651 Fax: +1 303 384 6430 Email: Web: capabilities.html

!     4 "$    != G\0DWX3<( 4            3 X"Z      4  5  } /~1         '      8

  ,      ' 


&       G\0DW }  3<(,  &      &     $ 3(7,   5       $                       &

   0                        &      ,      

3 "Z+'''X 7   '  'XSRQW7HLMLQ)LOPV,  &   /1 ,5     '&   "))   '



   0";     '   G\0DW 3<(  }        &

%,)))33    ,' '

&  8                


      '        , '   0   &  ' 1:+   & 


       &         &  &              '  &

 G\0DW}3<(       &  

 ' ,&      

&       $& &   'G\0DW3<(}       6    &          

     ,  %,))) 3        

         G\0DW}3<(     /:     '   &    8


  8      ,   


 }&   2. = Â&#x20AC;:   615 *)**Â ,  

' 615*"; 615*Â&#x201A;%)  }  5     

$   #"$%!  *&.[.,(9

   &   &   

   '       0       

8     0

        ,  &   0    ,         $          ,G\0DW3<(}       


 " P hot ovol t ai cs Int er nat i onal


     3 \                 -   $                         ,        ' '

                  ' #       '   ' &           ,     , '  Â&#x192;            '            "  4    &   ,       

  &       &       '           

   ,              ,&      + '            & ");),      '         &      +    ,   &             , '       '     

                    '    6    8  ,     8  '          '



%    6   &


  9  &               8 &           & '     & &


      /    &

                , '           6  '  

 ,              $ &  ,     '    ,

  &  &  



!   4     '   '               ,     ' 1             '


    8    '    &

 #        &       0          ' &       

     &            &

 + , 0     



 &   '   '        



& ,    ,'       /:



  #  "))",      /:        &   ' 1 /  6  +   1/6+!       '     /:     "Â&#x201E;  12       &")")1 / 7

 3  1/73! 46

www. pv -t e ch .o rg

  1/6+       ' &


    & 9    =   & ")"),             '

& & ' &      ' '    & 

              $   ' &       +,   

                        &  '       '     /:  > 0  ,                ,   ,   &

      ,      ' '       

      '          '

 &      $ 

/:  '              ` 

8  +  }+~Â ;@% ~Â ;@%,'  &   $        '   

       ' 1:+

`  $  ,     $    ,  &

  '            '          ,  ' 0

      &    ` > 0 &                0 &


 `     ,'   /& }Â&#x201A;Â&#x201A;- ,'    &          /:         ,      


 2   '     &             /: &  ;)Â&#x201E;6    ,       

 ,      &

   +   }  +~ Â ;@% ~ Â ;@%  


8         /&  } Â&#x201A;Â&#x201A; -   

&     $   +

                       &                   ,' '



&  & "), '

         /:  &                  -  0

 &        ,   &


  /: 6  



      && '     '          54" "4     +   0       '           +   } ");, '    Â&#x20AC;:         /:   +


         &     ', '   +  }");      

'           '             ' 0

     &            +  }");, & '     ' ,   2:,     '   $        &

    &     0  

 &      $          &      '    

   $Â )Â&#x2026;5 @;Â&#x2026;5'             &      +   } "); 


    / ,1       ,  Â&#x2020;'     ,              & ,                     '

&         >          '                  4   &              '  , '     ,   &      ,'    &     '     6 

              Â&#x192;    '         ,

   .13     +        & 


    &    ,        ' Â&#x2021;

   .   ]  1   ] :    +  }         5    



P hot ovol t ai cs Int er nat i onal


Expectations for the UK solar market

Emma Hughes, News & Features Editor, Photovoltaics International

ABSTRACT On April 1st 2010, the UK government’s Department of Energy and Climate Change (DECC) officially launched its renewable energy policy. The document includes the Carbon Reduction Commitment Energy Efficiency Scheme (CRC EES), designed to improve public and private sector organizations’ energy efficiency; and the generous feedin tariff (FiT) incentive, which pays 41.3p/kWh of solar photovoltaic energy generated. This article will look at the expectations for the UK solar photovoltaics market following the government’s policy launch. The paper will focus on the impact of the UK’s late arrival to the renewable energy market; why the FiT is so incremental for successful growth; what the expectations are for the development of the UK solar PV market as well as an investigation into whether the UK is really ready for this level of change.

Householders who install small-scale solar panel systems in the UK are now eligible to receive up to £1,000 a year – tax-free for 25 years – for the electricity they generate under the new government Clean Energy Cashback scheme, known more widely as the feed-in tariff. Government figures reveal that any UK resident who installs a typical 2.5kW PV system at their existing residence will initially be paid 41.3p per kWh generated [1]. Former Energy and Climate Change Secretary Ed Miliband outlined that such a set-up would result in a reward of up to £900 in the first year on top of a £140-a-year saving on energy bills (UK elections have taken place since this paper was penned). This policy was long awaited in the UK, as its residents became more aware of neighbouring European countries’ renewable energy success. However, the optimistic figures released by the government have been met with a great deal of uncertainty coupled with a severe lack of education in the sector. It remains to be seen just how successful the UK solar photovoltaics market will be, considering its infancy.

The UK feed-in tariff The generous financial incentive that is the UK feed-in tariff has set the market off to a good start. Its 41.3p/kWh rate is high enough to really push the investment in solar energy in the country. But who is really going to benefit from this incentive? The initial costs – an average installation spend of £8,000£12,000 – are simply too high for the average citizen to afford. The £8,000-£12,000 payout secures the customer full installation – including the site assessment, the modules, the inverter and the installer’s charge, yet this is still a large sum to pay out upfront. The varying issues and concerns surrounding this aspect of the market will be discussed later. The Department of Energy and Climate Change (DECC) claims that this initial


www. pv -t e c h .o rg

Courtesy of


Figure 1. C21e and C21t tiles installed on South Yorkshire Housing Association houses in Rotherham, England. cost is justified by the rate of return on investment (ROI), which it claims is 5-8% on a well-sited 2.5kW installation [2]. It also claims that a UK solar generator could ear n up to £1,000 a year, meaning that their initial costs would be earned back in a maximum of 12 years. Considering that a solar system usually has a lifetime of 25 years, the owner then earns up to £1,000 a year for an extra 13 years, marking an estimated profit of £13,000.

Size matters Since releasing the FiT policy guidelines, the DECC has faced the question of who can actually afford to install solar power systems, as the high expense of installation prevents solar from being available to the masses. A spokesperson at the DECC was quick to assure us that low-interest finance options will soon be available to cover these costs: “There are already signs that the finance market is looking to develop products in this area so that future FiT revenues can be used to pay off upfront loans.” UK installer Solarcentury has reported a fourfold increase in sales enquiries since the FiT was initially announced in February. Thus, supposing the funds are available, the UK residential market

looks set for moderate growth. But there is a problem. In the case of solar installations, size does indeed matter. Many UK residents’ homes may not have a well-positioned south-facing roof; furthermore, in the event that a resident does have such a roof, it may not be large enough to house the 2.5kW system required for the full ROI. In this case, they will not earn the full payback, and may even lose out on their investment altogether. Taking this into account, it looks as though the residential market will see success, but it may take a while to get going. However, one aspect of the solar market that looks set to experience growth is the farming sector. Given the large building sizes on most farms in the country, UK farmers are almost guaranteed to receive the ROI that has been quoted in all of the DECC documents. There is already evidence that the solar farmland market is beginning to pick up. Jonathan Scurlock, Chief Adviser, Renewable Energy and Climate Change at the National Farmers’ Union, sees the potential for this market: “For many farmers and landowners, it now makes environmental and financial sense to consider installing solar PV on farm

Gala Networking Dinner sponsored by



This comprehensive 2-day conference including a mini

together as one industry to make sure that the 250MW

exhibition will provide exclusive UK PV industry insights.

target is achievable. An exclusive Gala Networking

Challenges of supply, cost and the rapid growth in

Dinner, sponsored by Trina Solar, will be held as an

demand from the UK have created the need for an

accompanying event to the conference, thus opening

accelerated maturation of the solar power industry.

the opportunity for further networking and business

Solar Power UK 2010 will be the first step in ensuring

development in an elite environment.

that all the major players in the UK market can come

CONFERENCE DETAILS AND PROGR AMME ONLINE // REGISTER NOW! ht tp tp: p : //w w w. solarp p owerp o werp p or /SPUK10 ORGANISED BY





Photovoltaics International


Gala Networking Dinner

buildings.” Scurlock agrees that farm buildings present a perfect opportunity for the installation of PV, and that it is most likely that the FiT will indeed drive the expected uptake. “Introduction of feed-in tariffs has resulted in a flood of interest in on-farm generation from farmers and technology providers, and substantial growth in an infant UK industry can be expected over the next five years,” he says. However, while Scurlock has already seen a great amount of interest from farmers in PV, he also admits that there is a lack of education in connection with the policy, explaining, “Some details of guidance are still lacking for applicants to the scheme.”

Following the leader Residential and farm building figures are reminiscent of the solar success of solar market-leader Germany. The early indications for the UK market – highlighting a fairly successful residential market and an even more successful farm building installation rate – draw parallels with the fully established German market. Dr Henning Wicht, iSuppli’s Senior Director and Principal Solar Analyst, says that 40% of the PV installations in Germany are residential and two-thirds of the 50% commercial rooftop installations in the country are on farm buildings. Wicht predicts that the UK market will mirror the success in Germany due to this striking similarity.

Supply and demand Modules

attractive option, as there is currently nothing else available on the UK market. Secondly, inverters have an average expected lifetime of 10 years. This is reflected in the product warranty, which on average is between five and 10 years. Based on the 25-year average life expectancy of PV modules, the customer could potentially be faced with having to replace the system inverter twice, costing well into the thousands of pounds on top of the initial £8,000£12,000 already spent. Such a scenario has to be considered when calculating the ROI, potentially extending the time it takes to earn back the initial expense. This becomes all the more aggravating when one considers the prices across Europe in countries such as Germany and Italy, which are significantly lower in comparison to the pound sterling. Due to the current lack of knowledge on these products in the UK, most installers and customers will be unaware of this price comparison.

Inverters In the UK, at the moment there is really only one inverter option widely available – the Fronius inverter. Fronius is not a market leader, yet it is charging a premium price for its inverters for use in the UK. Typically, a Fronius inverter will cost anywhere between £1,170 and £17,100 for systems ranging from 1,300Wp to 40kWp. Considering that this price is included in the initial installation expenditure, it is quite reasonable. There are, however, two other factors at play here. Firstly, the conversion efficiency loss of Fronius inverters is approximately 4-5%. If we compare this inverter with one of the market leaders, such as SMA’s Sunny Boy range, one notices that the price and conversion efficiency loss are not necessarily in synch. The Sunny Boy has a conversion efficiency loss of 2%, making it far more efficient than the Fronius model. The pricing of the SMA inverters (which are certified for use in the UK but not widely available) is around £800-£1,700. In comparison, the SMA option seems preferable, being cheaper as well as more efficient, but UK customers are being railroaded into buying the less

When the price is right

Courtesy of Sundog Energy Ltd.

At time of writing, just over a month after the government’s passing of the UK FiT rates, there is an unfortunate shortage of top brand solar equipment. In the case of solar photovoltaic modules, there are two options: cheaper, less efficient modules, or over-inflated high efficiency modules. However, the choice of which product to use usually lies with the installer rather than the customer. If the installer arrives

with low-cost, low efficiency modules, such as the 170W Yingli modules offered by one installer in the UK, the system is unlikely to be any bigger than 1.2kW based on the average size of UK residents’ roofs. The customer will then have a system installed which is unlikely to ever earn them back the £8,000£10,000 they will have paid out initially. Yet if the installer opts for the expensive high efficiency modules, such as BP Solar’s range, or the Sanyo HIT module, one can expect to pay the very high-end price for the whole installation, costing £12,000 and approximately £6/W. This, according to iSuppli, is 50% more than customers in Germany pay for exactly the same products.

Figure 3. Integrated solar PV on a new build domestic home in south Scotland.


www. pv -t e c h .o rg

It is necessary at this stage to consider why the products available in the UK are so limited, and why the prices of these products are so inflated in comparison to the more successful European markets. Many European PV leaders, including France, Italy and most significantly Germany, will experience cuts to their FiT rates in 2010. For Germany, the cuts will take effect from July 1st, meaning that the market at present is booming as projects are installed while the incentives are high. The UK market is being further impacted as global inverter and module suppliers try to meet demand for their products as a result of this market boom. On one hand, the UK has launched its financial incentives at the worst possible time, as the German market hits its peak and installations are plentiful. High quality products can be sold at low prices, as the demand is certain. For the UK, distributors are still unsure of the market’s stability, and so will charge the maximum for their products, and only supply what is not needed throughout Europe. On the other hand, this could be a positive start for the UK market. Since the UK is only one month into benefiting from financial incentives for the installation of solar PV, a rush to install at this point is not expected. As the July 1st deadline approaches for Germany, it can be predicted that installations in Europe will slow down, quite possibly by a significant amount. As this decrease starts to take effect, the manufacturers – and, of course, the installers – working in Germany will begin to turn their focus onto the UK market. By this point, the country should be more aware of the FiT incentives available, and indeed more educated in the workings of the technology needed in the UK in order to promote success.

High efficiency PV inverters. More value to the solar energy.

Photovoltaic modules perfectly sealed and bonded

IF SERIES PV inverters 50 years experience in design and manufacturing of industrial power systems allowed us to develop the new PV Inverter IF. A flexible design which combines technology and experience to suit all requirements of each installation.

Bonding of frames

Bonding and potting of junction boxes

Bonding of back rails

GENERAL CHARACTERISTIC -Power from 20 to 500 kW -Efficiency more than 96% -European directives -RS-485 data communication -Robust and reliable -After-sales service

OPTIONS -Remote monitoring system -Preventive maintenance services -Integral solution in cabinet -Guarantee extension -Commissioning -Training courses

Stabilisation of frames

Sealants, adhesives and encapsulants are only a minimal expense factor in photovoltaic modules. But they have a considerable influence on the life time and on the durability of the photovoltaic module against climatic influences. Applying the premium products of OTTO ensures you to benefit from more than 40 years of expert knowledge. OTTO – always a good connection!

Paseo del Circuito 10 · E-20160 Lasarte-Oria, Gipuzkoa · Spain T: +34 943 376 400 · F: +34 943 371 279 · ·

Hermann Otto GmbH 83413 Fridolfing • Germany Phone: 0049-8684-908-0 info@otto - w w w.ot to - c hemie.c om

Courtesy of Sundog Energy Ltd.

Figure 4. Integrated solar PV retrofit on a home near York, England.

Light at the end of the tunnel Even with some uncertainty surrounding the UK solar market, expectations for its success are high. Ash Sharma, Research Director and Report Analyst at IMS Research, predicts that the UK’s feedin tariff will significantly accelerate the market, regardless of product availability and price. “Quite simply, the UK PV market will not grow to any substantial volume without large subsidies. The low insolation levels and high PV system prices would prevent any major uptake in the UK,” said Sharma. He foresees that the UK could see 40-60MW of installed PV in 2010, in comparison with the 5MW installed in 2009 when there was no FiT available. “From a demand point of view, I think UK consumers and investors are generally quite savvy and the FiT will be viewed very favourably. Many in the country invest heavily in property and the average age of a house-buyer in the UK is much lower than countries like Germany – [those in the UK] like investing money in property and adding PV systems will be an extension to this.” Although Sharma is optimistic about the growth of the UK market, predicting that 150MW of installations is achievable by 2011, he does recognize that the lack of inverters available could slow its progress. But despite this potential pitfall, IMS Research’s widespread study of other markets makes him confident that the UK’s solar growth will soon accelerate as he has seen the evidence of how the dynamics can change rapidly once full government support is given. “All indications in the medium term are positive: the FiT is reasonably generous and supports systems up to 5MW. And although the UK is not the sunniest country in Europe, it does have a reasonable amount of free roof-space and an appetite for investment,” he said.

Independent solar expert Michael Pitcher from BFC Solutions, an associate consultancy, also acknowledges the potential of the UK solar market. “Countries like Germany and Belgium have embraced PV and are already reaping the rewards, but solar panels perform very well in the British climate so there’s no reason why the UK shouldn’t also harness the benefits of solar energy,” he said. Energy companies such as npower and E.ON have also begun to show optimism for the future of the solar photovoltaics market. By May 1st, one month after the FiT took effect, npower was already seeing an 80% increase in customer interest in solar energy. Louisa Gilchrist, solar expert for npower, said, “It’s fantastic to see feed-in tariffs generating so much interest with homeowners and the scheme should be applauded for energizing the solar industry in the UK.” E.ON has also released positive feedback from customers in connection with the use of solar energy; however, the company also recognizes that many UK customers need assistance when choosing to install a PV system, due to the lack of knowledge on the products, installers and information available. As a result, E.ON launched the SolarSavers scheme, which provides a one-stopshop service for installing and generating solar power. Andrew Barrow at E.ON said, “Now that the UK is working to cut its CO 2 emissions down, we think it is important to offer residents the opportunity to be a part of this themselves.” Energy regulator Ofgem has also just released the first official set of figures relating to the growth of the UK solar market. These figures reveal that 11.266MW of solar photovoltaics have been installed since April 1st (for the period ending August 3rd 2010). From April 1st, there were 409 PV installations in total (0.979MW); for the same period in May they reached 942 (2.290MW); in June, 1,406 (3.524MW), and by July the figures had climbed to 1,753 (4.592MW). This progressive growth is also good news for the future of solar in the UK.

The lack of education does not lie solely with the customer, but also with the installers. This is something that will have to change dramatically before the market is able to progress into a noteworthy league. Sharma reinforces this point: “This will be one of the key restraints for the UK market. A great deal of education will be needed to allow the market to develop. It is also likely that a number of integrators from Europe will set up subsidiaries in the UK to target this market.” The migration of experienced European PV companies is highly likely as the majority of installers currently working in the UK are not PV specialists. While this is not a criticism of the UK installers, as the market was not nearly large enough to warrant UK-based s p e c i a l i z e d i n s t a l l e r s b e f o re t h e introduction of the FiT, it will be a great benefit to the market if experienced PV companies begin to move into the country, setting up subsidiaries in order to accelerate the uptake.

Conclusion The UK photovoltaics market certainly has some obstacles to overcome before it matures. While the country was late in receiving its FiT policy, this incentive is now set to drive the market forward in 2010. Its generous payback, combined with further gover nment financial support, will significantly increase the amount of PV uptake in the UK. While there is a lack of product availability in the country, there are positive signs of improvement, as European suppliers start to focus on the UK’s growing market, and foreign installers begin work in the country. Industry professionals predict a drop in product cost and an increase in product availability, and also foresee that an emphasis on customer as well as company education will occur within the coming months. The significant parallels with market-leader Germany only reinforce the experts’ predication that the UK solar market is set for success.

References [1]

Ignorance is not bliss All things considered, there is still this recurring theme of education, or indeed the lack thereof. At present, there is a severe gap in awareness of the subsidies available, of what products should be used and how they should be installed.


Department of Energy and Climate Change (DECC) report [available online at: en/content/cms/news/pn10_010/ pn10_010.aspx]. ‘The UK Renewable Energy Strategy 2009’, Department of Energy & Climate Change.

Solar Power UK Conference 2010 is the first official event of the Solar Power Group, the voice of the Solar Power industry in the UK. The conference theme, "Enabling the UK Solar Market for 2011," will focus on challenges of supply, cost and the rapid growth in demand for solar in the UK. Speakers include Alan Whitehead MP, leader of the PRASEG initiative; Rob Jarman, director at the National Trust; Ray Noble of the REA and Nigel Fox from the National Grid. An exclusive Gala Networking Dinner sponsored by Trina Solar will be held as an accompanying event to the conference, thus opening the opportunity for further networking and business development in an elite environment.

Don’t miss this solar PV highlight of 2010 for the UK market and register today or before the 10th September to benefit from the Early Bird Discount!


www. pv -t e c h .o rg

^!;*;_Z %`$q!;0 $q`;3x5$*;"3 !;3" q,*q*q0z    

'  1      '      

             ' &       1  0                   8  ,1           &  '           

     #      &  Â&#x192; , 1          &

     & '    

!;*;_Z 5{;*5_ #  ??@, 1    &     '                                         $         ,

$   ,  ,      , 

  ,  &&  ,       ,


 1  &         8

         '$', 1     &      '          ,  $,       & 8

     '     0       

"`;5#!q; 1             

    ' &          

=      8       , 


www. pv -t e ch .o rg



        1         '> 0 &



0*|q,q*qZ 1                         ' '       

 "; 0 ";Â&#x2C6;;* 0 ;*Â&#x2C6; " %&& Â&#x2C6;  $   





%q3"*q!qZ             8          


3;5#* `qZ   ,          '  1    

,         $  '   


'  $   

)- &  


$         '    0  8  1   &  8            



'  1        8     Â&#x192; $ $         6    ' ,                    

!;"`;_q=%0 %` !{;3 `{ ` Â&#x2039;    $  

                   `       8    $ ` 18     &       

 ` 7 $    $   ' 

` Â&#x160;  & '        ` .  ' '     ` / &

 & $       615*"; 615*Â&#x201A;%)&  1   


'       $ $   ` 5  '   ` Â&#x2039;      &  


%`$q_!#%;3`=%q%% 1        '     '             

'    ` Â&#x2030;   '  `      ` /  ` # ` Â&#x2039;


` Â&#x2030; '    ` #       

^     )    ~ 

) (Â&#x20AC;)     z `% `! 55$*;"3 >    '        

'    ,1    #4-Â&#x2030;44      

 "))      &             '  '       /:   .+7    Â&#x2030;+6(6(Â&#x160;51(Â&#x2030;11   '   

 '$'      ,'   '  '      

%`$q!}!q! * %%q% ! 1                      '        1                 =               &

       8  Â&#x2039;


'   $&$ ,      &

  $        1  '             ,        

    '  '     &        '    '     '      1        



P hot ovol t ai cs Int er nat i onal


The PV-Tech Blog By Tom Cheyney

CI(G)S spells success: Sulfurcell, Ascent, Solar Frontier gear up for commercial push

Photo: Sulfurcell

The growing market momentum of copper-indium-gallium(di)selenide (CIGS) thin film and its CIS cousins continues to be a trend worth noting, as execution on volume-production gameplans continue to take place across one of the most promising emergent PV module sectors. At the recent Intersolar North America show, Sulfurcell’s U.S. sales rep Boris von Bormann told me that the German CIS (the “S” stands for sulfur in this case) company started shipping sample modules to select American rooftop-oriented customers in October 2009. He expects the domestic UL certifications to be finalized soon for the framed 1.3 x 0.66m panels. The modules are produced at the firm’s 75MW capacity plant near Berlin, which von Bormann said has a 2010 output goal of about 20MW, with 35-40MW projected for 2011 – all the while with continuous improvements in efficiency and wattage ratings. Clever improvements on the balance-of-systems and installation side have also led to the company’s systems achieving 98% of the power output of a crystalline-silicon-based system on the same amount of roof space. The modules may bear a “Made in the USA” label one day, since Sulfurcell (which counts Intel Capital among its investors) is considering building a fully-integrated production facility in the U.S., according to von Bormann. A CIGS company already draped in red, white, and blue – flexible module maker Ascent Solar – has ramped the first 6-8MW of its 30MW production plant in Colorado. Head honcho Farhad Moghadam told me that they have most of the tools on the factory floor for 20MW, with a few incremental/additional systems coming in the near future. In terms of laser scribe and lamination gear, there’s already more than 30MW ready to go.

He also noted that modules have been submitted for IEC certification, with the UL process “right behind it.” The IEC tests should be completed soon, which once completed, would allow the company to start shipping product to the BIPV market in earnest. An intriguing emerging sector that we discussed was transportation, where flex modules like Ascent’s could be used on automobiles to help power ventilation systems to keep the cars cool in hot climes. Another more esoteric application would involve putting the conformal PV panels on tops of big-rig trucks or even trains. Moghadam said that with larger trains, one could install as much as 5-6kW of PV per rooftop, which when multiplied by many cars is the equivalent of a commercial/industrial-sized


www. pv -t e ch .o rg

power system. He sees this as a very lucrative specialty business, with 100MW of product sold into the subsector generating the revenue equivalent of 300MW of modules sold into more mainstream applications. On the opposite side of the globe, Showa Shell subsidiary Solar Frontier is a great example of a non-Chinese Asian company trying to penetrate/dominate the U.S./Americas and European markets with a non-crystalline silicon product line. Greg Ashley, the newly hired COO/VP for the region, enthused profusely about how SF was “a strong, bankable company” with “the best technology [he’s] encountered in the last nine years.” Ashley said that small commercial shipments have been made to the U.S. already, with some “fairly large projects” to be announced soon to provide the uptake for what he called “a major commitment/allocation” of panels to the U.S. and the rest of the region. The source of those CIS (here, the “S” stands for selenide) modules is the company’s fast-growing manufacturing complex in Miyazaki, where its third production facility is ramping. Deputy GM Ichiro Sugiyama told me that equipment came into the new 900MW-rated megafactory in July, and the first modules are scheduled for shipment in Q1 2011. Sugiyama-san described the basics of SF’s own CIS process. The back-contact is standard-issue sputtered molybdenum. The absorber layer also involves sputtering, with a precursor deposition step followed by selenization. He admitted that the company does use a little bit of gallium, making the stack in effect a “CIgS” semiconductor. Because of Japan’s ban on cadmium in manufacturing, SF came up with an alternative to the usual CdS buffer layer, choosing a chemical bath deposition of zinc-sulfide hydroxide that produces the 200-300nm-thick film. The top contact/TCO relies on a more standard chemical cocktail – zinc oxide. The technologist, who’s based at SF’s Atsugi Research Center, told me that no antireflective coatings are used in the films or on the module glass. The company’s approach to scribing employs a combination of mechanical and laser methods, noting that they could go “100% laser eventually.” He admitted this would be “interesting” because of lasering’s reduced scribe losses and ability to produce lines of narrower widths. The production of a bigger, 1.255 × 0.977m module (set for market release early next year) forced the company into optimization mode, given the challenges of adjusting uniformities to the larger surface area – although Sugiyama-san would not divulge any specific uniformity or yield numbers. He did say that inline process inspection and control are prevalent throughout the production line, including for incoming glass. The company’s smaller, 1.235 × 0.641m modules are already certified for the U.S. market, and the new megaplant itself will be UL-certified by next year. Median conversion efficiencies off the manufacturing line range from 11% to 11.5%, according to the Japanese exec, and the roadmap calls for those numbers to reach 14% (as measured from “frame edge to frame edge”) by 2014. This column is a revised version of a blog that originally appeared on Tom Cheyney is North American Editor for the Photovoltaics International journal and writes news and blogs for

THE 2010 THIN-FILM ANNUAL Your definitive Hardback Guide to th the World of Thin Films in 2010

ORDER ONLINE TODAY for â&#x201A;Ź35 or buy at our EUPVSEC stand (L2/H3/D1) for the reduced price of â&#x201A;Ź25!

The Rise of Thin-Film Solar Technology is the definitive handbook and resource for professionals working or interested in this exciting sector of the solar photovoltaics industry. This volume compiles authoritative technical and market-related articles written by over 70 industry experts from the pages and databases of the Photovoltaics International quarterly and the website. A comprehensive contact directory and product reviews offer indispensable resources for anyone determined to understand and navigate the thin-film PV segment.

Sign n u p online for your our free T hi h n-Film Insights Newslet slette etter err !

Photovoltaics International

lite Recruitment Daily News

SolarLeaders Te l e v i s i o n


Contact: Photovoltaics International | Trans-World House | 100 City Road | London | EC1Y 2BP | United Kingdom | | | +44 (0) 207 871 0123


“After evaluating competitive solutions in the marketplace, we concluded that Eyelit continues to be the right choice for Solyndra’s manufacturing software needs going forward.”

“I’m impressed with Eyelit’s capability and flexibility as compared to other MES solutions I’ve used in PV.” Ascent Solar Technologies, Inc. March 5, 2010

Solyndra June 3, 2010

“We’ve worked with and reviewed other solutions and concluded that Eyelit is the best solution for our new wafer plant.” Deutsche Solar AG February 25, 2010

“Eyelit provides us with the feature set needed to expand our production, along with a critical cost of ownership advantage.”

Calisolar February 12, 2009

“We wanted a solution that meets our development needs today and one that we will need as we move to production.”

Innovalight July 11, 2008


PVI Lite Vol 5  

A booming PV market has seen a flurry of market forecasters make upward revisions to PV installation figures for 2010. Across the supply cha...