High-efficiency cell production — just like trekking to the North Pole

Want to know which among the new “high-efficiency” solar cells is the “real deal?” Don’t look at the press-release numbers, suggests Coherent’s Finlay Colville — start by examining their journey in terms of process flows and production timelines.

by Finlay Colville, Coherent Inc.

March 31, 2010 – You’ve just made it to the North Pole. Well done! It’s time to write those memoires. You consult your logbook for inspiration, and what transpires is not a story about planting your flag at the North Pole, but the journey to get there: the years of planning and teamwork, tortuous detours to avoid pressure ridges, unforeseen challenges and how you overcame them through adversity to triumph.
Click to Enlarge
During the early 20th century, intrepid explorers frequently reported reaching the North Pole. Most were not believed when they had no compelling evidence from the journey to support their claims. Sometimes, there were no trained navigators in the party. Others announced arrival in super fast timescales which didn’t stack up.

Thankfully, reports of high-efficiency cell production within the PV industry involve less human attrition, but they do share a journey and an end point. During the past 12 months, announcements of high-efficiency crystalline silicon (c-Si) cells have unquestionably been in vogue, with press releases issued almost on a weekly basis. And since all high-efficiency concepts demand new equipment from alternate process flows, the supply chain eagerly digests each one in an attempt to forecast capex trends and new product development.

Consequently, this leads to questions being asked by existing and would-be tooling suppliers:

  • Which are demonstrator cells for trade shows?
  • Which have evolved from low-volume pilot lines but may have unacceptable process times for industrial implementation?
  • Which require additional capex/opex costs that simply cancel out incremental efficiency gains?
  • And finally, which are the real deal — in volume production with high yield and viable cost structures?

One way to compare the different high-efficiency cells is to examine — not simply efficiency data by way of a lookup table based upon press releases or datasheets — but what’s known about the journeys the various manufacturers took, and how willing they have been to talk (albeit without divulging anything competitive) about the technical challenges they have overcome. And this applies equally to the tool supply chain offering high-efficiency equipment and processes (both turn-key line suppliers and standalone process tool manufacturers). Is equipment being proposed really qualified for prime-time industrial roll-out, or available to see only within the pdf of a CAD file?

High-efficiency cell concepts have been understood for some time, but primarily at the lab level (with the exception of production lines realized at Sunpower, Sanyo, and BP-Solar). What’s not been obvious throughout the industry, and especially down through the equipment supply chain, is how to transition them into an industrial context with acceptable cost. What new process equipment should be used? Who offers this? How does an industrial process flow differ when 6-inch wafers are employed and high throughput is required? Does existing tooling look different within next-generation production lines? Are changes necessary downstream during module interconnection? And just when is this capex inflection point likely to happen within the industry, and with whom initially?

History provides some useful guidelines in our quest to understand the high-efficiency landscape and supporting capex changes. The period from 2006-2008 was characterized by strong market growth, demand outstripping supply, and healthy margins at the cell production stage. High-efficiency cells simply didn’t get prioritized across the board. Yes, they were on the radar of nearly all cell makers, there was strong participation within collaborative R&D projects, and a host of demonstrator cells were announced and exhibited at trade shows. But production lines didn’t match this enthusiasm, nor was immediate economic payback directly obvious. Standard cells with moderate efficiencies reigned supreme. And legacy tool suppliers were happy to fill their order books and enjoy unprecedented growth as a result.

Moreover, short-term roadmaps for c-Si cell production (championed by leading European cell and equipment manufacturers) were geared towards cost reduction mainly through economies-of-scale (GW fab-talk) while sticking to tried-and-tested standard c-Si cell types. Widely published efficiency enhancement roadmaps (incorporating all the high-efficiency cell process changes) were on much longer 10-year timelines, and often complemented by projected transition points to thinner wafers or upgraded metallurgical-grade (UMG) silicon, both of which have simply been pushed out (or postponed) as wafer raw material costs have plummeted.

Two things changed these perspectives during 2008-2009. Firstly, cell makers in China and Taiwan ramped up capacities and realized economies of scale for standard cell types, at lightning speed and with unprecedented success. Then oversupply kicked in and took effect. By year-end 2009, high-efficiency cell production had indeed climbed up the agenda with all cell makers and — perhaps more crucially — down through the equipment supply chain. Interestingly for most of the industry, this occurred by necessity, driven by market dynamics, and not by adhering to timelines spelled out by roadmaps previously advocated by the research community.

By the end of 2009, c-Si cell production could be broadly split into two camps: directly competitive with Chinese and Taiwanese standard cell types and low-cost production, or differentiated by way of an alternate high-efficiency cell concept. Today, however, just a few months on, and with oversupply being forecast to continue during 2010 and 2011, neither of these in isolation is likely to suffice for very long. Leading cell producers will almost certainly have to achieve both high efficiency and low cost to remain competitive, within timescales far sooner than roadmaps had suggested just 12 months ago. And it is this premise that forms the basis of recent high-efficiency cell announcements from cell producers whose strategies hitherto were firmly rooted in low-cost manufacturing, vertical integration, and economies of scale.

Return, then, to the burning questions raised earlier. Which high-efficiency cell concepts are the real deal? And what constitutes a meaningful capex forecast today for the burgeoning equipment supply chain? Here we refer to our North Pole analogy. When reading announcements on high-efficiency c-Si cells — both for cells produced and equipment being promoted for the new production lines — try to look beyond the end point (typically an efficiency number or press release) and into how much is known about the road taken (journey).

That said, however, the past 12 months has shown us that rulebooks are there to be broken. Also, as process tooling and standardization mature (e.g. via turn-key high efficiency selective emitter lines, new stand-alone equipment tools, or integrated front/back-end solutions) the scale of the challenge (time to market) may diminish considerably, with less in-house tool and process learning needed.

Perhaps some final words from Roald Amundsen after his 1911 Antarctic exploits will help us out: “Victory awaits those who have everything in order; luck, people call it. Defeat is certain for those who have neglected to take the necessary precautions in time; this is called bad luck.” Good luck with the journey!

Finlay Colville is director of solar marketing at Coherent Inc., Santa-Clara, CA, USA. E-mail: [email protected]; phone: +44-7802 238-775.