Solar Reality Check

By Michael P.C. Watts
There are plenty of papers at conferences talking about new patterning, materials and process approaches to solar cells. While listening, I wondered if there was any realistic chance for a new technology. The solar cell business is exponentially growing but has become rather gruesome over the last year or so…is there any real opportunity to insert new technologies ?

This is the first of a series of blogs to try and answer this question. If you hate serials, the complete analysis is available on my web site at www.impattern.com.

If the goal is to be a leader in the solar business, what is needed ?
Understanding the opportunty starts from the underlying drivers of the industry. The “Moore’s law” for solar below, shows how pricing has scaled with experience — roughly halving for each order of magnitude in cumulative production, without any revolutionary technology drivers.

Historical solar module pricing trends showing the price dropping with manufacturing experience, from IRENA

The pricing trend is almost exclusively an experience and volume manufacturing story. It shows the hiccup in 2009 when polysilicon supplies were limited and prices rose, letting thin film in. At the same time, as prices have been dropping, the market has been growing at 30% CAGR, and will probably do so through 2020. The cell suppliers may be whining about recent price trends, the reality is that there is nothing unexpected about these trends.

The vast majority (>80%) of this capacity is conventional polycrystalline silicon (polySi), with very little technology differentiation. Everyone sees the same trend and and they are expanding, so total available capacity is roughly 2x demand. The bottom line this is an undifferentiated, overcapacity, rocket ship to a huge long-term opportunity— very difficult to survive unless you have huge resources and unlimited stamina. The recent high profile failures at Suntech CI, Solyndra US, and Bosch GE are all examples of the challenges. The fact that Samsung are committing to solar should be a warning to all…they have an impressive record in Flash, LCD, and LED’s.

The graph above also shows price data for CdTe thin film cells. At the same installed capacity, and hence the same learning and volume savings, thin film is 3x lower in price than polySi. The year is also shown, in 2011 the prices of thin film and polySi were very similar. My takeaway is that if a short term advantage can be developed to support a larger commitment, there is a long term sustainable advantage to thin film cells.

The market is roughly 50 GWatt in 2013, and the market leaders have 5% market share or 2.5 GWatt. I would submit that a company needs to be 1/10 of the leaders to be a significant player, and to build 0.25 GWatt requires 3 to 4 production lines. If the learning is kept in-house, the graph above suggests a 2X cost disadvantage over the leader. Therefore, I would submit that to break in to the leaders you need 3 to 4 lines and significant technology or cost advantage just to level the playing field.

What opportunities are there to replace polySi ?
There are 3 ways to displace polySi: Improve internal efficiency, increase wavelength range of solar absorption, decrease cost by manufacturing with the highest possible productivity.

The polySi cell has an efficiency of around 20% and a theoretical maximum of 30%, so improvement of 0.3X is a possibility.

Silicon theoretically could utilize 30% and so misses about 70% of the suns energy, so there is up to a 2X opportunity in use more of the suns output.

The polySi cell is manufactured on 150mm wafers. Thin film cells use much less semiconductor material, and can be manufactured as meter sized cells resulting in significant cost savings that can be sustained as learning improves with more installations.

The bottom line is there are opportunities. Next week I will talk about paths to take advantage of them.

Anyone who hates serials can read the whole analysis today at my Web site www.impattern.com.