The State of the Art in Photovoltaic Technology

Martin Voelker
CRES Jefferson County Chapter

“Thin film,” “organic,” “amorphous,” “nano” – we’ve all seen announcements about new efficiency records for such new types of solar cells and gotten our hopes up. Once commercially viable, wouldn’t they revolutionize, nay: solarize! our energy landscape?

Not so fast: The man who stands between those claims and reality is Keith Emery, and testing solar cells is his life’s work. Having started at the National Renewable Energy Lab decades ago when it was still called the Solar Energy Research Institute (SERI), Emery established the PV cell performance characterization laboratory and now manages its Photovoltaic Cell and Modeled Performance Characterization group.

He spoke to the JeffCo chapter of CRES in March 2015.

Early on, it had become obvious that independent testing standards were imperative to verify that what worked in one researcher’s lab was reproducible, and would still produce the rated power reliably once cells had been assembled into modules and were exposed to the harsh outdoors: day or night, dry or wet, enduring impact from hail as well as massive temperature swings from minus 40°C (-40°F) to just below boiling at 90°C (194°F).

After testing for such defined real world stresses, it turns out that crystalline silicon continues to rule supreme. Crystalline silicon cells leave everything else far behind with around 90-95% of commercially manufactured PV; silicon as a material is easy to work, it’s technologically mature, and has a giant industry behind it which drives costs down. Today’s single layer cells all register above 15% conversion efficiency, and substantially higher efficiencies are achieved in more expensive yet profitable multi-junction cells used at utility scale in solar concentrator designs.

Initially, manufacturers of more exotic cell types were hopeful that efficiencies around 10% were sufficient to make a profit, such as thin film and amorphous silicon integrated into roof shingles. But what kills them – quite literally, looking at the bankruptcy rate in this industry – is the “balance of system cost.” This dollar figure arises once all the “packaging” is taken into account: glass enclosures, antireflective coatings, metal frames, mounting hardware, cabling and inverters.

Bottom line: unless a PV system is at least 15% efficient, it will not survive in the marketplace. The cost pressure is high, because PV competes with fossil fuels which are subsidized directly and indirectly, and because the residential market deals with additional “soft costs” for permitting, electricians, inspectors and installers. In the U.S. this typically doubles the cost of the panels. Bringing down these “soft costs” would do much more than any potential efficiency gains on the innovation side of the equation.

But before new cell types even get to a the stage where system cost can be assessed they need to prove their mettle in more basic ways, and Emery warns that until that point they’re merely “pie in the sky.” For instance, whenever one hears about “organic” or “nano” cells one needs to realize that these cells are typically very unstable and quickly degrade under real world conditions. The nano route, for instance, attempts to increase the surface area to better trap light, but that same feature turns into a bug because these surfaces will easily clog and go from ‘Great!’ to ‘Garbage!’

How well specific cell types perform and how they have improved over the years can be viewed on the famous “Record Cell Efficiency”
chart, also known as the “spaghetti chart,” generated and continuously updated by Keith Emery’s measurement lab.

It makes sense that Keith is the “Keeper of the Chart,” given that his lab is averaging about 200 cell calibrations and 250 module measurements per month. In addition, the group provides hundreds of on-demand PV measurements and solar simulation for research and industry clients.

NREL’s continuously updated efficiency chart is located here:

The above caveats about new PV technology must not obscure the fact that the industry has been booming for years and still is: prices are down and sales are up in what in the U.S. is a $150 billion-a-year market. The Dollar-Per-Watt Cost – a DOE term identifying the levelized cost of electricity – which a decade ago stood at $8/watt is now at $2/watt.

While this is good news for utilities and consumers, this low price is actually below manufacturing cost, and companies everywhere (including in China which, arguably, overbuilt their manufacturing capacity) are losing money. Inevitably, many companies will be headed for bankruptcy. Those with the deeper pockets should be able to ride it out, including most likely the two biggest U.S. players, Ohio’s First Solar and California’s SunPower, because they’re backed by Walmart and France’s oil giant Total, respectively.

But what about the bigger context, such as the U.S. Department of Energy’s goal of 1 terawatt solar by the year 2020? Unfortunately, says Emery, that goal won’t be reached for another decade (i.e. by ~2030) as current trends project only a third of a terawatt by 2018. This is bad news, meaning that PV and other renewable energies won’t be fast enough to replace a sufficient amount of fossil fuel generation to have a significant impact on CO2 emissions, and by consequence, climate change.

Then again, what keeps the U.S. from reaching terawatt scale is neither an issue of industrial capacity nor technical limits. As the example of Germany has shown, political will expressed in financial incentives are all it takes to roll out PV big time. And unlike Germany, our nation enjoys plenty of sunshine – literally ’til the cows come home.

As Keith Emery put it: “We don’t really need an engineering breakthrough; it’s all a matter of markets and of policy.”

Which is why organizations like CRES need to build public support for a swift shift to renewable energy through education and advocacy.

NREL on solar PV:
NREL on PV testing and analysis: