Earlier this week, Dresden-based Heliatek claimed a world organic solar cell efficiency record: 9.8%.
Although the record setting cell was 1cm2 and made on glass, the firm is entirely focussed on producing large sheets of solar cells on flexible plastic film that will be integrated into the fabric of buildings - known as BIPV, building-integrated Photovoltaics.
"We have already shown we can, in the lab, get 8.5-9% efficiency on PET foil," CEO Thibaud Le Séguillon told Electronics Weekly. Compared with glass, "we only loose a little bit of photocurrent because of optical things like reflections".
The firm span out of the Technical University of Dresden and the University of Ulm in 2006.
One of its key technologies are oligomers that convert photons to electron-hole pairs.
Oligomers are short chains of molecular building blocks, or atoms. In this case the oligomers are thiophenes which repeat five atom rings (see diagram). 
The length of the chain, and the atomic combinations attached to its ends, tune the photon harvesting capability.
The Heliatek thiophene converts most of the visible spectrum - 440-750nm.
"We still don't harness deep blue, and deep red to infra-red, so there is a path to higher efficiency," said Le Séguillon.
To effectively harvest photons, the Heliatek cell is actually two identical cells deposited one on the other - called a tandem solar cell (see diagram).
At the moment, both cells have the same absorption spectrum. "For final optimisation, we should split these absorbers," said Le Séguillon.
So why not three cells?
"We have created triple junctions. Currently, with two layers, we are harvesting almost all of the photons in the absorber range," he said.
Organic solar cells need multiple layers to function.
One of Heliatek's successes, according to Le Séguillon, is that all of its layers can be deposited at temperatures low enough to be compatible with substrates made from PET (Polyethylene terephthalate - drinks bottle plastic).
All deposition, except for the transparent conductive oxide layer (see diagram) which has to be sputtered, is through simple vacuum deposition.
............ The bracket indicates the organic vacuum deposited layers, which in total are 0.25μm thick and weigh 1g/m2
This requires a long heated source to provide the material vapour that will condense and set on the film.
And here the firm aims to benefit from the burgeoning OLED industry which uses similar techniques.
"OLED has gone there anyway, we can really profit from this. In principle, we are compatible with Gen8 [display fabs]. Gen5 is already available to OLEDs with linear source over 1m long. In Korea, they are already working with 3m sources," said Le Séguillon.
The sputtered transparent electrode is currently ITO - indium tin oxide - which may be replaced with cheaper zinc oxide.
Currently the firm works on 200x200mm PET foil and smaller glass substrates.
If you are aiming at plastic, why work on inherently more benign glass at all?
"Lifetime is significantly better on glass, and we have got to be able to differentiate between intrinsic and extrinsic effects," said Le Séguillon. "Thermal stability and life on glass comply already with building integration requirements. On polymer, I think we will solve the issues with our partners in next year."
And it cells on glass are good for the company's profile.
"It is always good to have record efficiencies, at moment we are ~1% more efficient on glass," he added.
One of the reasons that glass is more benign, is that it completely seals out oxygen and moisture that can damage the stack - although corrosion is already far less of an issue with solar cells than it is with OLEDs where delicate calcium electrodes are sometimes needed.
"Electron transport in OLEDs requires calcium-level electrodes. In photovoltaics, everything happens at 1eV lower, approximately at the aluminium level, so we don't have anything more oxidisable than aluminium," said Le Séguillon. "Put an unprotected OLED in oxygen, and you can see black spots on the cathode within minutes. Our cells without protection have a shelf life of 500 hours."
In this case, shelf life is quoted as the time taken for efficiency to drop to 80% of its initial value.
"Once they are protected, our latest cells are down to 95% of initial at over 1,000 hours at 85°C, and we have earlier generation cells now at over 6,000 hours at 85°C and still above 90% .
85°C operation is a tough hurdle for organic electronics. The company claims its stack can survive 120°C for an hour.
Ultra-violet light from the sun is ironically the other big killer of organic photovoltaics.
Although the stack is photo-stable "without any self-destruction" to light in its absorption band, and resistant to 'soft' UV (350-400nm), said Le Séguillon, "hard UV [200-250nm] needs to be filtered out. We need some kind of UV protection. This is readily available".
In 2012-14, the firm's technology road map calls for 5-8% efficient solar modules with a continuous outdoor life of 2-8 years, in 2015-17 this should be 8-11% modules with 8-20 years and, starting in 2018, over 12% and over 20 years.
First production will be of complete solar modules on PET foil in mid-2012.
The products will be for charging portable devices and designed so that they can be sandwiched between sheets of glass for added protection when required.
A second, larger, production line is slated for mid 2014, making building-integrated photovoltaics for windows and facades - although not roof-top arrays which would mean direct competition with Chinese solar firms.
The reason Heliatek thinks it can win on windows and facades is that its cells can be semi-transparent and look like tinted windows.
"We can offer 70% of efficiency with 20% transparency," said Le Séguillon.
One niche roof-top target is on lightweight industrial roofs where the low expected module weight - 0.5kg/m2 - may make organic the only practical option.
Building integrated: What semi-transparent solar cells might look like.