Sharp Corporation's next-generation solar cell technology that enters volume production is expected to be based on its long-term development of both c-Si and a-Si thin-film cells with efficiencies touted to be over 25%.

The hydrogenated amorphous (a-Si:H), N-type monocrystalline silicon (c-Si) heterojunction back contact (HBC) technology was announced with conversion efficiencies of 25.1% by Sharp in Japan in late April 2014, relatively close to Panasonic Corp’s announcement earlier that month of 25.6% efficiency for its latest HIT cell development, also along the hybrid lines now pursued by Sharp. 

Sharp had demonstrated the combination of a back contact structure with heterojunction architecture in early 2012 that had a conversion efficiency of 21.7%. 

Since then rapid incremental improvements have been made, culminating in the a-Si:H layer deposited on both sides of the wafer generating over 25% conversion efficiencies. 

Interestingly, both PV manufacturers have now surpassed the 25% record set by researchers at University of New South Wales (UNSW) back in 1999, though then the production costs would have been significantly higher than the competitive levels these hybrid HJ designs are achieving today. 

Junichi Nakamura from Sharp’s Energy System Solutions Division, responsible for developing high efficiency silicon solar cells, has been Undertaking Sharp’s PV technology conference circuit this year to tout the company’s breakthrough efficiencies.

Nakamura presented on the latest conversion efficiencies of its hybrid heterojunction cells in June 2014 at the 40th IEEE Photovoltaic Specialist Conference and recently followed up with a similar presentation at last week's EU PVSEC event in Amsterdam.

Details of the Sharp cell include short circuit current density (Jsc) of 41.7 mA/cm2 and open circuit voltage (Voc) of 736 mV. The high Voc is typical of cells benefiting from the surface passivation qualities of a-Si:H layers deposited on both sides of the wafer, while a fill factor (F.F.) of 0.819 was achieved, indicating series resistance issues or high leakage current characteristics had been contained, a problem regularly encountered with such cell architectures. 

The processing steps Nakamura focused on in his latest presentation suggested strongly that the hybrid HBC technology was now robust and ready for commercialisation. 

Electronic backsheet

Another indication of the manufacturability was shown in the use of a printed-wiring backsheet connecting the cell electrodes directly to the backsheet wiring lines. Not only does this reduce losses from the cell but provides the key manufacturing step of handling low-cost ultra-thin N-type mono wafers. New handling automation equipment and lamination tools would be required but these are already available and proven in the electronics surface mount (SMT) sector. 

Next step

However, until EU PVSEC neither Sharp nor Panasonic had been forthcoming in detailing when the latest cell developments would enter volume production. 

After the session in which Nakamura gave his presentation, PV Tech asked him about the technology entering production soon. 

“Production feasibility studies are already underway,” noted Nakamura. “No decision has been made yet to start production.”

However, he said he hoped a decision would be forthcoming soon.

Although HBC cells are still considered complex and high cost, significant efforts across the technology supply chain have been focused on reducing process complexity, boosting cell efficiency in anticipation that the technology will gain wider adoption and enter volume production. 

Companies such as Sharp would benefit from the higher efficiencies by meeting increasing demand in certain residential and commercial rooftop markets that require both improved efficiency in low-light conditions as well as overall higher efficiencies due to limited rooftop space, notably in Northern Hemisphere markets with reduced feed-in tariffs and increasing emphasis on self-consumption. 

Nakamura concluded his presentation by noting that the 25.1% cell efficiencies had been achieved in only 3.5 years of development with 26% efficiencies possible with current processes.