The SiliconPV conference brought together many of the leading researchers in photovoltaics from Europe and beyond to discuss some of the key opportunities that need to be taken advantage of for solar to stay on its path to representing a much larger share of the world’s energy over the coming years.

And discussions at the event were held with some urgency. This year’s siliconPV coincided with the publication in Science of a paper authored by many of the leading figures in the global solar industry and research community, titled “Photovoltaics at terawatt scale: Waiting is not an option.”

Drawing on this publication in his opening speech, conference chair Arthur Weeber of Dutch research organization TNO noted that the world will need to have 75 TW of PV installed by 2050, and that it should be defined by the EU as a technology of high strategic importance in order to accelerate efforts to bring full value chain manufacturing back to the region.

Weeber went on to note that industrially, solar is at a turning point with the current generation of PERC technology set to disappear almost entirely by 2025. And the n-type technologies replacing it are a big step forward, but we’ll need modules with at least 30% efficiency to reach 2050 renewable energy goals.

Industry ties

The conference proceeded on this urgent note, and a close link between the PV researchers and industry was far more evident than it has been in previous years, where the transfer to larger scale felt like little more than a dot on the horizon.

Much of the work presented this year, however, focused on processes and solutions that would be suitable for industrial use. Those where it was less obvious were invariably followed by a barrage of questions on larger device size, industrial applicability, material usage and more.

In some ways, the situation is reversed – with researchers having to keep up with moves the industry has made more rapidly than expected. Gallium doping as a solution to mitigate light-elevated temperature induced degradation (more on this later) was taken up by virtually the whole industry within about a year, and researchers are still at work examining possible broader impacts. Meanwhile, many laboratories seem still to be working on older cell formats of 166 mm or smaller, while commercial formats now are almost all based on 182 mm or 210 mm products.

Materials matter

Sticking with the trend of industrial awareness, SiliconPV this year revealed a strong focus on ending solar’s reliance on rare or expensive materials that don’t lend themselves to the industry’s model of growing scale and shrinking costs.

In Thursday morning’s session on heterojunction technology, four out of six presentations looked at ways to reduce or remove indium from the materials list, with a range of innovative solutions on offer: Tristan Gageot of CEA Ines spoke of challenges in slimming down the indium tin oxide layer from 20 nm to 15, 10 or even 5nm. And Anamaria Steinmetz of Fraunhofer ISE presented a cell that could do away with indium entirely. Both of these researchers rely on adding a nanocrystalline silicon layer to the stack to achieve this, which remains a promising approach but one that’s difficult to implement in a large-scale deposition process.

Already better documented is the need to cut silver out of all types of solar cells. Copper, or silver coated copper, have seen much progress as replacements. It was interesting to note this year that more work focused on aluminium pastes, which have the advantages of further industrial development, and that they can be used in the same screen printing processes cell makers have long used.

We’re far from done with degradation

In highlight sessions on the first morning of the conference, Wolfram Kwapil of Freiburg University described light elevated temperature induced degradation (LETID) as “the phenomenon of the PERC era, and Manchester University’s Tarek Abdul Fattah outlined the current state of the art – we know that the LETID mechanism involves hydrogen  atoms, and is affected by properties of the dopant, but little else is agreed upon.

The switch to gallium doping certainly makes cells less susceptible to performance loss from LETID, but cannot be said to eliminate it completely. Various alterations to the firing process in cell production have also been shown to improve LETID performance, but there’s still work to be done.

Fully explaining that mechanisms behind LETID has become a major goal for PV researchers. Also presented was a range of sophisticated models to explain the performance loss as well as sophisticated imaging techniques aiming to track the behaviour of hydrogen atoms in the material.

Bram Hoex, from the University of New South Wales, also shared new work on potential-induced degradation, noting that this has been an increasing problem in the past few years thanks to bifacial modules and a degradation mechanism particularly affecting the rear side. Hoex warned that newer TOPcon and HJT modules may also be more susceptible to PID than the current generation of modules and outlined a new testing protocol to detect it early on.

Tandems are the future (but still not quite the present)

In his keynote speech at this year’s conference, Rutger Schlatmann of Helmholtz Zentrum Berlin noted that PV efficiency has increased on average by 0.6% per year since the first cells were made in the 1960s. He described perovskite-silicon tandem cells as “the missing link” to keeping the technology on this trajectory, further noting that the knowledge to do this was very much present in Europe.

A later session on research into tandem cells revealed more of the earlier mentioned closer link to industry. Much of the work here focused on further optimization of silicon cell processing for integration with another top cell, with the addition of a nickel-oxide layer mentioned by a few researchers as a potentially useful innovation.