While solar cell manufacturers are finally expanding their production capacities, they are still reluctant to upgrade to the next level of cell technologies. This may not be apparent from the 2015 International Technology Roadmap for Photovoltaic (ITRPV) report. According to the 6th edition of the roadmap – which was released in spring 2015 and updated in July – in 10 years, the current standard back surface field (BSF) based cells would only account for about 20% of mass production, while PERC type cells would dominate, leaving sufficient space for other advanced cell architectures, such as back-contact cells, PERT, hetero-junction and the very high-end silicon based tandem cells. While this seems to be a pretty positive outlook, the 2014 ITRPV roadmap expected these levels to be attained 1 or 2 years earlier.
That does not mean developments got stuck. In fact, average cell efficiencies for all p-type solar cells categories in 2014 have exceeded the levels expected in the 2014 roadmap – by 0.3 to 0.5% absolute. It means that developments continue to be mainly driven by optimization of existing processes and practices – like it has been in the last few years. Companies have tried to delay capex spending as long as possible after the last round of expansions resulted in huge overcapacities.
P-type increasing beyond expectations
Passivated Emitter Rear Cell (PERC) technology seems to be the primary beneficiary as companies have to add few more machines to upgrade a traditional line to produce PERC cells. However, the efficiency potential of PERC – around 0.5% absolute – cannot be found in ITRPV’s document. Because, the 2015 roadmap has coupled PERC with traditional BSF cells for its efficiency evaluation. It rather differentiates according to wafer types – monocrystalline, multicrystalline high performance multicrystalline and quasi monocrystalline. This entire group of p-type cells have exceed the ITRPV group’s expectations. While the 5th edition anticipated multicrystalline cells to reach 17.7% in 2014 and incrementally improve to 18.8% by 2021, the 2015 roadmap confirms 18% has already been achieved last year and a further 1% absolute gain is expected in the next 4 years. This is about 0.3% to 0.5% absolute higher than the levels forecasted for high performance multi last year. However, solar cells based on casted crystalline materials are now expected to surpass the 20% efficiency mark for the first time by 2022, which means that high performance multi would reach that level nearly 3 years earlier than projected last year. P-type monocrystalline cells, that were at 20% last year, would gain by 2% absolute efficiency in the next 10 years.
PERC getting delayed
Although it cannot be identified by how much, a portion of this performance gain can be attributed to manufacturers now quickly embracing PERC architecture for their p-type cells – for both current and new capacity additions. The 2015 roadmap also expects PERC technology to gain a considerable market in the near future – about 10% in 2015 and to double its presence to 20% in the next 2 years and then attain a considerable share of about 35% by 2019. However, the previous roadmap has anticipated these levels to be reached 1 year earlier. While the 5th edition expected PERC to contribute 50% of cell production in 2024, the latest study forecasted no significant gain from 2019 onwards. Now ITRPV believes that the standard BSF cells will lose by about 60% absolute – from today’s 80% to 20%.
N-type hype over?
When it comes to cell structured based on n-type, three types of cell architectures were looked at – standard n-type, hetero-junction and back contact cells. ITRPV was over-optimistic about the efficiency development of standard n-type cells. The 2014 estimate was – with an efficiency base of slightly above 20% in 2013 – touching 21% in 2014 and ultimately surpassing 24% by 2024. The current roadmap estimates efficiency improvements for standard n-type structures by 1% in 2017 over the last year’s base of 20% and by another 1% absolute by 2019. The n-type cells would reach 23% in 10 years, with its market share expected to grow consistently, from basically 0 to about 10%. In the 2014 ITRPV roadmap, n-type cells were not given an own category for market shares, while heterojunction structures were not part of the efficiency roadmap.
The high-end of n-type
Hetero-junction structures have the second best efficiencies in the 6th ITRPV roadmap, progressively increasing from 21.5% in 2015 to 23% in 4 years, and finally reaching 25% by 2025. The market penetration of this cell technology, which is currently only offered in large quantities by Panasonic from Japan, closely follows the footprints of PERT (Passivated Emitter, Rear Totally Diffused) cells, from about 3% in 2015 up to close to 10% in a decade.
However, back contact cells, primarily produced by SunPower Corp., will remain the efficiency champions. From an already very high level of 23% in 2015, their efficiency is expected to increase to 24% in 2019 and to 26% in 2025. As for their market shares, ITRPV has considerably reduced its forecast over 2014, now estimating 4% in 2015 and more than doubling to close to 10% in 2019. From here, another 6 years are required to record the same growth, which would lead to a 20% share in 2025. Interestingly, silicon tandem cells have for the first time entered the technology roadmap. This species, promoted by Martin Green from the University of New South Wales (UNSW) in Sydney at conferences for several years, are expected to enter mass production in 2019, but would remain a niche, capturing a share of 5% by 2025.
Bifacial cells getting visible
While all the above cell concepts are light sensitive on one side, ITRPV foresees bright prospects for bifacial cells, which can absorb light from both sides. Starting from 2014, the 2015 ITRPV publication expects that the market opens for bifacial type cells, steadily becoming prominent by 2017 with a share of about 10% and then attaining 20% in 2025. The 2014 roadmap assumed bifacial technology would account 15% of cell production in 2024.
You can argue about the category bifacial as they are also n-type cells. However, since not all bifacial cells will be integrated into bifacial-type modules – which essentially have a glass-glass configuration or employ transparent backs sheets – the proportion of bifacial modules will be smaller than for cells, notes ITRPV.
In fact, bifacial cells seem to be the only advanced cell architecture that enjoys an increased market acceptance, more than what was anticipated in last year’s roadmap. All other cell technologies – including the most simple PERC structures – are spreading at slower rates than expected. And indeed you could see that at recent solar trade fairs – every module manufacturer had a glass-glass module on display, and several of them in combination with bifacial cells. On the other hand, efficiency improvements, especially for all p-type cells is taking place at much faster pace. That means cell makers continue to be able to increase the performance of the PV device just through optimization, which justifies their reluctance to invest into technology upgrades.