Multi vs. mono Part 10: Why is multi cell lagging behind mono in PERC process, LID and LeTID


As addressed in previous articles, high efficiency cell process technologies are mainly aiming at improvement of three prospects:

1, Cell surface passivation, to reduce light induced carrier recombination;

2, Light trapping process, to reduce reflection;

3, Additional doped layer at back, to increase Voc and Isc;

PERC process is so far a good complementary strategy on top of traditional AL-BSF cell lines to realize these improvements. It’s relatively cheap and can be quickly installed. In practice cell on silicon substrate, no matter mono or multi, with above techniques sufficiently applied, would be able to reach efficiency over 20%. However what we see is that PERC process has been widely applied among mono cell lines since 2015, while multi still largely stagnant. One of the obstacles is efficiency degradation at both cell and module products.

While people have highly recognized the impact of LID, we might have ignored another effect LeTID (Light elevated Temp. Induced Degradation). Researches show that LeTID is somewhat a multi PERC specific problem and its mechanism and solution remain yet unknown.

LID of mono PERC cell was initially large but measures have been developed to effectively alleviate it. Cause of LID effect was originally composed of two facts, one is boron-oxygen recombination effect resulting in deterioration of MLT, the other recombination of boron-iron with the same result. Large improvement of polysilicon purity has helped almost diminish boron-iron forming. But B-O effect still remains as a major threat to cell efficiency, due to the fact that oxygen content in mono wafer hasn’t largely declined over past years. People have developed measures at cell process rather than wafer to migitate the impact. Basically it’s a technique with combination of light illumination and annealing process. Hydrogen atoms are driven from passivation films at cell surface into silicon to form B-H recombination, which doesn’t capture minority carriers. As result, a passivation effect is generated and LID largely reduced.

Data shows that the best LId reduction equipment (from Europe) can lower LID by 80%, from 5-6% down to 1-2%. Recently there are some factories appearing on the market claiming that they’ve totally eliminated LID effect of mono PERC cell, yet we might need more time to watch it.


LeTID effect of multi PERC cell. Researchers found that when multi PERC cells got higher temp. after light illumination, a new type of degradation would appear which is larger (6-10%) than not only traditional AL-BSF (1-2%) but also mono LID (5-6%). In addition, there are some features of LeTID which make it a greater headache than LID to deal with. For example, it takes much longer time than LID for LeTID to appear under monitor devices (100-300 hours), and more baffling is that longer time again of illumination (400-800hours) would make degradation back to recover. It yet remains a myth for most research institutes regarding what’s caused this effect, as researchers soon found it not correlated to B-O recombination, nor to surface passivation features. As result, neither process technique nor equipment have appeared on the market to resolve LeTID problem, convincingly.

It’s reported that the two cell makers, Hanwha Q-Cells and Solarworld, have developed proprietary techniques to tackle LeTID effect on multi PERC cell (or, in general any multi cell, I haven’t got clear message by completing this report). Both companies have been silent regarding details of the technology. Hanwha Q-cells have been advocating its high efficiency cell process Q.ANTUM, which seems unlike a LeTID specific technique as it covers all c-Si wafers including direct wafer. As for Solarworld, it announced focusing on mono cell PERC technology back to Feb. this year. And only three month later it declared bankruptcy.
It seems multi cell factories need more time to fully adapt to PERC process. Or on the other hand, it’s made it more important for them to search any alternatives, for example black silicon, which would largely realize PERC functions and yet bring up more advantages to cost prospect(integrated with DW slicing).

By Xiaodong(Bond) Wang. Xiaodong is a solar energy enthusiast and a regular contributor.
Disclaimer: The views above are strictly those of the author and do not reflect those of any other organisation or individual or this publication.