While the global PV market has been through sharp pricing and demand peaks and troughs over the last five years, there are signs that the high-conversion-efficiency market segment is benefiting from stable demand and returning relatively stable margins. LONGi CEO Zhen Guo Lee spoke exclusively to pv magazine in Tokyo to share the companys dramatic predictions for cost reductions in mono wafer supply.
pv magazine: Japan has often been described as a market with a demand for high efficiency PV modules, what impact does that have on your supply of mono wafers to the market?
Zhen Guo Lee: The Japanese market is very important for us. The leading markets for LONGi are Taiwan, Korea, the U.S. and Europe. Japan accounts for approximately 10% of our exports and we export around 75% of our total production.
To what do you attribute this high level of demand for mono products in Japan?
The Japanese residential rooftop market demands high efficiency. The new build market segment is particularly strong. As high as 80% of new houses in Japan now come with PV, that translates to around 10,000 p.a. This will soon grow to 20,000 p.a. We are confident that the residential market in Japan will continue to grow, but the utility scale market in Japan is still the major source of demand.
One trend in PV manufacturing that is gathering momentum is the rollout of PERC technology. How does that affect mono wafer supply?
With the same equipment investment, the efficiency payoff with PERC application with mono wafers is much higher. While with multicrystalline, the efficiency increase is only 1%~3%, with mono something like a 5% increase can be achieved. This can deliver 290 W modules, while multi PERC can only reach around 270 W.
Looking to the other side of the mono supply business, we know that polysilicon prices reductions have been a big driver of overall PV costs falling. Increases is fluidized bed reactor polysilicon production is helping to push prices down even further. Are there issues in securing poly of sufficiently high quality for mono ingot and wafer production?
As high as 85% of polysilicon being produced today can be used for mono ingot pulling and then wafer production, so it is less of an issue than some think. Another assumption the solar industry often works under is that mono ingot pulling is much more expensive than it is for multicrystalline [ingots]. This is changing.
Several years ago mono ingot costs were high. Four years ago the non silicon costs for mono ingot pulling was as around $30/kg, today it is below $10/kg. My prediction is that in 2018 it will be around $5/kg which is very close to the cost of multi ingot production today.
Sawing costs, for wafer production, are also coming down fast. Five years ago wafer sawing for mono wafer production cost $0.50/piece. Today it is $0.25/piece and in 2018 it will be less than $0.15/piece.
What are the reductions in sawing cost a result of?
Diamond wire technology has made a major impact. And for multi wafer production, diamond wire is much more difficult to apply, therefore mono has a competitive advantage. The problems multi wafer production can encounter is wire breakage and problems with texturing the wafers. Diamond wire sawing was used in 60% of mono wafer production at the end of 2014, by the end of this year it will be 95%.
Which diamond wire equipment and material supplier do you work with?
LONGi is mainly using sawing equipment and diamond wires from Japan at present. We are evaluating diamond wire sawing technologies from EU and the U.S.
It is no surprise that you are pretty bullish about the potential for cost reductions in mono wafer production, but do you expect market share to grow? If yes, what is the reason?
Yes I do. Today mono wafer is about 25% of the total PV wafer supply. In four years time this will grow to as high as 50%. There are two major reasons for mono market share increase. Firstly, production cost of mono module will decrease rapidly, and it will equal to that of multi modules in 2018. Secondly, the operating temperature of mono module is relatively lower under natural light illumination, which result in more power generation per Watt. Therefore, mono PV system will bring more ROI to investors.
Weve spoken about PERC, but what about other efficiency boosting or high efficiency technologies like heterjunction solar cells. What impact will this have on the wafer market?
With heterojunction technology the cleaning and texturing of wafer surface are extremely important. This means that producers have to remove organic particles and metallic contaminants sufficiently, and get the cell texturing process right. LONGis R&D Surface Cleaning Group has been working on several projects to meet these challenges. We are doing this because we believe high efficiency concepts will increase the demand for mono wafers in the market.