[Berlin, Germany]
Sample me this? Production risks are back end loaded.
Sunways CEO Roland Burkhardt’s take on Solarvalue.
First, let’s cover what went well at the Solarvalue AG (XETRA:SV7) Analyst Conference. As you can see in the above photo, the conference had good analyst attendance including man about solar Jesse Pichel, Vice President and Sr. Research Analyst, PiperJaffray. Maximillian Fischer, Solarvalue Investor Relations, provided a short overview and was followed by Solarvalue CTO (Chief Technology Officer) Dr. Julio Bragagnolo. Alas, from here things did not always go up and to the right.
Solar Grade Silicon (SGS) lab samples
Solarvalue should deliver SGS lab sample material on the order of 10’s of kilograms before year end. The SGS sample material will be from a US lab not Ruše. The same physical process is used to make the sample material but it is not identical to the production process flow.
About 250 kilograms of SGS are required to cast a typical multicrystalline silicon (mc-Si) ingot, so 10’s of kilograms of material are insufficient for verification on existing production lines using customer manufacturing process flows. Way back in Secretive solar start-up, Solarvalue AG, reveals solar grade silicon production plans through joint venture bid for TDR-Metalurgija in Ruse, Slovenia, I said:
Solarvalue will need to develop a stable process and deliver sample production SGS material to solar wafer and cell partners for qualification.
These lab samples are preproduction and only an indication of what is possible in production. One analyst asked the key question. How do you know this process will scale in production?
Cross Examination on SGS Material Impurities
Following the presentation, Research Analysts asked Dr. Bragagnolo probing questions about the quality of the SGS material, its usefulness for making solar cells, and the process. Dr. Bragagnolo engaged with the Research Analysts in a round-robin discussion of one-upsmanship regarding the impurity levels in the Solarvalue SGS material.
Mr. Pichel triggered this exchange by citing information about Timminco having 5ppm (parts per million) Boron (B) and Phosphorus (P) impurity levels or better in their Solar Grade Silicon.
First Dr. Bragagnolo clarified the high quality silicon metal (metallurgical silicon) from the arc furnace would have below 1ppm of Boron and Phosphorus and sub 5ppm of other impurities. When pressed if this level of Boron could be used to make a working silicon solar cell, Dr. Bragagnolo claimed the SGS impurity target for Boron was under 1ppm or 0.1ppm.
However, in supporting these targets, Dr. Bragagnolo cited the old work at Solarex and did not have or present Solarvalue lab analysis data for even the lab sample material and admitted Solarvalue has not fully tested the silicon made with the process. In essence, Solarvalue has performed a book calculation regarding material made with their process to date.
Production Equipment Status
Solarvalue is working with a US engineering firm to translate their process and plant concept into the detailed engineering of a number of production components that will be assembled in Ruše, Slovenia. In reviewing the process flow, an iterative collaboration between Solarvalue and their engineering partner has resulted in improving the process to 30% material conversion efficiency from metallurgical silicon (mg-Si) to SGS albeit slipping the engineering schedule.
Lab samples aside, Solarvalue’s approach of going straight to production and bypassing a pilot line results in the process being developed and fine tuned on the production line. Any process issues discovered at this point will be more expensive to correct, resulting in further production start up delays, and idle capital production assets.
If it is Thursday, it must be M1
M1 is the arc furnace being retrofit for metallurgical silicon (mg-Si) production. No schedule was provided. Solarvalue has changed plans from the M5 arc furnace to M6 and now to M1. These changes are related to Solarvalue’s challenging relationship with W&P Profil CEO Alojz Cajnko who now controls M5 and M6. M1 is the largest yet oldest of the three arc furnaces. I also believe it will be more difficult to convert to silicon metal production than M6. No wonder Solarvalue will start production with purchased silicon metal.
As a result, Solarvalue has lost the strategic benefit of using their arc furnace to produce high quality silicon metal for initial SGS production. Analysts were concerned about the old furnace equipment introducing contaminants into the silicon metal. Knowing John Mott, I believe the Solarvalue team will handle this issue during the retrofit.
Sunways
At the Sunways AG (FRA:SWW) Analyst Conference, an analyst asked CEO Roland Burkhardt about Sunways’ reasons for partnering with Solarvalue and questioned Solarvalue’s competence in silicon production as his take away from the Solarvalue Analyst Conference the previous day.
Translating and paraphrasing Mr. Burkhardt’s response, he visited Slovenia to verify the existence of the arc furnace. It did indeed exist, and he left convinced Solarvalue has the people and expertise to make metallurgical silicon. Whether Solarvalue could produce usable (for solar cells) Solar Grade Silicon via their refinement process was an open question. He summarized by saying the concept was possible given the Solarvalue team’s related experience in particular with a similar process at Solarex. Concluding, Mr. Burkhardt said, We’ll believe it when we see it.
From the Sunways perspective, the Solarvalue relationship (Solarvalue and Sunways to cooperate on solar cell development) is an inexpensive call option on Solarvalue’s ability to deliver Solar Grade Silicon useable for mc-Si solar cells.
There was much more analyst interest in the Sunways LDK deal (Sunways Reactor-Wafer reversal deal with LDK Solar). And Monday’s announcement, Sunways AG enters polysilicon production business (Deutsch), indicates Sunways is developing a portfolio approach to silicon sourcing.
Additional Solarvalue Points
- The Solarvalue process is targeting 20% initial yield of SGS from mg-Si inputs and expects to improve the yield to 30% although no timeframe was given. The balance of silicon is sold as commodity mg-Si.
- Solarvalue claims they will have about 2400 metric tons (at 20% yield) of SGS production capacity in place by the end of 2008 and produce less than 1000 metric tons of SGS in 2008. It stuck me the SGS produced in 2008 is very similar to the numbers discussed last year in Solarvalue discusses plans and process representing a one year delay.
- The final casting of multicrystalline silicon (mc-Si) ingots by the solar wafer customer is the last step of the process.
- Even though Solarvalue will use hydroelectric power from a nearby power plant, the electricity cost is not low since Slovenia net imports electricity to meet demand.
- Capex (capital expenditures) was pegged at less than €50 Million.
- No pro forma financials were presented.
While they may have been honest and forthright, Dr. Bragagnolo made a number of statements that were unsettling in their delivery. For example, Dr. Bragagnolo said:
The process will work, what is unknown and constitutes a risk is when it will work.
It’s a time issue. There’s always risk - that’s life.
Here is the Solarvalue presentation from the Analyst Conference at the 8. Forum Solarpraxis.
In Solar Light Flashes: November 15, 2007, I said I would cover:
International Solar Silicon Production – an Overview
John R. Mott, COO of Solarvalue Proizvodnja d.d., Ruse (Slovenia)
Dr. Bragagnolo substituted for Mr. Mott. I missed the later panel discussions so I may not be writing a post about this.
I don’t expect any Research Analysts to initiate coverage of Solarvalue as a result of this Analyst Conference.
GP Note: In case there was any question after the post, Solarvalue AG: Why has the stock declined?, I do not own any shares of Solarvalue AG stock and have never traded them.
Labels: Solarpraxis, Solarvalue, Sunways
Intersolar 2008
SEMICON West 2008 / Intersolar North America 2008
21st NREL Industry Growth Forum
9. Forum Solarpraxis
