24 August 2010
First, I must state that I do not fully agree with all the parallels drawn between the silicon solar cell manufacturing industry and the semiconductor or electronics assembly industries. The solar cell manufacturing industry is definitely not a “drop-in” career for semiconductor engineers. The solar industry, however, is following many of the same trends and developments that the two other industries pursued as they matured several years ago.
“The solar cell manufacturing industry is definitely not a “drop-in” career for semiconductor engineers.”
The following are trends and developments now playing out in a big way as the solar industry moves out of its juvenile stage. These will give us important clues as to the success factors for the future:
- The silicon solar cell manufacturing industry is rapidly moving to lower cost countries such as China. This is actually a bit of a surprise to me as the industry is still dependent on government subsidies and I would have thought that the politicians would insist on more local production. To be clear, there will be plenty of local production as well, but the biggest production volumes will take place in lower cost countries.
- The cost of materials and equipment is dropping fast.
- The prices of solar cells and panels are coming down quickly.
- The volume of solar cells as panels is increasing dramatically.
- Competition among manufacturers will continue to intensify.
- Demand is volatile. (This is particularly true for the solar industry, until the price becomes competitive with coal, as government subsidies are subject to change along with general economical dynamics.)
- Production equipment is becoming steadily better and cheaper.
- Material science is improving significantly.
- Production and process expertise is improving very quickly.
I believe that the improvements in production and process expertise will contribute to helping the industry meet parity with coal quicker than otherwise expected. Process improvement leads to higher cell efficiency, which in turn helps drive down the cost per kWh.
“I believe that the improvements in production and process expertise will contribute to helping the industry meet parity with coal quicker than otherwise expected.”
Currently many production lines are purchased as complete lines, if not complete factories. A team of experts will commission the lines, enable them to achieve a certain level of performance and then basically say, “Don’t touch!” Even for the cell manufacturers that do purchase individual equipment and commission the lines themselves, there is currently a relatively limited understanding of the process capabilities. This is about to change, and the material suppliers and equipment vendors will contribute strongly to such process improvement.
Let’s use the metallization (firing) process for silicon solar cells as an example. The wafer’s thermal profile (time vs. temperature) has a significant impact on cell efficiency. The “sweet spot” of this thermal process depends on numerous variables such as the properties of the wafer, processes upstream of the firing furnace, properties of the firing material, the furnace and more. Furthermore, the thermal process is dynamic, meaning that over days and weeks, it drifts as the furnace lamps age, preventive maintenance introduces changes, wear and tear in the furnace, the wafer characteristics vary, etc.
Developing and maintaining an optimized thermal process will lead to higher cell efficiency. In a research project conducted by Heraeus, the cell efficiency increased by an incredible 0.51% simply by identifying the optimal wafer profile. Other case studies with actual manufacturers report 0.15% efficiency increases through thermal process optimization.
Because there are tens of millions of possible furnace setups (zone temperatures and conveyor speed), the process development will identify the appropriate process window (range of optimal profiles) and corresponding furnace setpoints. Once the process window is defined, it is a simple matter of transferring the identical process to all the production lines regardless of equipment age or brand name. Each furnace may have a different setup, but they all yield the same optimal profile on the wafers. Additionally, it is quick to adjust the process back into its “sweet spot” when it drifts. It needs to be quick as throughput is paramount to the profitability of the production lines. Actually, the process optimization does include optimizing the furnace throughput. This will, however, only lead to a real throughput gain when the furnace is the bottleneck in the production line.
”Since the production cost has not changed, virtually all of this revenue growth flows to the bottom line.”
The semiconductor manufacturing industry saw a tremendous increase in yield when the industry started maturing. To a large degree, it achieved this through better process control and improved production equipment and material. As the silicon solar cell manufacturing industry matures beyond its juvenile stage, it finds itself on the threshold of a similar development. It will move at warp speed because of the tremendous profit potential that will be unleashed. A leading furnace manufacturer recently calculated that increasing the cell efficiency by 0.10% for a modern 25 MW line would increase the revenue by more than $1/4 million per line per year. Since the production cost has not changed, virtually all of this revenue growth flows to the bottom line. This baby is growing up fast!
Editor's Note: For another solar industry executive's perspective, see "The Solar World Appears to Be Having Fun Again" by John Knowles, Chairman of DEK Solar.
About the Author
Bjorn Dahle is President of KIC, a company located in San Diego, California that makes thermal process tools for improving the efficiency and quality of solar cells during the manufacturing process.
Copyright © 2010 Novus Media Today, LLC. All Rights Reserved.