In 2006, global c-Si PV production capacity amounted to 5.5 GW, roughly 10% of todays production capacities. Multicrystalline silicon solar cells had, on average, conversion efficiencies of 14% to 15% at that time, whereas monocrystalline cells achieved an efficiency in the range of 16% to 17%. One of the pioneers of the PV equipment industry, located in Hohenstein-Ernstthal, Germany, was a company called Roth & Rau, which has since been acquired by Meyer Burger of Switzerland. Roth & Rau had invented a PECVD tool for the deposition of SiN:H on the front side of the solar cell to provide the anti-reflective coating. This tool became known in the industry as the SiNA-machine.
In those early days the engineers at Roth & Rau were convinced that higher conversion efficiencies of solar cells would be required in order to drive down the cost of electricity generated from solar power. In order to manufacture higher efficiency cells, it was clear that additional process steps would have to be applied to the cell coating process. A tool that was designed to enable various different treatments of the cell appeared to be well suited to spur demand from solar cell manufacturers. Based on their proven SiNA-platform, Roth & Rau in 2006 presented their Multiple Application Inline Apparatus (MAiA) tool to the market.
Besides obviously being able to deposit SiN:H-layers, this new equipment was designed to also enable the deposition of AlOx-layers on the wafer as well as SiOx or a-Si deposition. Furthermore, it could be used for plasma activated chemical etching, an interesting feature for the surface treatment and texturing of solar cells.
At that time global c-Si cell production capacities doubled every year and the new manufacturers focused on a rapid expansion trying to avoid any technology risks that could potentially delay their production ramp up. The key driver was the maximization of material yield and an increase in Wp. So while the new MAiA tool proved to be interesting for testing new technologies in pilot lines, there was no widespread adoption of this tool at commercial production scale.
When global PV demand stagnated in 2009, PV equipment vendors saw their tool shipments drop by 50% year-over-year, and most cell manufacturers drew the conclusion that this was not the time to introduce new technologies.
By the second half of 2010, driven by the surge in PV installations in Germany and Italy, the next investment cycle kicked in. Yet triggered by the hiccup in global PV demand in 2009, all the prices for raw materials and components along the PV value chain had gone into a slide, resulting in the collapse of the price for solar cells and modules observed in the market.
So while on one hand manufacturers were still adding new capacities to satisfy a rising demand for PV power (with demand mainly coming from European end markets at that time), on the other hand it was of paramount importance to the companies to conserve cash in order to guarantee the survival of the firm in the ensuing price war that unfolded in those days.
Again, most manufacturers shied away from engaging in new technologies, instead trying to pass the pricing pressure from the market to their own suppliers. This era lasted for two to three years during which time every supplier to the PV industry had to surrender (almost) all of their margins, and many suffered significant losses just to stay in business. For the equipment suppliers these were the hardest times as demand for new production equipment fizzled down to a tiny trickle and existing orders were cancelled when just a few months ago the cell manufacturers had asked the equipment vendors to do anything possible to reduce the delivery times of the new equipment.
Dealing in hard times
In order to support the cost reduction efforts of the cell manufacturers, the equipment vendors kept innovating in order to reduce the cost of ownership that ultimately determines the processing costs of a solar cell. We had to remain flexible within our technology roadmap, recalls Thomas Hengst, Head of Customer Relations for Cell Technologiesat Meyer Burger and a veteran of the PV equipment industry. Since cell and module prices declined at a rate of more than 30% p.a. in 2011 and 2012, it was virtually impossible to come up with savings in the production process that could have compensated for these price declines.
It took until the end of 2013 for the price decline for PV components to level off. Ironically, the stabilization was at least in part due to measures imposed by the U.S. and the EU to end the price war in the market. Both European and U.S. regulators had come to the conclusion that the price levels for PV modules observed in their respective markets in 2013 no longer correlated with the production costs and therefore antidumping measures should be enacted.
The forced stabilization of prices gave manufacturers the opportunity to realign their costs with market prices. While profitability levels reached by 2014 were not overwhelming, they allowed the companies room to breathe and refocus on the medium and long-term steps required for a brighter solar outlook. As the price pressure in the market relaxed, customers began to optimize and evaluate their production processes, said Hengst. At that moment our ongoing investment into our R&D truly paid off. We had continued to implement improvements to our MAiA system and had upgraded it to our generation 2.1 platform. As a consequence we were able to offer our customers PERC equipment with a throughput capacity of more than 3,400 wafers per hour while at the same time reducing the footprint of the machine as well as the operating costs.
Benefits for PERC
Hengst admits that the cost reductions achieved between 2010 and 2014 by the suppliers of the precursor gas for the AlOx, as well as the improvements in the pastes and the lowered CoO of the laser tools necessary for the PERC cell technology, were important factors in further lowering the production costs for cell manufacturers. By the end of 2013 all the efforts of the supplier industry fit together. Considering an upgrade to the PERC technology made strong economic sense for cell manufacturers, adds Mirko Meyer, Product Portfolio Manager at Meyer Burger.
For monocrystalline solar cells, the conversion efficiency can typically be increased by a full percentage point through the application of our PERC technology upgrade package, says Meyer Burgers senior process integration specialist Thomas Grosse. This corresponds to an added value per cell of more than 5%. The increased processing costs (including depreciation) through the additional MAiA tool and the required laser equipment amounts to around $0.035 per cell, so that the cell manufacturer can achieve an economic benefit of around $0.05 per cell at current market prices.
Running at full utilization with more than 25 million cells per year, the MB-PERC upgrade would have a payback time of less than one year for a monocrystalline cell production line, says Grosse. The efficiency improvement for multicrystalline cells is less pronounced, on average improving the cell efficiency by around 4% relative. Since the upgrade costs and thus the additional processing costs per wafer are identical for mono as well as for multicrystalline cells the resulting benefit for a multicrystalline cell manufacturer is to be found in the higher power class of the modules produced. So it is no surprise that the adoption rate of the PERC upgrade technology among monocrystalline cell producers has been significantly higher than among multicrystalline cell manufacturers. Meyer Burger believes that of the 8 10 GW of PERC production capacity ordered by customers since the end of 2013, the split between mono and multicrystalline cell manufacturers could be in the area of two thirds to one third. Besides demonstrating to cell manufacturers that upgrading their production lines to PERC made sense economically, Meyer Burger had to overcome a further challenge before orders for their upgrade tools really took off.
While many cell manufacturers supply cutting-edge technology products, only a few are able to do the pioneering work of taking a new technology from pilot line scale to full-scale industrial production, acknowledges Hengst. With our customers we go through this process individually and adapt our processes to meet their manufacturing and market requirements. The head of customer relations is proud of the achievements that Meyer Burger has reached together with its customers: Our increased order intake is clearxAdvertisementproof that technologically and economically upgrading to our MB-PERC process makes sense.
We offer one of the most cost efficient upgrade options with our MB-PERC technology on our MAiA platform. We enable our customers to further utilise their existing equipment profitably and efficiently. With the current orders on hand, Meyer Burger has surpassed its budgeted production capacity for all of 2015. In its 1H 2015 analyst call, Meyer Burger management revealed that it will quadruple the shipments of MAiA tools in 2H in order to meet the increased demand. To enable such a steep ramp up in production, Meyer Burger has hired more than 80 temporary workers for its Hohenstein-Ernstthal facility. PV equipment suppliers have not had the need to add personnel at such a rate in quite a while.
To the contrary, in recent years many PV equipment manufacturers, including Meyer Burger, had to optimize their production capacities and workforce due to lack of demand for new equipment. The German PV equipment suppliers that are organized in an interest-group of the VDMA underscored the fact that on a broader scale, the PV equipment industry is still faced with a difficult market environment with a recent update to their member statistics. The number of employees of the member companies has declined continuously since peaking in 2011. By the middle of 2015 these companies employed just 60% of the workforce they had four years ago.
Against this backdrop, the surge in order intake experienced by Meyer Burger over the past 12 months is encouraging.
Are PERC cells prone to LID?
With so many hopes attached to the PERC upgrade demand, it is no wonder that signs hinting at a possible new cell degradation mechanism, particularly pronounced in PERC cells, are taken seriously. Catchy headlines surrounding this topic had some industry observers worried that perhaps the investment activity in PERC upgrades might die off if no remedy for this problem could be found.
Asked what he makes of these discussions regarding detrimental degradation on PERC cells, Hengst remarks, We have looked into this issue closely with our customers and made extensive tests on serial production runs of modules made with PERC cells, both mono and multi. After more than a year of testing we have identified the relevant parameters to observe when selecting the wafer material to manufacture PERC cells. We have also found recipes for a treatment of the cells post-firing that can significantly minimize the degradation effect. But it remains a fact: The susceptibility of PERC cells to the hotly debated light-induced degradation (LID) is closely related to the casting or pulling process of the ingot from which the wafers are cut. An ingot with poor material specs will always deliver a cell that is susceptible to LID. Our suggested treatment helps manufacturers to minimize the detrimental impact but it all depends on the quality of the material. In Hengsts view, the evolution of cell manufacturing follows a typical pattern in which higher efficiency cells have tighter specifications for input materials. He doesnt view the observed degradation process as a show-stopper but rather as a case where the wafer suppliers will have to adapt to the new specifications and the cell manufacturers will have to consider some additional treatment of the cells to ensure that the increased cell efficiency gained through the rear-side passivation remains active throughout the lifetime of the module.
From an economic standpoint, the observed LID-effect is a challenge for multicrystalline cell manufacturers. With an economic benefit of $0.013 per cell achievable through the PERC process upgrade for a multicrystalline cell (taking into account increased costs) there isnt much headroom to spend on a post-firing treatment of the cells, nor for higher quality wafers. So the focus will be to raise quality standards for wafer production across the board and ensure any necessary post-treatment does not add more than $0.002 $0.003 per cell to the overall cost.
Thomas Grosse shared some results of the test runs with properly treated wafers compared to untreated wafers. As can be seen from the graph above multicrystalline PERC cells treated according to the recommended recipe by Meyer Burger show an LID that is on par with standard Al-backsurface multicrystalline cells.
The good news for equipment suppliers is that PERC upgrades are not the sole driver of demand. We see an increase in new cell production capacities, confirms Hengst. We have received orders for new SiNA machines corresponding to more than 2 GWp of new production capacities over the past 18 months. Some legacy production lines are being retired and we observe a net expansion of cell production capacities. With PV picking up in Latin America and the Middle East, demand will soon reach a level to justify the investment in production capacities there. So perhaps it is not too optimistic to assume that the PV industry is at the beginning of a new investment cycle. The equipment suppliers for their part will certainly not cease with their efforts to herald the next broad-based investment cycle in the solar industry aimed at making solar even more affordable than it is today.