Last week, Hanergy Solar Group Limited (Hanergy) provided the first glimpse of its strategy for its recent investments in thin-film CIGS production, having completed the acquisitions of previous CIGS hopefuls, Solibro, Miasolé and Global Solar.
In outlining its rationale for the continued focus on thin-film manufacturing expansions, Hanergy stated:
“…the group believes that it is of utmost importance that the Group is to develop for and deliver to its customers new CIGS turnkey lines so as to establish the group’s track record and market position in the CIGS turnkey line market, in order for the group to continue leading the global turnkey line business.”
Hanergy’s latest CIGS plans relate to new manufacturing capacity for CIGS, amounting to a staggering 5.25 GW of new CIGS capacity.
The plans don’t stop there either. Further expansion of a-Si turnkey manufacturing capacity is also scheduled (a mix of both a-Si/uc-Si and a-Si/a-SiGe) that would result in a cumulative (nameplate) capacity for a-Si of 4.75 GW.
Therefore, Hanergy’s combined a-Si and CIGS based plans would amount to 10GW of thin-film capacity. In effect, this sees a redistribution of the original (and previously amended) framework contracts of 2010 and 2011 (that were based upon the different a-Si absorber type capacity), to a new combination of a-Si/uc-Si, a-Si/a-SiGe and CIGS variants.
Equipment supply from Apollo
Aside from the sheer scale of these nameplate capacity expansions, the other takeaway from Hanergy’s revised thin-film plans is the continued faith that Hanergy places in its equipment supplier subsidiary (formerly known as Apollo Solar Energy Technology Holdings, and before that Apollo Precision).
Indeed, while there was significant risk in having Apollo deliver 10GW of a-Si based turn-key production lines to Hanergy (from a somewhat limited track record of proven production knowhow), Hanergy’s latest CIGS capacity plans appear to represent a carbon copy of its previous a-Si framework supply contract using its in-house equipment subsidiary.
Of the original a-Si based plans, approximately 1.7GW of nameplate a-Si capacity has been recognised as shipped (and signed off) by Hanergy across a range of different sites within China. Only part of this 1.7GW is running in mass production today.
Apollo to become the leading PV equipment supplier in 2013
The new framework plans by Hanergy to add 5.25 GW of turnkey CIGS lines, delivered by Apollo, are subject to a host of caveats, preconditions and shareholder agreement. The plans (simply for production equipment/services and excluding land and buildings/facilities) amount to US$8.6 billion.
Even a small portion of this $8.6 billion being recognised as PV equipment revenues by Apollo would probably be enough to keep Hanergy (or more specifically its Apollo PV equipment segment) as the leading PV equipment supplier (by PV tool revenue recognition) as far out as 2015.
While 2013 has yet to unfold (and final PV equipment revenues will not be known fully for a further five months), the strong likelihood is that Hanergy/Apollo will be the leading PV equipment supplier in 2013. Applied Materials held this title from 2008 to 2011 and Meyer Burger enjoyed a brief stint in pole position for 2012.
CIGS equipment supply to Hanergy
Similar to the changes in the past for turn-key a-Si based equipment, it is possible that subsequent alterations to the framework CIGS capacity plans will emerge. During the early delivery phase of a-Si capacity from Apollo to Hanergy, Hanergy also invested in a turnkey a-Si production line from an external line supplier.
Therefore, while the initial plans for new CIGS capacity are based upon Apollo expanding knowhow from turnkey a-Si to turnkey CIGS equipment, the prospects for US-based, European and Japanese CIGS process tool suppliers may not be as bleak as it appears today.
When Apollo started out in the turnkey a-Si business, the technology was very similar to an old deposition concept that was originally pioneered by US-based EPV. The manually loaded vertical batch chamber approach (with layers deposited sequentially) had also been championed by Solar Thin Films, Energo Solar, and NanoPV (and others), with varied success in volume manufacturing. Within China, GS Solar developed an analogous CVD process, and this was subsequently inherited by Apollo.
Indeed, Apollo’s turnkey a-Si lines have been a combination of in-house tool manufacture (CVD, PVD) and outsourcing other tools (from both foreign and domestic tool suppliers). Therefore, for CIGS, it may be more appropriate to consider Apollo as the turnkey line provider, not the turn-key line manufacturer.
Turnkey CIGS line supply
However, Hanergy may choose to create a brand new CIGS equipment supply segment, running alongside its a-Si turn-key line group. Being a turnkey a-Si equipment supplier in the PV industry has been difficult over the past few years; however, this has been nothing compared to the challenges for a turnkey CIGS equipment supplier.
Aside from single equipment suppliers delivering small pilot CIGS lines for R&D activities, there has only ever been one turnkey CIGS production line (using the more familiar turnkey line terminology) shipped within the solar PV industry. This was Centrotherm’s 30MW CIGS line shipment to Sunshine PV in Taiwan back in 2008.
Getting the turnkey CIGS line qualified at Sunshine proved to be a significant challenge for Centrotherm, and before Sunshine had fully ramped the 30MW line, it was upgraded to 45MW (nameplate) capacity in 2011. Sunshine has yet to run the line at full capacity, and Centrotherm has subsequently closed its CIGS turnkey line division.
The home-build success of First Solar & Solar Frontier
The only two thin-film PV manufacturers with (tier one) bankability, volume manufacturing knowhow and efficiency/yield metrics that are PV industry competitive are, of course, First Solar and Solar Frontier. Besides the differentiated business models that each of these companies has (compared to all other thin-film producers), it is no coincidence that each of First Solar and Solar Frontier has a 20-year R&D/technology track-record.
Each of these companies was also instrumental in developing the knowhow, IP and production equipment specifications for their thin-film technology variants. The relationships with their preferred tool suppliers have been a key feature behind the success of First Solar and Solar Frontier’s production lines.
For every other thin-film producer, Hanergy included, this is currently setting a massive barrier to entry; aside from capital required and having a credible route to market. Bypassing the technology learning knowhow curve is not a luxury afforded to new CIGS proponents today, in sharp contrast to the c-Si entrants over the past few years.
CIGS process flow options
Another pending issue regarding Hanergy’s 5.25GW of CIGS capacity aspirations relates to the type of CIGS that would comprise this total figure. The current phasing is divided into 650MW, 700MW, 1.2GW and finally 2.7GW. How much will be Solibro CIGS, or Miasolé CIGS or Global Solar CIGS?
CIGS is still a headline banner for a wide range of technologies, from substrate type to process flow to panel size and efficiency level. Assigning sub categories at the technology level does help to simplify the CIGS landscape: glass or flexible R2R, sequential sputtering or co-evaporation, vacuum or non vacuum, etc.
Figure 1 divides the technologies used across the top-60 CIGS companies over the past 10 years into six different process flows. Each of glass and flexible R2R is split into sequential/sputtering, co-evaporation and other, with the other grouping including spraying, printing or electro-deposition.
For simplicity, the previous exploits of Solyndra are grouped within the glass/evaporation category, although there could easily be a case made for Solyndra’s CIGS technology being separated as a stand-alone, bespoke approach.
Figure 1 divides the 60 CIGS companies into one of these six flow variants, and shows the relative popularity of the different approaches by number of companies that have sought to ramp to mass-production for each. More than half of the CIGS companies so far have chosen either glass/sputtering (such as Solar Frontier) or glass/evaporation (such as Solibro).
In addition to Solibro, the two other Hanergy CIGS companies also have different flows compared to market-leader Solar Frontier. Miasolé is based upon flexible/sputter and Global Solar on flexible/evaporation.
Therefore, all of the Hanergy CIGS companies have a different process flow arrangement (substrate and/or deposition) to Solar Frontier whose technology has its origins dating back to the R&D of Shell Solar and Siemens Solar.
Figure 2 shows the split of the collective CIGS production, across each of the six generic technology types. The glass/sputter approach accounts for 78% of CIGS megawatt production levels to the end of Q3 2013, and within this dominant share, Solar Frontier has produced 92% of all glass/sputter CIGS. The collective efforts of more than 20 CIGS companies that comprise the flexible categories have produced just 6% of CIGS output until now.
Forecasting CIGS capacity expansion activity
However, process flow categorisation is simply one of many risk factors assigned to any thin-film CIGS expansion plan at the gigawatt status. Any flow type that has been confined to R&D or pilot lines (or is using PV equipment suppliers that have limited track record) has to be assigned a higher risk profile, when forecasting the impact on the equipment suppliers’ order books or end-market supply/demand.
Figure 3 shows the latest NPD Solarbuzz analysis for CIGS capacity, going out five years to the end of 2018. The data shown is adapted from the recent PV Equipment Quarterly report, and is supplemented by the latest plans of Hanergy for CIGS equipment. The capacity shown in the baseline case corresponds to the most-likely scenario for CIGS capacity, with the data showing ramped/annualized mass production capacity (not nameplate).
The baseline forecast is contingent upon additional capacity being added by the leading CIGS manufacturers in the next few years, and is subject to its own subset of risk factors. Also shown in Figure 3 is an upside scenario for CIGS capacity that is based upon capacity plans of CIGS challengers that are currently tracked by NPD Solarbuzz, but excluded from forecasting based upon their risk profiles.
While Hanergy has yet to identify the proposed phasing of the 5.25GW of new CIGS capacity, a steady ramp from 2015 is assumed in the (high-risk) upside scenario within Figure 3. At 5.25GW, Hanergy obviously dominates the upside scenario presented. The graphic confirms the scale of the ambitions underpinning Hanergy’s plans for CIGS.
Under the baseline scenario, CIGS capacity increases from 1.54GW at the end of 2013 to approximately 4GW at the end of 2018. However, factoring in the at-risk expansions (dominated by Hanergy), the upside shows more than twice this capacity level at 10.8GW.
Regardless, however, of the risk assigned to the upside, this potential will continue to attract the attentions of any CIGS equipment supplier. Hanergy’s CIGS plans alone amount to US$4.3 billion of investment into new CIGS capacity/services.
Thin film unquestionably remains the challenging technology in the PV industry to c-Si and this is unlikely to change over the next five years. Therefore, any company seeking to come into the PV industry to ship PV modules (such as Hanergy or TSMC) either competes head on with Chinese c-Si suppliers, or goes for the thin-film route.
The efficiency progress of First Solar and Solar Frontier continues to add to this optimism for thin-film challengers. As such, it is likely that a certain amount of new investments (Hanergy and others) will flow into the PV industry over the next five years.
What impact this has on the end-market is a different question. Interestingly, none of the leading c-Si module suppliers appears to have any defensive strategy against the threat of thin-film. This is due to several reasons, but is largely based upon the phase during 2009-2011 when thin-film competition was being taken seriously by c-Si companies and the fact that many of the investments into thin-film then did not end up directly competing with tier one c-Si strategies for market-share.
However, with Hanergy now dangling a US$4.3 billion carrot in front of CIGS PV equipment suppliers, there is certain to be renewed optimism for any tool supplier that had previously placed tools with Solibro, Miasolé or Global Solar, in addition to the Japanese tool suppliers that have now accumulated essential production knowledge through recent Miyazaki fab expansions.
And as such, the wildcard tag assigned to Hanergy’s thin-film aspirations is likely to remain until such times as volume product is shipped to the end market, either in the form of a-Si or CIGS based panels.