New Hampshire, USA — Flexing its muscles yet again, thin-film solar PV leader First Solar has quietly acquired GE’s similar solar intellectual property portfolio, but questions linger about whether and when the company will see the benefits.
The deal includes both a specific module purchase commitment plus a longer-term commitment with agreed-upon pricing “over an extended period of years,” according to First Solar CEO Jim Hughes during the company’s 2Q13 earnings results. GE, meanwhile, will supply inverters for First Solar’s global deployments, technology acquired through French firm Converteam, and it will seek to sell solar PV into power plants alongside wind and thermal assets. (Such an example already exists, with GE helping Invenergy outfit adjacent solar and wind farms at its Grand Ridge projects in Illinois, a GE spokesperson pointed out.) Financially, GE also receives 1.75 million shares of First Solar common stock, making them one of FSLR’s top-10 shareholders — a roughly $82 million value at yesterday’s market close, but currently down to about $70 million as First Solar’s stock slumped in reaction to its other financial and project announcements (more about that on the next page).
GE and First Solar had been trading CdTe efficiency records lately. Both agree that thin-film solar PV and especially cadmium telluride (CdTe) have lots of room to improve conversion efficiency and better compete with silicon PV.
GE’s CdTe technology is “distinctly different” from First Solar’s CdTe process but nonetheless it is “consistent with our manufacturing platform,” Hughes said in the call. Both sides, though, have yet to “fully determine how we best integrate their technology into our roadmap, determine the proper sequencing in terms of upgrading of equipment and what it means in terms of our profits,” he said. An update on that evaluation likely won’t come until First Solar’s Analyst Day next spring, but he confidently stated that within two years some of GE’s CdTe technology would be incorporated into First Solar’s modules. GE currently has 19.6 percent cell efficiency in a research cell, beating FSLR’s best lab cell mark by nearly a full percentage point, he noted, and GE’s >450 issued patents and pending applications in CdTe effectively doubles FSLR’s portfolio.
We asked First Solar for further clarification about the companies’ CdTe methodologies and how they might be combined, but the company declined to comment.
First Solar has long been one of the bellwethers in the solar PV sector, and certainly the far-and-away leader in CdTe, with technology specifically geared toward large-scale solar deployments. (Solar Frontier holds a similar position on the CIGS side of thin-film PV). First Solar also was one of the first solar upstream manufacturers to extend further downstream into project development, creating a captive pipeline for its products.
But not everyone agrees this was a good deal for First Solar. “[It] appears defensive,” writes Credit Suisse analyst Patrick Jobin in a research note. It’s the second recent acquisition by First Solar (following Tetrasun and its C-Si technology) for solar PV technology that’s “relatively early-stage,” he points out, which perhaps suggests that the company’s “core technology is not cost-competitive in today’s low-poly environment.” Deutsche Bank analyst Vishal Shah agrees, writing that GE’s CdTe IP likely won’t materially help First Solar’s efficiency marks until 2017, and that the rest of the partnership likely will amount to merely “a few 100MWs of negotiated volume contracts.”
With this deal, GE effectively bows out of the CdTe segment that it joined in 2011, with dreams to scale it up into a multibillion-dollar business alongside its wind business. Its planned $300 million, 400-MW factory in Aurora, Colorado was put on ice last summer, though, and now GE will “discontinue the build-out” of the plant entirely and seek to lease the space, according to a company rep. The original operation in Arvada is going away as well, with 50 employees affected, and future research going through GE’s operations in New York.
The acquisition of GE’s CdTe technology IP was only a portion of First Solar’s busy day yesterday. Garnering more attention was its softer-than-expected 2Q13 financial results and lower guidance for the rest of the year.
Second-quarter revenues were $520 million, down significantly due in part to delays in selling its ABW project, though this are expected to come through later this year. For the full year, First Solar also lowered its outlook for sales and profits, as it holds on to two projects (Solar Gen and Macho Springs) to try and sell them after construction and improve their economics, and stomachs delays in some approval processes for its Antelope Valley project.
On the technology side, First Solar touted a new back-contact technology to help reduce module degradation and improve lifetime energy production. TetraSun, its recent c-Si acquisition, is being moved up to the planning stages for a cell manufacturing line with qualification and initial shipments planned for 2H14. And First Solar’s Series 3 Black modules have passed Thresher tests and long-term sequential tests by TUV Rheinland, only one of five industry module manufacturers to do so.
First Solar’s production in 2Q13 was 389 MW (DC), with factories running at about 75 percent utilization, an overall module manufacturing cost-per-watt of $0.67/W and a best-line mark of $0.62/W, and top conversion efficiencies of 13.3 percent as of the beginning of August. The company thinks it can get that best module line cost down to $0.57/W by year’s end, assuming that the new back-contact process will improve module efficiencies to 14 percent as hoped.
On the project side, First Solar made one more announcement yesterday, acquiring a 1.5-GW portfolio of projects from Element Power for an undisclosed amount, spanning a number of U.S. markets (California, Arizona, Texas, Georgia, North Carolina, Colorado, Louisiana and Illinois) as well as an entry into Mexico’s Sonora state. (Here’s a handy map for reference.) Deutsche Bank’s Shah, though, noted that these “very early-stage” projects and probably amount to “less than $50 million” and won’t improve the company’s numbers much. CEO Hughes confirmed in the Q&A call that most of these projects are strategic locations with interconnections and don’t include PPAs. The company’s project pipeline has about 1.5 GW in mid- to late-stage progress and about 6.6 GW with a 12-24 month development cycle.
Asked during the conference call Q&A about the global utility-scale solar market opportunities, Hughes projected a wide range of “mid-50s” to 60-65 GW by 2016, with up to two-thirds of that in utility-scale. “People may revisit their estimation of distributed generation versus utility scale generation” upon further analysis of “significantly higher capacity factors on the utility scale side,” Hughes said. “For a given investment and for a given space within the dispatch order for solar to contribute to reducing the peak, you will get more total electricity out of facilities with a higher capacity factor.”
Lead image: Woman on bench and sunset, via Shutterstock