Overcapacity has caused financial problems in the past few years, so the solar industry has been somewhat less focused on implementing technology advances - but today things are looking up and research is being rivived.
The solar market was technology driven until 2012, but then experienced a collective tightening of belts as the industry went into overcapacity in 2012 and 2013.
Much of the photovoltaic manufacturing sector saw profitless prosperity the past two years, but for the last two quarters of 2013, manufacturers finally began to see a return to profitability – and 2014 heralds the most profitable year for manufacturers since 2010, according to analysts at GTM Research, a Greentech Media company based in Cambridge, Mass.
Optical (left) and IR (right) image of photovoltaic module. Photo courtesy of Dirk Jordan/NREL.
Curiously, the upturn is not predicted to be caused by any staggering breakthroughs in materials, harvesting methods and/or fabrication techniques, but, instead, will be driven by strong demand growth and stable pricing.
“We expect a strong end-market performance this year, with global solar installations in 2014 growing by 25 percent year over year. Most of this growth will be driven by continued strong growth trends in China, Japan and the US – as was the case in 2013 – but we are also seeing sizable volumes of solar being deployed in markets where even a couple of years ago there was almost no solar being installed at all, in many cases with few or no subsidies, which is very exciting,” said Shyam Mehta, a lead upstream solar analyst at GTM Research. “These emerging markets include Latin/South America (Mexico, Chile) and the Middle East (particularly [United Arab Emirates] and Saudi Arabia).”
Thin-film buzz moves to silicon
In the past, thin-film-based panel production was considered the technology that would gain market share over the industry-standard silicon-based solar cells, but this never happened, and silicon-based cells are gaining market share today, according to industry analysts NPD Solarbuzz of Santa Clara, Calif. “Companies are currently trying to take all costs out of manufacturing, rather than using technology as a means to stay profitable,” said Finlay Colville, vice president and team leader.
But that’s not to say that promising new avenues in solar technology are not being explored, particularly as they relate to silicon-based cells. As solar companies are finally able to “sell out” again, they are looking at expanding capacity, said Sarah Kurtz, Photovoltaic Reliability Group manager at the US Department of Energy’s National Renewable Energy Laboratory (NREL) in Golden, Colo.
“The majority of this increased capacity is currently going into silicon, but the expansion provides an opening for new technologies,” Kurtz said. “While low cost is still a key driver, there is an increased focus on reliability and quality control at both the module and system levels.”
Although a relatively mature technology, there is still scope for silicon technology to improve, namely in increasing efficiencies and decreasing cost. Some key advances include a recent conversion from p-type to n-type technology.
Historically, p-type was the norm, but advances in n-type allow a higher tolerance to common impurities, resulting in high minority carrier diffusion lengths, when compared with p-type. Furthermore, n-type crystalline does not suffer from the boron-oxygen-related light-induced degradation (LID) that is common on p-type Czochralski silicon.
IHS Solar of Santa Clara, Calif., a leading provider of diverse global market and economic information, forecasts that n-type applications will mature and find their way into mainstream manufacturing.
A close view of a silicon crystalline solar cell. Photo courtesy of Dennis Schroeder/NREL.
“Looking forward, n-type will be mostly a monocrystalline technology growing from 5 percent of the total cell production capacity to approximately 32 percent by 2020,” said analyst Jon-Frederick L. Campos of IHS. “The two key n-type mono technologies in focus currently are interdigitated back-side contact (IBC) and [heterojunction intrinsic thin-layer] (HIT).”
Exploitation of HIT technology has seen an increase since Sanyo’s HIT patent expired, allowing various companies to look into this approach to reach high efficiency. These technologies (or very similar variants) are being used by mainstream players such as SunPower Corp. (IBC) of San Jose, Calif., Panasonic (HIT) of Kadoma, Japan, and Yingli Solar of Baoding, China (HIT).
IHS expects IBC technologies to reach efficiency levels of 24.8 percent by 2020. Today’s most efficient monocrystalline cells achieve an average efficiency of 21.7 percent (Sanyo) through HIT and are predicted to reach 24.1 percent by 2020.
When it comes to the main growth driver in the PV industry, multicrystalline silicon wafers are still the star and will likely remain p-type. Multicrystalline, specifically, meets the requirements of commercial PV for both cost-effectiveness and large-scale production.
Solar cells, such as this wafer of multicrystalline silicon produced by CaliSolar, are tested at NREL’s One-Sun Solar Simulator. Photo courtesy of Dennis Schroeder/NREL.
“Generally, from a technology standpoint, cell efficiencies based on multicrystalline depend on the crystallization process, the solar cell processing sequence on the crystallization process, and a mutual adjustment of both processes,” Campos said. “The enhancement of the efficiency of multicrystalline cells may help to further broaden the demand and installation of solar applications worldwide.”
Bad news for organic PV
The excitement for technologies often referred to as third generation has seemingly died away in recent months, with much of the activity surrounding organic and dye-sensitized cells now confined to the research laboratory. Unfortunately, issues regarding efficiency, lifetime, scalability and cost remain key barriers to entry.
Canadian Solar Inc. was named the 2012 winner of the Solar Project of the Year Award at POWER-GEN International’s Projects of the Year awards gala in Orlando, Fla., for its 148-MW solar park PV project in southern Brandenburg, Germany. Photo courtresy of Canadian Solar Inc.
The second- and third-generation (i.e., thin-film solar cells) options are no longer seen as the preferred route to low-cost manufacturing in the industry because the cost to manufacture crystalline silicon-based modules has fallen so much, said NPD Solarbuzz’s Colville.
The focus for the future seems to be on improving and optimizing existing first-generation technologies, with many manufacturers cautious about releasing plans for any new capital expenditures.
Indeed, NPD Solarbuzz reports that any new lines being installed are typically using optimized process flows for mono- and multicrystalline silicon ingot to module production. “This environment is expected to continue during 2014, with any new technologies unlikely to see strong adoption before 2015 at the earliest,” Colville said. “There remain many R&D-based variants being explored, but none has yet to evolve out of niche adoption.”
This roof-mounted PV system at Georgia Tech’s Aquatic Center was constructed using multicrystalline silicon solar modules. It generates an energy output of about 5 MWh annually. Photo courtesy of NREL.
The reluctance by manufacturers to invest in capacity increases or new equipment on the whole is a cause for concern, noted Mehta at GTM Research.
“China’s new PV industry regulations encourage M&A [mergers and acquisitions] instead of ‘greenfield’ expansions,” Mehta said. “This is a bit concerning because it does not incentivize the capital-equipment sector (which has traditionally driven most of the technological innovation in the industry) to continue investing in advanced technology platforms and processes, which could hurt the industry further down the line, since there is still much work to be done to reduce solar electricity costs even further.”
A possible solution may lie in innovative new financing mechanisms for solar projects that could attract a different class of investors to consider solar. Securitization of downstream solar assets aim to encourage institutional investors, such as master limited partnerships (MLPs), real estate investment trusts (REITs) and “yield cos.”
“There is still a lot of work that needs to be done to make this happen. This means driving greater transparency as regards the long-term performance and reliability of components (modules and inverters), more standardized project screening practices, and better after-sales-service infrastructure and ancillary services for components,” Mehta said. “Ultimately, this is part and parcel of becoming more ‘mainstream’ – that is, in enabling solar’s transition from a relatively small, nascent, immature sector to a more stable, mature and sophisticated industry.”
Top fabrication advances
Fluidized bed reactor (FBR) technology for polysilicon production is garnering a lot of interest for its potential to produce solar-grade silicon at significantly lower cost than the currently dominant Siemens process, which uses CVD (chemical vapor deposition). The REC group, based in Oslo, Norway, is producing FBR silicon at a cash cost of $12 to $12.50/kg, compared with $16 to $20/kg for large Siemens plants, according to GTM Research. REC also has estimated that, in China, its FBR technology could be produced at a cash cost of $7.90/kg, which would be a disruptively low cost.
Example of a frameless module. Photo courtesy of Lumos.
Glass-glass and frameless modules, as opposed to standard aluminum-framed modules, are expected to increase market penetration in 2014 and 2015.
Diamond wire saws use diamond-coated sawing wire to enable cutting speeds up to three times faster than the current industry standard (steel wire cutting in a slurry medium). IHS Solar expects diamond wire to gain significant market share by 2016 and a majority share by 2020. By switching to diamond wire saws, SunPower Corp. has reduced silicon consumption to 4.2 g/W, a 15 percent reduction over the past year. The company is now using wafers at – or thinner than – 135 µm.
A glass-glass module. Photo courtesy of ET Solar.
Replacing silver contacts with lower-cost copper equivalents reduces silver paste consumption by about 45 percent, enabling cell efficiency gains on average of 0.2 to 0.25 percent. These gains translate to an average module power gain of 3 to 5 W. Hanwha SolarOne Co. Ltd. of Shanghai has integrated its Project EstarII cell technology into its production processes; the company plans to leverage this technology to gradually increase the cell efficiency of its multicrystalline silicon wafers to more than 18.5 percent by the end of 2015.
Passivated emitter and rear contact (PERC) is under heavy R&D and is in line for large-scale deployment by many major Asian manufacturers. Metal wrap through (MWT) is gaining technological superiority through development by PV manufacturers such as Canadian Solar and JA Solar.
Example of a frameless module. Photo courtesy of Solaria.
A glass-glass module. Photo courtesy of Sunovation.