Aging Gracefully: How NREL Is Extending the Lifetime of Solar Modules
NREL-Led Consortium Studies Why Solar Modules Fail in the Field and How To Extend Their Life
What makes for a good solar module? A few things are obvious: high energy yield, low cost, and reliable in the field.
Reliability plays a huge role in the lifetime costs and performance of solar modules and systems. These high-tech semiconductor devices must continue generating electricity for 30 to 40 years of sun, wind, hail, snow, and heat.
We expect modules to slowly degrade and produce slightly less electricity over time as they are exposed to outdoor conditions over the years. A major question in the solar energy industry is exactly how much we should expect solar modules to degrade each year (generally 0.5%–1%) and when they will eventually degrade so much that they no longer produce adequate power (often about 20% loss from their original output) or become unsafe.
For modules built today, it is probably 30 years. Each additional year makes the cost of electricity from that module cheaper and means we will need to mine or recycle fewer raw materials to reach our clean energy goals. Could research push that age of retirement to 50 years?
Launched in November 2016 with funding from the Department of Energy’s (DOE’s) Solar Energy Technologies Office (SETO), the Durable Module Materials (DuraMAT) Consortium is a multi-laboratory consortium led by the National Renewable Energy Laboratory (NREL), with Sandia National Laboratories and Lawrence Berkeley National Laboratory as core research labs. Additional researchers from multiple universities, solar companies, other national laboratories, and an industry advisory board provide perspectives from across the solar energy community.
After five years of researching solar module reliability and awarding $30 million in high-impact projects, DuraMAT was awarded an additional $36 million by SETO for six more years of funding starting in 2021, as the consortium continues its focus on five core objectives intended to accelerate a sustainable, just, and equitable transition to zero-carbon electricity generation by 2035.
Solar Modules: Where We Have Been and Where We Are Going
Photovoltaic (PV)—meaning they convert light to electricity—modules have existed in their modern form since the middle of the 20th century, but the technology has seen explosive growth over the last two decades. And the next two decades promise even greater growth for solar technologies.
“If solar is going to expand and become this ubiquitous technology that we have across our power system, on our houses—and be responsible for 40% of our electricity generation—old technologies are not enough,” said Teresa Barnes, a senior researcher at NREL and director of DuraMAT. “PV modules need to be made more efficient, less expensive, and more sustainably at much larger scale. But we also need to know that these new modules—whether they’re new module designs or new cell technologies like bifacial or tandem cells—will perform predictably in the field.”
DuraMAT is exploring ideas that could extend solar module lifetime up to 50 years. And it is looking at new variations of module and cell technologies, such as bifacial modules that also collect reflected light on their backsides, or new, high-efficiency cells that require advanced packaging to survive for longer than 30 years.
To better understand how modules fail, DuraMAT has developed accelerated stress tests based on the environmental conditions seen in different climates. These tests are paired with powerful materials science forensics (think CSI but for degraded PV modules) and detailed physics modeling of those failures to better understand what causes module degradation, with the ultimate goal of predicting when they will fail. To top it all off, DuraMAT collects the resulting data in a central, shared data repository and applies its insights to develop new, creative approaches to improve module durability.
The ultimate goal is to better predict how new materials and module designs will perform, building confidence that they will last for more than 30 years in the field, despite our lack of long-term field data for new technologies. Field data shows that older PV technologies are durable. DuraMAT is applying that knowledge to make more accurate predictions about newer technologies.
Sweating the Degradation Details
One of DuraMAT’s most celebrated successes is its application of combined, accelerated stress testing. Traditional stress testing subjects solar modules to a range of stressors—such as heat, humidity, or sunlight—but only one, or perhaps two, at a time. However, some of the failures seen in fielded modules are not easy to reproduce in these traditional stress tests, possibly because outdoor conditions stress modules in combination—heat, light, and voltage often occur together on sunny days, or wind and rain during a storm. DuraMAT researchers have found that stressors often need to be applied in combination to get field-relevant results more quickly.
While combined stress testing is not an entirely new idea, DuraMAT has taken it to a new level. In controlled chambers at NREL’s Outdoor Test Facility, PV modules are subjected to multiple stressors, such as extreme temperatures (both hot and cold), being drenched in water, and ultraviolet light exposure to simulate in a few weeks or months what happens outside over years.
Other tests are meant to simulate other stresses, such as how years of wind exposure could expand cracks in PV cells (see video below). DuraMAT then pairs that information with computer modeling and microscopic materials analysis from solar modules that failed in the field to better understand the mechanisms that drive these failures.
Early-Career Scientists Prove It Works
One such effort was led by a team of early-career scientists—DuraMAT places an emphasis on opportunities for early-career researchers. The team combined expertise and strengths from several national laboratories to develop a method to predict which backsheet materials would crack in the field based on accelerated testing. The industry experienced a fairly large batch of module failures (approximately 10 gigawatts) due to a new backsheet material that was widely used between about 2010 and 2015. This material started cracking after a few years in the field, despite passing all of the industry’s standard qualification tests.
A backsheet is the bottom layer of a solar module that encloses the back of the module and is often made from polymer (plastic) materials. This layer provides critical electrical insulation and mechanical integrity to a module, and the material’s failure forced PV developers to replace modules with the “bad” backsheet. (The PV industry also has several well-established “good” backsheet materials that have lasted for decades.)
Using the known good and bad backsheets enabled the DuraMAT team to develop a procedure to validate new test sequences. Combining that sequence with advanced materials analysis techniques, the team was able to understand why backsheets of the “bad” material were failing on both a chemical and mechanical level. By comparing the samples that failed under combined stress tests with failed modules from the field, the team of early-career researchers validated that the stress-test failures match this type of field failure. Now the team is examining other types of module materials and designs, including the screening procedures for new backsheet material development and studies of modules with glass backsheets.
“DuraMAT, in the way that it’s structured, incubates early-career scientists in a unique way,” said Laura Schelhas, who participated on the team as an early-career researcher at SLAC National Accelerator Laboratory and has since moved to NREL as executive director for the second phase of DuraMAT. “DuraMAT allows early-career researchers to try their hand as the principal investigator on projects, and it gives them a taste of reporting, project management, staffing, and budgeting that really speaks to career development for less-experienced researchers. The backsheet project was a great example of how that works—we got a lot of highly collaborative publications. The early-career researchers were invited to conferences to present their work, and some have transitioned to staff positions within NREL and other labs.”
What’s Next for DuraMAT?
After starting from scratch five years ago, the next, six-year phase of DuraMAT is off to a strong start. Many projects having already been awarded from the $36 million in total funding available for the consortium’s work in the second phase.
“A lot of our research continues to focus on reliability and durability in the commercial technology portfolio,” said Barnes, when asked where DuraMAT is headed. “We are now shifting our emphasis towards predictive testing and modeling methods that will enable us to assess reliability more quickly and more accurately in new technologies. Solar needs to keep improving, and product development cycles can be a lot faster than reliability testing cycles. We need to find a way to assess reliability and durability at the speed of product development as the industry scales up rapidly.”
It is a challenging goal, but the DuraMAT community is now aiming to begin predicting module lifetime and how that could shape the materials supply chain for solar modules. Driven by physics of failure and physics of degradation mechanisms, there will be more focus on predictive lifetime modeling, allowing for further research and possible commercialization of modules with 50-year lifetimes.
“We are trying to shift into a reliability research mode where we’re directly targeting modules that last 50 years,” Barnes said. “We’re very focused on high-energy-yield modules and making those in a sustainable way. We know there are going to be big material and energy impacts from ramping up deployment as fast as we need for the energy transition. But our question is, ‘How can we do that in a way that’s environmentally sustainable and in a way that our supply chain can keep up?’”
“We’re aiming for the same high quality of research but at a greater quantity in this phase,” Schelhas said.