While the Paris climate accord has delivered new momentum to the renewable energy industry, including the solar PV and storage sectors, DNV GL argues that new approaches to assuring the reliability of these technologies will be required if the Paris targets are to be met. DNV GL says the focus in developing and deploying new technologies has been on efficiency and cost while material reliability over 20 or 30 years needs to be ensured.
DNV GL has made its case public in a new position paper: Advanced materials in renewable energy tackling the reliability challenge.
Trade-offs between availability, cost and performance may be made, but in all cases long-term reliability is a key requirement for materials used in the energy industry, said Liu Cao, a researcher at DNV GL Research & Innovation and lead author of the position paper.
DNV GL has explicitly singled out perovskite semiconductor, superhydrophobic PV module coatings and new battery chemistries as two areas in which new approaches to testing and long-term performance modeling is required. Perovskites have made headlines among the PV world, with rapid efficiency gains being achieved – from, 3.8% to 20.1% within five years, DNV GL notes.
However, questions as to the stability of perovskite material do remain and is widely considered to be the major hurdle before the technology can be deployed by the global solar industry at scale. DNV GL highlights both the chemical and thermal instability of perovskite semiconductors as being particularly problematic.
The DNV GL paper also addresses superhydrophobic and self-cleaning module coatings as technologies that have the potential to deliver significant array output gains, particularly in very cold and hot and dusty environments. Such coatings could reduce the need for module cleaning, by preventing either snow, ice or dust build up, respectively.
Despite the potential of superhydrophobic and self-cleaning coatings, DNV GL notes that the strength and durability of such coatings has yet to be demonstrated. The fragile roughness structure can be irreversibly damaged, inevitably leading to a gradual loss… [of] superhydrophobicity and self-cleaning attributes, the paper explains.
More generally, the DNV GL team has highlighted four key insights in its position paper:
1. Single average degradation rate is an inadequate metric of long-term performance;
2. Qualification tests are insufficient for lifetime assessment;
3. Accelerated laboratory tests may not reveal all the degradation mechanisms;
4. Real-time monitoring is valuable but unable to predict lifetime alone.
DNV GL suggests that empirical models be coupled with a fundamental understanding of degradation, and rich and increasingly ubiquitous sensor data be deployed in predictive models. It offers its BatteryXT model as an example in which predictions for estimating battery performance and can be used alongside historical data and statistical analysis.