This is a guest post by Graham Palmer, author of the book Energy in Australia: Peak Oil, Solar Power, and Asia’s Economic Growth. A short bio is given at the end of the post.
With rapidly declining costs and a favourable EROI, solar photovoltaics (PV) is often assumed to be an irresistible “disruptive technology” that is on a preordained trajectory. This faith is typified in Paul Krugman’s recent article in the New York Times in which he imagines that “drastic cuts in greenhouse gas emissions are now within fairly easy reach”. Yet in two recent posts Solar Scotland and The efficiency of solar photovoltaics, it is apparent that all may not be as simple as it seems. In my book, Energy in Australia: Peak Oil, Solar Power, and Asia’s Economic Growth, published as part of the SpringerBriefs series, I elaborate on some of these issues and try to draw out some of the nuances.
Since the EROI of global oil supply is taken at round 10:1 to 18:1 and on a downward trend, but solar PV is taken at around 10:1 up to 60:1, and rising, it could easily be concluded that the net-energy available from PV is superior to current oil production, and may actually begin to approach the very high net-energy figures that used to be available in the Golden Era of oil gushers in Texas and Louisiana.
Yet, intuitively it is not at all obvious that PV has the same value to society vis-á-vis fossil fuels. A nation’s PV fleet could be turned off for a week and few would notice. Nor would it adversely affect system reserve margins or loss-of-load assessments. Yet even minor disruptions to gasoline supplies, natural gas, cancelled trains, or the internet can have a major effect on daily life. The curious thing is that the literature on PV life-cycle analyses seems to readily accept these high numbers without questioning what these figures really mean. Yet high-school science and maths students are regularly taught to question whether their answers “make sense”.
It is only recently that more rigour has been applied to trying to understand the figures, leading to Prieto and Hall’s (2013) examination of the large-scale deployment of PV in Spain through 2009 and 2011. Coincidentally, I was researching a paper on PV, which was published in Sustainability Journal and posted on BraveNewClimate shortly afterwards. Both of us came to similar conclusions on EROI (between 2 and 3), which is significantly lower than commonly quoted figures, and below the critical minimum EROI required for society (Hall et al 2009).
The difference between my analysis and Prieto and Hall’s is that I started with data from a published paper that used conventional boundaries, but I broadened these to provide a more real-world inquiry, whereas Prieto and Hall conducted a comprehensive bottom-up analysis for Spain. As Prieto shows, there are a myriad of societal costs that are simply ignored in conventional PV-LCA analyses, and that solar PV is completely dependent upon the fossil fuelled system in which it is embedded.
At the heart of the PV-EROI issue is the methodological guidelines established by the IEA-PVPS program (Fthenakis, V., et al. 2011). The guidelines establish the boundaries and methodology for PV-Life Cycle Analyses, and provide a coherent and consistent framework for comparing PV systems. The issue is that results calculated with the constrained IEA-PVPS boundaries are being used to compare PV with other energy sources, leading to a gross overestimation of their true value to society. The critical issue of intermittency is ignored, the system boundary for PV panels is truncated to exclude upstream energy costs, and many other important system-based factors are deemed to lie beyond the standard boundaries.
Similarly, the often assumed idea that a “suite of renewables” with smart-grids and electric vehicles to achieve some sort of “optimized synergy” is frequently overstated. It is well established that geographical smoothing, along with “technology-diversity” smoothing can improve the statistical performance of integrated systems, but cannot deal with the “big gaps” events, particularly during winter.
Similarly, combining electric vehicles with solar PV seems like a great idea at face value, but how would it work at a system level? Will motorists want to “fill up their tank” during the middle of the day at peak tariffs, and sell back to the grid at night? How will the vehicle get recharged so it has full range by the morning? What happens to charging during winter? Is there a business case for cycling a large proportion of system energy through EV batteries and what amount of degradation is acceptable given that batteries are cycle limited? The batteries remain the weakest link in the marketability of EV’s, and car manufacturers are more interested in optimising battery capacity for driving range and longevity. (Instead, the more obvious synergy is between baseload, which can provide low cost and predictable year-round off-peak power for charging when most vehicles are parked at home, whilst underpinning the load factor for baseload generators in the event of the electrification of the motor car fleet.)
Intermittency is readily accommodated in conventionally powered grids but absent balancing from large-scale hydro, integration limits of intermittent energy remain unknown beyond about 20% of grid penetration. Low cost storage is key, but the paradox of storage is perhaps best summed up by John Morgan when he states
.. the idea that advances in energy storage will enable renewable energy is a chimera – the Catch-22 is that in overcoming intermittency by adding storage, the net energy is reduced below the level required to sustain our present civilization.
Nonetheless, intermittent sources of power can still play a useful role in fuel savings, emissions abatement, and trade balance (for example, the case of wind in Portugal) but at a cost. Treating PV as an extension of, rather than as a substitute for, the fossil fuel enterprise enables a more productive discussion of PV’s niche role in electricity generation. For example, with a small amount of distributed storage, solar PV could provide a potentially valuable contribution to network support in summer-peaking grids. And in isolated grids reliant on high cost diesel or gas, PV provides a valuable fuel-saving role.
EROI is important because it is a foundational issue that can’t be remedied with markets, price signals, or support mechanisms. In the case of PV, the technology is perfectly functional – improvements in efficiency or cost of panel won’t substantially improve the “extended EROI” – it is the system-wide issues needed to remedy integration that erase the gains. A wider understanding of these issues is sorely needed.
Graham Palmer is a professional engineer with qualifications in electronic, industrial and renewables engineering, working in the HVAC (heating ventilation and air conditioning) industry in Melbourne Australia. He is also an independent part-time researcher in the sustainable energy field, contributing reports, articles, and papers in the energy efficiency and sustainable energy fields. His research interests include understanding the broader societal consequences of energy, climate policy and technology.