Comment & Analysis
Given the current controversy over the reduction in support for renewables, particularly photovoltaic generated electricity, I write on behalf of the Renewable Power Committee of IMechE to set out some basic facts on this form of power generation.
Photovoltaics (PV) play an increasing role in UK energy supply. The DECC central forecast estimates that the UK is likely to reach 10 to 12GW installed capacity of PV by 2020 and 2036 GWh was generated in 2013, yet there are still some who believe that PV systems have a high carbon footprint and require more energy than they produce during their lifetime. The facts which follow are based on a review carried out by Committee members of a number of objective studies on the subject.
Life cycle assessments (LCA) typically provide data on the environmental impact of a product including the carbon footprint and the embodied energy that has gone into raw material extraction, manufacture, operation and disposal of a system. Hundreds of research papers have been written studying the life cycle analysis of PV covering a range of technologies and reporting a wide variation in outcomes. This variability in the performance outcomes is due to a number of assumptions including the solar radiation, system lifetime, balance of system components and installation type, conversion efficiency, cell type, manufacturing process and location, as well as differences in LCA methods. Systems installed in the UK are either crystalline silicon (mono and multi-Si) or thin film (amorphous silicon (a-Si), cadmium telluride (CdTe) and copper indium gallium diselenide (CIGS)), with the vast majority being crystalline silicon.
In 2012 the US National Renewable Energy Laboratory led a comprehensive literature review and harmonisation study on the impact of PV which has helped to create a more coherent picture of PV systems. This study screened 397 research reports for crystalline PV and 109 reports for thin-film PV written between 1980 and 2009. Following the application of rigorous criteria, research was identified that best represented the carbon footprint of the technologies. Harmonized median figures for ground mounted and roof-top installations for crystalline PV were estimated at 40 gCO2/kWh for Mono-Si and 47 gCO2/kWh for multi-Si with irradiation at 1700 kWh/kWp/yr and figures varied between 14 and 27 gCO2eq/kWh for thin film systems under irradiation conditions of 2400 kWh/kWp/yr. In comparison the solar resource in the UK is considerably lower varying between 1200 kWh/kWp/yr in the South West to 800 kWh/kWp/yr in northern Scotland. Using the same assumptions transposed to the UK, therefore, this would result in figures of 68 and 80 gCO2eq/kWh for moni-Si and multi-Si respectively and between 34 and 65 gCO2eq /kWh for thin film. As a comparison, CCGT generation has a much larger footprint of between 488-600 gCO2eq/kWh.
A recent UK Climate Change Committee report, “Reducing the UK’s Carbon Footprint”, focusing on more recent studies between 2000-2009, estimates slighter lower emissions for new installations in the UK at around 55 gCO2eq /kWh for crystalline technologies and 30 gCO2eq /kWh for CdTe based on ranges quoted between 40-70 gCO2eq /kWh for mono Si, 45-85 gCO2eq /kWh for multi-Si and 20-45 gCO2eq /kWh for CdTe. Another literature review carried out by the International Energy Agency also quotes lower figures.
The energy aspect of the LCA is often expressed in terms of the energy payback time (EPBT) i.e. the operational time it would take to produce the amount of energy required for its production. For PV systems the IEA report gives the 2008 EPBT as 1.7 and 0.8 years for crystalline silicon and CdTe respectively under southern Europe conditions (not including recycling). Under UK conditions this equates to energy payback of 2.9 years and 1.3 years. PV panels have an expected lifetime of 25-30 years (and most tier-1 manufacturers would say 40+years) so will produce considerably more energy than was used to produce them. A more recent report, commissioned in Germany, reported EPBT in a range from 0.55 to 1.3 years for Nuremburg, or between 0.9 years and 2.2 years for the UK.
The figures quoted above all depend on location of manufacture since the carbon intensity of grid power varies. For example, the carbon footprint (CO2/kWh) is likely to double if manufacture is in China compared to Europe while the EPBT increases by at least 30%. Even with these increases the carbon footprint of PV is still well below that of CCGT generation and the energy produced is still many times that needed for its own production.
It is also important to note that there are continuous improvements in PV systems arising from manufacturing efficiencies, wafer thickness and conversion efficiencies so the figures can be expected to reduce still further. For example, standard 60 cell modules were rated at 235Wp in 2012, 275Wp in 2014 and are predicted to be 320Wp by 2016.
It is recognised that greater solar PV deployment is not without its challenges. For instance, there are still uncertainties related to large volumes of embedded PV generation and its impact on grid system balancing and harmonic generation. Careful planning is needed to ensure the appropriate use of land and buildings. Further information on this can be found in DECC’s Solar roadmap1. The advent of economical electricity storage would alleviate the grid system balancing issue, allowing generation to be time-shifted to match demand.
I hope that this short report will dispel some of the myths surrounding this form of power generation.