Net Zero Carbon: Is Solar PV Now the Elephant in the Room?
"The burning question: In the modern landscape does solar PV still payback its embodied carbon footprint?
Embodied Carbon Payback of Solar PV
Embodied carbon is the carbon footprint to make a product. It arises throughout the supply chain and cuts across geographies. It therefore gives us a true picture of the carbon intensity to manufacture a product.
When it comes to Solar Photovoltaics (PV), it is well established that they can have a high embodied carbon footprint. However, this has traditionally been offset by the savings in electricity from the national grid.
Traditionally the carbon emissions of generating UK electricity have been high. Not too long-ago coal and gas fired power stations dominated the fuel mix. That has rapidly changed with the rise of renewables, particularly wind, and biomass has transformed the carbon footprint of the national grid.
In fact, the carbon footprint of UK electricity has reduced by a massive 45% in just 4 years (2015 to 2019). It also continues to decarbonise and is expected to make a significant contribution to the UKs aspiration to be net zero carbon.
The burning question is therefore, in the modern landscape does solar PV still payback its embodied carbon footprint? If so, how many years does it take?
Or is solar PV now the elephant in the room?
With the rise of net zero carbon buildings, this is now a prominent question.
Let's take a look.
Embodied Carbon of Solar PV
There are many different types of solar PV. Despite this, crystalline PV has been dominant, with over 90% of the market share.
Crystalline PV is can be further separated into mono-crystalline, which has a higher efficiency, and polycrystalline, which has a slightly lower efficiency. For this article, we will take a look at mono-crystalline PV.
The literature on embodied carbon of PV tends to report results in kg CO2e per kWh generated. The current author considers this as a unsuitable metric for solar PV in the UK when comparing to embodied carbon, particularly in the current landscape, considering that electricity from the national grid is decarbonising fast. It is also highly dependant upon the amount of energy generated that it offers particularly poor applicability across geographies. The amount of energy generated from a solar panel in the south of Spain is naturally significantly higher than the same system in the UK. Even within the UK the variation in output is large.
The metric works for other energy generation technologies and allows for comparison of solar PV with other technologies. However, for the embodied carbon payback of solar panels we needed a more location and time specific approach.
Instead, we have used data on the embodied carbon per kWp or per m2 of solar panel, which offers greater accuracy to specific locations. Using this approach, we have used several references on the embodied carbon of mono-crystalline PV [IEA, 2015; ecoinvent V3; M. Ito, 2011]. There are many other references, but we found that most are based upon the same background data.
The average embodied carbon in those references was 2,560 kg CO2e per kWp.
This now needs to be compared with the amount of energy generated.
Electricity Generation of Solar PV
The amount of electricity generated by solar PV is naturally dependent upon the location and how it has been installed, e.g. orientation, pitch, shading, …etc.
In regards to the UK, the Centre for Alternative Technology (CAT) state that many UK PV systems could be expected to generate 700-900 kWh per kWp. In fact, a well installed system in sunny parts of the south of the UK could generate over 1,000-1,100 kWh per kWp. Outputs therefore vary notably.
The International Energy Association (IEA, 2015), estimate that the average yield of PV in the UK at optimal angle in urban areas is 920 kWh per kWp, which they state is before an annual degradation of electricity generated of 0.7%. This is the annual electricity generation that we will use for the purpose of this article.
Each year the PV system will generate electricity, thereby reducing the load on the national electricity network. This saves carbon emissions each year over the operational lifetime of the system, which is generally taken to be 25-30 years.
Next we need to compare to the carbon saved in operation.
Carbon Emissions of UK Electricity – Grid Decarbonisation
According to Defra’s GHG emissions factors for company reporting for 2019, UK electricity has an all scopes carbon emissions of 0.316 kg CO2e per kWh. What is often not appreciated, is that this is based on the UK’s fuel mixture for 2017. The emissions factors are always 2 years out of date, due to the time it takes to compile the data. This electricity will also undergo grid decarbonisation.
The rate of grid decarbonisation in the future can naturally only be estimated. In order to estimate the rate of grid decarbonisation, data from the National Grid’s Future Energy Scenarios, FES, (National Grid, 2018) on their decarbonisation in 2 degrees scenario has been overlaid with Defra’s GHG emissions factors for company reporting. This scenario has been used here to estimate emissions of electricity generation in the future.
For clarity, only the emissions from scope 2 were decarbonised. The well to tank emissions were assumed to remain constant to give a conservative approach. This provides the estimated carbon profile of UK electricity below:
Embodied Carbon Payback Time of Monocrystalline Solar PV – High Electricity Output
Now we have all the data to estimate the embodied carbon payback of monocrystalline PV for the UK. The final key parameter is the system install date. Due to the rapid scale of grid decarbonisation the year of installation is important.
The embodied carbon payback results are shown below:
The results show that even with monocrystalline PV systems with the highest expected generation of 1,100 kWh per kWp per year and system installed in 2020 it could take 15 years for the embodied carbon to payback - That is around half of the expected system lifetime of 30 years.
It would also be around 2035 before the embodied carbon is paid back, which means if anyone is hoping the achieve net zero carbon by 2030 they will need to give this due consideration.
That same system, with an output of 1,100 kWh in year one, if installed in 2025, would now have an estimated embodied carbon payback of 23 years. That’s right, the 5 years difference in install date has shifted the embodied carbon payback by 8 years – this is because of the grid decarbonisation. Those 5 extra years are far in the future, where the carbon emissions from a kWh of UK electricity are expected to be low - very low. As a result the amount of carbon saved per year in the far future is equally low.
Another way of representing the data is a carbon gain (or loss) over the anticipated 30 year lifetime, e.g. how much carbon does the system payback. This is shown below, indexed to zero being the embodied carbon paid back in full.
Anything on the negative area of the chart is expected to not payback its embodied carbon in full. A figure of zero gives a full payback of embodied carbon, but no benefit beyond that. We really therefore want to see systems operating in the higher areas of the chart. These results only go up to 0.5, which would be equal to a carbon gain ratio of 1.5, giving 50% more carbon savings over 30 years than the embodied carbon of products – this is not a particularly impressive carbon gain when compared with other technologies.
Embodied Carbon Payback Time of Monocrystalline Solar PV – More Typical Case
So what about a more typical UK PV system? Using the previously established figure of 920 kWh / kWp per year gives the results below:
Here are some insights for monocrystalline PV:
Discussion - Implications for PV and Net Zero Carbon
These results are significant and have profound implications for net zero carbon buildings – particularly those that are aspiring for net zero by 2030. It brings into question the scope of net zero carbon. If an organisation is aspiring to be next zero carbon by 2030, it raises difficult questions for those not including embodied carbon as part of their definition of net zero.
It presents a strong case study for considering embodied carbon through whole life carbon assessments as part of net zero carbon.
Can we just source lower embodied carbon monocrystalline panels? Perhaps we can, but there is a chronic lack of embodied carbon data published by the manufacturers. In fact, we did not find manufacturer specific data, such as an EN 15804 Environmental Product Declaration (EPD) for solar panels.
It is important to highlight that these results are specific to mono-crystalline PV in the UK. They would not apply to other types of PV, although polycrystalline would likely have a similar result profile. The results certainly do not apply to other countries. Each country should be assessed as its own case. The key parameters are the amount of solar irradiation and how carbon intensive is the electricity network. High irradiation and a carbon intensive electricity supply, e.g. as is the situation in Australia, will naturally give a very different embodied carbon payback period. Those offer the ideal application for solar PV.
In the context of the UK, it is clear that the embodied carbon of solar PV is now an important parameter. At the same time as the grid decarbonises the embodied carbon of solar panels will only become even more prominent.
The Big Question – What About Wind Turbines?
A natural question is what about wind turbines, do they suffer from the same long embodied carbon payback times?
In short, no they don’t. The embodied carbon of a large wind turbine can be expected to payback in around 1 year. Over such a short timescale grid decarbonisation is not a factor.
To raise another pertinent question, aren’t renewables part of the UK’s national grid mix anyway? Which gives rise to the decarbonising electricity grid.
Well yes, they are - but solar PV is not a particularly large contributor on a national scale. Between 2015 and 2019 the UK reduced the carbon intensity of electricity by 45% per kWh. This was predominantly driven by wind and biomass, combined with a reduction in reliance on coal. Solar PV had a smaller part to play in that decarbonisation.
What Next for PV?
It is important to highlight that there are other types of photovoltaics with a lower embodied carbon - in some cases significantly lower.
According to the IEA (2015) a cadmium-telluride, CdTe, based PV system would have an embodied carbon 60% lower than a monocrystalline PV system for each kWp.
For whatever reasons, CdTe based PV haven’t really taken off in the UK, but perhaps it’s time to investigate the feasibility of other types of solar PV. As a side note, the element cadmium is known to be highly toxic. There are studies that show cadmium-telluride is less toxic than the cadmium in isolation. PV laminates are also encapsulated. That said, it offers a good example of why true sustainability is difficult to achieve and requires trade-offs to balance one issue with another.
It leaves some burning questions, will other types of PV take over, or will crystalline PV clean up its act? Who knows, for now we know that the embodied carbon of PV needs to be taken seriously and warrants further research.
This should be taken to avoid serious carbon leakage, especially in the context of net zero carbon.
What do you think? Share your thoughts on social media using the links below, or consider subscribing to our Newsletter.
IEA, 2015. Life Cycle Assessment of Future Photovoltaic Electricity Production from Residential scale Systems Operated in Europe, 2015. http://www.iea-pvps.org/index.php?id=314&eID=dam_frontend_push&docID=2391
Masakazu Ito, 2011. Life Cycle Assessment of PV systems. Available from https://www.intechopen.com/books/crystalline-silicon-properties-and-uses/life-cycle-assessment-of-pv-systems
National Grid, 2018. Future Energy Scenarios 2018, “Decarbonisation in Two Degrees”, from Table 3.2, http://fes.nationalgrid.com/fes-document/
Links to other Knowledge Base pages:
Sign Up to Newsletter