As solar gets cheaper, it must get greener too

Print Friendly, PDF & Email

A new solar standard in France has developers competing on more than just price – they must also compete on their project’s environmental sustainability.

Print Friendly, PDF & Email

In recent months, news about the solar energy industry was dominated by headlines describing record-breaking tariffs, reflecting the fast declining price of solar electricity. With COP22 taking place in Marrakech, Morocco this week, the hope is that further commitments to reducing emissions will lead to further growth in the global solar power portfolio.

Indeed, Australia’s announcement this week that it plans to ratify the Paris Agreement on climate change by the end of the year is a most promising sign. The current growth in the pipeline of large-scale solar projects, supported by the Australian Renewable Energy Agency, is also a major step in the right direction.

However, while the focus has been on the attractive economics of utility-scale solar, it is often taken for granted that solar energy is inherently environmentally sustainable and that its carbon credentials don’t require scrutiny.


The fact is that, even solar power plants have an environmental footprint on a lifecycle basis. For instance, Concentrated Solar Power (CSP) has a footprint of 20g of CO2 per kilowatt-hour (kWh) of electricity produced, in addition to consuming vast amounts of water. Similarly, PV power plants also have carbon footprints which, on a lifecycle basis, can range from 12g/kWh for a facility using First Solar’s thin film modules, to as much as 24 g per kWh – for one using multi-crystalline silicon panels.

According to the European Commission Product Environmental Footprint methodology, upstream processes generate between 80 per cent to just over 95 per cent of the emissions in a PV power plant’s lifecycle. This includes the extraction of raw materials, production of semi-conductor materials, manufacturing of modules and Balance of Systems (BoS) components, and construction.

While the carbon footprint is of paramount importance in the context of decarbonising the energy system, other impacts created by the manufacturing of components of a PV system – such as acidification, eutrophication, abiotic resource consumption and particulate matter emissions – are also important.

The comparatively high carbon footprint of multi-crystalline silicon panels is a direct result of the energy-intensive processes required to refine silicon and ‘grow’ the ingots that are sliced into wafers. It is also a direct consequence of the fact that the manufacturing of commoditised multi-crystalline silicon panels had largely moved from markets where electricity has a low carbon intensity, to countries that often rely on coal and other carbon-intensive forms of electricity generation. As a result, many of today’s commodity PV modules may come with a heavy environmental price tag.

While the Energy Return on Energy Invested (EROEI) equation demonstrates how quickly a PV module can recover the energy required to produce it, it often tends to be treated as an added frill, as opposed to an important metric. In Europe, this is largely due to the legacy of subsidy-driven tariffs, which paid for solar electricity without necessarily looking into the carbon footprint of the facilities that generated it. Since the goal was to replace carbon-intensive generation with clean energy, one could argue that 24 g per kWh for a multi-crystalline silicon power plant is still a significantly better deal than having almost 1,000 g of CO2 emitted for the same amount of electricity generated by coal.

However, as we transition to tender-based renewable energy programs, regulators in Europe and beyond now have an opportunity to build on the gains made in reducing the overall environmental impact of their power generation portfolios. New policies that make it necessary for independent power producers to competitively bid for generation licenses, allow governments to ensure that they’re nurturing a renewable energy sector that is both, cost and carbon competitive.

How? France has effectively pioneered a new benchmark for the rest of the world to follow. While most competitive tenders focus on price and technical compliance, France’s Agence de L’Environnement et de la Maîtrise de l’Energie (ADEME) has designed a system to ensure that the country derives maximum value – and not just in the financial sense of the term – from its utility-scale solar energy investments.

A recent tender for 200MW of solar power placed a 15 per cent evaluation premium on projects that used PV modules with low carbon footprints. In other words, developers that wanted to win a Power Purchase Agreement in France needed to compete on more than just price – they needed to compete on the environmental sustainability of their projects.

While some may say that the policy has a hint of trade protectionism about it, I would argue that it actually helps the country ensure that its renewable energy program is as environmentally sustainable as it can possibly be.

While it is true that this excludes a number of lower tier panel manufacturers that don’t have the financial firepower to invest in low-carbon manufacturing, the policy effectively protects France from utility-scale deployments of more carbon-intensive PV panels, without sacrificing the economic competitiveness of its program. This is a clear demonstration of how evaluating the environmental footprint can be effectively embedded in the competitive process.

France’s progressive policy on reducing the carbon footprint of its solar energy program serves as a precedent for other countries to follow. The tools – such as France’s competitive tender structure and a new European initiative to introduce ecolabels for PV modules – already exist.

And as PV is established as the lowest priced electricity, it is time to level the playing field further and include the environmental externalities of electricity generation in this equation. We need to ensure that we do not repeat the mistakes of the past 150 years of fossil fuel use.

Andreas Wade is First Solar’s Global Director for Sustainability.

Print Friendly, PDF & Email

  1. Brad Sherman 3 years ago

    A disappointing article in some respects. The 24 g emitted for solar versus 1000 g for coal seems to have neglected the embedded energy of the coal-generating station (bit of steel there, I suspect) as well as all the upstream emissions related to coal extraction and transport. 1000 g is the ‘typical’ value just for the combustion of the fuel. A fairer comparison would be 0 g emitted for solar versus 1000 g emitted for coal.

    The embedded energy calculations presumably also assumes a certain energy mix in the manufacturing process. What happens to this assumption when all energy comes from zero-emission sources? I reckon it becomes even more negligible.

    Apparently, a 98% reduction in emissions isn’t good enough.

    Let’s not take our eye off the ball and let the perfect become the enemy of the good.

    • JeffJL 3 years ago

      Good point Brad.

      Compare apples to apples and call out all who do not (even if, as in this case, it underestimates the benefits of renewable energy.

    • The wind lobby under my bed 3 years ago

      There is no way of telling whether embodied impacts for coal were included or not unless you look at the source document. However, an LCA will show you that the amount of steel and concrete used for a coal fired power plant is a very small part of the overall footprint (because the emissions are so high during operations). The range in emissions intensity of coal fired power plants is quite significant as well, and would dwarf the effect of embodied emissions. Therefore, 1 kg/kWh is a valid approximation for coal in order to compare it to renewables.
      BTW, using 0 g emitted for solar would not be a fairer comparison. Yes, we should compare apples with apples, but that means all alternatives should include embodied emissions of plant and equipment. That’s exactly the point the author was trying to make in differentiating various renewable technologies. Setting them all to zero is incorrect and prevents you from making more informed decisions.

      • Brad Sherman 3 years ago

        Are you saying that the embodied emissions for a coal-fired plant are less than those for, say, an equivalent wind generating capacity (after allowance for the different capacity factors)? As I read your comment it sounds like you’ve just rounded the embodied emissions for coal down to zero and I would like to make sure I understand your point.

        Given that emissions from fuel combustion alone can range from 900 – 1300 t-CO2e per MWh (for common existing power plants) depending on the type of coal and the combustion technology I don’t see an embodied emission of 24 g / kWh as of much practical relevance. Especially when any LCA analysis used to generate such a figure has itself made assumptions about the energy mix used to build the technology. For example, aluminium from a VIC or NSW or QLD smelter (effectively 90%+ coal) would have a much higher embedded CO2 content compared to aluminium from Quebec (effectively 100% hydro).

        LCAs are useful for broad comparisons of technologies but not so helpful when it comes to choosing between two specific instances of those technologies. It’s a lot like catchment modelling for pollution control – CM provides useful comparisons at the catchment scale but it is generally unable to help decisions at the farm scale. Once you start talking differences of less than 3 or 4 % you are way below the noise level of the analysis method which I suspect has error bars of 10-20% on it – but this is just an educated guess.

        From my perspective, embodied emissions are a 2nd or 3rd order consideration when it comes to environmental impact and focusing on the 0th order impact that is emissions from fuel combustion is a more useful comparison.

        • The wind lobby under my bed 3 years ago

          Sorry for the slow reply Brad.
          No, I wasn’t saying the embodied emissions for coal are less (or higher) than for wind. I didn’t round down the emissions to zero either. All I was trying to say is that using 1 kg CO2e/kWh is a reasonable proxy for black coal power plants. Whether this includes embodied (and other upstream) emissions or not is less relevant as the variation is typically well below the variation due to fuel and technology variation (as you already indicated in the second post).
          My key point was that you should try to look at all relevant emissions (combustion as well as embodied/upstream). Focusing on “0th order” impacts alone risks shifting environmental impacts to the supply chain, rather than solving them.
          I fully agree with your comment that minor differences (e.g. 3-4%) are not relevant due to uncertainty, but when comparing two different types of solar panels, the differences could well be significant. I think that’s what the author was trying to get at. Nevertheless, I also agree with you that for now a “98% reduction in emissions” is more important than getting to 99% reduction.

  2. Ed 3 years ago

    Perovskite cells (e.g. Dyesol/Oxford) will have a much lower emissions footprint that Silicon PV. as they don’t require the same level of heat during manufacturing.

  3. neroden 3 years ago

    So I’ll buy my solar panels from Tesla, who are promising to produce them 100% using electrical power (including the heat) and are getting the electrical power from Niagara Falls (though I’m sure they’ll add solar panels in the future).

    The only meaningful line in this article is this one:
    ” It is also a direct consequence of the fact that the manufacturing of
    commoditised multi-crystalline silicon panels had largely moved from
    markets where electricity has a low carbon intensity, to countries that
    often rely on coal and other carbon-intensive forms of electricity

    So as China puts solar panels up, the commodity solar panels will get greener and greener.

  4. JHM 3 years ago

    The supply chain of the whole global economy emits about 333 g CO2 per US$ of GDP. Fortunately this carbon intensity is falling about 1% pa. You can slander any green technology you like by dredging up supply chain emissions. So what?

    As solar cuts its costs to manufacture it is indeed cutting it is indeed cutting supply chain emission. A Watt of PV used to require over a $1 of resources, and no it needs less than half that. So that is roughly a cut of half of the upstream emissions.

    But more importantly a Watt of solar can produce over 40 kWh over its lifetime. So if it is displacing 1000 g per kWh from coal, that’s a reduction of some 40 kg CO2 per Watt of solar. This lowers the carbon intensity of the supply chain of everything.

    So the key issue is to dive 333 g per $ GDP down to 0 g. In a carbon neutral world, we will still be making solar panels, and the embedded carbon will be zero.

  5. Sir William 2 years ago

    The Los Angeles Times calculated that Elon Musk’s three companies, Tesla Motors, SolarCity, and SpaceX, combined have received a staggering $4.9 billion in government support over the past decade. As Kerpen noted: “Every time a Tesla is sold . . . average Americans are on the hook for at least $30,000 in federal and state subsidies” that go to wealthy Tesla owners. This is crony capitalism at its worst.

    So….If certain US Representatives have made this commitment for you, when a vehicle from a foreign country arrives in the Los Angeles Port, which is All Electric and get the Equivalent of 300MPG, is this “commitment” the reason why 3 different US agencies say….”Get it out of here….or we will order it Crushed” ?

    Under the Obama Administration……

    NYTimes : One of World’s Most Efficient Vehicles Unable to Enter U.S.

    One of the world’s most efficient vehicles and an X Prize winner has been unable to enter the United States since Jan. 25, held up at the Port of Los Angeles at the city’s international airport by the Environmental Protection Agency, United States Customs and the Transportation Department, people associated with the vehicle said on Thursday.

    It uses an AC Propulsion drivetrain similar to the one in the Tesla Roadster and manages the equivalent of 300 m.p.g at 75 m.p.h. It won first place in the Alternative Tandem category in the 2010 Automotive X Prize, a competition created to foster the commercialization of efficient vehicles capable of more than 100 miles per gallon.

    When the US subsidizes US electric car companies with Billions, does it mean Climate Change will have to be put on hold, because the US government will block real solutions from entering the country, based on minute technicalities ?

Comments are closed.

Get up to 3 quotes from pre-vetted solar (and battery) installers.