Solar panels repay their energy 'debt': study

December 6, 2016 by Mariëtte Le Roux
Solar panels, which convert sunlight into electricity, are a key player in the fast-growing renewable energy sector, which also includes water- and wind-generated electricity

The climate-friendly electricity generated by solar panels in the past 40 years has all but cancelled out the polluting energy used to produce them, a study said Tuesday.

Indeed, by some calculations, the so-called "break-even point" between dirty input and clean output may already have arrived, researchers in the Netherlands reported.

"We show strong downward trends of environmental impact" of solar panel production, the team wrote in the journal Nature Communications.

The study sought to address concerns that solar technology may be using fossil fuel energy in the panels' manufacture, and emitting , faster than it was able to offset.

The authors found that for every doubling in solar capacity installed, energy used to produce decreased by 12-13 percent, and dropped by 17-24 percent, depending on the material used.

Solar panels, which convert sunlight into electricity, are a key player in the fast-growing renewable energy sector, which also includes water- and wind-generated electricity.

Unlike energy from fossil fuels such as oil, coal and gas, the generation of electricity by so-called photovoltaic (PV) panels does not release planet-harming carbon dioxide.

Solar panel capacity grew sharply, on average, by 45 percent per year from 1975 to reach 230 billion watts (Gigawatt or GW) in 2015.

In 1975, there were fewer than 10,000 solar panels around the world, compared to about a billion today, study co-author Wilfried Van Sark of Utrecht University in the Netherlands told AFP.

By the end of 2016, "we would have some 300 GW installed"—about 1-1.5 percent of global electricity demand.

Falling costs

Over an average lifespan of 30-odd years, a PV system pays back the energy that was used in producing it "multiple times," said the study authors.

Looking at data since 1976, the researchers calculated that on a global scale, solar energy's "debt was likely already repaid in 2011" for both energy input and greenhouse gases.

Even on the least optimistic data, break-even point will be reached at the latest next year for net energy, and in 2018 for greenhouse gases, they said.

The photovoltaic effect, by which certain materials convert the photon particles in sunlight into energy, was first identified by French physicist Edmund Bequerel in 1839.

The first photovoltaic battery was built in 1954 but was too expensive for widespread use.

The technology was used in the 1960s to generate power on spacecraft, and only started taking root on Earth in the 1970s.

From 1975, costs decreased by about 20 percent for every doubling in capacity, the study found.

In 1976, one would have paid about $80 (75 euros at today's rates) for one Watt-peak (Wp) unit, compared to about 64-67 US cents today.

Explore further: High renewable electricity growth continued in 2015

More information: Nature Communications, nature.com/articles/doi:10.1038/ncomms13728

Related Stories

High renewable electricity growth continued in 2015

November 23, 2016

The 2015 Renewable Energy Data Book shows that U.S. renewable electricity grew to 16.7 percent of total installed capacity and 13.8 percent of total electricity generation during the past year. Published annually by the National ...

High efficiency rating shines on Panasonic solar panel

October 13, 2015

How to get more electricity from a given area of solar panels—this is an important question that scientists explore as companies focus on the future of solar panels. Inhabitat said Panasonic has developed the most efficient ...

Yale engineer to build 'hot' solar cells

September 22, 2014

Associate professor of electrical engineering Minjoo Larry Lee has been awarded $2,540,000 to develop dual-junction solar cells that can operate efficiently at extreme temperatures above 750 degrees Fahrenheit. In addition ...

Recommended for you

Scientists write 'traps' for light with tiny ink droplets

October 23, 2017

A microscopic 'pen' that is able to write structures small enough to trap and harness light using a commercially available printing technique could be used for sensing, biotechnology, lasers, and studying the interaction ...

When words, structured data are placed on single canvas

October 22, 2017

If "ugh" is your favorite word to describe entering, amending and correcting data on the rows and columns on spreadsheets you are not alone. Coda, a new name in the document business, feels it's time for a change. This is ...

Enhancing solar power with diatoms

October 20, 2017

Diatoms, a kind of algae that reproduces prodigiously, have been called "the jewels of the sea" for their ability to manipulate light. Now, researchers hope to harness that property to boost solar technology.

19 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

ForFreeMinds
2.2 / 5 (10) Dec 06, 2016
If it takes 40 years for solar cells to generate the energy used to produce them, then it's not a good investment in generating energy. People who invest in businesses, usually want their money back in just a few years.
gkam
3 / 5 (8) Dec 06, 2016
"Over an average lifespan of 30-odd years, a PV system pays back the energy that was used in producing it "multiple times," said the study authors."
KenC
5 / 5 (2) Dec 06, 2016
The figures seem to be all over the place. What are they saying?

I thought it takes about 2 years for a well placed panel (correct angle, no shade) to generate the energy to offset the energy needed to produce it. Is that not generally true?
fmfbrestel
5 / 5 (7) Dec 06, 2016
The figures seem to be all over the place. What are they saying?

I thought it takes about 2 years for a well placed panel (correct angle, no shade) to generate the energy to offset the energy needed to produce it. Is that not generally true?


That is true for a panel made today. This paper is comparing ALL of the energy used to produce EVERY SINGLE solar panel that has EVER been produced, VS the energy payback from ALL of those panels.

Being that every panel produced in the last 12-24 months is still negative, plus the early panels cost MUCH more energy to produce, leads to the TOTAL energy balance for ALL solar panels to be around equal now, or 2011, or 2018 (depending on the assumptions).
RealScience
4.6 / 5 (11) Dec 06, 2016
If it takes 40 years for solar cells to generate the energy used to produce them...

It doesn't.
Silicon panels now reach energy payback in less than two years.

To understand why we are only reach overall payback now, consider that installations have been growing at 1.4x per year, or doubling every two years. So by the time that this year's panel has paid itself off, we have installed two more that haven't yet paid them selves off, and by the time they pay themselves off, we've installed four more.

So any given panel will pay back its 'energy debt 10 to 20 times over (depending on the type of panel and where it is installed), and it is merely that the solar installations are growing so fast and thus most panels are too new to have payed back their energy debt - an effect of looking at an industry whose doubling rate happens to match its payback period.

Reaching payback is thus merely a sign that energy payback has shortened to below the doubling rate.
BackBurner
5 / 5 (1) Dec 06, 2016
It's still an overly optimistic report though. A complete solar installation depends on controllers, inverters, batteries and on-demand power generation. The cost of all those components need to be considered.

My system includes 36 panels, 2 Sunny Island Controllers, 3 inverters, an automated transfer switch, a manual transfer switch, an essential loads panel, 8 lead acid deep cycle batteries (the ones that are $390 each) and a 15kW propane generator, because the sun doesn't always shine.

The cost of solar energy can't be reduced to the cost per Watt each panel delivers. This is a naive report if the author has quoted it accurately.

Having said that, my experience is the system, which cost out at ~$80,000 in 2007, paid for itself in 5 years.
antialias_physorg
4.5 / 5 (6) Dec 07, 2016
A point that is often overlooked: The 'payback' has a (positive) cocaine effect the more panels are installed. Because with a higher percentage of solar in the energy mix more of the energy for new solar panels will not be coming from fossil fuels but from those panels already installed...until, eventually, the 'payback' time will be zero, because no fossile fuels were used in their manufacture at all.

On a related note: I seem to remember an art project where a contraption was shown that fuse sand in the desert (as a first-step demonstrator how an autonomous, mobile, solar-powered factory could continually build solar panels in the desert)
Eikka
3 / 5 (2) Dec 07, 2016
Silicon panels now reach energy payback in less than two years.


Question: is that adjusted for energy source, or just the plain process energy needed to make them?

Point being that the difference between the two is a factor of 3-4. If you're agnostic about where the energy comes from and treat it as a simple abstract, you don't know how much energy was used to generate -that- energy.

That has a great importance, because if the energy is paid back in two years and the lifespan is 30 years, the EROEI of the solar panel is 15. If the energy payback by the actual source is 6-8 years then the EROEI drops to 4-5

Now, the trouble is that an EROEI below 6 may be net positive in energy, but it's not socially sustainable because at that point our energy expenditure measured in GDP starts to shoot up. Too much relative effort is spent on producing energy and our ability to maintain an advanced economy necessary to produce the high-tech stuff starts to suffer.
Eikka
3 / 5 (2) Dec 07, 2016
My system includes ... 8 lead acid deep cycle batteries


The Energy Stored On energy Invested (ESOI) of a lead acid battery is approximately 2. That costs you half as much energy as you can ever put through the batteries, which puts your payback WAY back in reality.

http://news.stanf...513.html

It's another hidden cost, since the lead plates in batteries are low tech, cheap stuff that is mined and manufactured with dirt-cheap coal power somewhere in China, which is currently the largest producer at 1.4 million metric tons a year.

http://www.indexm...uct=lead
Eikka
3 / 5 (2) Dec 07, 2016
Because with a higher percentage of solar in the energy mix more of the energy for new solar panels will not be coming from fossil fuels


In theory, but how much energy is spent on converting solar electricity into a form that is useable in the actual manufacturing process?

For example, from solar power to diesel fuel for a dump truck, or how about solar power into concrete, glass and steel?

Can a solar panel ever produce enough energy to produce all the chemicals, materials and fuels needed for its own manufacture?
Eikka
2.3 / 5 (3) Dec 07, 2016
Looking it up, the modern way of making polysilicon panels cheaper with less energy involves a process of Chemical Vapor Deposition (CVD) where Silane gas (SiH4) is passed over a plate. The production of Silane involves "metallurgical grade silicon, hydrogen, and silicon tetrachloride"

https://en.wikipe...al_grade

Metallurgical grade silicon is commercially prepared by the reaction of high-purity silica with wood, charcoal, and coal in an electric arc furnace using carbon electrodes.


The reaction goes SiO2 + 2 C → Si + 2 CO

That is to say, you need carbon to make a solar panel. Now the question is, how do you produce carbon with the electricity from the solar panel you just made? How do you close the loop and how much energy does that take?

Of course you could just burn trees for carbon, but that puts you back to the 1800's and the first energy crisis when steel manufacturing was threatening all the forests in Europe.
Phys1
3.7 / 5 (3) Dec 07, 2016
Can a solar panel ever produce enough energy to produce all the chemicals, materials and fuels needed for its own manufacture?

Yes, if I believe what has been posted here, in 2 years.
Eikka
2.3 / 5 (3) Dec 07, 2016
Can a solar panel ever produce enough energy to produce all the chemicals, materials and fuels needed for its own manufacture?

Yes, if I believe what has been posted here, in 2 years.


How, when half of the processes don't even exist yet beyond the laboratory desk?

Ignoring the points where processes don't exist, such as manufacturing steel with solar power, suppose the round-trip efficiency from solar electricity to a new solar panel is roughly 25%. That would mean the solar panel has to operate for 8 years instead of 2 years to make its own replacement, and therefore again the EROEI drops form 15 to 3.75 which is below useful.

antigoracle
1.8 / 5 (5) Dec 07, 2016
Can a solar panel ever produce enough energy to produce all the chemicals, materials and fuels needed for its own manufacture?

Yes, if I believe what has been posted here, in 2 years.

LOL
You are sure earning that Retard of the Decade award.
antialias_physorg
4.2 / 5 (5) Dec 07, 2016
How, when half of the processes don't even exist yet beyond the laboratory desk?

Whichg processes did you have in mind?

Manufacturing steel with solar power is no different than manufacturing steel with other power (and where exactly do you need steel for solar power?)
That "25%" number is just pulled out of your ass. How exactly do you imagine such a number is an argument for...anything? It'd be really interesting to know under what heading of 'logic' that should be filed? Crazy logic? Delusional logic? Truthiness logic?
Eikka
3 / 5 (2) Dec 07, 2016
Here's an interesting paper about the minimum EROI needed to survive:

http://www.mdpi.c...1/25/htm

Basically, the absolute minimum surplus you need, assuming energy is used in an optimal manner, and there are no great disasters or wars, is about 3:1.

One part is used to obtain more energy, one part is used to sustain yourself, and one part is needed to grow the next generation. That kind of society does not grow or shrink, because it has only enough energy to keep going with what it already has. It has no luxury beyond the basic necessities because it exists on the knife-edge of survival.

To survive against global environmental change and other risks, and develop new technology and evolve, a society needs a much greater surplus. According to the estimates, hunter-gatherers survived on a 10:1 ratio. The Roman Empire perhaps 20:1, the early 20th century societies went up to 200:1 when oil was first discovered and now we're dropping back towards 3:1
Eikka
3 / 5 (2) Dec 07, 2016
Manufacturing steel with solar power is no different than manufacturing steel with other power (and where exactly do you need steel for solar power?)


For the big boat that carries the solar panels from China to Germany, of course. Or the factory itself, or the nuts and bolts that you use to fasten the panels onto your roof...

I used steel as an example, because just as with metallurgical silicon, the manufacture of steel involves carbon to reduce iron ores to metal, and then carbon again as additive to obtain the desired mechanical properties out of the material. That carbon currently comes from coal, and the question is how do you go from solar electricity into a lump of C for the process..

That "25%" number is just pulled out of your ass. How exactly do you imagine such a number is an argument for...anything?


It's an example of principle - I'm not arguing that the figure is real. I'm posing the question: what do you think that percentage is?
Eikka
3.7 / 5 (3) Dec 07, 2016
On a related note: I seem to remember an art project where a contraption was shown that fuse sand in the desert (as a first-step demonstrator how an autonomous, mobile, solar-powered factory could continually build solar panels in the desert)


That's just cargo cult engineering.

Melting a bunch of sand on the ground doesn't demonstrate how one could make solar panels because all you get is a bunch of dirty opaque glass. It isn't even a first step - it's just a misunderstanding of how solar panels are made.
gkam
2.3 / 5 (9) Dec 07, 2016
While with the consulting company (1978), we asked to do a study to see if nuclear powerplants generated as much energy in their lifetime as it took to make them. They responded we would be refused any more government contracts if we did the study.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.