The World Has a Serious Coal Problem

…but not because of climate change

Photo by Amir Arabshahi on Unsplash

World coal production faces a long slow decline after more than a decade of stagnation. Since we use two-thirds of that supply to produce electricity, we need to build alternatives, fast. But can wind, solar or nuclear indeed replace, let alone outlast fossil fuels? What if solar panels were just a more clever way to burn coal?

Thank you for reading The Honest Sorcerer. If you value this article or any others please share and consider a subscription, or perhaps buying a virtual coffee… At the same time let me express my eternal gratitude to those already support my work — without you this site could not exist.

Last year just 20% of all the energy consumed worldwide was in the form of electricity, the rest came directly from burning fossil fuels. That, however, is only half of the story. Literally. After including all the coal, oil and gas combusted in power plants, the energy actually consumed by electric grids around the world jumps to a whopping 228 Exajoules (1), or 41% of all energy produced worldwide. In case you were wondering what’s the hype around cleaning up the electricity sector, smart grids and “renewables”, look no further for an answer.

World Energy Supply in Exajoules, 2024. Electricity values have been converted from TWh to EJ, to provide a basis for direct comparison. Waste heat produced by other sources (e.g.: internal combustion engines) are not included. Chart: own work. Data source: Statistical Review of World Energy

The argument goes: if we could somehow generate all our electricity from low-carbon sources, we could prevent 182 Exajoules worth of fossil fuels from being burned. That is, a little more than a third of all fossil fuels consumed worldwide, with a comparable amount of CO2 emissions to boot. Truth to be told, yes, combusting coal oil and natural gas is that inefficient: out of that 182 Exajoules consumed by thermal power plants around the globe only 66 Exajoules (2) worth of electricity ends up being fed into power lines above. (The difference comes from waste heat generated by thermal power plants.)

What rarely, if ever, gets a mention is what happens to those electrons once they leave the power plant. Electric grids, as I alluded to in my last article, are incredibly wasteful. The US grid, for example, loses 59% of the power before it ever reaches the customer. I don’t have similar data for the rest of the world, but I believe it is fair to say, that many regions in the global South do even worse than that, while the relatively newly built Chinese grid is likely to do a better job. Calculating with a 40% loss as a global average, however, still leaves us with a dismal 22% fuel-to-socket efficiency when it comes to burning fossil fuels for power. Compare this to the idyllic setting courtesy of solar panels mounted on rooftops. No waste of fuel, no transmission losses, no emissions. Wonderful, isn’t it? Well, yes, unless you scratch the shiny surface of this story just a little.

The fundamental problem

While “renewables” do come with much less CO2 emissions throughout their lifecycle (compared to a gas or coal fired power plant), this is not the main reason for their widespread adaption. The fundamental issue for the world economy is not the release of carbon dioxide. At least not just yet. Climate change has just reached a level where it started to negatively effect industrial output. The complex self-adaptive system we call the global economy is much more concerned with year-over-year growth, long term detrimental effects be damned. And this is where solar panels offer a unique solution to a serious problem not often discussed in mainstream media.

Cumulative growth in energy supplied by fossil fuels. Note how coal is just slightly above where it was a decade ago. Chart: own work. Data source: Statistical Review of World Energy

The fundamental problem of the world economy is that real, meaningful growth requires a comparable increase in energy consumption (3). The economic super-organism, hell-bent on growth, seems to have an insatiable hunger for energy. That, however, on a planet where 91% of all energy supplied eventually comes from fossil fuels, means an ever increasing demand for coal, oil and gas. Now, the question poses itself: what to do if the growth in the global supply of fossil fuels slowed down to a mere 1% a year, as it did? How to maintain growth in such an environment? (Forgive the super-organism for not seeing the wood for the trees and not realizing that there is no infinite growth on a finite planet.)

The situation is even worse when viewed from the perspective of coal: energy production from the oldest fossil fuel in use has been on a flat plateau since 2011. This flatlining — and most likely peaking — of world coal production is a much bigger deal than it looks at first sight. Even China, the world’s largest consumer by far and the country having the fourth largest coal reserve in the world is struggling to keep up with demand. Thanks to its insatiable hunger for the black rock it barely has 35 years of coal left, indicating that the world’s largest economy by energy consumption faces an imminent peak and decline of production. Coal prices in China have increased five-fold since 2014, and doubled for the rest of the world over the same time period as a result of a surge in demand and in face of limited supply growth. Despite the extreme price hikes during and after the health crisis and the proxy-war in Europe world coal supply failed to grow meaningfully over the past four years.

Despite that dire outlook, coal provided 34% of all electric power fed into the grid worldwide as of 2024, consuming 68% of all available coal supply globally that year. That number for China, the world’s largest manufacturing hub, is even more staggering: coal being responsible for 58% of all electricity supplied. Since the black rock remains essential to many other applications (including but not limited to metallurgy, steel making and cement manufacturing) the world needed to find more efficient ways of using this valuable, but highly polluting resource. Keep burning two thirds of it in thermal power plants is definitely not one of those.

Global power generation and emissions (2000–2024). Renewables, and natural gas came only as an addition to a still increasing coal and natural gas consumption in the electricity generation sector. Source: EMBER

The question is thus the following: How to convert coal to electricity more efficiently? Rapid advancements in polycrystalline silicon manufacturing provided us with the answer: let’s turn that coal into solar panels first, then use those panels to generate more power. According to the Global Electricity Review 2025 assembled by EMBER, last year alone solar added more than twice as much global electricity generation as any other source in 2024. What’s not to like? And if it cleans up the air in cities while making citizens believe that their climate is being saved at the same time it’s all the better.

Calculating with even a modest 10:1 energy return on investment (EROI) solar makes a tremendous amount of sense. Investing just 1 unit of coal power into making aluminum, silicon, glass etc. with a return of 10 units in the form of electricity over the lifetime of a panel sounds like a great deal. (Plus it gives a huge boost to a whole range of industries, contributing to economic growth). Hence the generous subsidies handed out for both making and installing solar panels, and China becoming the world’s number one manufacturing hub for photovoltaics.

With that said, while “tackling the climate crisis” remains a noble goal, solar panels offer no solution to that predicament. The key advantage of “renewables”, biofuels, geothermal and hydro lies in the fact that these sources of power can use low-grade energy as an input. Sunlight, a gentle breeze or rainfall accumulating in reservoirs behind a dam are very diluted forms of energy, something which cannot be used to melt metals, nor to move a thousand tons of cargo across continents without being concentrated first. This is why a square kilometer covered by windmills, solar panels or hydro provides very little energy compared to a lump of coal or a few barrels of oil, and thus are unlikely to replace fossil fuels.

On the other hand this low-grade energy comes “for free” — you only have to pay for the machines converting that low-grade energy into high-grade electricity. From this perspective “renewables”, biofuels, geothermal, tidal hydro and even nuclear (4) are in effect an effort to harness those forms of energy which are otherwise not so valuable to the broader economy, in order to save high-grade energy provided by fossil fuels for other uses — including the manufacture of said energy technologies. Alternative sources of power can thus serve in an auxiliary role only, where the bulk of material transformation and long distance transportation continues to be done by fossil fuels.

Material reality

From a material perspective “renewables” are no more sustainable than coal oil and gas. In fact there is nothing “renewable” about them, other than the clever marketing, making people believe that somehow these devices harness free energy and thus must be clean and green. In reality, solar panels are made from aluminum (framing), glass (making up more than half of their weight), and of course silicon (where the magic happens). There is of course internal wiring from copper and silver paste, applied to silicon wafers, among other more rare metals (such as cadmium, indium, gallium etc.). The tiny problem with these materials is that they require high heat, carbon atoms and massive amounts of stable electricity to produce. In other words: fossil fuels.

Aluminum does not occur naturally: it is found in an ore called bauxite, mined using giant excavators and delivered by heavy duty trucks and dumpers (all running on diesel fuel, of course). Refining that ore into pure aluminum requires lots of electricity. It takes 17,000 kWh of power to manufacture 1 tonne of aluminum, using a process which already runs at a 95% efficiency with not much left to improve. The problem is, that this process requires a stable supply of power to be so effective, intermittent electricity from wind and solar won’t suffice.

Glass, making up 76% of a typical solar panel’s weight, is primarily made from quartz sand, which is needed to be heated up to roughly 1700°C (3090°F) in a furnace — usually burning natural gas. Sand of course isn’t carried by the wind into the factory either; a fleet of trucks and excavators burning diesel are needed to do the job. (Again, trying to make glass by wind and solar is highly impractical, to say the least.) The same raw material, quartz, is used to make the silicon wafers themselves, doing the magic of converting sunlight into electricity. This quartz needs to be rather pure, however — beach sand won’t cut it — and thus must come from a few special mines around the world. (What happens when — and not if — these mines deplete is another story.) Perhaps I don’t need to repeat myself explaining how that quartz ends up in a refinery, but shedding some light on how those batches of pure white quartz crystals are turned into 98% silicon might be helpful here:

A typical batch contains around 453 kg of gravel and chips and 250 kg of coal. As an electric current passes through the electrodes on the lid of the furnace, it forms an arc that generates heat up to 4,000°F or 2,350°C. The high temperatures trigger a reaction where the oxygen is removed, leaving behind silicon and carbon monoxide. The entire reduction process takes approximately six to eight hours.

Again, since stable high currents cannot be provided by solar panels over six to eight hours without interruption, one needs a thermal power plant (or at least a nuclear reactor) to power this process… Plus a quarter ton of coal, of course, to provide the necessary carbon atoms for reduction. (By the way, the same goes to making pig iron, where coal is not just a source of high heat, but also a reducing agent removing unwanted oxygen atoms from iron ore.) And this was just the first step in making silicon wafers: this 98% pure metallic silicon then needs to be evaporated and condensed during the Siemens process (taking up an additional 60,000 kW/ton), then the resulting 99.999% pure silicon must be crushed and melted again in a special crucible to produce the silicon used in photovoltaic panels. And these are just some of the materials involved in making solar panels. Silver, copper, germanium etc. all have their supply chains involving tons of fossil fuels in every step of their lifecycle.

Solar panels, nor any other technology for that matter, work indefinitely. Contrary to media claims, recycling photovoltaics at the end of their 25 year lifespan is far from being solved. Millions of tons of these high-tech miracles could end up as toxic waste, leaching heavy metals into the groundwater. And even if we could reclaim 90% of their raw materials (which in many cases is technically impossible or simply does not worth it), we would still need to keep mining and making virgin materials. Not only to make up for that 10% which is lost during the recycling chain, but also in order to guarantee the structural strength and electrical properties of these materials.

Aluminum, glass, silicon etc. are rarely, if ever, used in their pure forms: additives and alloying elements are added to varying degrees to improve these material’s technical characteristics such as yield strength and heat resistance. Mixing up aluminum scraps from various sources thus results in a low grade alloy with unpredictable properties showing great variability from one batch to the other. The recycled metal content is thus usually kept below a certain ratio in manufacturing; depending on the intended use of the newly created alloy, usually at or below 10–30%. The circular economy, recycling materials over and over, is thus nothing but a fairy tale: as long as we keep manufacturing stuff we will need new materials and we will need to get rid of a lot of old junk in the process.

“Renewables” and nuclear share an important feature in this regard. Solar panels and many components of wind turbines are in fact not that different from Uranium fuel rods: mined, processed and delivered by fossil fuels, only to be discarded as waste at the end of their lifecycle. And while recycling remains a theoretical possibility, refuse from both “renewables” and nuclear remain a huge headache: the former due to their immense quantity and the latter due to its radioactivity. Both technologies are the product of a fossil fuel age with no viable method of being produced or reprocessed in the absence of coal, oil and natural gas.

And not only that. Both “renewables” and nuclear — together with hydro and geothermal power — rely on a finite amount of easy-to-mine minerals from metal ores to high quality quartz. Once these energetically and economically affordable resources are out — together with the energetically affordable portion of fossil fuels — it will be impossible to continue with the production of not just solar panels and uranium fuel rods, but literally everything else as well.

Conclusion

Despite claims to the contrary low-carbon sources of energy cannot become a lasting success. The deployment of “renewables” nuclear, hydro etc. will continue to take an enormous amount of high heat, material transformation and carbon atoms into the foreseeable future. In other words: fossil fuels. Neither “renewables” nor nuclear offer a truly sustainable alternative to coal oil and gas, as all of these technologies rely on a mining and metallurgical industry fueled almost entirely by fossil fuels. These much touted “clean” and “green” sources of power are thus just as time- and resource-bound as the very fuels they aim to replace.

In addition to that, and thanks to the intermittent nature of wind and solar power, thermal coal plants also have to be built alongside new photovoltaic projects, and a massive material and energy investment will also have to be made in batteries and the grid itself. Trying to replace coal with “renewables” will thus not only take up a lot of valuable land (5), but will certainly involve a lot more mining, smelting, manufacturing, delivery and maintenance, all powered by fossil fuels of course. The widespread adoption of low-carbon sources of power, however, will still provide a little extra boost to world energy supply, much needed for growth.

What the wide-scale adaption of “renewables” and the subsequent reduction of fossil fuels’ share in electricity production could achieve is to free up coal, oil and gas for more mining and manufacturing. Since the super-organism will continue to grow in the meantime — this time giving birth to even more power-hungry AI data-centers — more and more renewable technologies will have to be deployed to keep up with rising demand. And while the relative share of fossil fuels will continue to drop in the electricity mix — if ever so slowly — this won’t mean we will be able to ditch coal. Or oil. Or gas. These valuable but polluting energy resources will increasingly have to be used to build more solar panels and other gadgets, instead of being burned in coal or gas fired power plants. The end result will be the same though: emissions up, reserves down — with even more gimmicks produced, compared to business as usual.

As we have seen above, what seemed to be a great idea on paper, has brought along a number of other expenditures. The list now includes power hungry AI energy innovations in China, for example, as part of an effort to manage an ever more complex grid. As for the EU and the US, wind and solar projects did not help to stop the relentless rise in the price of electricity, only added another source of instability. The price of a kilowatt grew by 25% for the industrial sector in the US since 2020, meanwhile European companies now have to pay 50% more for the same amount of power.

Cumulative changes in electricity generation compared to 2014. Chart: own work. Data source: Statistical Review of World Energy

While the still growing economies of China and other non-OECD countries were able to put up with rising costs and managed to increase the capacity of their electric grid by as much as 51% over the past ten years, the US and EU found themselves in an energy quagmire. How long this trend can continue is anyone’s guess. One thing seems to be sure: affordable mineral and energy resources are not infinite. Neither in Europe or America, nor in China. As prior growth in coal production turns into a decline in China, and as deindustrialization continues to gain momentum in both Europe and America, there will eventually come a point beyond which it will become impossible to replace, re-manufacture, recycle all those solar panels reaching the end of their lifecycle. At that point, however, there will be no return to coal fired power plants either, as whatever coal remains will be unaffordable for the economy to extract. And then, at the final hour of our industrial civilization, our children will look up on those dusty panels and say: nice try.

Until next time,

B

Thank you for reading The Honest Sorcerer. If you value this article or any others please share and consider a subscription, or perhaps buying a virtual coffee… At the same time let me express my eternal gratitude to those already support my work — without you this site could not exist.

Notes:

(1) An exajoule, or one quintillion joules, is enough to bring 3 cubic km of water to boil. It’s an amount of energy equal to 239 megatons of TNT (or nearly sixteen thousand Hiroshima bombs).

(2) I calculated with 34% thermal efficiency for coal, and 40% for natural gas. While newer, combined cycle gas power plants can and do achieve 55% efficiency, there are still a large quantity of older generation gas plants out there with a much lower thermal efficiency. As for the mix of fossil fuels used: coal provided 34% of all electric power fed into the grid worldwide in 2024, eating up 68% (or 122 EJ) of all available coal supply (165 EJ). Natural gas was used for 22% of all electricity generated, consuming 42% of global natural gas supply (63 of 149 EJ). Low carbon energy (nuclear, hydro, renewables) provided 41% of all electricity supply worldwide (47 EJ), with the remaining 3% coming from other sources including oil, statistical differences and sources not specified elsewhere (e.g. pumped hydro, non renewable waste and heat from chemical sources). Data source: Statistical Review of World Energy

(3) Unless we talk entirely fictional GDP growth — as it is the case with OECD countries, whose economies ostensibly grew by 56% since 2014, even as their energy consumption shrank by 3% over the same time period. Since our trucks, ships, cars, smelters, power plants, factories operate on mature technologies developed and perfected decades ago (as I explained here already), that growth is more attributable to the financialization of the economy, than real economic growth lifting people out of poverty.

(4) While nuclear is not considered renewable, and rightly so, it still operates on a low heat power source since PWRs (the most common reactor design) operate on a 315–375 °C core temperature. While certainly hot enough to boil water, industrial processes (such as glass, cement and steel making) often require temperatures above a 1000 °C, a range not even experimental gas cooled reactors can provide.

(5) It is also worth noting here, that “renewables” have much lower capacity factor than thermal power plants. That is, the ratio of the electrical energy produced by a generating unit for the period of time considered compared to the electrical energy that could have been produced at continuous full power operation during the same period is much lower than for fossil fuels. In case of solar this ratio of installed capacity vs actual electricity generated is 13% worldwide, and 25% for wind. That means, that we have to install 8 times more solar and 4 times more wind turbine capacity than what we actually need for a steady 24/7 supply of power. This, of course, necessitates a comparable amount of battery (or other) storage to store and retrieve power when needed. Data source: Statistical Review of World Energy