Solar power has begun to transform the world’s energy system

Not to mention that 20 years is enough time to develop a native industry of solar panel manufacturers. The issue with oil is it requires a constant flow of resources from specific locations in the world that are blessed by geography. Solar power has much less of that going on.

It's possible, but you may have noticed that out of the ≈200 countries in the world, over the last 20 years, about 180 of them have completely failed to develop a native industry of solar panel manufacturers, and about 100 of them have completely failed to develop a native industry of anything, continuing their agrarian and resource-extraction economies more or less as they have been for centuries, just with imported Chinese cellphones. People in those countries often blame the rich countries for keeping them down, for example by selling them goods at lower prices than their domestic production of those goods, and they're not completely wrong, but in many cases the dynamics preventing them from escaping that equilibrium are mostly internal.

Hypothetically, yes, such a blockaded country could develop a native industry of solar panel manufacturers in 20 years, and that industry would have an easier time traveling up the learning curve on the domestic market without having to match the prices of the Chinese hyperscalers. But in about 90% of cases they would fail to do so, for the same reasons the US still doesn't have any high-speed trains 60 years after the Shinkansen entered service and still doesn't have a moon base 56 years after Neil Armstrong.

> for the same reasons the US still doesn't have any high-speed trains 60 years after the Shinkansen entered service and still doesn't have a moon base 56 years after Neil Armstrong.

So.. lack of demand and ROI?

Energy independence and HSR are indeed poor metaphors for each other.

In the U.S. one can travel coast-to-coast faster and cheaper in a car than they can by rail. Then, of course, there is air travel. That is to say, there are alternatives.

A country completely dependent on foreign solar panels could develop non-solar alternatives. Or they could just surrender. So of course they also have alternatives. But this is existential whereas HSR is not. So, yes, it's a pretty poor comparison.

> Energy independence and HSR are indeed poor metaphors for each other.

It's not a metaphor. You're reasoning very sloppily. The absence of high-speed rail in the US is caused by a societal breakdown in technological and economic development. That breakdown also causes other effects. One of those effects is that over the last 20 years the US not only failed to develop a native industry of solar panel manufacturers; it lost the world-leading native industry of solar panel manufacturers that it already had. There's no strong reason to believe that a blockade would reverse that breakdown rather than accelerating it.

> In the U.S. one can travel coast-to-coast faster and cheaper in a car than they can by rail.

Yes. That's because the US doesn't have high-speed rail, even 60 years after the Shinkansen went into service. If the US did have high-speed rail, one would be able to travel coast-to-coast faster and cheaper by rail than they could in a car. And the difference is not small.

The fastest trains on the Beijing–Shanghai high-speed rail line average 290km/h, about 3–4 times faster than a car in the US and 50% faster than even the fastest Autobahn car speeds. The peak speed is 350km/h, but as in a car, some time is wasted speeding up and slowing down at stops at the beginning and end of the trip, and along the way.

The higher speeds also lower costs; https://www.trip.com/trains/china/route/beijingnan-to-shangh... tells me that the 1300-km trip currently costs US$22 for one person, which works out to about 1.7¢ per km. In the US, driving a car typically costs 70¢ per mile https://www.irs.gov/tax-professionals/standard-mileage-rates which is 43¢/km. So driving a car the same distance would not only take 3–4 times longer, it would cost 25 times as much.

https://www.youtube.com/watch?v=uBUYDvu9XgU&t=15m25s reports that a year ago they paid US$92, which would be 7¢/km, so either trip.com is lying, they were taking a higher class of service, or the price has dropped precipitously. It looks to me like coach-class airline seating, but https://en.wikipedia.org/wiki/Beijing%E2%80%93Shanghai_high-... tells me that when the service launched there were three classes of service.

Maybe in China cars are cheaper, in which case driving would only cost 10 times as much, I don't know. But it clearly isn't going to be as cheap as taking the high-speed train.

A consequence of the US's deficits in transportation is that a large fraction of the mental energy of its professional and intellectual classes is devoted to operating cars in traffic rather than to developing vaccines, improving Wikipedia, creating video games, or even selling ads.

60 years is a long time in terms of technological development. 60 years after the Wright Brothers achieve controlled powered flight in 01903 was 01963, when both the US and USSR had orbited cosmonauts, and the Apollo Program was well underway. 60 years after the first stored-program computer was delivered in 01949 (either the EDSAC or the secret Manchester Baby) was 02009, when Intel and AMD were shipping billion-transistor six-core processors. A wealthy country not being able to deploy the already existing technology in that time frame shows that it's experiencing not slow technological and economic development but slow collapse.

Can you build an industrial plant to build the panels only using solar power?

How and from where do you source the necessary primary materials for such an endeavor?

If you try to answer those questions you will see that you are bullshiting yourself.

Transport is heavily dependent on infrastructure. If you have a train between city A and city B but you can get to and from the train station without a car, this is not going to work.

But markets are far from the only mover here. Regulation, lobbying, habits... Also I guess the US would feel ashamed for not building their trains themselves in the first place, they would probably have to buy them abroad... So "proudness" is probably a factor here too.

High speed trains in the part of Europe I know are very well utilized and even a bit too crowded to my taste (still way better than planes - allow working easily with table, walking, no absurd wait times waiting in line at the airports, arrive much nearer to my actual goals...).

I have tried, actually. Maybe you can share what you've learned when you, hypothetically, investigated the question yourself. I'll start.

You do need materials, but you can source the materials anywhere on Earth; it's just a question of how expensive it is to refine them. Every element occurs as an impurity in every rock at some level. When you can import them freely, some deposits are uneconomic.

For building a plant to refine silicon, things like platinum and iridium, which are very scarce in most rocks, are very helpful. But they aren't ingredients in the solar cells themselves. Solar cells themselves are made of silicon, aluminum, silver†, lead, and tin, with trace quantities of phosphorus (or arsenic) and boron. These are mounted to "ultra-white" glass, which is made of silicon again, oxygen, sodium, calcium, and trace amounts of manganese. The mounting is done typically with EVA, which is mostly a hydrocarbon with a little oxygen in it.

The total amount of these materials is surprisingly small. The silicon wafer (2.33g/cc) is about 100μm thick, and the glass (2.5g/cc) is typically 2.5mm thick (3.2mm is "ultra thick"). So a square meter of solar panels, rated at some 200W, contains 6.3kg of glass (mostly oxygen and silicon) and 0.23kg of crystalline silicon, plus much smaller amounts of other materials.

So raw materials aren't a constraining factor unless you're living on a barge or a space station or something. Knowhow, organization, discipline, cooperation, etc., are the constraining factors. Sadly, those are in short supply almost everywhere.

______

† Silver is used for large conductive strips on the surface of the silicon; it can be replaced with copper at a significant loss of efficiency. There is already pressure to do this because the raw-materials cost of silver accounted for about 10% of the wholesale cost of current PV modules last time I checked, and about 10% of global silver production went into PV modules. Since then production has increased and PV prices have dropped.

Yes as I suspected you actually have no clue what you are talking about. You are listing stuff like a recipe as if you can just shop around for those things.

Silicon production is an energy intensive process; you need 11-13 kWh per kg of silicon produced. Technically it's a process using electrodes and thus raw electricity so you could source it from renewable. But that's in theory you need large amount of predictable power for a long time and on demand, which is not at all what the renewables have been so far.

Then if you look into aluminum production you will see that it requires carbon electrodes, that are made in ovens continuously heated to up to 1300°C for hours on end (about 20h per anode). They do not shut down those ovens since it takes multiple WEEKS to get to temperature, and they use natural gaz as the fuel. It's not clear if we could even make an alternative using purely raw electricity that would have enough power density. The aluminum production process itself requires megawatts levels of energy, usually you need a 500MW substation. Most plants are built next to a power plant, usually coal or nuclear. At the current 200W/m2 efficient level for solar panel, you would need about 2 500 km2 of solar panel to get that much power.

Glass production also requires a lot of dense energy. It typically uses gaz for heating but maybe they can figure out an industrial process to electrify it, currently not the case anyway. We are talking about megawatts level of energy again and a glass furnace cannot ever be shut down during its 15-20 years lifetime, so it's not like intermittent renewable are an option.

And I'm not talking about the various mining operations, necessary to get the raw stuff which is basically running almost exclusively on fossil fuel (but at least some of it can be transitioned to electric).

So now, I have to say 2 things: - firstly, my question was obviously rhetoric, the answer for anyone who has studied the subject is clearly no. But that requires an understanding that isn't surface level. - secondly you are clearly an arrogant asshole who thinks he knows shit when he clearly doesn't. But I'll let you live in your fantasy world where you can have industrial production with just electricity from solar panels.

Thank you for sharing what you know. I'm aware of these issues. My personality is irrelevant to them; they are what they are whether the person talking about them is an arrogant asshole or not. I'm puzzled as to why you thought I'd be interested in discussing whether or not I'm an arrogant asshole, really!

Your calculation of solar capacity is off by a factor of a million; 500 megawatts at 200W/m² is 2.5 km², not the 2500 km² you say (the size of Yosemite National Park), which would be 500 terawatts, roughly 30 times current world marketed energy consumption. The same magnitude of error in the other direction would have led you to claim that an aluminum smelting plant requires 500 watts, less power than a household blender.

You also forgot to divide by the capacity factor; 200W/m² is the nameplate capacity, what the square meter produces in full sun, not the year-round average, which is closer to 30W/m², depending on factors like latitude, clouds, and tracking. (That increases the estimate from 2.5km² to 17km², 1700 hectares or 7 sections, the area of the city of Los Altos, California, or a quarter the area of the Curonian Spit park in Kaliningrad.)

These basic errors suggest that either you are not fully aware of the extent of your knowledge, or you are knowingly exaggerating it.

It seems like your primary objection is the intermittency of solar energy, which can be straightforwardly solved with BESS; even without lithium resources, either liquid metal batteries or nickel–iron batteries are an adequate resource anywhere in the world. Sodium-ion batteries are another scalable form of BESS that does not depend on scarce elements; a 200MWh utility-scale sodium-ion battery came online a year ago in Qianjiang: https://www.energy-storage.news/first-half-world-largest-200... but plausibly nobody outside of China knows how to do this.

There are straightforward solutions to the problems you're describing, even without BESS; many haven't been developed beyond the lab scale because they aren't economically competitive with the established approaches you're describing. In a hypothetical blockaded country, those alternatives wouldn't be competing with cheap fossil fuels. In practice, though, BESS is plenty.

Silicon purification to solar grade is not simply an electrolytic process, as you incorrectly imply; it requires a series of refinement steps to become PV-grade silicon.

In the case of glassmaking, the necessary technology is already well developed. An all-electric glassblowing pilot plant entered production last year in Cognac: https://www.youtube.com/watch?v=FuK8f4cB7Ps. And you can buy off-the-shelf glassmaking furnaces for mass production: https://www.hornglass.com/products/melting-furnaces-and-equi...

Electrically heated furnaces are more controllable and versatile, which is why they are universally used in laboratory glassmaking. Unlike the case with aluminum, fossil fuels are nothing but trouble for glassmaking; limited adiabatic flame temperatures, glass-batch contamination from fuel impurities, and the unfortunate necessity to vent flame-fired furnace to the atmosphere are problems glassmakers have had to overcome in order to use cheap energy from fossil fuels, not benefits.

Carbon is probably the only possible electrode material for aluminum production, although zirconia has been suggested. The net reaction is Al₂O₃ + 3C → 2Al + 3CO, consuming about 700kg of carbon per tonne of aluminum produced. Fortunately such small quantities of carbon are not difficult to obtain, and in extremis it would even be bearable to obtain them via direct air capture; we're talking about hundreds of grams of carbon per 300-watt solar panel, so a single tree contains enough carbon to smelt the aluminum for a megawatt or so of panels.

Mining is almost entirely electrified already; attempting to run fossil-fuel machinery in an underground mine shaft, or even an indoor warehouse, poses the kind of risk of asphyxiating workers that is normally considered unacceptable except in, for example, Russia. Gargantuan strip mining machinery like the Marion 8750 is largely electric for the same reasons that diesel locomotives are electric.

Thank you for a productive, if gratuitously insulting, exchange of views!

My comment at https://news.ycombinator.com/item?id=44461370 discusses the energy economics of atmospheric carbon capture.