The Speculist: The (Solar) Singularity is Near


Live to see it.

« When I Grow Up... | Main | Reader's Choice Video 1 »

The (Solar) Singularity is Near

One of the reasons I don't lose sleep over Peak Oil is that there is such a broad range of alternative energy sources under development. The list includes, but is not limited to, the following:

Nuclear Fission
Concentrated Solar
Ethanol -- from switchgrass, cornstalks, etc.
Ethanol -- from waste
Methanol -- from coal
Synthfuel -- from coal
Synthfuel -- from shale
Synthfuel -- from tar sands
Biodiesel -- from waste
Biodiesel -- from algae
Nuclear Fusion

Progress is being made on all of these fronts. And if oil shoots up to $200, $300, $400 per barrel over the next couple of years, we can expect interest in these (as well as funding applied to them) to skyrocket.

Let's look at just the second and third items on the list, the two major forms of harnessing energy from the sun. What we normally think of as "solar energy" is the application of photovoltaic technology -- turning the sun's power directly into electricity. "Concentrated solar" power, AKA solar thermal energy, involves concentrating and capturing heat from the sun, which is then used to create steam and move an electricity-producing turbine.

We wrote about the tremendous promise of concentrated solar power just a few weeks ago, so I won't rehash all that here. Suffice it to say that, even if photovoltaic technology had hit some kind of peak of its own, meaning that we wouldn't expect much more from it than what we're getting now, concentrated solar would remain as a major potential energy source that we have barely even begun to exploit.

But the truth is that photovoltaic solar energy is far from any peak. Ray Kurzweil has repeatedly stated his assessment that solar energy is on a Moore's-Law-style trajectory of its own, and that all the worlds energy could be supplied by solar in as little as 20 years. So if Moore's Law is leading us to The Singularity, is this acceleration of solar power capability leading us to a solar singularity?

Some probably wouldn't like that term, seeing as it could make the whole question as to what exactly we mean by "singularity" even murkier than it currently is. But it has a ring to it, doesn't it?

Solar Singularity.

Anyhow, if we are going to get to the point where solar really does (or even could) supply all the world's power within a couple of decades, we are obviously going to have to see:

Accelerating progress in solar energy technology culminating in a fundamental shift in how the world's energy needs are met.

And that, then, can be how we define the solar singularity. It seems unlikely that it could be confused with any other kind of singularity, doesn't it?

We talked briefly on the most recent FastForward Radio about how we would know when we've reached the solar singularity. One suggestion was "when solar is cheaper than anything else." Another was "when they don't even bother to drill any more." Those are both good candidates. But how could we ever get to that point?

I don't have an answer for the latter possibility, but for the former (via FuturePundit), a company called Sunrgi says we'll be closing in on it within the next couple of years:

Washington D.C., USA--A new patents pending solar energy system will soon make it possible to produce electricity at a wholesale cost of 5 cents per kWh (kilowatt hour). This price is competitive with the wholesale cost of producing electricity using fossil fuels and a fraction of the current cost of solar energy.

XCPV (Xtreme Concentrated Photovoltaics), a system that concentrates the equivalent of more than 1,600 times the sun’s energy onto the world’s most efficient solar cells, was announced today by Sunrgi, a solar energy system designer and developer, at the National Energy Marketers Association’s 11th Annual Global Energy Forum in Washington, DC. The technology will enable power companies, businesses, and residents to produce electricity from solar energy at a lower cost than ever before.

“Solar Power at 5 cents per kWh would be a world-changing breakthrough,” said Craig Goodman, president, National Energy Marketers Association. “It would make solar generation of electricity as affordable as generation from coal, natural gas or other non-renewable sources, without requiring a subsidy.”

“In a little more than a year we were able to develop and successfully test XCPV,” said Robert S. (Bob) Block, co-founder and Sunrgi principal. “We expect the Sunrgi system to become available for both on- and off-grid power applications, worldwide, in twelve to fifteen months.”


If they can get the cost of electricity production down to the level they're talking about -- and that's an enormous if -- then oil doesn't have to get any more expensive for people to start adopting this technology. The singularity could be very near, indeed.

All we have to do is figure out a good way to run cars off solar power. It could be time to start thinking about a battery or hydrogen fuel cell singularity.

Bring 'em on!

UPDATE: Meanwhile, some students at MIT claim they can concentrate the sun's power 1000 times using inexpensive and easily-available components. This CSP / thermal approach could bring heated water and electricity to remote -- and even not-so-remote -- communities in the developing world. For the people who may eventually be on the receiving end of such technology, the sudden availability of electricity may be singularity enough.


Hi Phil,

The caveat in all these analyses "Solar Power at 5 cents per kWh" is that fossil fuels have already subsidised, energy-wise, these technologies for which these calculations are made. The chips, the glass or metal which will be used as the focussing mirror, the copper to connect them, the material used for covering - all of them were created during a time when energy was cheap and those assumptions have inherently entered this guys calculations. Shale oil was supposed to become viable at $30 oil. It is still struggling with $120 oil. I'm not sure how much stock you put with Energy Returned on Energy invested analysis, but it is worth giving a look.

Heinlein said it.

It's raining soup. What we need is a decent bucket.

Solar is a really great idea especially after dark. In cloudy weather and in the wintertime.

Once solar is cheaper than coal or nukes there is no doubt we can shut down those evil electrical generating plants.

In the mean time it might also be wise to look into this:

WB-7 First Plasma


There's a little problem here: energy storage.

Energy must be stored for periods when there isn't enough (or any) sunlight. Even worse, it must be stored in a transportable form for automotive transit (the largest user of petroleum today in the US).

The last time we had an energy shock, all sorts of ideas were tossed around about energy storage. Nothing happened. Hydrogen, flywheels, hydroelectric (gravity) storage, etc.

Today, people seem to be ignoring the problem.

Hydrogen is waved about as a magical solution, never mind that it is highly inefficient, hard to store in high energy density, and would require trillions of investment in a national infrastructure in order to use it.

Batteries are likewise imagined to solve the problem. But Moore's law hasn't been working there. We have been working on electric cars for over 100 years, and we still aren't that much better off than the original lead-acid batteries. The best battery technology today (in terms of specific energy density) has a density of 1/40th that of gasoline! And, batteries are way more expensive than gas tanks. They also require a large infrastructural investment to be viable (charging your car 8 hours for 100 miles of use is not very practical).

If oil stays this high, the dramatic economic pain will cause people, hopefully, to demand that we use the power we have - coal and nuclear and offshore oil - and to heck with CO2.

That's not as irrational as it seems. Today we are engaged in an extreme application of the precautionary principle, strangling energy production at enormous cost in order to avoid a mild climate change (if the completely untestable and questionable models are to be believed at all).

Re: MIT,

Motors are low reliability items. Look at the Zome Works idea - fluid flows from a heated tank to a cooled tank to aim the collector. No moving or electrical parts.

1 sq meter = 100 sq ft. the collector had better come in at 50¢ or less a sq ft to make it economical. Because the rest of the money is going to have to go into high temp solar cells and thermal collectors.

Oh, yeah. It will need to last 20 years. It will have to withstand wind loads. Rain. Snow. Hail storms. Lightning. etc.

Nice idea. It will need some refinement.

If the MIT guys were really bright they would be working on energy storage. That is the # 1 pitfall of most alternatives.

Solar Singularity.
"When companies in the sunny part of the country no longer have customers, only power providers."

Another was "when they don't even bother to drill any more." Those are both good candidates. But how could we ever get to that point?

Nope. You forget, we use oil for other things besides gassing up the fliver. Even after we convert all our vehicles and power stations to [something else], we'll still need to grease the wheels. It will be a long time before we have no use for oil.

That being said, the real advantage to switching to solar or some similar energy source is the distribution cost. By some estimates it costs a gallon of gasoline's worth of energy to delivery each gallon of gas to the pump. There's an unavoidable energy cost in pumping, transporting, and refining the raw stock as well as transporting it to the filling station. Solar delivers itself: you put a converter device on your roof (PV, thermal, etc.) and flip a switch inside to use it. Father Sun takes care of everything but the wiring. You don't have to beg foreign governments to let you have some of theirs, you get enough locally to take care of your house/city/state/nation. Your "solar singularity" will be quite dramatic when it reaches the tipping point in 10-15 years and everything starts to change.

One thing the "we're running out of oil" crowd seems to have in common with the "oil is the only real solution" crowd is their certainty that some problems just can't be solved.

I should clarify that I've got nothing against using oil. Personally, I'm in favor of (carefully and responsibly) drilling for oil anywhere it's available, and making use of coal. My stepping-off point for this particular blog post was alternatives to oil. I don't think we'll switch from oil or coal primarily because of climate change or peak oil either one -- I think we'll do it because better alternatives become available. It's been observed that we didn't leave the stone age because we ran out rocks; I think the same principles are in play now.

I suppose I could also "clarify" that I'm okay with nuclear, but seeing as it was the first energy source I named, and seeing as I have written in favor of it multiple times times over the past few years I'm not sure how much clearer I can get.

Storage is a big problem, but not as hopeless as John and M. Simon would have us believe. The blog post about CSP linked above talks about some very encouraging developments in storing solar heat, which can then be converted to electricity.

And, rather than just waving hydrogen about as a magic solution, some folks are taking steps towards seeing if it will work. And there are efforts underway to take on potentially bigger challenges than fueling cars.

Maybe those less-than-bright MIT students will heed Simon's advice and get to work on the storage problem. But even if they don't, there are a lot of different people approaching it from a lot of different angles. Flywheels are one possible alternative to batteries for large-scale power storage. For powering cars, we may yet end up using oil or synthetic oil substitutes made from shale or coal. And we may combine those with methanol and ethanol (made from sane, non-food sources) to round out our selections.

None of which means that we have to keep emitting CO2 at the same or a greater rate -- seeing as research continues into how to burn oil and coil more cleanly than before. Still, since I'm sure objections can be raised to all of these ideas, we must assume that progress in any of these areas (or some unexpected area) is quite simply impossible.

Nuclear fission for terrestrial-based electricity: yes !

Nuclear fusion for any form of energy: no ! The energy required to maintain the electromagnetic 'bottle' containing the fusion reaction offsets any energy which may be obtained from the (very difficult to control) reaction.

Nothing wrong with Solar perse. There are warts. The first being energy density per sqmeter. There is only about 1.4kw/hr/m2 falling on the earths surface. Which brings me to my second point. When guys like Kurzweill talk about a singularity in solar energy they are really talking about raising the efficiency from its current 15% level. The cap is still 1.4kw/hr. That is quite different from the seemingly limitless expansion of compute power that most associate with Moore's Law.

H2 today is generated from gas stocks. To do so via electrolysis is to incur a 30% entropy loss in the conversion process. Does not make much sense under those terms. As for storage of solar, well if you go to megascale why not use underground salt domes? Pump the air in, store it, release it against an air turbine to extract the energy? In the same vein, there is already an airpowered car that is near release. No pollution from the car and is suited for urban needs where distances are not great.

Having a viable solar/nuke substructure will help mitigate against future gas hikes. But we have to begin soon. Otherwise we just remain hostage to the middle east.

I don't understand this talk of a solar "singularity".

Even if we could make technology that's 100% efficient at capturing the sun's energy at the Earth's surface, it would take something like four square miles of solar collectors (solar panels, concentrating mirrors, lenses, whatever) to replace a single 1 GW power plant.

That's a heck of a lot of land -- and a heck of a lot of equipment to be built, placed, and maintained. And really a rather large and dramatic ecological footprint.

What, exactly, makes that superior to even the worst power plant in current use?

What if ultracaps (EESTOR) pan out?

Then we're all set for full electrics powered by utility solar.

HALF LIFE!!?!?!?!

Post a comment

(Comments are moderated, and sometimes they take a while to appear. Thanks for waiting.)

Be a Speculist

Share your thoughts on the future with more than


Speculist readers. Write to us at:

(More details here.)



Powered by
Movable Type 3.2