I strongly agree, but I see the government's decision in favor of hydrogen as simply an inevitable consequence of how governmental decisions are made. This is part of why I am so unimpressed by calls for more governmental R&D funding. Politicians fundamentally can't make the funding decisions sensibly, say by sending moderate funding to thousands of projects instead of poisonous amounts of funding to a few projects. To fund rationally would require that they admit uncertainty in a manner incompatible with being the sort of overconfident alpha male who can get elected.

/rant on/
Batteries are cursed because the only people who really understand them are electrochemists. In development, they are black boxes and can only be trouble shot by destroying them. Worse, the chemical system effectively remembers every abuse it is ever exposed to: themal, electrical,vibrational, etc. If the development chain does not care for them all along the way, promising systems can look impractical. My experience has been that mechanical and electrical engineers hate these systems because there is so little available to modify, adjust and tweak. Hands on engineers really hate these system because the development cycle is so long. For example, it may only take a few seconds to propose, perform and evaluate a software experiment. It may only take a a few minutes to propose, perform and evalue an experiment on a hardware system. It make only take a few hours to propose, perform and evaluate an experiment on a mechanical system like an engine. However, it can take days, weeks or months to perform and evaluate an experiment on a battery!
Additionally, fuel cells are the perfect pork barrel for engineers. There are complex controls, complex air & humidity handling systems, complex electrical power flows and converters plus electrochemical parts to make things interesting! Batteries, on the other hand, really only have one component that is designed in one department or supplier. Fuel cells are cool to watch and see. Batteries are plastic boxes with two shiny termials. For corporate or public politicians, fuel cells are much easier to rally broad support and spread the spoils!
/rant off/

A major problem with batteries is not necessarily capacity but rather cost. You can increase capacity by buying more batteries. You can deal with limited lifespan by buying new batteries. But that costs money. The solid state batteries may have more capacity and longer lifespan, but they will also cost more.
Large battery banks are a maintenance hassle, so any reduction of maintenance costs would be welcome.
Fuel cells are not a replacement for batteries but should be thought of more as generators.

I hope batteries do improve, they need roughly 200% more capacity with no price increase. Even 100% would allow electric cars. But great improvements have been announced weekly for decades without working out. No one has wanted better batteries more than the Pentagon and their funding hasn't made the magic work.

As to storing power from solar cells in batteries? I think flywheels will take over this area. They are scalable and limited mostly by the strength of materials. Modern composites are many times stronger than titanium - amazing.

Another method, for big scale, is using solar electricity to pump water up hill during the day. Hydroelectic power is then produced conventionally. The beauty of this method is simplicity - you don't have to convert the solar electicity to alternating current or invent anything new. The drawback is low efficiency.

Fuel cells: my guess - providing hydrogen in massive quantity will prove tough indeed.

Any ideas why flywheels aren't more prominant already? There was an interesting Discover Magazine article on them about 7 years ago. It seems to me that cheap nanotubes, if they are ever developed, should allow Really powerful flywheels. Sufficiently powerful that even if they can be used for hydrogen storage they won't need to be.
I agree that batteries aren't fun to R&D, and that this is a problem. I genuinely think that one of the biggest barriers to Drexlerian nanotechnology is that the necessary research is basically boring compared to less promising nanoscience based approaches, though the very biggest barrier is obviously the shortage of incremental spinoffs.

Michael Vassar,

Flywheels are not more widely used for the same reason photovoltaics, cogeneration, and wind are not more widely used: They cost too much.

Spiny Widgmo,

Thanks for that insight. I find it very plausible that what you describe about engineers is at work influencing the relative effort that goes into fuel cells versus batteries. Unfortunate.

Randall, of course they cost more now, or they would be ubiquitous, but cost doesn't seem to prevent spending on PV, wind, and cogeneration to nearly the same degree.
Also, wind is economical in some places, and becoming more economical with time. I wonder whether China will begin manufacturing of wind-energy systems in the next decade. That could make a big impact. Anything that makes manufacturing cheaper should make wind cheaper.
PV is hideously cost-inefficient, but solar heating, for instance, if cost effective far more frequently than it is used. So are many energy saving features. Markets tend towards efficient equilibria != markets are always at equilibrium.

great article. the point about funding going to fuel cells is very important, because it really is a political decision. Sadoway is the type of expert who really knows. this is a materials problem that in my opinion should be solved through open solicitation funding from the federal gov't, probably sbir's - that takes most of the decisions out of politicians' hands.

the best material to make flywheels is (and probably always will be) steel. high strength steel. in addition to costs there is the practical problem of flywheels not being used on anything moving because all that kinetic energy will act like a gyroscope.

i recall reading about a power company in western mass that owned two reservoirs at two different elevations and stored energy exactly as is described here. i wish i had a reference for you - it's like wind, very efficient where nature allows.

Good comments by all.

There is nothing inherently costly about flywheels; magnetic bearings are expensive in low volumes.

Flywheels are best in static applications where they can be bunkered. They contain a lot of energy and can be bombs if disturbed by car wrecks, etc.

Vertical flywheels wouldn't hinder turns. Maybe my mechanical engineering is rusty. But see above.

Steel is certainly cheap but just isn't strong enough to compete as newer materials fall in price.

It isn't that flywheels are great. I just think batteries are horrible. They wear out, demand all sorts of polluting chemistry, and (the worst sin of all) aren't making much progress.

I personally advocate nuclear. Big time! But if folks won't buy nukes then avoid batteries to smooth the highly variable output of solar or wind.

Three comments on flywheels.
1) For mobile applications, until material science can guarantee very high safety factors in the design of high energy density flywheels, you will not see them used. They will consume mass and volume either in the flywheel or the containment vessel, destroying the energy density. I can't find references, but the rumor mill in the area I work in has circulated with stories of high energy desnity flywheels suffering fairly spectacular catastrophic failures. Imagine converting a large fraction of the kinetic energy stored into a high pressure, velocity gas & flywheel schrapnel mixture trying to escape a containment vessel. These classes of failures are undetectable and uncorrectable. I think UT has done work on flywheels that fail gracefully. Otherwise, all you can do is build everything with a high confidence of quality and inspect regularly.

2) For stationary applications, the parasitic drains, noise and energy volume density can be deal breakers. I imagine there is also some concern about catastrophic failures. The destructive potential of large flywheels storing significant energy is no less than the equivalent amount of energy stored behind a dam. I'd feel much safer living withing a half a mile of dam than a half a mile of a flywheel energy storage complex.

3) If everyone is excited about flywheel & batteries for energy storage of unrealiable sources of power like solar and wind, I think a centralized facility that uses water for energy storage makes the most sense. It will be expensive and increase the complexity of distributed power to require local energy storage to smooth out loads and generation. Centralized energy storage helps solve a problem for traditional energy producers also. They typically suffer from the need to bring inefficient peak load sources on line to deal with late afternoon power demands and a lack of demand in the early morning. The economic benefits do not justify constructing energy storage for most areas, however should distributed generated with wind and solar become a reality, centralized energy storage may be a viable economic prospect.

Spiny Widgmo

For stationary apps, steel underground should be pretty safe. Dig in somewhat deep if you'r worried, put it in old mine-shafts for instance. Anyway, why should stationary apps fail unless they are on a fault zone?
For mobile apps, how about building out of carbon fiber today and carbon nanotubes tomorrow, in freely floating fiber form (no epoxy resin = high surface area). Use a pair to compensate for gyroscopic effects, and instead of containing them in a solid containment vessle, set up a system to ignite them upon release so the explosion is no worse than a gasoline explosion. OTOH, the bomb spinoffs could be ugly. Don't think about the truck-bombs :-(

Re:Michael

Any flywheel is going to be vulnerable to earthquakes and other jars. Sophisticated electronics could probably alleviate this by pulsing the magnetic bearings to dampen movement. But FW won't happen in mobile applications - people just won't buy in - I wouldn't myself.

My earlier remark that bunkers are practical for fixed installations was based on the idea that flywheels can be sized to fit any terrain and they will be in rural areas anyway - that is where huge banks of solar or wind generators would be.

As to igniting them - a failing flywheel will fracture (I think) faster than a flame front could travel. Interesting thought though. Last night I realized that implosion might neutralize the explosion. It is used to optimize nuclear weapons and for so-called "active armor" on tanks. It is just a thought - the cost, complexity, and public relations would kill it even if it worked.

I am still fond of the idea of pumping water up hill when the wind blows or the sun shines. Then using conventional hydroelectric generation. But the locality must have suitable elevation gradients and room for reservoirs.

Returning to batteries. Even though I dislike them they are quite good partners in hybrids because they can supply a lot of power for a short time. And while capacity growth seems stalled, they have become much more durable. That really counts in cars and trucks.

May be there is another direction solid-state battery technology must take. Not all areas are being explored. Thinking out-of-the box is more of an answer than trying to apply current technology in another package. A reusable, hardened battery that has no recharge memory and does not rely on chemistry or Maxwell’s equations to work…

you needc to develop a super resistor for storing large potential in caps.
ie:

a wall of packed nutrons.

PS:
The serface area of a sphere increases when devided into smaller spheres.
the constant is a shape :) not a speed.
energy is mass taking up more space than it used to.
why does it do this ??

again:
the constant is a shape, not a speed

if you think mass is none sphereical look closer or father away.
gravity concures.

Ciao,
Dov